European Journal of Health Psychology Issue 1, 2019

Volume 25 / Number 1 / 2019

Volume 26 / Number 1 / 2019

European Journal of

Health Psychology

European Journal of Health Psychology

Editor-in-Chief Claus Vögele Associate Editors Verena Klusmann Arnold Lohaus Britta Renner Christel Salewski Silke Schmidt Heike Spaderna

Focal psychodynamic psychotherapy – an evidence-based method “This book provides scientific evidence for an approach that has a long-term impact on one of the most vexing psychiatric problems – anorexia nervosa.”

Jacques P. Barber, PhD, ABPP, Professor and Dean of Gordon F. Derner School of Psychology at Adelphi University, NY, USA

Hans-Christoph Friederich / Beate Wild / Stephan Zipfel / Henning Schauenburg / Wolfgang Herzog

Anorexia Nervosa Focal Psychodynamic Psychotherapy 2019, xvi + 124 pp.

US $39.80 / € 31.95 ISBN 978-0-88937-554-3 Also available as eBook This manual presents an evidencebased focal psychodynamic approach for the outpatient treatment of adults with anorexia nervosa, which has been shown to produce lasting changes for patients. The reader first gains a thorough understanding of the general models and theories of anorexia nervosa. The book then describes in detail a three-phase treatment using focal psychodynamic psychotherapy. It provides extensive hands-on tips, including precise assessment of psychodynamic themes and structures using the Operationalized Psychodynamic Diagnosis (OPD) system, real-life case studies, and

www.hogrefe.com

clinical pearls. Clinicians also learn how to identify and treat typical ego structural deficits in the areas of affect experience and differentiation, impulse control, self-worth regulation, and body perception. Detailed case vignettes provide deepened insight into the therapeutic process.

A final chapter explores the extensive empirical studies on which this manual is based, in particular the renowned multicenter ANTOP study.

Printable tools in the appendices can be used in daily practice. This book is of interest to clinical psychologists, psychotherapists, psychiatrists, counselors, and students.

European Journal of

Health Psychology Volume 26 / Number 1 / 2019

Editor-in-Chief

Editorial Office

Claus Vögele, Université du Luxembourg, FLSHASE Campus BELVAL, Maison des Sciences Humaines, 11, Porte de Sciences, L-4366 Esch-sur-Alzette, Luxembourg, Tel. +352 46 6644-9755, E-mail ejhp@uni.lu Nicole Knoblauch, Université du Luxembourg, 11, Porte de Sciences, L-4366 Esch-sur-Alzette, Luxembourg, Tel. +352 46 6644-9755, E-mail ejhp@uni.lu

Associate Editors

Verena Klusmann, Hamburg, Germany Arnold Lohaus, Bielefeld, Germany Britta Renner, Konstanz, Germany

Christel Salewski, Hagen, Gemany Silke Schmidt, Greifswald, Germany Heike Spaderna, Trier, Germany

Editorial Board

Urs Baumann, Salzburg, Austria Elmar Brähler, Leipzig, Germany Birte Dohnke, Schwäbisch Gmünd, Germany Michael Eid, Berlin, Germany Heike Eschenbeck, Schwäbisch Gmünd, Germany Toni Faltermaier, Flensburg, Germany Dieter Frey, München, Germany Edgar Geissner, Prien am Chiemsee, Germany Nina Knoll, Berlin, Germany Carl-Walter Kohlmann, Schwäbisch Gmünd, Germany Thomas Kubiak, Mainz, Germany

Friedrich Lösel, Cambridge, UK Mike Martin, Zürich, Switzerland Franz Petermann, Bremen, Germany Wolfgang Schlicht, Stuttgart, Germany Silke Schmidt, Greifswald, Germany Urte Scholz, Zürich, Switzerland Ralf Schwarzer, Berlin, Germany Andreas Schwerdtfeger, Graz, Austria Monika Sieverding, Heidelberg, Germany Wolfgang Stroebe, Utrecht, The Netherlands Petra Warschburger, Potsdam, Germany Jürgen Wegge, Dresden, Germany

Publisher

Hogrefe Publishing, Merkelstraße 3, 37085 Göttingen, Germany, Tel. 0551 99950-0, Fax 0551 99950-111, E-mail publishing@hogrefe.com North America: Hogrefe Publishing, 7 Bulfinch Place, Suite 202, Boston, MA 02114, USA, Tel. (866) 823-4726, Fax (617) 354-6875, E-mail publishing@hogrefe.com Regina Pinks-Freybott, Hogrefe Publishing, Merkelstraße 3, 37085 Göttingen, Gemany, Tel. +49551 99950-0, Fax 0551 99950-111, E-mail production@hogrefe.com Hogrefe Publishing, Herbert-Quandt-Straße 4, D-37081 Göttingen, Germany, Tel. +49551 99950-900, Fax 0551 99950-998, E-mail zeitschriftenvertrieb@hogrefe.com Melanie Beck, Hogrefe Publishing, Merkelstraße 3, 37085 Göttingen, Germany, Tel. +49551 99950-423, Fax 0551 99950-111, E-mail marketing@hogrefe.com ISSN-L 2512-8442, ISSN-Print 2512-8442, ISSN-Online 2512-8450

Production Subscriptions Advertising/Inserts ISSN Copyright Information

Publication

Subscription Prices

Payment

Electronic Full Text Abstracting Services

Ó 2019 Hogrefe Publishing. This journal as well as the individual contributions and illustrations contained within it are protected under international copyright law. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without prior written permission from the publisher.

All rights, including translation rights, reserved.

Published in 4 issues per annual volume. European Journal of Health Psychology is the continuation of Zeitschrift für Gesundheitspsychologie (ISSN 0943-8149), the last annual volume of which (Volume 24) was published in 2016.

Calendar year subscriptions only. Rates for 2019: Institutions – from US $326.00/ 249.00 (print only; pricing for online access can be found in the journals catalog at hgf.io/journals2019); Individuals – US $120.00/ 94.00 (print & online). Postage and handling – US $16.00/ 12.00. Single copies – US $86.00/ 67.00 + postage and handling.

Payment may be made by check, international money order, or credit card, to Hogrefe Publishing, Merkelstr. 3, D-37085 Göttingen, Germany. US and Canadian subscriptions can also be ordered from Hogrefe Publishing, 7 Bulfinch Place, Suite 202, Boston, MA, 02114, USA. The full text of European Journal of Health Psychology is available online at http://econtent.hogrefe.com and in PsycARTICLES. Abstracted/Indexed in Social Sciences Citation Index (SSCI), Social Scisearch, Journal Citation Report/Social Sciences Edition, PsycINFO, PsycARTICLES, PSYNDEX, Scopus, IBZ, and IBR. 2017 Impact Factor 0.448, Journal Citation Reports (Clarivate Analytics, 2018)

European Journal of Health Psychology (2019), 26(1)

Ó 2019 Hogrefe Publishing

Contents Original Articles

Short Reports

Psychometric Properties of the French Brief Resilience Scale Ingo Jacobs and Antje Horsch The PREVIEW Study: Supporting Behavior Change in an International Intervention Study Among Participants With Pre-Diabetes Maija Huttunen-Lenz, Sylvia Hansen, Thomas Meinert Larsen, Pia Christensen, Mathijs Drummen, Tanja Adam, Moira A. Taylor, Elizabeth Simpson, Jose A. Martinez, Santiago Navas-Carretero, Teodora Handjieva-Darlenska, Sally D. Poppitt, Marta P. Silvestre, Mikael Fogelholm, Elli Jalo, Roslyn Muirhead, Shannon Brodie, Anne Raben, and Wolfgang Schlicht

10

Symptoms of Muscle Dysmorphia Between Users of Anabolic Androgenic Steroids With Varying Usage and Bodybuilding Experience Marc Ashley Harris, Tina Alwyn, and Michael Dunn

21

Psychological Predictors of Fatigue, Work and Social Adjustment, and Psychological Distress in Rheumatology Outpatients: A Short Report Faith Matcham, Sheila Ali, Katherine Irving, and Trudie Chalder

25

News and Announcements Meeting Calendar

Ă“ 2019 Hogrefe Publishing

1

30

European Journal of Health Psychology (2019), 26(1)

Original Article

Psychometric Properties of the French Brief Resilience Scale Ingo Jacobs1,2

and Antje Horsch3,4

1

Department Natural Sciences, Medical School Berlin, Germany

2

Department of Psychology, Sigmund Freud University Berlin, Germany

3

Department Woman-Mother-Child, Lausanne University Hospital, Lausanne, Switzerland

4

Institute of Higher Education and Research in Healthcare, University of Lausanne, Lausanne, Switzerland

Abstract: The Brief Resilience Scale (BRS) is a reliable and valid assessment of the self-perceived ability to bounce back or recover quickly from stress. The current study translated and validated the French version of the BRS (BRS-F) in a sample of N = 220 midwives. In a confirmatory factor analysis, the unifactorial model fitted acceptably to the data. High levels of Tucker’s φ implied that the component loadings of the BRS-F and of the original BRS are almost equal. The BRS-F demonstrated good levels of reliability and meaningful correlations with mental health symptoms and burnout. The resilience-mental health difficulties link was fully mediated through emotional exhaustion.

Thus, the BRS-F is a psychometrically sound assessment of self-perceived resilience, which is now available to researchers and clinicians in French-speaking contexts. The results also suggest that the BRS-F is relevant for use by healthcare professionals who may benefit from interventions aimed at increasing their resilience.

Keywords: Brief Resilience Scale, burnout, mediation, mental health difficulties, midwives

Over the past two decades, resilience has become a highly popular concept in psychological and medical science (Chmitorz, Kunzler, et al., 2018; Reich, Zautra, & Hall, 2010). Resilience captures a person’s ability to return quickly to the previous level of functioning despite experiences of significant adversity (i.e., bounce back or recover quickly from stress; Carver, 1998; Smith, Tooley, Christopher, & Kay, 2010). The construct of resilience can be divided into medical resilience (i.e., the objective physical recovery following illness or injury) and psychological resilience (i.e., the subjective recovery after adverse stressful events, which might include illness; e.g., Chmitorz, Kunzler, et al., 2018; Reich et al., 2010). In this paper, we adhere to psychological resilience. By emphasizing the ability to bounce back, resilience differs from related concepts, such as thriving (moving to a superior level of functioning following difficult experiences), adaptation (adjusting to a new, stressful situation), or psychological resistance (not becoming stressed or ill in the face of adversity; Carver, 1998).

Although some authors view resilience as a fixed, stable trait, resilience is usually regarded as an outcome or process in response to difficult experiences that is shaped by interactions between individual resources and one’s environment, leading to varying levels of resilience across the life-span (Windle, Bennett, & Noyes, 2011). The latter perspective on resilience is consistent with Smith et al.’s resilience model (Smith et al., 2008, 2010). In this model, the Ó 2019 Hogrefe Publishing

ability to regain homeostasis after experiencing significant stress is considered as a personal resource that is susceptible to change (e.g., by intervention). Smith et al. (2010) regard the belief that one possesses this resource as an important prerequisite for actually being able to recover quickly from stress. They assume that resilience selfefficacy develops when people, who are sufficiently equipped with coping resources, learn via experience, example, or encouragement that they are able to quickly restore their homeostasis. Unlike highly stable personality traits, resilience self-efficacy shows only moderate to high rank-order stability over intervals up to 6 months (e.g., Rodríguez-Rey, Alonso-Tapia, & Hernansaiz-Garrido, 2016). In this study, we operationalize resilience as the belief to be resilient (for simplicity, we continue to refer to resilience instead of resilience self-efficacy).

A review on resilience scales concluded that most resilience scales actually assess resources that likely promote resilience and resistance to illness (Windle et al., 2011), which is somewhat removed from the original construct.

For example, the Connor Davidson Resilience Scale (CDRISC; Connor & Davidson, 2003) assesses personality characteristics (e.g., tolerance of negative affect, personal competence, positive acceptance of change) that may act as protective factors and contribute to a resilient outcome.

However, it is incorrect to use these measures of resilience resources as direct indicators of resilience. Smith et al.

European Journal of Health Psychology (2019), 26(1), 1–9 https://doi.org/10.1027/2512-8442/a000022

2

(2008) accordingly showed that resilience remains specifically and negatively linked to emotional distress and physical symptoms when overlap with resilience resources (e.g., CD-RISC score) was statistically controlled for, whereas the converse held not true. Moreover, Lai and Yue (2014) showed that resilience mediates the effects of resilience resources (i.e., optimism, self-esteem) on physical health.

These findings imply that resilience is a more proximate predictor of health outcomes than the broader protective factors that promote one’s resilience.

The Brief Resilience Scale (BRS; Smith et al., 2008) assesses psychological resilience as a unitary construct.

Three positively and three negatively worded BRS items ask for one’s self-perceived ability to bounce back from stress (i.e., resilience self-efficacy) rather than one’s actual ability. Drawing on four samples of undergraduates, cardiac patients, and women with and without fibromyalgia, Smith et al. (2008) provided evidence for the reliability of the BRS in terms of Cronbach’s alpha (α .80), for its structural validity as a unifactorial scale by the means of Principal Component Analysis (PCA), for its known-group validity (e.g., group differences between women with and without fibromyalgia), for its convergent validity with measures of resilience resources, optimism, social support, and active coping, as well as for its discriminant predictive validity for physical symptoms, perceived stress, anxiety, depression, negative affect, and fatigue. In a recent review on resilience inventories (Windle et al., 2011), the BRS was the only scale that asked directly for one’s ability to recover from stress and it belonged to the three scales that received the highest psychometric ratings. Among others, the BRS has been translated into Dutch (Leontjevas, de Beek, Lataster, & Jacobs, 2014), Spanish (Rodríguez-Rey et al., 2016), Brazilian Portuguese (De Holanda Coelho, Cavalcanti, Rezende, & Gouveia, 2016), German (Chmitorz, Wenzel, et al., 2018), Chinese (Lai & Yue, 2014), and Malaysian (Amat, Subhan, Jaafar, Mahmud, & Johari, 2014). However, no French version of the BRS is available yet. Using a French-speaking sample of midwives, the current study thus aimed to introduce the French Brief Resilience Scale (BRS-F).

Principal component analyses of BRS items consistently extracted one eigenvalue > 1.00, implying a unidimensional structure (Amat et al., 2014; Lai & Yue, 2014; Smith et al., 2008). Confirmatory factor analyses (CFAs) of the BRS either supported two highly correlated first-order factors representing positively and negatively phrased items (Rodríguez-Rey et al., 2016), a unifactorial model based on five BRS items (De Holanda Coelho et al., 2016), or a major resilience factor along with a minor method factor for negatively worded items (Chmitorz, Wenzel, et al., 2018). We therefore expected that one major factor fits acceptably to the BRS-F data (Hypothesis 1a, H1a). We also European Journal of Health Psychology (2019), 26(1), 1–9

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

hypothesized that the component structure of the BRS-F and of the original BRS (Smith et al., 2008), the Chinese BRS (Lai & Yue, 2014) and the Malaysian BRS (Amat et al., 2014) can be considered as equal (Hypothesis 1b, H1b).

In prior research, the BRS demonstrated good levels of reliability with estimates of Cronbach’s α usually exceeding .80 (e.g., Amat et al., 2014; Rodríguez-Rey et al., 2016; Smith et al., 2008; for an exception see Lai & Yue, 2014).

However, coefficient α suffers several limitations and yields poor estimates of the reliability under some circumstances.

Thus, we will also report coefficient omega (ω; McDonald, 1999), which might be more revealing regarding the reliability of the BRS-F. In Chmitorz, Wenzel, et al. (2018), coefficient ω for the German BRS was ω = .85. We hypothesized that the BRS-F shows an adequate level of reliability that is comparable to the reliability of the original BRS (Hypothesis 2, H2).

Patient-care professionals are more vulnerable than other professionals to develop mental health difficulties (Aust, Rugulies, Skakon, Scherzer, & Jensen, 2007). Their mental health problems are linked to high quantitative, emotional, sensorial, and cognitive demands at work, a high rhythm of work, and a demand for hiding emotions (Aust et al., 2007).

Midwives frequently experience their job as stressful and conclude that lack of work resources and poor organization cause the most stress (Knezevic et al., 2011). The empathic nature of the caring relationship itself may also contribute to emotional suffering and mental health problems in midwives (Leinweber & Rowe, 2010; Sheen, Slade, & Spiby, 2014). More than two-thirds of midwives in Australia and over 95% of midwives in the UK had been exposed to a traumatic event at work (Leinweber, Creedy, Rowe, & Gamble, 2017; Sheen, Spiby, & Slade, 2015), such as managing traumatic births and perinatal loss (Sheen et al., 2014). Being frequently exposed to work-related stressors can cause mental health problems in midwives, such as anxiety (Muliira, Sendikadiwa, & Lwasampijja, 2015) and posttraumatic stress disorder (PTSD; Leinweber et al., 2017; Sheen et al., 2015). Resilience might thus be highly relevant for midwives, who are facing various work-related stressors on a regular basis. The ability to bounce back is thought to promote one’s mental and physical health, as has been confirmed in several populations (e.g., Gloria & Steinhardt, 2014 in postdoctoral research fellows; Lai & Yue, 2014 in Chinese undergraduates; Leontjevas et al., 2014 in Dutch residents of a nursing home rehabilitating unit; Rodríguez-Rey et al., 2016 in Spanish adults; Smith et al., 2008 in US undergraduates, cardiac and fibromyalgia patients), but not in midwives so far. In general, the work-related mental health of midwives is still understudied and this study thus aimed to fill an important gap (Favrod et al., 2018). We expected that the BRS-F score Ó 2019 Hogrefe Publishing

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

will correlate negatively with midwives’ anxiety, depression, and PTSD symptoms (Hypothesis 3a, H3a), which would bolster the criterion validity of the BRS-F. Burnout develops as a prolonged response to chronic interpersonal stressors on the job that consume, exceed, and exhaust one’s personal and social resources. It consists of three dimensions: an overwhelming emotional exhaustion, depersonalization (or feelings of cynicism), and (low) personal accomplishments (Maslach & Leiter, 2016). Burnout is a prevalent phenomenon among midwives (e.g., Hildingsson, Westlund, & Wiklund, 2013; Sheen et al., 2015) and resilience might prevent midwives from burning out.

In two samples of nurses, resilience resources (i.e., CDRISC) correlated negatively with emotional exhaustion and depersonalization, and positively with personal accomplishments (García-Izquierdo, de Pedro, Ríos-Risquez, & Sánchez, 2017; Zou et al., 2016). Research on health care practitioners also showed that resilience correlated positively with personal accomplishments and negatively with emotional exhaustion and depersonalization (Riley, Mohr, & Waddimba, 2018). These findings might generalize to midwives. We therefore expected that midwives’ BRS-F score will correlate negatively with emotional exhaustion and depersonalization, and positively with personal accomplishments (Hypothesis 3b, H3b).

Burnout, an occupation-specific dysphoria, is separable from depression, which is a more broadly based mental health problem. Depression and burnout are thus distinct, yet empirically related concepts (Maslach & Leiter, 2016).

Burnout is also negatively linked to one’s psychological health (e.g., García-Izquierdo et al., 2017) and positively linked to general psychological distress (e.g., Zou et al., 2016), and PTSD symptoms (e.g., Sheen et al., 2015). Given that higher levels of resilience are supposed to be linked to lower levels of burnout, which are in turn likely linked to better mental health, it was expected that burnout would mediate the resilience-mental health difficulties link (Hypothesis 3c, H3c).

Methods Participants and Procedure Recruitment took place at two university hospitals in the French-speaking part of Switzerland. During staff meetings and by the distribution of flyers, all midwives working at both hospitals were invited to participate. Staff accessing the anonymous online survey found a detailed information sheet before giving informed consent. The survey consisted of seven inventories (results of two inventories are not reported here) and took about 30 min to complete. The

Ó 2019 Hogrefe Publishing

3

ethics committee of the Canton de Vaud approved this study (study nr: 237/2013). Of the 280 eligible midwives, N = 220 participated (78.6% response rate). Results of this survey unrelated to the present study have been reported in Jacobs, Charmillot, Soelch, and Horsch (2018).

Measures The Brief Resilience Scale (BRS; Smith et al., 2008) is a 6item questionnaire designed to assess resilience as selfperceived ability to bounce back or recover quickly from stress. Each item is rated on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). The three negatively phrased items 2, 4, and 6 were recoded with the result that a higher score indicated a higher degree of resilience. The original BRS showed good levels of internal consistency and test-retest reliability, and adequate factorial, convergent and discriminant validity (Smith et al., 2008). The BRS was translated into French using forward-backward translation and cultural adaptation (Wild et al., 2005). A native French-speaking mental health professional, familiar with terminology of the concept measured by the questionnaire and knowledgeable of the English-speaking culture conducted the forward translation. An independent translator, whose mother tongue was English and who had no knowledge of the questionnaire, then translated the questionnaire back to English. For both of these steps, emphasis was placed on conceptual and cultural, rather than linguistic equivalence (literal translation). The last author (AH), fluent in both English and French languages, then compared both, the original, as well as the translated English version, and discussed any problematic words or phrases that did not completely capture the concept addressed by the original items. Finally, the translated questionnaire was piloted in the target population (three midwives). No final adaptations were required.

The Hospital Anxiety and Depression Scale – French version (HADS; Bocerean & Dupret, 2014) assesses anxiety and depression with two 7-item subscales. Each item is scored from 0 to 3, with higher scores indicating greater anxiety or depression. In the current study, ordinal Cronbach’s α for the anxiety and depression subscales were .78 and .77, respectively. The HADS was chosen for this study because it has been widely used (e.g., Smith et al., 2008) and is a well-validated short screening questionnaire.

The Posttraumatic Stress Disorder 7-item Symptom Scale (PTSD-7; Breslau, Peterson, Kessler, & Schultz, 1999) is a brief screening scale for DSM-IV PTSD. It measures five symptoms from the avoidance and numbing symptom cluster and two symptoms from the hyperarousal cluster using a dichotomous yes/no response format. In the current study, coefficient α based on tetrachoric correlations was good,

European Journal of Health Psychology (2019), 26(1), 1–9

4

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

Table 1. Descriptive statistics, coefficient α (in parenthesis), and correlations between study variables 1

2

3

4

1. Brief Resilience Scale

(.84)

2. Depression

.24***

(.77)b

3. Anxiety

.19**

.64***

(.78)b

4. PTSD-7

.27***

.37***

.29***

(.83)b

5

5. Emotional exhaustion

.34***

.37***

.37***

.49***

(.87)

6. Depersonalizationa

.23***

.23***

.23***

.25***

.50***

7. Personal accomplishments

.33***

.31***

.17**

.30***

.39***

6

7

(.69) .37***

(.66)

Mean

3.53

5.31

8.38

1.94

19.13

4.64

32.23

Standard Deviation

0.74

3.65

4.07

1.77

9.58

3.94

5.42

Notes. N = 219–220. aThree scores were altered in order to reduce the impact of univariate outliers (see Tabachnick & Fidell, 2014, p. 111). bCoefficient alpha is based on polychoric or tetrachoric correlations. **p < .01, ***p < . 001 (two-tailed).

α = .83. The PTSD-7 was included in this study because it is a validated short screening questionnaire for PTSD. The French Maslach Burnout Inventory (MBI; Dion & Tessier, 1994) captures three core dimensions of burnout (Maslach & Jackson, 1981): emotional exhaustion (i.e., feeling exhausted and emotionally overextended by one’s work; 9 items), depersonalization (i.e., impersonal response toward recipients of one’s service; 5 items), and personal accomplishment (i.e., feeling competent and successful in one’s work; 8 items). Each item is rated on a 7-point scale (1 = never to 7 = every day). The French MBI showed good psychometric properties (Dion & Tessier, 1994). In the present study, Cronbach’s α of the three MBI subscales ranged from .66 to .87 (for descriptive statistics see Table 1).

Data Analyses In order to test the fit of the unifactorial model, the alternative two-factorial model (Rodríguez-Rey et al., 2016), and the alternative method-factor model (Chmitorz, Wenzel, et al., 2018), item-level CFAs were performed using EQS 6.2 (Bentler, 2006). The CFAs were based on the standard covariance matrix and robust maximum likelihood estimation (Satorra & Bentler, 2001). The robust w2-statistic was complemented by four fit indices (cf. Brown, 2006): the comparative fit index (CFI), the Tucker-Lewis Index (TLI; CFI & TLI: acceptable fit .90; good fit .95), the root mean square error of approximation (RMSEA; reasonable fit .08; close fit .05), and the standardized rootmean-square residual (SRMR; acceptable fit .08; good fit .05). For model comparisons, we also used Akaike’s Information Criterion (AIC; the model with the smaller AIC fits better). Next, an item-level PCA was conducted, 1

one component was retained, and the loading vector was compared with the respective vectors obtained for the original BRS (Smith et al., 2008), the Chinese BRS (Lai & Yue, 2014), and the Malaysian BRS (Amat et al., 2014). The level of congruence was evaluated with Tucker’s ϕ. LorenzoSeva and ten Berge (2006) suggested that 0.85 φ 0.94 indicates a fair similarity, while φ 0.95 implies that the two components compared can be considered as equal.

Coefficient ω was estimated with the free software JASP version 0.9 (JASP Team, 2018).

In order to establish criterion validity, correlations between the BRS-F score, mental health, and burnout variables will be presented. Finally, a resilience-burnout-mental health difficulties mediation model with three parallel mediators (depersonalization, emotional exhaustion, personal accomplishments) was tested. The outcome variable was created by aggregating the z-scored anxiety, depression, and PTSD-7 subscale scores. Mediation analysis was carried out in IBM SPSS 22 and PROCESS (Hayes, 2013) using ordinary least squares regression analysis. The 95% bias-corrected confidence intervals of the indirect effects were based on 5,000 bootstrap resamples. Significance of an indirect effect was implied when the 95% CI precluded zero. In all analyses, an a priori significance level of α = .05 was chosen.1

Results The factor structure of the French Brief Resilience Scale Mardia’s normalized estimate of multivariate kurtosis was 7.82, indicating the need for the robust Satorra-Bentlerscaled w2-test statistic. The unidimensional model failed

The current study was not preregistered. In order to increase transparency, data of the current study are available upon individual request to the first author.

European Journal of Health Psychology (2019), 26(1), 1–9

Ó 2019 Hogrefe Publishing

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

5

Table 2. Descriptive statistics, corrected item-total correlations (rit), component loadings (PCA), and standardized factor loadings and explained variance (R2) in the CFA of the BRS-F items Range

rit

Loadings (PCA)

Loadings (CFA)

R2

0.43

1–5

.57

.70

.63

.40

0.43

0.58

1–5

.65

.77

.71

.51

0.95

0.63

0.12

1–5

.64

.77

.71

.50

3.54

1.07

0.51

0.51

1–5

.51

.65

.57

.32

5. En général je traverse les moments difficiles sans trop de difficulté.

3.31

1.00

0.36

0.75

1–5

.62

.75

.70

.49

6. J’ai tendance à prendre beaucoup de temps pour me remettre des revers dans ma vie.

3.51

1.08

0.46

0.67

1–5

.69

.81

.76

.58

BRS-F items

M

SD

Skew

1. Je tends à rebondir rapidement après des moments difficiles.

3.95

0.89

0.87

2. J’ai du mal à traverser des événements stressants.

3.43

1.01

3. Je me remets facilement d’un événement stressant.

3.44

4. Il est difficile pour moi de revenir brusquement à la réalité quand quelque chose se passe mal.

Kurtosis

Note. PCA = principal component analysis; CFA = confirmatory factor analysis.

to fit perfectly to the data, SB-w2 (df = 9) = 18.51, p = .030.

However, three fit indices signaled a good model fit (CFI = .97, TLI = .95, SRMR = .04), whereas the RMSEA = .07 indicated a reasonable model fit (AIC = 0.51). All loadings were substantial in size, the mean loading was .68 (range: .57 – .76; see Table 2). The amount of variance that the factor contributed to the items ranged from R2 = .32 to .58.

Thus, the unifactorial model fitted well to the data, the factor structure was meaningful and well-defined with four loadings exceeding .70. In the two-factorial model (Rodríguez-Rey et al., 2016), positively and negatively phrased items loaded on the respective factors and both factors were allowed to correlate. This model reached a good model fit, SB-w2 (df = 8) = 17.80, p = .023, CFI = .97, TLI = .95, SRMR = .04, and RMSEA = .08, but it did not yield a better fit than the unifactorial model in terms of the w2-difference test, ΔSB-w2(df = 1) = 0.59, p = .444, and it was even inferior, as indicated by a slightly greater AIC, AIC = 1.80. In the method-factor model (Chmitorz, Wenzel, et al., 2018), a general resilience factor and an uncorrelated method factor with loadings on all negatively phrased items were specified. This model showed a good model fit, SB-w2 (df = 6) = 13.45, p = .031, CFI = .98, TLI = .97, SRMR = .03, and RMSEA = .08, but it showed a slightly poorer fit than the unifactorial model in terms of the AIC, AIC = 1.45. Moreover, the method factor was poorly defined (standardized loadings: λ2 = .05, λ4 = .86, and λ6 = .13). Taken together, the present data provided support for the unifactorial model, which yielded a good model fit and it fitted comparably or even slightly better to the data than the alternative, less constrained two-factorial and method-factor models (H1a confirmed).

A PCA on the BRS-F item scores extracted one eigenvalue > 1.00 (i.e., 3.32) suggesting one component to retain. Ó 2019 Hogrefe Publishing

This component accounted for 55.3% of the variance in the BRS-F items. The mean loading was .75 (range: .65–.81; see Table 2). Similar to the CFA results, item 4 showed the lowest and item 6 the highest loading. According to Tucker’s φ, the PCA loadings depicted in Table 2 were highly congruent with the loadings reported in Smith et al. (2008) for the original BRS (samples 1–4: φ = 0.99, 0.99, 1.00, and 0.98, respectively), in Amat et al. (2014) for the Malaysian version of the BRS (φ = 1.00), and in Lai and Yue (2014) for the Chinese version of the BRS (Hong Kong data: φ = 0.97; Nanjing data: φ = 0.98). Thus, the component structure of the BRS-F equals the component structure of the original BRS and its Chinese and Malaysian versions (H1b confirmed).

Descriptive Statistics and Reliabilities In the current sample, Cronbach’s α of the BRS-F was α = .84. Drawing on the F-test to compare coefficients α from independent samples (Feldt, Woodruff, & Salih, 1987), coefficient α of the BRS-F did not significantly differ from the respective coefficients α obtained for the original BRS (Samples 1–3; Smith et al., 2008) and for the Spanish BRS (Rodríguez-Rey et al., 2016), all p .11. However, it was significantly smaller than the respective estimates of α obtained in the fourth US-sample (Smith et al., 2008) and in the Malaysian sample (Amat et al., 2014), and it was significantly larger compared to estimates of α in both Chinese samples (Lai & Yue, 2014), all p < .01. Coefficient ω was .84, which is almost identical with the estimates of ω obtained for the German BRS (Chmitorz, Wenzel, et al., 2018). This indicates that a high proportion of test variance was due to a general resilience factor. Taken together, European Journal of Health Psychology (2019), 26(1), 1–9

6

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

Figure 1. Standardized path coefficients in the resilience–burnout–mental health difficulties mediation model (N = 219). ***p < .001.

the results support the reliability of the BRS-F (H2 confirmed).

For all six items, participants used the full range of the five-point scale. Item means ranged from M = 3.31 to 3.95 suggesting moderate item difficulties and items were slightly skewed and kurtotic (see Table 2). All corrected item-total correlations were satisfactorily, ranging from rit = .51 to .69. Thus, the BRS-F also showed good distributional properties and satisfactory item-total correlations.

Correlational Analysis and Mediation Analysis Midwives who indicated poorer resilience also indicated more symptoms of anxiety, depression and PTSD, more emotional exhaustion and depersonalization, and less personal accomplishments (for correlations see Table 1), thus providing support for the concurrent validity of the BRS-F (H3a and H3b confirmed).

The standardized path coefficients for the resilienceburnout-mental health difficulties mediation model are shown in Figure 1. The total effect of resilience accounted for 8.8% of the variance in mental health difficulties, R2 = .09, F(1, 218) = 21.05, p < .001. When the three core dimensions of burnout entered the model, a total of 29.1% of variance in mental health difficulties was explained, R2 = .29, F(4, 215) = 22.06, p < .001. Only a significant partial effect for emotional exhaustion was found, β = .43, p < .001. The direct effect for resilience, β = .11, p = .076, and the specific effects for depersonalization, β = .01, p = .89, and personal accomplishments, β = .12, p = .070, were not statistically significant. The bias-corrected 95% CI for the total indirect effect, ab = .18 [ .27, .12], precluded zero suggesting significance. A significant portion of the indirect effect was mediated via emotional exhaustion, ab = .15 [ .23, .09]. Depersonalization, ab = .002 [ .03, .04], and personal accomplishments, ab = .04 [ .09, .002], did not act as significant specific mediators. The pattern European Journal of Health Psychology (2019), 26(1), 1–9

of a significant total effect, a non-significant direct effect and a significant indirect effect are consistent with full mediation (H3c confirmed).

Discussion This study translated and validated the French version of the BRS (BRS-F) in a sample of midwives. The BRS operationalizes resilience as the self-perceived ability to bounce back (Smith et al., 2008). The unifactorial model showed an acceptable to good model fit, thus confirming the notion of resilience as a unitary construct (De Holanda Coelho et al., 2016; Smith et al., 2008). The less constrained alternative two-factor model (Rodríguez-Rey et al., 2016) and the method-factor model (Chmitorz, Wenzel, et al., 2018) did not outperform the unifactorial model. This finding might reflect actual differences in the meaning of resilience between cultures (i.e., German, Spanish, Swiss, Brazilian) or populations (e.g., representative samples, convenience samples, midwives). More stringent research using multi-group CFA based on comparable, sufficiently sized samples from diverse cultures is needed to test the factorial invariance of the BRS across cultures. If factorial invariance can be established, results obtained with the BRS from different cultures can be reliably compared. The present results confirmed at least that the component structure of the BRS-F is comparable with the structure of the original BRS and its Chinese and Malaysian versions. Finding evidence for an equal component structure is a first important step in showing that the structure of a scale may generalize across different cultures and languages (McCrae & Costa, 1997).

The BRS-F also showed good levels of reliability. Cronbach’s α in our study was comparable to those of the original BRS (Samples 1–3 in Smith et al., 2008) and the Spanish BRS (Rodríguez-Rey et al., 2016). However, α was significantly smaller than the αs obtained in the US sample 4 (Smith et al., 2008) and in the Malaysian sample Ó 2019 Hogrefe Publishing

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

(Amat et al., 2014), and larger compared to both Chinese samples (Lai & Yue, 2014). We therefore concluded that it is situated in the middle range of previously reported estimates of α for different versions of the BRS. Moreover, the high level of coefficient omega indicates that the BRS-F total score estimates a latent resilience factor that is common to all six items at a high precision. The high level of precision is comparable with results obtained for the German BRS (Chmitorz, Wenzel, et al., 2018).

Criterion validity was established by negative correlations between the BRS-F score and depression, anxiety, PTSD symptoms, emotional exhaustion, and depersonalization, and a positive correlation with personal accomplishments. Although, as far as we know, shown for the first time in a sample of midwives, the latter findings are in line with research on healthcare practitioners showing substantial correlations between resilience and core dimensions of burnout (e.g., Riley et al., 2018). The negative correlations between resilience and mental health variables have not been previously shown in midwives. They are consistent with the more general notion that the ability to bounce back promotes one’s mental and physical health (e.g., Gloria & Steinhardt, 2014; Lai & Yue, 2014; Leontjevas et al., 2014; Rodríguez-Rey et al., 2016; Smith et al., 2008, 2010).

Finally, this study demonstrated as far as we know for the first time that the relationship between midwives’ resilience and mental health difficulties was fully mediated by their emotional exhaustion. This finding integrates and extends prior research on healthcare professionals showing that resilience resources relate negatively to emotional exhaustion (Zou et al., 2016), and that emotional exhaustion relates positively to mental health difficulties (GarcíaIzquierdo et al., 2017), to general psychological distress (Zou et al., 2016), and to PTSD symptoms (Sheen et al., 2015). A resilient person is more likely to restore homeostasis in the face of occupational stress, thus feels probably more recovered from it, and experiences lower exhaustion as a consequence, which in turn lowers his or her risk for mental health difficulties. Stated differently, emotional exhaustion is a central mechanism that links resilience to mental health problems.

Our results indicate that the BRS-F is appropriate and relevant for use by midwives who may benefit from interventions aimed at increasing resilience (see Chmitorz, Kunzler, et al., 2018, for a review) to enable them to cope with the day-to-day emotional demands and stressors of their work.

For example, workshops based on stress inoculation techniques may be helpful (Meichenbaum, 1977). More recently, Grant and Kinman (2012) developed interactive workshops that cover stress management skills, such as relaxation and time management, as well as sessions that enhance competencies linked to resilience, such as emotional intelligence, reflective practice, social awareness, Ó 2019 Hogrefe Publishing

7

and empathetic skills. Another idea is the use of challenging patient scenarios that do not fit within learned “rules” as an integral part of ethics teaching as a way of developing resilience (Howe, Smajdor, & Stöckl, 2012).

This study has several limitations: First, the crosssectional design prevents causal interpretations of the results. Second, data were collected using an internet survey, which may raise questions regarding the quality of the data. However, in previous studies paper-and-pencil and internet-based data collection methods resulted in equivalent data (e.g., Weigold, Weigold, & Russell, 2013).

Third, findings are based on self-report data, which might have biased the results (e.g., socially desirable responding).

Fourth, the present study draws on a sample of midwives, which may limit the generalizability of the results. In future research, the reliability and validity of the BRS-F should thus be tested in the general population as well. Fifth, although the observed negative associations between resilience and the mental health indicators provide support for the criterion validity of the BRS-F, the full mediation of the resilience-mental health difficulties link might somewhat mitigate the liability of the criterion analyses. More studies including a broader spectrum of health outcomes (e.g., externalizing pathology, physical symptoms, healthrelated quality of life) are needed to further substantiate the criterion validity of the BRS-F. Finally, the total effect of resilience and the effect of personal accomplishments on mental health difficulties in the mediation model barely missed statistical significance. Both findings might reflect insufficient statistical power of the current study for detecting small effects. More research with larger samples is needed in order to test whether resilience remains weakly and specifically related to mental health difficulties in the presence of the burnout core dimensions, and whether personal accomplishments establish a second pathway for mediation that might follow from its weak effect on mental health difficulties. Despite these limitations, the present study demonstrates the validity and reliability of the BRSF in French-speaking midwives and thus makes its accessible to researchers and clinicians in French-speaking environments. Future studies may replicate the current findings and aim to investigate the link between resilience and indicators of quality of care in different groups of healthcare professionals.

References Amat, S., Subhan, M., Jaafar, W. M. W., Mahmud, Z., & Johari, K. S. K. (2014). Evaluation and psychometric status of the Brief Resilience Scale in a sample of Malaysian international students. Asian Social Science, 10, 240–245. https://doi.org/ 10.5539/ass.v10n18p240 Aust, B., Rugulies, R., Skakon, J., Scherzer, T., & Jensen, C. (2007).

Psychosocial work environment of hospital workers: Validation

European Journal of Health Psychology (2019), 26(1), 1–9

8

of a comprehensive assessment scale. International Journal of Nursing Studies, 44, 814–825. https://doi.org/10.1016/j. ijnurstu.2006.01.008 Bentler, P. M. (2006). EQS 6 structural equation program manual.

Encino, CA: Mutivariate Software, Inc. Bocerean, C., & Dupret, E. (2014). A validation study of the Hospital Anxiety and Depression Scale (HADS) in a large sample of French employees. BMC Psychiatry, 14, 354. https://doi.org/ 10.1186/s12888-014-0354-0 Breslau, N., Peterson, E. L., Kessler, R. C., & Schultz, L. R. (1999).

Short screening scale for DSM-IV posttraumatic stress disorder. American Journal of Psychiatry, 156, 908–911. https://doi. org/10.1176/ajp.156.6.908 Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York, NY: Guilford Press.

Carver, C. S. (1998). Resilience and thriving: Issues, models, and linkages. Journal of Social Issues, 54, 245–266. https://doi.org/ 10.1111/j.1540-4560.1998.tb01217.x Chmitorz, A., Kunzler, A., Helmreich, I., Tüscher, O., Kalisch, R., Kubiak, T., . . . Lieb, K. (2018). Intervention studies to foster resilience – a systematic review and proposal for a resilience framework in future intervention studies. Clinical Psychology Review, 59, 78–100. https://doi.org/10.1016/j.cpr.2017.11.002 Chmitorz, A., Wenzel, M., Stieglietz, R.-D., Kunzler, A., Bagusat, C., Helmreich, I., . . . Tüscher, O. (2018). Population based validation of a German version of the Brief Resilience Scale. PLoS One, 13, e0192761. https://doi.org/10.1371/journal.pone.

0192761 Connor, K. M., & Davidson, J. R. T. (2003). Development of a new resilience scale: The Connor-Davidson Resilience Scale (CDRISC). Depression and Anxiety, 18, 76–82. https://doi.org/ 10.1002/da.10113 De Holanda Coelho, G. L. H., Cavalcanti, T. M., Rezende, A. T., & Gouveia, V. V. (2016). Brief Resilience Scale: Testing its factorial structure and invariance in Brazil. Universitas Psychologica, 15, 397–408. https://doi.org/10.11144/Javeriana.upsy15-2.brst Dion, G., & Tessier, R. (1994). Validation de la traduction de l’lnventaire d’epuisement professionnel de Maslach et Jackson [Validation of the translation of Maslach and Jackson’s Burnout Inventory]. Revue Canadienne des Sciences du Comportement, 26, 210–227. https://doi.org/10.1037/0008400X.26.2.210 Favrod, C., Jan du Chêne, L., Martin Soelch, C., Garthus-Niegel, S., Tolsa, J. F., Legault, F., . . . Horsch, A. (2018). Mental health symptoms and work-related stressors in hospital midwives and NICU nurses: A mixed methods study. Frontiers in Psychiatry, 9, 364. https://doi.org/10.3389/fpsyt.2018.00364 Feldt, L. S., Woodruff, D. J., & Salih, F. A. (1987). Statistical inference for coefficient alpha. Applied Psychological Measurement, 11, 92–103. https://doi.org/10.1177/014662168701100107 García-Izquierdo, M., de Pedro, M., Ríos-Risquez, M., & Sánchez, M. (2017). Resilience as a moderator of psychological health in situations of chronic stress (burnout) in a sample of hospital nurses. Journal of Nursing Scholarship, 50, 228–236. https://doi.org/10.1111/jnu.12367 Gloria, C. T., & Steinhardt, M. A. (2014). Relationships among positive emotions, coping, resilience and mental health. Stress and Health, 32, 145–156. https://doi.org/10.1002/smi.2589 Grant, L., & Kinman, G. (2012). Enhancing wellbeing in social work students: Building resilience in the next generation. Social Work Education, 31, 605–621. https://doi.org/10.1080/02615479.

2011.590931 Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis. A regression-based approach. New York, NY: Guilford Press.

European Journal of Health Psychology (2019), 26(1), 1–9

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

Hildingsson, I., Westlund, K., & Wiklund, I. (2013). Burnout in Swedish midwives. Sexual & Reproductive Healthcare, 4, 87–91. https://doi.org/10.1016/j.srhc.2013.07.001 Howe, A., Smajdor, A., & Stöckl, A. (2012). Towards an understanding of resilience and its relevance to medical training.

Medical Education, 46, 349–356. https://doi.org/10.1111/ j.1365-2923.2011.04188.x Jacobs, I., Charmillot, M., Soelch, C. M., & Horsch, A. (2018).

Validity, reliability, and factor structure of the Secondary Traumatic Stress Scale – French version. Manuscript submitted for publication.

JASP Team. (2018). JASP (Version 0.9) [Computer software].

Amsterdam, The Netherlands: University of Amsterdam.

Knezevic, B., Milosevic, M., Golubic, R., Belosevic, L., Russo, A., & Mustajbegovic, J. (2011). Work-related stress and work ability among Croatian university hospital midwives. Midwifery, 27, 146–153. https://doi.org/10.1016/j.midw.2009.04.002 Lai, J. C., & Yue, X. (2014). Using the Brief Resilience Scale to assess Chinese people’s ability to bounce back from stress.

Sage Open, 4, 1–9. https://doi.org/10.1177/2158244014554386 Leinweber, J., Creedy, D. K., Rowe, H., & Gamble, J. (2017).

Responses to birth trauma and prevalence of posttraumatic stress among Australian midwives. Women & Birth, 30, 40–45. https://doi.org/10.1016/j.wombi.2016.06.006 Leinweber, J., & Rowe, H. J. (2010). The costs of “being with the woman”: Secondary traumatic stress in midwifery. Midwifery, 26, 76–87. https://doi.org/10.1016/j.midw.2008.04.003 Leontjevas, R., de Beek, W. O., Lataster, J., & Jacobs, N. (2014).

Resilience to affective disorders: A comparative validation of two resilience scales. Journal of Affective Disorders, 168, 262–268. https://doi.org/10.1016/j.jad.2014.07.010 Lorenzo-Seva, U., & ten Berge, J. M. (2006). Tucker’s congruence coefficient as a meaningful index of factor similarity. Methodology, 2, 57–64. https://doi.org/10.1027/1614-2241.2.2.57 Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Organizational Behavior, 2, 99–113.

Maslach, C., & Leiter, M. P. (2016). Understanding the burnout experience: Recent research and its implications for psychiatry.

World Psychiatry, 15, 103–111. https://doi.org/10.1002/wps.

20311 McCrae, R. R., & Costa, P. T. (1997). Personality trait structure as a human universal. The American Psychologist, 52, 509–516. https://doi.org/10.1037/0003-066X.52.5.509 McDonald, R. P. (1999). Test theory: A unified treatment. Mahwah, NJ: Erlbaum.

Meichenbaum, D. (1977). Cognitive behaviour modification.

Cognitive Behaviour Therapy, 6, 185–192. https://doi.org/ 10.1080/16506073.1977.9626708 Muliira, R. S., Sendikadiwa, V. B., & Lwasampijja, F. (2015).

Predictors of death anxiety among midwives who have experienced maternal death situations at work. Maternal and Child Health Journal, 19, 1024–1032. https://doi.org/10.1007/ s10995-014-1601-1 Reich, J. W., Zautra, A. J., & Hall, J. S. (Eds.). (2010). Handbook of adult resilience. New York, NY: Guilford Press.

Riley, M. R., Mohr, D. C., & Waddimba, A. C. (2018). The reliability and validity of three-item screening measures for burnout: Evidence from group-employed health care practitioners in upstate New York. Stress and Health, 34, 187–193. https://doi. org/10.1002/smi.2762 Rodríguez-Rey, R., Alonso-Tapia, J., & Hernansaiz-Garrido, H. (2016). Reliability and validity of the Brief Resilience Scale (BRS) Spanish version. Psychological Assessment, 28, e101–e110. https://doi.org/10.1037/pas0000191

Ó 2019 Hogrefe Publishing

I. Jacobs & A. Horsch, Psychometric Properties of the BRS-F

Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507–514. https://doi.org/10.1007/BF02296192 Sheen, K., Slade, P., & Spiby, H. (2014). An integrative review of the impact of indirect trauma exposure in health professionals and potential issues of salience for midwives. Journal of Advanced Nursing, 70, 729–743. https://doi.org/10.1111/jan.12274 Sheen, K., Spiby, H., & Slade, P. (2015). Exposure to traumatic perinatal experiences and posttraumatic stress symptoms in midwives: Prevalence and association with burnout. International Journal of Nursing Studies, 52, 578–587. https://doi.org/ 10.1016/j.ijnurstu.2014.11.006 Smith, B. W., Dalen, J., Wiggins, K., Tooley, E. M., Christopher, P. J., & Bernard, J. (2008). The brief resilience scale: assessing the ability to bounce back. International Journal of Behavioral Medicine, 15, 194–200. https://doi.org/10.1080/ 10705500802222972 Smith, B. W., Tooley, E. M., Christopher, P. J., & Kay, V. S. (2010).

Resilience as the ability to bounce back from stress: A neglected personal resource? The Journal of Positive Psychology, 5, 166–176. https://doi.org/10.1080/17439760.2010. 482186 Tabachnick, B. G., & Fidell, L. S. (2014). Using multivariate statistics. Essex, UK: Pearson.

Weigold, A., Weigold, I. K., & Russell, E. J. (2013). Examination of the equivalence of self-report survey-based paper-and-pencil and internet data collection methods. Psychological Methods, 18, 53–70. https://doi.org/10.1037/a0031607 Wild, D., Grove, A., Martin, M., Eremenco, S., McElroy, S., VerjeeLorenz, A., & Erikson, P. (2005). Principles of good practice for the translation and cultural adaptation process for patientreported outcomes (PRO) measures: report of the ISPOR task force for translation and cultural adaptation. Value in Health, 8, 94–104. https://doi.org/10.1111/j.1524-4733.2005.04054.x

Ó 2019 Hogrefe Publishing

9

Windle, G., Bennett, K. M., & Noyes, J. (2011). A methodological review of resilience measurement scales. Health and Quality of Life Outcomes, 9, 8. https://doi.org/10.1186/1477-7525-9-8 Zou, G., Shen, X., Tian, X., Liu, C., Li, G., Kong, L., & Li, P. (2016).

Correlates of psychological distress, burnout, and resilience among Chinese female nurses. Industrial Health, 54, 389–395. https://doi.org/10.2486/indhealth.2015-0103 History Received June 15, 2018 Revision received January 21, 2018 Accepted January 23, 2019 Published online June 3, 2019 Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sections.

Conflict of Interest The authors are not aware of any conflict of interests.

ORCID Ingo Jacobs https://orcid.org.0000-0002-8458-1134 Ingo Jacobs Department Natural Sciences Medical School Berlin Calandrellistraße 1-9 12247 Berlin Germany ingojacobs@yahoo.de

European Journal of Health Psychology (2019), 26(1), 1–9

Original Article

The PREVIEW Study Supporting Behavior Change in an International Intervention Study Among Participants With Pre-Diabetes Maija Huttunen-Lenz1 , Sylvia Hansen1, Thomas Meinert Larsen2, Pia Christensen2, Mathijs Drummen3, Tanja Adam3, Moira A. Taylor4, Elizabeth Simpson4, Jose A. Martinez5,6, Santiago Navas-Carretero5,6, Teodora Handjieva-Darlenska7, Sally D. Poppitt8, Marta P. Silvestre8, Mikael Fogelholm9, Elli Jalo9, Roslyn Muirhead10, Shannon Brodie10, Anne Raben2, and Wolfgang Schlicht1 1

Department of Exercise and Health Sciences, University of Stuttgart, Germany

2

Department of Nutrition, Exercise and Sports, University of Copenhagen, Denmark

3

Department of Nutrition and Movement Sciences, Maastricht University, The Netherlands

4

School of Life Sciences, University of Nottingham, United Kingdom Centre for Nutrition Research, University of Navarra, Pamplona, Spain

5 6

CIBER Fisiopatología Obesidad y Nutrición (CIBERobn), Instituto de Salud Carlos III, Madrid, Spain

7

Department of Pharmacology and Toxicology, Medical University of Sofia, Bulgaria

8

School of Biological Sciences, University of Auckland, Human Nutrition Unit, New Zealand

9

Department of Food and Nutrition, University of Helsinki, Finland

10

Charles Perkins Centre and School of Life and Environmental Biosciences, University of Sydney, Australia

Abstract: Individuals at risk of Type 2 Diabetes are advised to change health habits. This study investigated how the PREMIT behavior modification intervention and its association with socio-economic variables influenced weight maintenance and habit strength in the PREVIEW study. Overweight adults with pre-diabetes were enrolled (n = 2,224) in a multi-center RCT including a 2-month weight-loss phase and a 34-month weight-maintenance phase for those who lost 8% body weight. Initial stages of the PREMIT covered the end of weight-loss and the beginning of weight-maintenance phase (18 weeks). Cross-sectional and longitudinal data were explored. Frequent PREMIT sessions attendance, being female, and lower habit strength for poor diet were associated with lower weight re-gain. Being older and not in employment were associated with lower habit strength for physical inactivity. The PREMIT appeared to support weight loss maintenance. Younger participants, males, and those in employment appeared to struggle more with inactivity habit change and weight maintenance.

Keywords: habits, behavior therapy, prevention, type 2 diabetes mellitus, weight loss

The most common form of diabetes is type 2 diabetes (T2D) (Guariguata et al., 2014; Tamayo et al., 2014). Typically, development of T2D is a gradual process over a number of years. The intermediate stage between normal glucose metabolism and T2D is described as pre-diabetes, characterized by impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or both. The study here gives insight into a part of the PREVention of diabetes through lifestyle Intervention and population studies in Europe and around the World (PREVIEW) lifestyle (diet and exercise) intervention (EU FP7 grant agreement no. 312057).

PREVIEW was an 8-center randomized controlled trial (RCT) whose aim was to identify an effective combination of diet and physical activity to decrease T2D risk in

European Journal of Health Psychology (2019), 26(1), 10–20 https://doi.org/10.1027/2512-8442/a000026

overweight participants with pre-diabetics (Fogelholm et al., 2017).

As a part of the PREVIEW RCT, participants’ efforts to change diet and physical activity habits were supported by a theory- and evidence-based behavior modification intervention PREview behavior Modification Intervention Toolbox (PREMIT; Kahlert et al., 2016). The PREMIT provided a stage-based approach (see Figure 1) to change health risk behaviors (e.g., Prochaska & DiClemente, 1992). Habit change was supported by modifying behavioral determinants such as motivation, volition (i.e., commitment to a particular action), knowledge, skills, and social support (Fishbein et al., 2001; Michie, Johnston, Francis, Hardeman, & Eccles, 2008) during preliminary Ó 2019 Hogrefe Publishing

M. Huttunen-Lenz et al., Supporting Health Behavior Change

11

Figure 1. Timeline of the PREVIEW RCT, PREMIT program and its objectives by session up to the end of Stage 3.

(Stage 1), preparation (Stage 2), and action (Stage 3) stages.

While the PREMIT behavior modification intervention was influenced by different theoretical models such as the Health Action Process Approach (Schwarzer, 2001), the Social Cognitive Theory (Bandura, 1996), and the SelfDetermination Theory (Ryan & Deci, 2000), the focus in here, however, was not in testing the assumptions of those theoretical models explaining habit formation. Several studies have shown that social-cognitive variables such as attitudes, risk perception, outcome expectancies, self-efficacy, action planning, and motivation have been associated with habit formation (e.g., Bandura, 1996; Renner & Schwarzer, 2005; Ryan, Patrick, Deci, & Williams, 2008). The focus here was to initially explore associations between the PREMIT, weight-maintenance, and socio-economic variables, then examine self-reported changes in habit strength (e.g., Gardner, Lally, & Wardle, 2012), and its association with weight-maintenance and the PREMIT participation.

Habits describe strong, situational, and unconsciously triggered predictors of automated behavior (Gardner, Corbridge, & McGowan, 2015; de Vries, Eggers, Lechner, Van Osch, & Van Stralen, 2014). Habits are formed through repeated performance of an action, and at the core of a habit is a cue-dependent automaticity, which is separate of its cause, that is, context-dependent repetition of habit (Gardner, 2012). Therefore, habit can be understood as the automaticity of a behavior rather than frequency of a behavior, which once formed does not need to be frequently performed (Gardner, 2012). Habit strength, on the other hand, refers to the strength of a process where an impulse to behave in a certain way is instigated upon Ă&#x201C; 2019 Hogrefe Publishing

encountering a setting in which the behavior has been performed in the past, that is, frequency of a behavior in a context (e.g., time, place, circumstance; Gardner et al., 2015; Labrecque & Wood, 2015; de Vries et al., 2014). Strong habit strength describes frequent performance of behaviors in stable contexts, while weak habit strength describes performance of behaviors infrequently or in unstable contexts (Labrecque & Wood, 2015). Strong habit strength may reduce the impact of intentions on behaviors, and thus habit strength may predict future behaviors (Ji & Wood, 2007; Williams, Wood, Collins, & Callister, 2015; Wood, Tam, & Guerrero Witt, 2005).

In the context of health behavior change, research has indicated that introducing small habit changes in daily routines show promise in implementing long-term behavior changes. In this way, after an initial learning stage, behavior becomes an automation (i.e., habit) requiring little cognitive effort to be performed (Gardner et al., 2012; Lally & Gardner, 2013). Further, research has suggested that habit strength is a strong predictor of health-related behaviors, including unhealthy snacking (Verhoeven, Adriaanse, Evers, & De Ridder, 2012) and physical activity (Phillips & Gardner, 2016).

Previously, higher levels of physical activity have been associated with factors such as greater educational achievement, perceived behavioral control, and better knowledge of the advantages of physical activity. Lower levels of physical activity, on the other hand, have been associated with factors such as older age, a higher body weight, and being in paid employment (Marques-Vidal et al., 2015; Mesters, Wahl, & Van Keulen, 2014). Factors such as younger age, European Journal of Health Psychology (2019), 26(1), 10â&#x20AC;&#x201C;20

12

higher body weight, and lower educational achievement have been associated with less frequent attendance at group sessions supporting habit changes (Goode et al., 2016). Influence of gender on physical activity habits, however, is less well understood (Duclos, Dejager, Postel-vinay, Nicola, & Quéré, 2015; Mesters et al., 2014).

The purpose of the analyses presented here was two-fold.

The first set of analyses examined whether patterns of associations between socio-economic characteristic and weight maintenance and participation in the PREMIT sessions during the preparation (Stage 2) and action (Stage 3) stages were similar to those reported in previous studies (e.g., Marques-Vidal et al., 2015; Mesters et al., 2014) despite the support offered by the PREMIT. It was hypothesized that weight regain after weight loss would be associated with less frequent PREMIT attendance and with higher habit strength for physical inactivity and poor diet. Also, it was hypothesized that socio-economic variables of younger age, being employed, and living in a household with children were associated with higher weight-regain and not attending the PREMIT sessions. The second set of analyses examined changes in habit strength for poor diet and physical inactivity during the preparation (Stage 2) and action stages (Stage 3) of the PREMIT and whether any changes were associated with weight maintenance. It was hypothesized that from the preparation to the action stage, habit strength for inactivity and poor diet would decrease. Finally, for physical inactivity habit strength only, as physical activity was emphasized during the PREMIT sessions, it was hypothesized that more frequent PREMIT attendance would be associated with lower habit strength after controlling for socio-economic factors of age, gender, employment, and living with children.

Methods Study Design The PREVIEW RCT was a 36-month intervention, which comprised two phases. Phase I comprised an initial 8-week low-energy diet (LED – Cambridge Weight Plan Ltd., Corby, UK) weight-loss phase for all participants, which was followed by a 34-month weight maintenance phase (Phase II) for those who had lost at least 8% of their initial body weight during Phase I. Weight-loss during the Phase I was achieved using meal replacement products providing approximately 800 kcal/day. Participants were not expected to change their physical activity habits during the Phase I weight-loss period. Prior to the start of the Phase II, eligible participants were randomized into different intervention arms, using a 2 2 diet and exercise European Journal of Health Psychology (2019), 26(1), 10–20

M. Huttunen-Lenz et al., Supporting Health Behavior Change

factorial design. The intervention arms were comprised from two dietary programs (high protein with lower dietary Glycemic Index (GI) diet or moderate protein with medium dietary GI diet) and two physical activity programs (highintensity physical activity or moderate-intensity physical activity). The full study protocol is published elsewhere (Fogelholm et al., 2017).

The PREMIT (behavior modification intervention) ran concomitantly with the PREVIEW RCT and contained four different stages (Figure 1). Stage 1 (preliminary) covered mostly the PREVIEW RCT Phase I (weight-loss). Stage 2 (preparation) started at the end of the PREVIEW RCT Phase I, and Stage 3 (action) covered the first 4 months of the PREVIEW RCT Phase II (weight maintenance). Stage 4 (maintenance) of the PREMIT covered the remaining 2.5 years of the PREVIEW RCT Phase II. Participants were expected to acquire new diet and physical activity habits during Phase II of the PREVIEW RCT to prevent weight regain. Development of new diet and physical activity habits was supported during Stages 2 and 3 of the PREMIT, while habit maintenance was supported during Stage 4 of the PREMIT. The PREMIT was not tailored for the different PREVIEW RCT arms; it offered the same support for each participant irrespective of their group allocation.

This paper reports on the PREMIT preparation and action stages for behavior change. During Stage 2 (preparation for behavior, i.e., habit change), behavioral determinants of intention (behavioral goals), outcome expectancies (beliefs about consequences of behavior change), and self-efficacy (perceived ability to change habits) were influenced.

Participants’ self-efficacy to perform new dietary and exercise behaviors was promoted and positive outcome expectancies following a behavior change were reinforced.

In Stage 3 (action-habit change), capabilities to act in longterm self-interest (self-regulation) were emphasized and self-regulation to perform the intended behaviors was endorsed. A number of behavior change techniques, such as improving knowledge of lifestyle choice consequences, action planning, setting behavioral goals, reinforcing successful behaviors, and planning solutions for behavioral barriers were employed (for further details, see Kahlert et al. 2016).

The PREMIT behavior modification intervention was delivered within group sessions by trained instructors with approximately 10–20 participants per group. Participants were generally allocated to group sessions according to the four PREVIEW intervention arms. Nevertheless, there was variability between the study sites. However, as support offered by the PREMIT was not tailored to match the PREVIEW RCT allocation, but to support weight maintenance and health behavior change in general, participants were considered as a single group independent of their PREVIEW RCT allocation. Ó 2019 Hogrefe Publishing

M. Huttunen-Lenz et al., Supporting Health Behavior Change

Participant Recruitment Participants were recruited to the PREVIEW RCT between August 2013 and March 2015 from the study sites in University of Copenhagen (UCPH), Denmark; University of Helsinki (HEL), Finland; University of Nottingham (UNOTT), United Kingdom; University of Maastricht (UM), The Netherlands; University of Navarra (UNAV), Spain; Medical University of Sofia (MU), Bulgaria; University of Auckland (UOA), New Zealand; and University of Sydney (UNSYD), Australia. Overweight and obese (Body Mass Index (BMI) 25 kg/m2) men and women aged 25–70 years with confirmed pre-diabetes and willing to be randomized into a weight-loss intervention were eligible for inclusion. Pre-diabetes was confirmed by an oral glucose tolerance test (OGTT) using the American Diabetes Association criteria (American Diabetes Association, 2011). Participant recruitment was undertaken by advertising in print and visual media, and by contacting primary and occupational health care providers for referrals. Before full screening, potential participants were pre-screened either by telephone or using an e-mail questionnaire. The Human Ethics Committees in each of the participating countries approved the study protocol, and all participants enrolled in the study provided written informed consent (Fogelholm et al., 2017).

Data Collection As shown in Figure 1, outcome measures were collected at the beginning of the PREMIT Stage 2 (at Clinical Investigation Day 2–Week 8) and at the end of Stage 3 (at Week 26), covering a period of 18 weeks. Data collection included both anthropometric (e.g., body weight and height), metabolic (e.g., HbA1c), and psychological variables. All psychological measurements were collected using standardized questionnaires. For non-English speaking countries questionnaires were translated into local languages. Accuracy of the translations were checked by back-translating the local versions into English and comparing them with the original English version.

Outcome Measurements Body Weight Body weight was measured while lightly clad at Week 8 (beginning of the PREMIT Stage 2) and Week 26 (end of the PREMIT Stage 3) and values were used to calculate percentage weight change.

Socio-Demographic Characteristics The European Social Survey and International Social Survey (ESS, 2015) was used to collect information about gender, Ó 2019 Hogrefe Publishing

13

age, educational achievement, marital status, people living in the household, and employment status.

PREMIT Attendance Attendance to the PREMIT behavior modification intervention was recorded separately for each session on a centralized database (OpenClinica open source software, version 3.1., OpenClinica LLC & Co., Waltham, MA, USA).

Habit Strength of Physical Inactivity and Poor Diet The habit strength questionnaire asked about physical inactivity (6 items – 3 for each sitting and inactive commuting habits) and unhealthy dietary (6 items – 3 for each high fat and snacking habits) behaviors (based on Ji & Wood, 2007; Wood, Tam, & Guerrero Witt, 2005). Questions inquired about both frequency (how often) and stability (context – time and either location or purpose) of behaviors.

Frequency of behaviors was evaluated on a scale from 1 (= never/almost never) to 7 (= very often/almost every day). Stability of the context regarding the time and either location or the purpose of behaviors was evaluated on a scale from 1 (= never/almost never) to 7 (= almost always at the same time/location/purpose). Habit strength was estimated by multiplying score for behavior frequency by score for stability of circumstances (time and either location or the purpose, i.e., max 7 7 7). Mean values for habit strength of inactivity (encompassing sitting and inactive commuting habits) and poor diet (encompassing high fat and snacking habits) were computed. Hence habit strength can range from 1–343. Lower scores reflect either infrequent performance or variable circumstances and higher scores reflect frequent performance in stable circumstances. Scale reliability was satisfactory. Cronbach’s alphas for habit strength at Week 8 were α = .74 for physical inactivity and α = .83 for poor diet. At Week 26 Cronbach’s alphas were α = .74 for habit strength of physical inactivity and α = .79 for poor diet.

Statistical Methods The analyses are based on the 1,569 participants who were eligible for participation in the PREVIEW RCT (Phase II) and for whom body weight data were available at Week 26. Participants were analyzed as one group without adjustments for study sites or the PREVIEW RCT treatment arms.

All analyses were conducted using IBMÒ SPSS Statistics Program version 23.

A number of participants were unsure of the level of educational achievement and used the “other” option with description, for example, “PhD”. Where possible, data were recoded within the correct level of achievement. Missing values were estimated for the constructs of habit strength of inactivity and poor diet. Missing values imputation European Journal of Health Psychology (2019), 26(1), 10–20

14

method was specified as automatic with maximum of 50 case draws and 5 sets of imputations were done. For the imputation model, age, gender, attendance to the group counseling sessions, and weight change percentage were included as predictor variables.

Outlying values were identified for weight change and habit strength as those 3.29 standard deviations above the mean value and removed, leaving 1,521 participants for the analyses. Before statistical significance-testing data, transformations were employed to improve data normality for habit strength of physical inactivity (SQRT; Howell, 1997; Tabachnick & Fidell, 2001). Habit strength for poor diet could not be successfully transformed and a nonparametric statistical test was used instead. Sensitivity analyses were conducted with the transformed, but not imputed, dataset and reported if discrepancies existed.

Descriptive statistical methods were employed to analyze participant characteristics. Bonferroni adjusted (p .01) chi-square tests analyzed the PREMIT group counseling session attendance by different participant characteristics.

Percentage of weight change was calculated as “weight Week 26 minus weight Week 8” and expressed as percentage (±). Weight change was not adjusted for the PREVIEW RCT treatment arm. Independent samples t-tests were used to compare percentage change in body weight between participant characteristics. Analysis of variance (ANOVA; Welch’s F) was used to examine association between weight change and number of group counseling sessions attended using p-value .025. Post hoc tests after significant Welch’s F were performed with Games-Howell correction using significance level of .01 with those who had attended all the available sessions forming the comparison group (Field, 2013).

For habit strength of poor diet non-parametric sign-test and for habit strength of inactivity dependent samples ttests were used to evaluate changes from the beginning of the PREMIT Stage 2 until the end of Stage 3. Spearman’s and Pearson’s bivariate correlations were conducted to evaluate associations between weight change and habit strength at the end of Stage 3. Sequential regression with 2 steps was used to estimate whether frequency of attending group counseling sessions increased variance explained in habit strength (inactivity-outcome variable) at the end of Stage 3 (step 1 age, gender, employment, living with children; step 2 the PREMIT attendance; categorical variables were dummy coded).

Bonferroni corrected p-value of .007 was used for all analyses that included habit strength for inactivity. Analyses with habit strength for poor diet used Bonferroni corrected p-value of .025. Effect sizes were calculated using Lenhard and Lenhard (2016) apart for the non-parametric sign-test for which the effect was calculated as r (Field, 2013). While statistical analyses were completed with European Journal of Health Psychology (2019), 26(1), 10–20

M. Huttunen-Lenz et al., Supporting Health Behavior Change

transformed data, for the ease of interpretation, means and standard deviations are presented for the nontransformed data.

Results Participant Characteristics Across all sites, 15,611 individuals were prescreened, 5,472 individuals were screened in clinic, and 2,326 were found eligible to participate in the study. Of these, 2,224 individuals began the LED phase, with a dropout rate of 9.2% between Week 0 (baseline) and Week 8. At Week 8, 1,857 participants were eligible to continue into the PREVIEW RCT Phase II (weight maintenance). Dropout rate between Week 8 and Week 26 was calculated as 15.5%.

Characteristics at baseline for the 1,521 participants included in this study are shown in Table 1. Most participants were female, married or in a civil partnership, of Caucasian ethnicity, and with high educational achievement.

Body weight change from the beginning of Stage 2 to the end of Stage 3 ranged from 13.3 kg ( 16%) to +17.0 kg (+20%), with mean change of +0.3 kg (SD = 4.0 kg), with over half (55%) of the participants having regained some body weight.

PREMIT Attendance, Socio-Economic Variables, and Weight Maintenance No consistent pattern between participant characteristics and the PREMIT attendance was found (Table 2). Overall, participation in the group sessions declined from 6th to 9th session, but recovered for the 10th session, which coincided with the end of Stage 3 of the PREMIT program and Week 26 attendance. This pattern, however, was not observed among participants from the non-Caucasian backgrounds.

Consequently, participants from the non-Caucasian backgrounds were less likely to have attended the last group counseling session during Stage 3 (w2(1) = 29.90, p .01).

ANOVA (Welch’s F) indicated that there was a statistically significant association between the number of the PREMIT group counseling sessions attended during the preparation and action stages, and weight change (% body weight); Welch’s F(5, 191) = 15.95, p .001, est. ω2 = .05.

The number of participants attending 0, 1, 2, 3, 4, or 5 times in the group sessions and M ± SD percentage of weight change are shown in Table 3. Post hoc comparisons using Games-Howell adjustment indicated significant difference in weight change (%) between those who attended all 5 sessions and those who attended 0–3 of the sessions. Examination of the mean values indicated that those who attended Ó 2019 Hogrefe Publishing

M. Huttunen-Lenz et al., Supporting Health Behavior Change

Table 1. Participant characteristics Participant characteristics (n = 1,521) Age* (years) (M ± SD)

53.4 ± 10.8

BMI (kg/m2) (M ± SD) Week 8 Week 26

30.5 ± 5.2 30.6 ± 5.4

Female

1,004 (66%)

Married or in civil partnership*

1,080 (71%)

Living in a household with at least 2 adults*

1,208 (79%)

Living in a household with at least 1 child* Ethnicity – Caucasian In paid employment (regardless of hours worked per week)*

309 (20%) 1,380 (91%) 925 (61%)

Level of educational achievement* Up to secondary education

237 (16%)

Secondary vocational education

269 (18%)

Higher vocational education

301 (20%)

University

592 (39%)

Other (including those with missing data)

122 (8%)

Note. BMI = Body Mass Index; M = Mean; SD = Standard Deviation. *As given at the start of the trial.

0–3 of the sessions regained body weight, whereas those who attended 4 or 5 of the sessions lost weight.

Pairwise comparisons with weight change percentage during Stages 2 and 3 of the PREMIT as dependent variable, showed that males regained significantly more weight than females, t(1,519) = 4.36, p .025, dCohen, unequal group size = .24. Those not married or in a civil partnership, t(1,519) = 2.67, p .025, dCohen, unequal group size = .15, living in one adult households, t(1,519) = 2.37, p .025, dCohen, unequal group size = .15, and not Caucasian ethnicity, t(1,519) = 2.27, p .025, dCohen, unequal group size = .20, also had a significantly higher percentage of body weight regain. Effect sizes for all comparisons, however, were small. Employment status, t(1,336) = .29, p > .025, or living in a household with children, t(1,519) = 1.95, p > .025, were not associated with body weight change.

Habit Strength and PREMIT Attendance Means and standard deviations (SD) for habit strength (poor diet and physical inactivity) are shown in Table 4.

Habit strength for physical inactivity had not significantly changed at the end of Stage 3, t(1,520) = 2.17, p .007.

Sign-test indicated that habit strength for poor diet had increased between the beginning of Stage 2 and the end of Stage 3 (Z = 32.04, p .025, r = .58), indicating a large effect size. The score for habit strength on average increased from a low (7.70) at Week 8 to significantly higher level at Week 26 (45.56). The low score observed at Week 8 was likely due to the recent completion of the Ó 2019 Hogrefe Publishing

15

low-energy diet (Phase I). While habit strength for physical inactivity at the end of Stage 3 was not significantly associated with weight change percentage, rx,y(1,521) = .10, p .007, habit strength for poor diet at the end of Stage 3 was associated with percentage weight change, ρ(1,521) = .25, p .025, though the effect size was weak. This suggested that increasing habit strength for poor diet was associated with higher weight regain percentage.

Sequential regression was used to examine whether attending the PREMIT group counseling sessions during the preparation and action stages predicted habit strength for inactivity at the end of the action stage, after effects for socio-demographic variables of age, gender, employment status, and living with children were controlled. After all the variables were entered into the model, overall R = .302 [F(9, 1,511) = 16.84, p .007] was significant, with adjusted R2 = .09 indicating that the model explained 9.0% of the variance in the habit strength for inactivity at the end of Stage 3. At step one, age, gender, employment status, and living with children were entered in the equation resulting in a significant R = .295 [F(4, 1,516) = 36.09, adjusted R2 = .09, p .007]. At step two, five categorical variables created for frequency of attending the group counseling sessions were included. Results indicated no significant change in the overall model; R = .295 [Change F(5, 1,511) = 1.40, R2 = .004, p .007]. Both unstandardized (B) and standardized (β) regression coefficients and intercept are shown in Table 5. Inspection of the coefficients indicated age and employment status were individually significant predictors of habit strength at Week 26, such that older age and not being in employment were associated with lower habit strength for inactivity.

Discussion Although healthy diet and physical activity are effective means to reduce the risk of developing T2D (Alouki, Delisle, Bermúdez-Tamayo, & Johri, 2016), many of those at risk struggle to achieve permanent habit changes and participate in interventions offering support with habit changes (Følling, Solbjør, & Helvik, 2015; Goode et al., 2016). Results from this study showed that, on average, participants had regained some weight during the PREMIT Stages 2 and 3. Higher weight regain percentage, however, was associated with being male, not married or in a civil partnership, living in one adult household, and of nonCaucasian ethnicity. In addition, more frequent attendance at the PREMIT sessions and lower habit strength for poor diet at the end of Stage 3 were associated with lower body weight regain. But habit strength for physical inactivity was not associated with body weight regain. Furthermore, European Journal of Health Psychology (2019), 26(1), 10–20

Notes. Participant characteristics as given at baseline. Participants with no data have been included in these categories. Significant differences between the groups at the level of p .01 are shown in bold.

361 (61%)

487 (82%) 688 (74%) 83 (59%) 1,092 (79%) 945 (78%) 230 (74%) 250 (80%) 925 (77%) 335 (76%) 840 (78%) 405 (78%) 770 (77%) 1,175 (77%) 10th (CID3)

396 (67%) 597 (65%)

541 (58%) 81 (57%)

82 (58%) 911 (66%)

821 (60%) 742 (61%)

802 (66%) 191 (62%)

160 (52%) 185 (59%)

205 (66%) 788 (65%)

717 (59%) 268 (61%)

297 (67%) 696 (64%)

634 (59%) 308 (60%)

323 (62%)

902 (59%) 9th

595 (59%)

994 (65%) 8th

671 (67%)

433 (73%)

428 (72%) 689 (74%)

618 (67%) 102 (72%)

101 (72%) 1,016 (74%)

949 (69%) 849 (70%)

906 (75%) 211 (68%)

202 (65%) 222 (71%)

248 (79%) 869 (72%)

829 (69%) 312 (71%)

341 (77%) 776 (72%)

739 (68%) 332 (64%)

364 (70%)

1,051 (69%) 7th

753 (75%) 1,117 (74%) 6th

Male Female All Session

720 (72%)

None

Caucasian

Ethnicity Children

At least one child One adults

Other adults

At least two adults Other Married/Civil partnership

Marital status Gender

Living in a household with

Table 2. Attendance to individual PREMIT sessions according to participant characteristics during preparation and action stages

Other

Paid employment

Other

M. Huttunen-Lenz et al., Supporting Health Behavior Change

Employment status

16

European Journal of Health Psychology (2019), 26(1), 10â&#x20AC;&#x201C;20

frequency of attending the PREMIT was not predicting habit strength for physical inactivity at the end of action Stage 3 over and above socio-economic variables, especially of age and employment status.

Overall, the majority of participants attended the PREMIT sessions during Stages 2 and 3, with over one quarter of the participants (28%) recorded as attending all the five sessions. Previously, attendance in weight-loss trials has been associated with characteristics such as age and level of educational achievement (Goode et al., 2016). However, in PREVIEW, no consistent pattern was observed between participant characteristics and attendance at group sessions.

Apart from participants of non-Caucasian ethnicity, participation in the group counseling sessions declined between the data collection points, but recovered for the final session of Stage 3, coinciding with Week 26. This was very likely due to attendance at clinic for blood tests and associated study measurements at this time point. Consequently, non-Caucasian participants were not only less likely to have attended the last group counseling session during Stage 3, but also potentially more likely to miss the clinic attendance for data measurements.

Habit strength for physical inactivity, measured here as a combination of behavioral frequency and context, was not significantly associated with weight change. Although not significantly different, the mean value for habit strength for physical inactivity at the end of Stage 3 was lower than at the beginning of Stage 2. As even small changes in the daily routines may enable long-term habit modification (Gardner et al., 2012; Lally & Gardner, 2013), small changes observed in habit strength for inactivity may indicate successful habit changes in future. Further research, however, should explore whether habit strength for inactivity continues to weaken.

Habit strength has been shown to be a strong predictor of behavior such as unhealthy snacking (Verhoeven et al., 2012). As expected, increasing habit strength for poor diet, that is, frequency and context of snacking and eating high fat food, was associated with higher weight gain. Contrary to expectations, habit strength for poor diet was significantly higher by the end of Stage 3 than at the beginning of Stage 2, indicating, on average, poorer diet habits. In retrospect, this result may not be surprising as at the start of Stage 2 participants had just finished an 8-week LED with meal replacement products. How the habit strength for unhealthy diet develops during the maintenance period of the PREMIT intervention needs to be explored to understand whether the results reported here mirror differences in participantsâ&#x20AC;&#x2122; eating habits during the LED and after, or whether there is a tendency toward poorer diet habits as the intervention progresses.

Previous research has indicated that variables such as age and being in paid employment are associated with level Ă&#x201C; 2019 Hogrefe Publishing

M. Huttunen-Lenz et al., Supporting Health Behavior Change

17

Table 3. Weight change during the 18-week follow-up period (preparation and action stages) and means, standard deviations, and results for pairwise comparisons between participants attending different amount of the PREMIT group counseling sessions Pairwise comparisons with attending all the sessions (5) Cohen’s d 95% CI Number of PREMIT sessions attended

N

SD

M Diff.

SE

p-value

Cohen’s d

Lower

Upper

21

3.13

3.42

3.39

0.78

.01

0.82

1.27

0.38

158

1.43

3.84

2.23

0.38

.01

0.49

0.68

0.31

None (0) One (1)

Weight change % (M)

Two (2)

201

1.71

4.33

2.51

0.38

.01

0.54

0.71

0.37

Three (3)

312

0.70

4.41

1.49

0.34

.01

0.32

0.47

0.17

Four (4)

403

0.29

4.42

0.50

0.32

.01

–

–

–

Five (5)

426

0.79

4.81

Note. M = Mean; SD = Standard Deviation; SE = Standard Error; CI = Confidence Interval.

Table 4. Means (M) and standard deviations (SD) for habit strength for poor diet and physical inactivity Time point Variable

Week 8 (M ± SD)

Week 26 (M ± SD)

n = 1,521

n = 1,521

Habit strength (Scale: 1 = Low to 343 = High) Physical inactivity Poor diet

Unstandardized Standardized coefficients Coefficients B β p-value

Model 139.44 ± 86.24

134.67 ± 83.58

7.70 ± 14.65

45.56 ± 43.56

Note. All data before data transformations.

of physical activity as well as with success in attending interventions promoting healthy habit changes (Goode et al., 2016; Marques-Vidal et al., 2015; Mesters et al., 2014). Here, predictors for habit strength for physical inactivity at the end of the action stage of the PREMIT were examined. Results indicated that after controlling for social economic variables of age, gender, living with children, and employment, frequency of attending in the PREMIT did not significantly add to the prediction.

Furthermore, of individual predictors, only age and employment were significant predictors for habit strength for physical inactivity so that younger age and being in employment were associated with higher habit strength for inactivity. While the association between employment and physical activity is in line with previous literature (e.g., Marques-Vidal et al., 2015; Mesters et al., 2014), association between younger age and physical activity was unexpected. In addition, neither gender nor living with children was associated with habit strength as expected. While it could be hypothesized that pressures of combining work and family life are particularly acute for younger participants, leaving limited capacity to concentrate on physical habit changes, living with children was not found to be associated with inactivity habit strength. Alternatively, younger adults might be less concerned about the future consequences of the T2D and perceive that it will affect them only in middle/old age. Thus, younger adults may Ó 2019 Hogrefe Publishing

Table 5. Unstandardized and standardized regression coefficients for socio-economic variables and attendance for the PREMIT sessions with habit strength for physical inactivity as dependent variable at Week 26

Constant

.007

13.559

Gender: male

0.266

0.032

.007

Age

0.033

0.091

.007

Household with children

0.238

0.025

.007

Employment other

1.907

0.239

.007

PREMIT attending 4 sessions

0.145

0.016

.007

PREMIT attending 3 sessions

0.137

0.014

.007

PREMIT attending 2 sessions

0.253

0.022

.007

PREMIT attending 1 sessions

0.648

0.051

.007

PREMIT attending 0 sessions

0.510

0.015

.007

Note. Statistically significant results at the level of p .007 are in bold.

be less motivated to change their habits and may require additional support in weight maintenance from the PREVIEW instructors.

While frequency of attending the PREMIT group counseling sessions was not an independent predictor of habit strength for inactivity, it should be emphasized that this result is not an indication that the PREMIT intervention does not influence habit strength for physical inactivity.

As participants had been preparing for behavior change during Stage 1 (weight-loss phase of the PREVIEW RCT, Figure 1), it is possible that attending these sessions may have had more influence on physical inactivity habit strength than attending the sessions during active physical activity changes (Stages 2 and 3). Stage 1, however, was not included in this report as analysis here was focused on understanding changes in social-cognitive variables during preparation and action stages of active behavior change.

Moreover, frequent attendance to the PREMIT group counseling sessions appeared to be associated with lower weight European Journal of Health Psychology (2019), 26(1), 10–20

18

regain. Although it was not possible to show a causal relationship between attending the group counseling sessions and weight change, results indicated that having the support offered during the PREMIT sessions may be useful in guiding and sustaining participants through successful behavior change. Results, however, suggested that especially younger participants and those in employment may struggle with weight-maintenance and physical activity habit changes, and may require special emphasis during the PREMIT session to ensure adequate support for successful behavior change.

The present study had a number of limitations. The multiple interactions with the PREVIEW RCT and the PREMIT made evaluation challenging. However, as the PREMIT intervention was also theory- and evidence-based, it was possible to evaluate whether social-cognitive variables developed as expected from the previous evidence base, correlated with group session attendance, or were associated with changes in objective physiological measurements such as weight. Effects of the different PREVIEW intervention arms were not controlled in this study. However, as the purpose of the PREMIT program was to support weight maintenance regardless of the intervention condition, results still provided insight into how a program such as the PREMIT can be utilized in supporting participants.

The results of the analyses are also only indicative, as they were based on causal relations between the PREMIT program and changes in the social-cognitive variables. In addition, habit strength was assessed as combination of frequency and stability (context) of behavior, which meant that the automaticity facet of the habit was not considered.

Current analyses also did not consider a range of other cognitive variables, such as self-efficacy and motivation, and their interactions with habit strength. Therefore, it is important that future research examines how habit strength interacts with other social-cognitive and environmental variables.

Conclusions Results did not indicate any clear pattern of socio-economic variables associated with the PREMIT attendance. Weight regain was found to be associated with less frequent PREMIT attendance, male gender, not being married or in a civil partnership, living in one adult household, and nonCaucasian ethnicity. Results also suggested mixed success in modifying poor diet and physical inactivity habits during the preparation and action stages of the PREMIT program.

Changes in body weight were not reflected in changes in habit strength for physical inactivity, but weight regain was associated with higher poor diet habit strength.

Socio-economic variables of younger age and being in European Journal of Health Psychology (2019), 26(1), 10–20

M. Huttunen-Lenz et al., Supporting Health Behavior Change

employment were found to be significant predictors of higher habit strength for physical inactivity. Changes in habit strength for inactivity were not associated with frequency of attending the PREMIT program. Taken together, the results indicated that while attending the PREMIT appeared to support participants in weight-loss maintenance, PREMIT instructors may need to ensure that in particular younger participants, those who are from nonCaucasian ethnic background, are male, and who are in employment receive adequate support for habit change and weight maintenance.

References Alouki, K., Delisle, H., Bermúdez-Tamayo, C., & Johri, M. (2016).

Lifestyle interventions to prevent type 2 diabetes: A systematic review of economic evaluation studies. Journal of Diabetes Research, 2016, 2159890. https://doi.org/10.1155/2016/ 2159890 American Diabetes Association. (2011). Standards of medical care in diabetes – 2011. Diabetes Care, 34(Supplement 1), S11–S61. https://doi.org/10.2337/dc11-S011 Bandura, A. (1996). Social cognitive theory of human development.

In T. Husen & T. N. Postlethwaite (Eds.), International Encyclopedia of Education (2nd ed., pp. 5513–5518). Oxford, UK: Pergamon Press. de Vries, H., Eggers, S. M., Lechner, L., Van Osch, L., & Van Stralen, M. M. (2014). Predicting fruit consumption: The role of habits, previous behavior and mediation effects. BMC Public Health, 14, 730. https://doi.org/10.1186/1471-2458-14-730 Duclos, M., Dejager, S., Postel-vinay, N., Nicola, S., & Quéré, S. (2015). Physical activity in patients with type 2 diabetes and hypertension – insights into motivations and barriers from the MOBILE study. Vascular Health and Risk Management, 11, 361– 371. https://doi.org/10.2147/VHRM.S84832 ESS. (2015). European Social Survey (2015) ESS Round 7 (2014/ 2015). London, UK: ESS. (Technical report). Retrieved from https://www.europeansocialsurvey.org/docs/about/annualreports/ESS-ERIC-Annual-Activity-Report-2014-2015.pdf Field, A. (2013). Discovering Statistics using IBM SPSS Statistics (4th ed.). New Delhi: SAGE Publications.

Fishbein, M., Triandis, H. C., Kanfer, F. H., Becker, M., Middlestadt, S. E., & Eichler, A. (2001). Factors influencing behavior and behavior change. In A. Baum, T. A. Revenson, & J. E. Singer (Eds.), Handbook of health psychology (pp. 3–17).

Mahwah, NJ: Erlbaum.

Fogelholm, M., Larsen, T., Westerterp-Plantenga, M., Macdonald, I., Martinez, J., Boyadjieva, N., . . . Raben, A. (2017). PREVIEW: Prevention of Diabetes through Lifestyle Intervention and Population Studies in Europe and around the World. Design, methods, and baseline participant description of an adult cohort enrolled into a three-year randomised clinical trial.

Nutrients, 9, 632. https://doi.org/10.3390/nu9060632 Følling, I. S., Solbjør, M., & Helvik, A.-S. (2015). Previous experiences and emotional baggage as barriers to lifestyle change – a qualitative study of Norwegian Healthy Life Centre participants.

BMC Family Practice, 16, 73. https://doi.org/10.1186/s12875015-0292-z Gardner, B. (2012). Habit as automaticity, not frequency. European Health Psychologist, 14, 32–36. https://doi.org/10.1111/j.15591816.2003.tb01951.x Ó 2019 Hogrefe Publishing

M. Huttunen-Lenz et al., Supporting Health Behavior Change

Gardner, B., Corbridge, S., & McGowan, L. (2015). Do habits always override intentions? Pitting unhealthy snacking habits against snack-avoidance intentions. BMC Psychology, 3, 8. https://doi. org/10.1186/s40359-015-0065-4 Gardner, B., Lally, P., & Wardle, J. (2012). Making health habitual: the psychology of “habit-formation” and general practice. The British Journal of General Practice, 62, 664–666. https://doi. org/10.3399/bjgp12X659466 Goode, R. W., Ye, L., Sereika, S. M., Zheng, Y., Mattos, M., Acharya, S. D., . . . Burke, L. E. (2016). Socio-demographic, anthropometric, and psychosocial predictors of attrition across behavioral weight-loss trials. Eating Behaviors, 20, 27–33. https:// doi.org/10.1016/j.eatbeh.2015.11.009 Guariguata, L., Whiting, D. R., Hambleton, I., Beagley, J., Linnenkamp, U., & Shaw, J. E. (2014). Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Research and Clinical Practice, 103, 137–149. https://doi.org/ 10.1016/j.diabres.2013.11.002 Howell, D. C. (1997). Statistical methods for psychology (4th ed.).

Stamford, CT: International Thomson Publishing.

Ji, M., & Wood, W. (2007). Habitual purchase and consumption: Not always what you intend. Journal of Consumer Psychology, 17, 261–276. https://doi.org/10.1016/S1057-7408(07)70037-2 Kahlert, D., Unyi-Reicherz, A., Stratton, G., Meinert Larsen, T., Fogelholm, M., Raben, A., & Schlicht, W. (2016). PREVIEW Behavior Modification Intervention Toolbox (PREMIT): A Study Protocol for a Psychological Element of a Multicenter Project.

Frontiers in Psychology, 7, 1136. https://doi.org/10.3389/ fpsyg.2016.01136 Labrecque, J. S., & Wood, W. (2015). What measures of habit strength to use? Comment on Gardner (2015). Health Psychology Review, 9, 303–310. https://doi.org/10.1080/17437199.

2014.992030 Lally, P., & Gardner, B. (2013). Promoting habit formation. Health Psychology Review, 7(sup1), 137–158. https://doi.org/10.1080/ 17437199.2011.603640 Lenhard, W., & Lenhard, A. (2016). Calculation of Effect Sizes.

Dettelbach, Germany: Psychometrica. Retrieved from https:// www.psychometrica.de/effect_size.html. https://doi.org/10.13140/ RG.2.1.3478.4245 Marques-Vidal, P., Waeber, G., Vollenweider, P., Bochud, M., Stringhini, S., & Guessous, I. (2015). Sociodemographic and behavioural determinants of a healthy diet in switzerland.

Annals of Nutrition and Metabolism, 67, 87–95. https://doi.org/ 10.1159/000437393 Mesters, I., Wahl, S., & Van Keulen, H. M. (2014). Sociodemographic, medical and social-cognitive correlates of physical activity behavior among older adults (45–70 years): A cross-sectional study. BMC Public Health, 14, 1–12. https://doi. org/10.1186/1471-2458-14-647 Michie, S., Johnston, M., Francis, J., Hardeman, W., & Eccles, M. (2008). From theory to intervention: Mapping theoretically derived behavioural determinants to behaviour change techniques. Applied Psychology, 57, 660–680. https://doi.org/ 10.1111/j.1464-0597.2008.00341.x Phillips, L. A., & Gardner, B. (2016). Habitual exercise instigation (vs. execution) predicts healthy adults exercise frequency.

Health Psychology, 35, 69–77. https://doi.org/10.1037/ hea0000249 Prochaska, J. O., & DiClemente, C. C. (1992). Stages of change in the modification of problem behaviors. In M. Hersen, R. M. Eisler, & P. Miller (Eds.), Progress on behavior modification (pp. 184–214). Sycamore, IL: Sycamore Press.

Renner, B., & Schwarzer, R. (2005). Risk and Health Behaviors – Documentation of the Scales of the Research Project: “Risk Appraisal Consequences in Korea” (RACK). Bremen, Berlin:

Ó 2019 Hogrefe Publishing

19

International University Bremen & Freie Universität Berlin.

Retrieved from http://www.gesundheitsrisiko.de/docs/RACKEnglish.pdf Ryan, R., & Deci, E. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68–78. https://doi.org/ 10.1037/0003-066X.55.1.68 Ryan, R. M., Patrick, H., Deci, E. L., & Williams, G. C. (2008).

Facilitating health behaviour change and its maintenance: Interventions based on Self-Determination Theory. The European Health Psychologist, 10, 2–5.

Schwarzer, R. (2001). Social-cognitive factors in changing healthrelated behaviors. Current Directions in Psychological Science, 10, 47–51. https://doi.org/10.1111/1467-8721.00112 Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics, international student edition (4th ed.). Boca Raton, FL: Allyn & Bacon.

Tamayo, T., Rosenbauer, J., Wild, S. H., Spijkerman, A. M. W., Baan, C., Forouhi, N. G., . . . Rathmann, W. (2014). Diabetes in Europe: An update. Diabetes Research and Clinical Practice, 103, 206–217. https://doi.org/10.1016/j.diabres.2013.11.007 Verhoeven, A. A. C., Adriaanse, M. A., Evers, C., & De Ridder, D. T. D. (2012). The power of habits: Unhealthy snacking behaviour is primarily predicted by habit strength. British Journal of Health Psychology, 17, 758–770. https://doi.org/ 10.1111/j.2044-8287.2012.02070.x Williams, R. L., Wood, L. G., Collins, C. E., & Callister, R. (2015).

Effectiveness of weight loss interventions – is there a difference between men and women: A systematic review. Obesity Reviews, 16, 171–186. https://doi.org/10.1111/obr.12241 Wood, W., Tam, L., & Guerrero Witt, M. (2005). Changing circumstances, disrupting habits. Journal of Personality and Social Psychology, 88, 918–933. https://doi.org/10.1037/0022-3514.

88.6.918 History Received August 16, 2018 Revision received March 26, 2019 Accepted April 2, 2019 Published online June 3, 2019 Acknowledgments The corresponding author (Maija Huttunen-Lenz) is the guarantor of this work and had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Lene Stevner assisted with advice on ethical issues, Good Clinical Practice and approval of the study protocol.

The additional contributors listed below assisted in conduct of the trial during recruitment, intervention, and/or data collection: University of Copenhagen, Denmark: Ulla Skovbæch Pedersen, Marianne Juhl Hansen, Bettina Belmann Mirasola, Maria Roed Andersen, Anne Wengler, Jane Jørgensen, Sofie Skov Frost, Eivind Bjørås, Grith Møller, Lone Vestergaard Nielsen.

University of Helsinki, Finland: Elli Jalo, Saara Kettunen, Laura Korpipää, Tiia Kunnas, Heini Hyvärinen, Heikki Tikkanen, Sanna Ritola.

University of Nottingham, United Kingdom: Elizabeth Simpson, Shelley Archer, Natalie Bailey-Flitter, Nicky Gilbert, Laura Helm, Sally Maitland, Melanie Marshall, Theresa Mellor, Grace Miller, Seodhna Murphy, Vicky Newman, Amy Postles, Jakki Pritchard, Maria Papageorgiou, Cheryl Percival, Clare Randall, Sue Smith, Sarah Skirrow.

University of Navarra, Spain: Blanca Martinez de Morentin Aldabe, María Hernández Ruiz de Eguilaz, Salomé Pérez Diez, Rodrigo

European Journal of Health Psychology (2019), 26(1), 10–20

20

San-Cristobal, Maria dels Angels Batlle, Laura Moreno-Galarraga, Alejandro Fernández-Montero, Marian Nuin, Javier Baquedano, Maria Eugenia Ursúa, Francisco Javier Martinez Jarauta, Pilar Buil, Lourdes Dorronsoro, Juana María Vizcay, Teodoro DuráTravé, and all general practitioners and nurses from the Navarra Health Services who collaborated in the recruitment of the participants.

Medical University of Sofia, Bulgaria: Nadka Boyadjieva, Pavlina Gateva-Andreeva, Georgi Bogdanov, Galina Dobrevska.

University of Auckland, New Zealand: Amy Liu, Lindsay Plank, Anne-Thea McGill, Madhavi Bollineni, Kelly Storey, Nicholas Gant, Jonathon Woodhead, Hannah Chisholm, Wonjoo Lee, Chelsea Cheah, Eric Hansen, Hacer Tekinkaya, Nadia Harvey.

University of Sydney, Australia: Kylie Simpson, Michele Whittle, Kirstine Bell.

We want to acknowledge all the additional people who have worked and are currently working for PREVIEW including trainees, post- and undergraduate students. Finally, a respectful thank you to all the study participants that participated in PREVIEW. Publication Ethics The study has been approved by the following ethics committees: University of Copenhagen (UCPH): The Secretariat of Research Ethics Committees for the Capital Region of Denmark. University of Helsinki (HEL): The Coordinating Ethics Committee of the Helsinki and Uusimaa Hospital district, Finland. University of Maastricht (UM): The Medical Ethics Committee for azM og Maastricht University (METC azM/UM), Netherlands. University of Nottingham (UNOTT): East Midlands – Leicester Central Research Ethics Committee, United Kingdom. University of Navarra (UNAV): Clinical Research Ethics Committee of Navarra, Spain. Medical University of Sofia (MU): Commission on Ethics in Scientific Research with the Medical University – Sofia (KENIMUS), Bulgaria.

University of Sydney (UNSYD): Ethics Review Committee at the Sydney Health Local District, Australia. University of Auckland (UOA): Northern B Health and Disability Ethics Committee, New Zealand.

European Journal of Health Psychology (2019), 26(1), 10–20

M. Huttunen-Lenz et al., Supporting Health Behavior Change

Authorship The PREVIEW project was designed by Anne Raben (University of Copenhagen, Denmark), Jennie Brand-Miller (University of Sydney, Australia), Margriet Westerterp-Plantenga (University of Maastricht, Netherlands), Mikael Fogelholm (University of Helsinki, Finland), Wolfgang Schlicht (University of Stuttgart, Germany), and Edith Feskens (Wageningen University, Netherlands).

The PREVIEW intervention study (RCT) for the adult participants was designed by Anne Raben (University of Copenhagen, Denmark), Mikael Fogelholm (University of Helsinki, Finland), and Thomas Meinert Larsen (University of Copenhagen, Denmark).

The PREMIT behaviour modification intervention was designed by Wolfgang Schlicht (University of Stuttgart, Germany), Daniela Kahlert (University of Stuttgart, Germany), and Annelie UnyiReicherz (University of Stuttgart, Germany).

It was Maija Huttunen-Lenz’s idea to examine changes in habits during Stages 2 and 3 of the PREMIT program.

All authors were engaged in the analysis and interpretation of data and the critical revision of the manuscript for important intellectual content. Finally, Maija Huttunen-Lenz drafted the manuscript.

Funding EU 7th Framework Programme (FP7/2007-2013) under grant agreement no. 312057, The National Health and Medical Research Council – EU Collaborative Grant, AUS, The NZ Health Research Council (14/191) and UoA Faculty Research Development Fund, The Cambridge Weight Plan kindly donated all products for the 8-week Low-Calorie Diet period.

ORCID Maija Huttunen-Lenz https:orcid.org/0000-0002-1034-1613 Maija Huttunen-Lenz Department of Sport and Exercise Science University of Stuttgart Nobelstraße 15 70569 Stuttgart Germany maija.huttunen@inspo.uni-stuttgart.de

Ó 2019 Hogrefe Publishing

The best ways to support the healthy development of children and adolescents and their families Kristin S. Mathiesen / Ann V. Sanson / Evalill B. Karevold (Editors)

Tracking Opportunities and Problems from Infancy to Adulthood 20 Years With the TOPP Study 2018, x + 272 pp.

US $49.80 / € 39.95 ISBN 978-0-88937-543-7 Also available as eBook The unique longitudinal study “Tracking Opportunities and Problems (TOPP)” began following nearly 1,000 children and their families in Norway in 1993. Few studies have ever accumulated such extensive information from such a large number of families.

Eight waves of data on many aspects of child and family life have been collected from children aged 18 months to 18 years. The TOPP Study has provided new knowledge about and insight into the precursors, developmental paths and predictors of both good adaptation and mental health problems of children, as well as into parenting and family relationships.

The editors have collated the key findings in three parts. Part 1 addressesthe mental health and development of children and adolescents.

Part 2 focuses on parents, looking at

www.hogrefe.com

individual parental and family-related factors, including parental couple relationships. Part 3 looks at methodological issues, including the sample, response rate and measurement and analytical approaches. Each chapter reviews the existing knowledge in these areas in relation to the TOPP findings and provides extensive reference lists for those who want to dig deeper.

This unique book provides thoughtprovoking insights into the TOPP findings to help guide therapeutic practice, to suggest new avenues of research, to inform teaching, and to shape policy planning and preventive actions. It is thus an invaluable resource for all professionals, researchers, educators, policy makers, and students working with children and adolescents and their families.

Hogrefe OpenMind Open Access Publishing?

It’s Your Choice!

Your Road to Open Access Authors of papers accepted for publication in any Hogrefe journal can now choose to have their paper published as an open access article as part of the Hogrefe OpenMind program. This means that anyone, anywhere in the world will – without charge – be able to read, search, link, send, and use the article for noncommercial purposes, in accordance with the internationally recognized Creative Commons licensing standards.

The Choice Is Yours 1. Open Access Publication: The final “version of record” of the article is published online with full open access. It is freely available online to anyone in electronic form. (It will also be published in the print version of the journal.)

2. Traditional Publishing Model: Your article is published in the traditional manner, available worldwide to journal subscribers online and in print and to anyone by “pay per view.”

Whichever you choose, your article will be peer-reviewed, professionally produced, and published both in print and in electronic versions of the journal. Every article will be given a DOI and registered with CrossRef.

www.hogrefe.com

How Does Hogrefe’s Open Access Program Work?

After submission to the journal, your article will undergo exactly the same steps, no matter which publishing option you choose: peer-review, copy-editing, typesetting, data preparation, online reference linking, printing, hosting, and archiving. In the traditional publishing model, the publication process (including all the services that ensure the scientific and formal quality of your paper) is financed via subscriptions to the journal.

Open access publication, by contrast, is financed by means of a one-time article fee (€ 2,500 or US $3,000) payable by you the author, or by your research institute or funding body. Once the article has been accepted for publication, it’s your choice – open access publication or the traditional model. We have an open mind!

Short Report

Symptoms of Muscle Dysmorphia Between Users of Anabolic Androgenic Steroids With Varying Usage and Bodybuilding Experience Marc Ashley Harris, Tina Alwyn, and Michael Dunn Department of Applied Psychology, Cardiff Metropolitan University, UK

Abstract: Anabolic Androgenic Steroids (AAS) usage has been repeatedly linked with a body image disorder called Muscle Dysmorphia (MD).

However, evidence for how this relationship manifests is currently inconclusive. This study focused on the extent to which MD may precipitate or perpetuate the use of AAS. Utilizing a cross-sectional design, the sample consisted of 57 users (current and lifetime) and 51 non-using (never used AAS) bodybuilders recruited from two bodybuilding dedicated gymnasiums located in Wales, UK. Significantly higher levels of MD were found in users compared to non-users across training experience duration (0–2, 3–5, and 6+ years). MD levels irrespective of AAS usage (users vs. non-users) were consistent between 0–2 and 3–5 years of training declining however between those of 6+ years training experience.

No differences were reported in symptoms of MD between users of AAS with varying lengths of AAS use exposure. This provides preliminary evidence suggesting MD may influence both initiation and maintenance of AAS use; however, neither regular gym attendance nor prolonged AAS usage may exacerbate MD symptoms.

Keywords: muscle dysmorphia, body-image, anabolic androgenic steroids, anaerobic weight resistance training

Muscle Dysmorphia (MD) defines a pattern of pathological thinking about a specific aspect of body image, whereby an individual is persistently and distressingly preoccupied that they are insufficiently muscular, even though they are far more muscular than average (Pope, Gruber, Choi, Olivardia, & Phillips, 1997). As a clinical definition, MD is a subcategory of body dysmorphic disorder, although focused on dissatisfaction with muscularity, rather than overall body image (Leone, Sedory, & Gray, 2005). Prevalence rates are around 5.9% in the general population (Bo et al., 2014); however, precise rates are currently unknown due to limited published studies and the taboo nature of the topic. The etiology of MD is currently unclear and is likely to be the expression of a variety of factors (Baghurst, 2008).

There is consensus however that MD has causal roots in the way contemporary media and Western culture emphasize the importance of male muscularity (Kanayama, Hudson, & Pope, 2012). Additionally, there also exists an established notion that sports participation and specifically bodybuilding itself may reflect a considerable risk factor for developing MD (Lantz, Rhea, & Mayhew, 2001; Olivardia, 2001). Ó 2019 Hogrefe Publishing

This assertion suggests that such highly competitive environments may also have a direct impact on an individual’s distorted perceptions of body image. However, it is important to consider the possibility of other putative societal, cognitive, and biological causal factors (Lantz et al., 2001) even if such models are speculative rather than scientific.

There is considerable evidence that MD shares a complex relationship with the use of anabolic androgenic steroids (AAS), although evidence for how this relationship manifests is inconclusive (Mitchell et al., 2017; Rohman, 2009). Kanayama, Barry, Hudson, and Pope (2006), Kanayama, Pope, Cohane, and Hudson (2003) showed substantially higher symptoms of MD in users compared to non-users, which were prominent in men with a long history of AAS abuse. Pope and Kanayama (2012) argue that while many attributes showed little association with AAS use, conduct disorders and body-image concerns showed strong associations. Cole, Smith, Halford, and Wagstaff (2003) found an increase in symptoms of “reverse anorexia” for both current and former users, when compared to a control group of non-using bodybuilders whereas European Journal of Health Psychology (2019), 26(1), 21–24 https://doi.org/10.1027/2512-8442/a000023

22

M. A. Harris et al., Muscle Dismorphia and Anabolic Androgenic Steroid Use

Davies, Smith, and Collier (2011) found no differences in MD between current and former users. A fundamental question remains as to whether AAS usage functions as a precipitating or perpetuating factor in MD manifestation.

The current study endeavored to explore this potential influence by measuring levels of MD in men with or without AAS experience, engaged in varying lengths of exposure to anaerobic weight resistance training environments.

Access to such data places us in a unique position to establish the influence of being surrounded by muscular male physiques in a real-world setting on the manifestation and development of MD. In view of the previously discussed literature, it is predicted that MD levels will be higher in novice trainers who use AAS compared to their non-user peers and that this difference will be more pronounced in trainers with more sustained gym exposure. Levels are predicted to be higher in those with more extended AAS experience supporting previous research.

Method Participants N = 108 predominately working-class male bodybuilders aged between 18 and 40 years were recruited from two bodybuilding dedicated gymnasiums located in Wales, UK (one city based, one rural based), using opportunistic and snowballing sampling techniques. Gym A was among the 30â&#x20AC;&#x201C;50% most deprived nationally for income, employment, health and education, among the 50% most deprived for access to services and housing, and among the 10% most deprived for community safety. Gym B was among the 30â&#x20AC;&#x201C;50% most deprived nationally for income, among the 50% most deprived for employment, health, education and access to services, and among the 10â&#x20AC;&#x201C;20% most deprived for community safety. The sample included 57 users (current and lifetime) and 51 non-using (never used AAS) bodybuilders. The inclusion criteria required participants to be frequent bodybuilders (3 sessions min per week), and to be male. This criterion was justified as regular exposure to the bodybuilding environment is likely to be critical for the desirability of AAS to become apparent, and as a body image explanation for AAS use has been developed and tested primarily on men.

Design Categorical measures were collected via four questions: Motivation for bodybuilding (physical appearance, physical fitness, sport, or strength); Bodybuilding experience (how long engaged in anaerobic exercise); European Journal of Health Psychology (2019), 26(1), 21â&#x20AC;&#x201C;24

User versus non-user status; and AAS use experience (if used, length of usage).

These measures were selected based on two previous studies which found the factors underpinning usage appeared to change over time (Harris, Dunn, & Alwyn, 2017), with body image concerns becoming more prominent as duration of AAS use increased (Harris, Dunn, & Alwyn, 2016). Symptoms of MD were measured using the Muscle Dysmorphic Disorder Inventory (MDDI) (Hildebrandt, Langenbaucher, & Schlundt, 2004). The MDDI is a 13-item measure of MD symptomology, with items rated on a 5-point Likert scale, providing a total of between 13 and 65. The MDDI questions pertain to cognition, emotions, and behavior including three core measures of MD symptomology: Drive for Size (thoughts of being smaller, less muscular, and weaker than desired); Appearance Impairment (negative beliefs about oneâ&#x20AC;&#x2122;s body and resulting appearance anxiety or body exposure avoidance); and Functional Impairment (behaviors related to maintaining exercise routines, interference of negative emotions when deviating from exercise routines, or avoidance of social situations), which have demonstrated good reliability (Cronbachâ&#x20AC;&#x2122;s Îą = .77â&#x20AC;&#x201C;.85; testretest reliability r = .81â&#x20AC;&#x201C;.87), and construct validity (Zeeck et al., 2018).

The mean MDDI score for the overall sample of 34.9 (38.1 and 31.7 for users and non-users respectively) was consistent with those reported by male bodybuilders in previous studies (Zeeck et al., 2018).

Results Data were collected from n = 57 users (current and lifetime) and n = 51 non-users (never used) with varying levels of training experience (see Figure 1) and for users, with varying lengths of usage experience (see Figure 2).

Analysis 1 Conductance of a two-way between-subject ANOVA with factors of anaerobic training duration [0â&#x20AC;&#x201C;2 years (n = 13 and n = 15 for users and non-users, respectively), 3â&#x20AC;&#x201C;5 years (n = 19 and n = 12 for users and non-users, respectively), and 6+ years (n = 25 and n = 24 for users and non-users, respectively)] and AAS use (user/non-user) revealed a significant main effect of training duration, F2,102 = 4.52, p < .05, Ρ2p = .09, and AAS use, F1,102 = 10.17, p < .05, Ρ2p = .15; however, No training duration AAS use Ă&#x201C; 2019 Hogrefe Publishing

User

45

Non-User

40 35 30 25 20 15 10 5 0 0-2 years

3-5 years

6+ years

Training Experience

Figure 1. Mean MD scores for users and non-users of AAS across years of training experience. Values = Mean ± Standard Error Mean.

interaction was reported, F2,102 = .78, p > .05, η2p = .01.

Simple pairwise comparisons showed significantly lower MD levels in those training for 6+ years (irrespective of user/non-user status) compared to those with 3–5 years training experience only (p < .05). Overall, a mean score of 38.1 (SD = 2.5) was reported by users compared a mean score of 31.7 (SD = 2.2) for non-users.

Analysis 2 MD levels did not differ statistically across duration [(0–2 years (n = 31), 3–5 years (n = 12), and 6+ years (n = 14)] of AAS exposure, F2,54 = .25, p > .05, η2p = .009.

Discussion The present study aimed to compare levels of MD between gym attendees who either use or do not use AAS. Subsumed within this core aim was to measure and compare the influence of sustained exposure (on both users and non-users of AAS) to other male physiques (as evident in a competitive anaerobic training environment) on levels of MD and between AAS users with varying lengths of AAS exposure.

Results suggested a more complex relationship between MD, bodybuilding, and AAS usage than previously suggested (cf. Pope, Khalsa, & Bhasin, 2017). It would appear that MD may compel vulnerable individuals to embark on an AAS regimen as users exhibit higher MD scores compared to non-users during the initial phases of bodybuilding. However, importantly, as no increase in MD in users and non-users develops over time, this suggests that accumulative exposure to a training environment (other male bodybuilders and possibly bodybuilding paraphernalia) Ó 2019 Hogrefe Publishing

Muscle Dysmorphia Score (Mean + SEM)

Muscle Dysmorphia Score (Mean + SEM)

M. A. Harris et al., Muscle Dismorphia and Anabolic Androgenic Steroid Use

23

45 40 35 30 25 20 15 10 5 0 0-2 years

3-5 years

6+ years

AAS Use Experience

Figure 2. Mean MD scores for users of AAS across years of AAS use experience. Values = Mean ± Standard Error Mean.

does not exacerbate self-concern over muscular inadequacies. Indeed, levels of MD appear to be mitigated in those with extensive gym experience irrespective of AAS usage following an initial period of MD stability. This challenges somewhat the argument that the development of MD arises due to unrealistic comparisons being made between the individual and those with socially constructed, desirable physiques. Similarly, the analysis revealed no differences in symptoms of MD between users of AAS with varying lengths of AAS use contradicting recent qualitative findings (Harris et al., 2016). Harris et al. (2017) have reported a similar finding regarding levels of intra-sexual competition in a competitive bodybuilding environment. Such findings are somewhat at odds with the widespread belief that continual exposure to media imagery of muscular physiques creates both MD and overly aggressive individuals. If it does, then this influence may be restricted to media-generated, 2-dimensional imagery but that influence does not extend to a real-world gym environment. Plausibly, MD symptomology (irrespective of its etiology) may predispose certain individuals to initiating a bodybuilding/AAS use regime, rather than a bodybuilding/AAS use regime increasing MD symptomology in certain at-risk individuals.

The current study utilized a cross-sectional design limiting any firm conclusions pertaining to the developmental nature of MD, as does the utilization of self-report measures of AAS. Future research would undoubtedly benefit from adopting a longitudinal approach. While a considerable body of future research is required surrounding the relationship between MD and AAS use, these findings suggest a treatment intervention aimed at addressing maladaptive body image concerns (i.e., MD symptomology) could be promising for reducing or preventing AAS use and a counter-intuitive intervention strategy may be to encourage individuals with body image concerns to take up weight-resistance exercise within a supportive social environment.

European Journal of Health Psychology (2019), 26(1), 21–24

24

M. A. Harris et al., Muscle Dismorphia and Anabolic Androgenic Steroid Use

References Baghurst, T. (2008). Characteristics of muscle dysmorphia in male football, weight training, and competitive bodybuilding samples.

Ann Arbor, MI: ProQuest. https://doi.org/10.1016/j.bodyim.

2009.03.002 Bo, S., Zoccali, R., Ponzo, V., Soldati, L., De Carli, L., Benso, A., . . .

Abbate-Daga, G. (2014). University courses, eating problems and muscle dysmorphia: Are there any associations? Journal of Translational Medicine, 12, 221. https://doi.org/10.1186/ s12967-014-0221-2 Cole, J. C., Smith, R., Halford, J. C. G., & Wagstaff, G. F. (2003). A preliminary investigation into the relationship between anabolic-androgenic steroid use and the symptoms of reverse anorexia in both current and ex-users. Psychopharmacology, 166, 424–429. https://doi.org/10.1007/s00213-002-1352-3 Davies, R., Smith, D., & Collier, K. (2011). Muscle dysmorphia among current and former steroid users. Journal of Clinical Sport Psychology, 5, 77–94. https://doi.org/10.1123/jcsp.5.1.77 Harris, M. A., Dunn, M., & Alwyn, T. (2016). A qualitative exploration of the motivations underlying anabolic-androgenic steroid use from adolescence into adulthood. Health Psychology Report, 4, 315–320. https://doi.org/10.5114/hpr.2016.

61669 Harris, M. A., Dunn, M. J., & Alwyn, T. (2017). Intrasexual competition as a potential influence on anabolic-androgenic steroid use initiation. Journal of Health Psychology. Advance online publication. https://doi.org/10.1177/1359105317692145 Hildebrandt, T., Langenbucher, J., & Schlundt, D. G. (2004).

Muscularity concerns among men: Development of attitudinal and perceptual measures. Body Image, 1, 169–181. https://doi. org/10.1016/j.bodyim.2004.01.001 Kanayama, G., Barry, S., Hudson, J. I., & Pope, H. G. (2006). Body image and attitudes toward male roles in anabolic-androgenic steroid users. American Journal of Psychiatry, 163, 697–770. https://doi.org/10.1176/appi.ajp.163.4.697 Kanayama, G., Hudson, J. I., & Pope, H. G. (2012). Culture, psychosomatics and substance abuse: The example of body image drugs. Psychotherapy and Psychosomatics, 81, 73–78. https://doi.org/10.1159/000330415 Kanayama, G., Pope, H. G., Cohane, G., & Hudson, J. I. (2003). Risk factors for anabolic-androgenic steroid use among weightlifters: A case-control study. Drug and Alcohol Dependence, 71, 77–86. https://doi.org/10.1016/S0376-8716(03)00069-3 Lantz, C. D., Rhea, D. J., & Mayhew, J. L. (2001). The drive for size: A psycho behavioral model of muscle dysmorphia. International Sports Journal, 5, 71–86.

European Journal of Health Psychology (2019), 26(1), 21–24

Leone, J. E., Sedory, E. J., & Gray, K. A. (2005). Recognition and treatment of muscle dysmorphia and related body image disorders. Journal of Athletic Training, 40, 352–359.

Mitchell, L., Murray, S. B., Cobley, S., Hackett, D., Gifford, J., Capling, L., & O’Connor, H. (2017). Muscle dysmorphia symptomatology and associated psychological features in bodybuilders and non-bodybuilder resistance trainers: A systematic review and meta-analysis. Sports Medicine, 47, 233–259. https://doi.org/10.1007/s40279-016-0564-3 Olivardia, R. (2001). Mirror, mirror on the wall, who’s the largest of them all? The features and phenomenology of muscle dysmorphia. Harvard Review of Psychiatry, 9, 254–259. https://doi.org/ 10.1080/hrp.9.5.254.259 Pope, H. G., & Kanayama, G. (2012). Anabolic-androgenic steroids.

In J. Verster, K. Brady, M. Galanter, & P. Conrod (Eds.), Drug abuse and addiction in medical illness (pp. 251–264). New York, NY: Springer. https://doi.org/10.1007/978-1-4614-3375-0 Pope, H. G., Gruber, A. J., Choi, P., Olivardia, R., & Phillips, K. A. (1997). Muscle dysmorphia: An unrecognised form of body dysmorphic disorder. Psychosomatics, 38, 548–557. https:// doi.org/10.1016/S0033-3182(97)71400-2 Pope, H. G., Khalsa, J. H., & Bhasin, S. (2017). Body image disorders and abuse of anabolic-androgenic steroids among men. Journal of the American Medical Association, 317, 23–24. https://doi.org/10.1001/jama.2016.17441 Rohman, L. (2009). The relationship between anabolic androgenic steroids and muscle dysmorphia: A review. Eating Disorders, 17, 187–199. https://doi.org/10.1080/10640260902848477 Zeeck, A., Welter, V., Alatas, H., Hildebrandt, T., Lahmann, C., & Hartmann, A. (2018). Muscle Dysmorphic Disorder Inventory (MDDI): Validation of a German version with a focus on gender.

PLoS One, 13, e0207535. https://doi.org/10.1371/journal. pone.0207535 History Received February 19, 2018 Revision received March 8, 2019 Accepted March 11, 2019 Published online June 3, 2019 Marc Ashley Harris Department of Applied Psychology Cardiff School of Sport and Health Sciences Cardiff, Wales CF5 2YB UK mharris2@cardiffmet.ac.uk

Ó 2019 Hogrefe Publishing

Short Report

Psychological Predictors of Fatigue, Work and Social Adjustment, and Psychological Distress in Rheumatology Outpatients A Short Report Faith Matcham1 , Sheila Ali2, Katherine Irving3, and Trudie Chalder1,2 1

Department of Psychological Medicine, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, United Kingdom

2

Chronic Fatigue Research and Treatment Unit, South London and Maudsley NHS Foundation Trust, London, United Kingdom

3

King’s College Hospital NHS Foundation Trust, London, United Kingdom

Abstract: This study aims to investigate the longitudinal relationships between psychological variables and follow-up levels of fatigue, work and social adjustment, and psychological distress in people with rheumatic diseases. The study is a prospective observational study. Patients attending rheumatology outpatient appointments completed a questionnaire during their hospital visit and were mailed a follow-up questionnaire either at their next appointment or via postal questionnaire. Multivariate linear regression models examined the association between baseline cognitive behavioral responses, personality, social support and acceptance and follow-up levels of fatigue, work and social adjustment, and psychological distress, adjusting for age, gender, disease duration, and the length of time between baseline and follow-up. A total of 108 patients completed the follow-up questionnaires. The biggest predictors of having high levels of fatigue at follow-up were increased baseline damage beliefs and behavioral avoidance. Behavioral avoidance at baseline also had a strong relationship with worsened work and social adjustment at follow-up. The biggest predictor of psychological distress at follow-up was a lack of fatigue acceptance at baseline.

Keywords: psychological, fatigue, distress, work and social adjustment, rheumatic diseases

Autoimmune rheumatic diseases (ARD), including rheumatoid arthritis (RA), ankylosing spondylitis (AS), and systemic lupus erythematosus (SLE), are characterized by pain, inflammation, and reduced physical function. Fatigue is experienced as a common and debilitating symptom in ARD (Stebbings & Treharne, 2010) and may contribute to the high prevalence of depression in this patient group (Matcham, Rayner, Steer, & Hotopf, 2013). Moreover, social participation has been recently highlighted as an outcome of key importance to patients with ARD (Trenaman et al., 2017).

There is a growing body of evidence demonstrating associations between cognitive and behavioral factors such as acceptance, catastrophizing and avoidance, and outcomes of fatigue and distress, as well as mixed evidence to suggest that social support may play a role in adjustment to illness (Treharne, Kitas, Lyons, & Booth, 2005). A recent

Ó 2019 Hogrefe Publishing

systematic review has highlighted the lack of prospective research in this field (Matcham, Ali, Hotopf, & Chalder, 2015). However there is some evidence to suggest that various psychological variables such as helplessness, affect, and coping strategies may be prospectively associated with fatigue (Davis, Okun, Kruszewski, Zautra, & Tennen, 2010; Treharne et al., 2008). We have previously reported the cross-sectional associations between cognitive behavioral coping responses, fatigue, sleep, and distress in patients with ARD (Ali, Matcham, Irving, & Chalder, 2017). The purpose of the current study was to extend these findings to evaluate the prospective association between psychological factors and follow-up fatigue, adjustment, and distress outcomes in patients with rheumatic diseases. Identifying the psychological variables associated with poorer fatigue, distress, and adjustment in ARD may provide an indicator of useful targets for intervention.

European Journal of Health Psychology (2019), 26(1), 25–29 https://doi.org/10.1027/2512-8442/a000024

26

F. Matcham et al., Predictors of Fatigue, Adjustment and Distress in Rheumatology

Materials and Methods Design and Participants This is a prospective observational study, examining associations between psychological variables and fatigue, work and social adjustment, and psychological distress at follow-up in patients with ARD attending an outpatient rheumatology clinic. Details of the recruitment procedure and baseline data collection are reported elsewhere, along with eligibility criteria for participation (Ali et al., 2017). In addition to participants reported in our previous study, who had diagnoses classified as connective tissue disease (CTD), spondyloarthropathy (SpA), or RA, here we include an additional 30 patients with “other” primary diagnoses of gout, sarcoidosis, and systemic sclerosis.

Of the 192 people with baseline data, 108 (56.3%) completed a follow-up assessment. The mean time between baseline and follow-up was 7.4 months (range: 1.2–19.5).

Patients who did and did not complete follow-up data did not differ by any baseline demographic, clinical, or psychological characteristics (available upon request). Table 1 shows the psychological and demographic characteristics of the sample.

Participants were asked to complete a questionnaire at their next clinic appointment, and postal questionnaires were sent to those who: (a) did not have a follow-up clinic appointment during the data collection period; or (b) could not be approached during their follow-up clinic appointment. Participants provided written informed consent for publication of deidentified information, and the procedure for data collection was approved by the London Dulwich Research Ethics Committee (10/H0808/135).

symptom focusing; catastrophizing; avoidance resting; damage beliefs; embarrassment avoidance; all-or-nothing behaviors. For each subscale, higher scores represent more extreme beliefs or behaviors.

Fatigue Acceptance Fatigue acceptance was measured using the 9-item Fatigue Acceptance Questionnaire (FAQ; Brooks, Rimes, & Chalder, 2011). Higher scores indicate lack of fatigue acceptance or reduced willingness to confront or control symptoms.

Follow-Up Assessment The primary outcomes of interest were fatigue, work and social adjustment, and psychological distress at follow-up.

More details about each questionnaire, including internal consistency and reliability have been reported previously (Ali et al., 2017).

Fatigue Fatigue was measured using the Chalder Fatigue Scale (CFS; Chalder et al., 1993). This is an 11-item scale which measures both mental and physical fatigue. Responses to the items are summed to calculate a total score out of 33, with higher scores representing greater levels of fatigue.

This measure has been reported to have an internal consistency of 0.92 (Chalder et al., 1993).

Work and Social Adjustment Work and social adjustment were measured using the Work and Social Adjustment Scale (WSAS; Mundt, 2002). This 5-item scale results in total scores ranging from 0 to 40, with higher scores indicating worsened social adjustment and work disability.

Baseline Variables In addition to providing demographic and clinical data, participants completed several psychological assessments at baseline. Measures were chosen to reflect both a conventional cognitive behavioral model and third wave therapies and more details about each questionnaire, including internal consistency and reliability, have been reported previously (Ali et al., 2017).

Psychological Distress Psychological distress was measured using the Hospital Anxiety and Depression Scale (HADS; Zigmond & Snaith, 1983). This is a well-validated measure of depression and anxiety, which can be used with the depression and anxiety subscales, or by combining the two subscales to create and overall measure of psychological distress, as in the current analysis. Higher scores indicate increased symptoms of psychological distress.

Cognitive Behavioral Responses Cognitive behavioral responses were measured using the Cognitive Behavioral Responses Questionnaire (CBRQ; Skerrett & Moss-Morris, 2006; Ryan, Vitoratou, Goldsmith, & Chalder, 2018). This measures beliefs about symptoms, and the behavioral responses to these symptoms. It was designed to examine unhelpful coping behaviors in relation to symptoms, including the subscales: activity avoidance; European Journal of Health Psychology (2019), 26(1), 25–29

Analysis Pairwise correlation analyses examined associations between variables at baseline to inform inclusion of predictors in regression models (Electronic Supplementary Material 1, ESM 1). Multivariate linear regressions examined linear associations between baseline psychological variables Ó 2019 Hogrefe Publishing

F. Matcham et al., Predictors of Fatigue, Adjustment and Distress in Rheumatology

27

Table 1. Mean and standard deviation (SD) scores for baseline psychosocial variables and follow-up fatigue (CFQ), adjustment (WSAS), and distress (HADS) Total (N = 192)

RA (N = 56)

CTD (N = 69)

SpA (N = 37)

Other (N = 30)

Demographics Age, M (SD)

49.4 (14.9)

53.7 (13.4)

47.0 (16.0)

49.7 (14.9)

46.1 (10.1)

Female gender, n (%)

148 (76.7)

44 (78.6)

75 (87.2)

19 (50.0)

10 (76.9)

White British/Irish

91 (47.4)

28 (50.0)

27 (39.1)

18 (48.7)

18 (60.0)

Other white

18 (9.4)

7 (12.5)

5 (7.3)

6 (16.2)

0 (0.0)

Black

41 (21.4)

11 (19.6)

23 (33.3)

1 (2.7)

6 (20.0)

Asian

17 (8.9)

7 (12.5)

6 (8.7)

3 (8.1)

1 (3.3)

9 (4.7)

1 (1.8)

0 (0.0)

6 (16.2)

2 (6.7)

1 (0.5)

0 (0.0)

1 (1.5)

0 (0.0)

0 (0.0)

15 (7.8)

2 (3.6)

7 (10.1)

3 (8.1)

3 (10.0)

14.5 (7.5)

13.6 (4.8)

Ethnicity, n (%)

Mixed ethnicity Other Unstated Years of education, M (SD)

14.5 (9.0)

17.2 (8.4)

12.8 (7.6)

Marital status Single

46 (24.0)

14 (25.0)

18 (23.1)

5 (13.5)

9 (30.0)

Married/cohabiting

95 (49.5)

26 (46.4)

31 (44.9)

25 (67.6)

13 (43.3)

Divorced/separated

23 (12.0)

7 (12.5)

9 (13.0)

2 (5.4)

6 (16.7)

Widowed

7 (3.7)

3 (5.4)

2 (2.9)

1 (2.7)

1 (3.3)

Unstated

21 (10.9)

6 (10.7)

9 (13.0)

4 (10.8)

2 (6.7)

Clinical characteristics 9.3 (9.4)

10.3 (9.2)

9.2 (10.4)

7.9 (7.3)

Pain (VAS 0–100)

Illness duration, years, M (SD)

38.5 (29.0)

44.3 (30.6)

32.9 (26.2)

40.5 (30.6)

38.6 (29.2)

8.1 (9.4)

Disease severity (VAS 0–100)

41.4 (28.3)

44.3 (30.2)

41.4 (26.7)

39.7 (30.4)

39.1 (29.8)

Baseline psychosocial variables, M (SD) Fatigue (CFQ 0–33)

18.8 (6.6)

18.7 (6.6)

19.3 (6.9)

16.8 (6.9)

20.0 (4.9)

Work social adjustment (WSAS 0–40)

16.0 (12.0)

16.6 (12.0)

16.0 (10.9)

14.7 (12.9)

16.7 (13.7)

Psychological Distress (HADS 0–42)

14.9 (9.3)

14.9 (9.3)

14.9 (8.2)

12.9 (7.9)

17.0 (8.5)

Fear avoidance (CBRQ 0–24)

11.3 (4.7)

11.6 (5.3)

11.2 (4.2)

10.5 (4.4)

11.8 (5.2)

Catastrophizing (CBRQ 0–16)

7.9 (4.0)

8.7 (3.7)

7.3 (4.2)

7.7 (3.9)

7.9 (4.2)

Damage beliefs (CBRQ 0–20)

11.2 (3.6)

11.6 (3.6)

11.1 (3.7)

11.0 (3.0)

10.6 (4.0)

9.7 (6.4)

9.2 (6.1)

9.8 (6.9)

9.5 (6.5)

10.6 (5.9) 12.9 (6.0)

Embarrassment avoidance (CBRQ 0–24)

13.1 (5.9)

13.9 (5.4)

13.6 (6.0)

11.3 (6.4)

All or nothing (CBRQ 0–20)

Symptom focusing (CBRQ 0–24)

7.9 (5.1)

7.5 (5.0)

8.4 (4.6)

7.7 (5.0)

8.0 (6.5)

Avoidance resting (CBRQ 0–32)

9.7 (6.3)

9.3 (5.3)

10.1 (6.4)

8.3 (5.9)

11.2 (8.0)

Fatigue acceptance (FAQ 0–54)

22.7 (15.0)

22.3 (16.5)

23.3 (14.4)

21.8 (14.0)

23.4 (15.3)

Follow-up data*, M (SD) Fatigue (CFQ 0–33)

16.6 (6.5)

16.4 (7.1)

18.1 (6.8)

13.8 (4.0)

16.5 (6.5)

Work social adjustment (WSAS 0–40)

13.9 (11.4)

14.2 (11.8)

14.6 (10.3)

10.4 (11.1)

15.3 (14.2)

Psychological Distress (HADS 0–42)

14.0 (7.9)

14.2 (8.6)

15.1 (7.2)

12.3 (6.1)

12.3 (9.6)

Notes. No significant between diagnoses group differences for psychosocial variables; total group used for main analysis. M = Mean; SD = Standard Deviation; VAS = Visual Analogue Scale; CFQ = Chalder Fatigue Scale; WSAS = Work and Social Adjustment Scale; HADS = Hospital Anxiety Depression Scale; CBRQ = Cognitive Behavioral Responses Questionnaire; FAQ = Fatigue Acceptance Questionnaire; CTD = Connective Tissue Disease; SpA = Spondyloarthropathy; RA = Rheumatoid Arthritis. *Complete follow-up data available for 90 participants for CFS, 94 participants for WSAS, 98 participants for HADS.

(cognitive behavioral responses: fear avoidance, damage beliefs, catastrophizing, embarrassment avoidance, symptom focusing, all-or-nothing behavior and behavioral avoidance; and fatigue acceptance) and follow-up levels of fatigue, work and social adjustment, and psychological distress, with models additionally adjusted for age and length of time between baseline and follow-up. Ó 2019 Hogrefe Publishing

Results Psychosocial Variables and Fatigue, Adjustment, and Distress Table 2 shows the results of linear regression models examining the relationship between baseline psychological European Journal of Health Psychology (2019), 26(1), 25–29

28

F. Matcham et al., Predictors of Fatigue, Adjustment and Distress in Rheumatology

Table 2. Models examining linear associations between baseline psychosocial variables and follow-up fatigue (CFQ), adjustment (WSAS), and distress (HADS) Fatigue (CFQ) B (SE) Fear avoidance

0.23 (0.20) –

Damage beliefs

Beta

p

0.16 0.27 –

–

Work/Social Adjustment (WSAS) 95% CI 0.18, 0.64 –

B (SE) 0.36 (0.30) –

Beta

p

0.15 0.24 –

–

95% CI 0.25, 0.97 –

Psychological Distress (HADS) B (SE)

Beta

p

95% CI

0.11 (0.19)

0.07 0.56

0.50, 0.27

0.35 (0.28)

0.18 0.21

0.20, 0.90

Embarrassment avoidance

0.11 (0.16)

0.10 0.50

0.22, 0.44

0.26 (0.25)

0.14 0.30

0.24, 0.76

0.15 (0.15)

0.12 0.32

0.15, 0.44

Symptom focusing

0.19 (0.19)

0.18 0.30

0.56, 0.18

0.13 (0.29)

0.07 0.65

0.71, 0.45

0.07 (0.19)

0.06 0.69

0.30, 0.45

All-or-nothing

0.43 (0.18)

0.33 0.02

0.06, 0.79

0.59 (0.28)

0.26 0.04

0.02, 1.16

0.41 (0.17)

0.28 0.02

0.01, 0.75

Fatigue acceptance

0.07 (0.07)

0.15 0.36

0.08, 0.21

0.23 (0.12)

0.29 0.05

0.00, 0.46

0.14 (0.07)

0.27 0.04

0.00, 0.27

Notes. Models additionally adjusted for age and length of time between baseline and follow-up. B = coefficient; SE = Standard Error; Beta = Standardized β; p = p-Value; 95% CI = Confidence Intervals.

variables and follow-up fatigue, work/social adjustment, and distress. Assessment of multicollinearity between predictor variables revealed high levels of multicollinearity (variance inflation factors of 4) between catastrophizing, behavioral avoidance and damage beliefs for fatigue and work/social adjustment outcomes, and catastrophizing and behavioral avoidance for psychological distress outcomes, leading to these variables being omitted from the respective analyses.

The largest effect sizes were seen for the relationships between high levels of baseline all-or-nothing behavior and fatigue at follow-up. High levels of baseline all-ornothing behavior and fatigue acceptance were associated with increased levels of work/social adjustment and psychological distress at follow-up. Results of the analysis did not alter after multiple imputation methods for missing data (available upon request).

Discussion This small-scale prospective study identified several psychological variables which are targetable to improve fatigue, adjustment, and distress outcomes in ARD. All-ornothing behavior, defined as being over-active then extensively resting, was consistently associated with increased long-term fatigue, work/social adjustment, and distress.

Other cognitive processes such as lack of fatigue acceptance (attempting to control uncontrollable fatigue) were found to be widely associated with work and social adjustment and psychological distress.

This research supports previous findings highlighting the role that psychological variables can play in influencing functional outcomes in rheumatic diseases (Matcham et al., 2018). These findings also highlight that interventions

European Journal of Health Psychology (2019), 26(1), 25–29

targeting these variables may prove beneficial to improve these outcomes. Cognitive behavioral therapies have been widely used to manage fatigue and distress in rheumatic diseases (Hewlett et al., 2011) and similar approaches have also demonstrated efficacy in manualized and online formats (Dures & Hewlett, 2012).

The findings of this study highlight variables which could be effectively targeted in rheumatic diseases to improve a range of outcomes of importance to patients. In particular, the association between low fatigue acceptance and an increase in all outcomes suggests that the effectiveness of a third wave therapy such as acceptance and commitment therapy (ACT), should be evaluated in this patient group.

ACT does not attempt to get rid of fatigue and/or distress but invites people to open up to unpleasant feelings and move toward valued-based goals.

The main limitation of the current study is its small sample size, limiting opportunity for adjusting multivariate models for a wider range of confounding variables. Effect sizes may be imprecise due to a lack of statistical power.

However, this study makes a step toward increasing the evidence base in this field, filling gaps in the literature highlighted in previous systematic reviews (Matcham et al., 2015). The sample used is also a highly heterogenous group of people approached in clinic with minimal selection criteria applied, and may therefore be more representative of typical rheumatology outpatients. Replicating these results with a larger sample size would provide more robust estimates of effect size.

Fatigue, work and social adjustment, and psychological distress are outcomes of importance to patients (Stebbings & Treharne, 2010), which are also associated with worsened disease activity and physical function (Matcham et al., 2018). This paper highlights the psychological variables which may be useful interventional targets to improve these outcomes. Ó 2019 Hogrefe Publishing

F. Matcham et al., Predictors of Fatigue, Adjustment and Distress in Rheumatology

Electronic Supplementary Material The electronic supplementary material is available with the online version of the article at https://doi.org/10.1027/ 2512-8442/a000024 ESM 1. Table (.docx) Correlations between baseline predictor variables and outcome variables.

References Ali, S., Matcham, F., Irving, K., & Chalder, T. (2017). Fatigue and psychosocial variables in autoimmune rheumatic disease and chronic fatigue syndrome: A cross-sectional comparison.

Journal of Psychosomatic Research, 92, 1–8. https://doi.org/ 10.1016/j.jpsychores.2016.11.002 Brooks, S. K., Rimes, K. A., & Chalder, T. (2011). The role of acceptance in chronic fatigue syndrome. Journal of Psychosomatic Research, 71, 411–415. https://doi.org/10.1016/j. jpsychores.2011.08.001 Chalder, T., Berelowitz, G., Pawlikowska, T., Watts, L., Wessely, S., Wright, D., & Wallace, E. P. (1993). Development of a Fatigue Scale. Journal of Psychosomatic Research, 37, 147–153.

Davis, M. C., Okun, M. A., Kruszewski, D., Zautra, A. J., & Tennen, H. (2010). Sex differences in the relations of positive and negative daily events and fatigue in adults with rheumatoid arthritis. The Journal of Pain, 11, 1338–1347. https://doi.org/ 10.1016/J.JPAIN.2010.03.009 Dures, E., & Hewlett, S. (2012). Cognitive–behavioural approaches to self-management in rheumatic disease. Nature Reviews Rheumatology, 8, 553–559. https://doi.org/10.1038/nrrheum.

2012.108 Hewlett, S., Ambler, N., Almeida, C., Cliss, A., Hammond, A., Kitchen, K., . . . Pollock, J. (2011). Self-management of fatigue in rheumatoid arthritis: A randomised controlled trial of group cognitive-behavioural therapy. Annals of the Rheumatic Diseases, 70, 1060–1067. https://doi.org/10.1136/ard.2010.

144691 Matcham, F., Ali, S., Hotopf, M., & Chalder, T. (2015). Psychological correlates of fatigue in rheumatoid arthritis: A systematic review. Clinical Psychology Review, 39, 16–29. https://doi.org/ 10.1016/j.cpr.2015.03.004 Matcham, F., Davies, R., Hotopf, M., Hyrich, K. L., Norton, S., Steer, S., & Galloway, J. (2018). The relationship between depression and biologic treatment response in rheumatoid arthritis: An analysis of the British Society for Rheumatology Biologics Register. Rheumatology, 57, 835–843. https://doi.org/10.1093/ rheumatology/kex528 Matcham, F., Rayner, L., Steer, S., & Hotopf, M. (2013). The prevalence of depression in rheumatoid arthritis: A systematic review and meta-analysis. Rheumatology, 52, 2136–2148. https://doi.org/10.1093/rheumatology/ket169 Mundt, J. C. (2002). The Work and Social Adjustment Scale: A simple measure of impairment in functioning. The British Journal of Psychiatry, 180, 461–464. https://doi.org/10.1192/ bjp.180.5.461 Ryan, E. G., Vitoratou, S., Goldsmith, K. A., & Chalder, T. (2018).

Psychometric properties and factor structure of a long and

Ó 2019 Hogrefe Publishing

29

shortened version of the Cognitive and Behavioural Responses questionnaire. Psychosomatic Medicine, 80, 230–237.

Skerrett, T. N., & Moss-Morris, R. (2006). Fatigue and social impairment in multiple sclerosis: The role of patients’ cognitive and behavioral responses to their symptoms. Journal of Psychosomatic Research, 61, 587–593. https://doi.org/ 10.1016/j.jpsychores.2006.04.018 Stebbings, S., & Treharne, G. J. (2010). Fatigue in rheumatic disease: An overview. International Journal of Clinical Rheumatology, 5, 487–502. https://doi.org/10.2217/IJR.10.30 Treharne, G. J., Kitas, G. D., Lyons, A. C., & Booth, D. A. (2005).

Well-being in rheumatoid arthritis: The effects of disease duration and psychosocial factors. Journal of Health Psychology, 10, 457–474. https://doi.org/10.1177/1359105305051416 Treharne, G. J., Lyons, A. C., Hale, E. D., Goodchild, C. E., Booth, D. A., & Kitas, G. D. (2008). Predictors of fatigue over 1 year among people with rheumatoid arthritis. Psychology, Health & Medicine, 13, 494–504. https://doi.org/10.1080/ 13548500701796931 Trenaman, L., Boonen, A., Guillemin, F., Hiligsmann, M., Hoens, A., Marra, C., . . . Bansback, N. (2017). OMERACT Quality-Adjusted Life-Years (QALY) working group: Do current QALY measures capture what matters to patients? The Journal of Rheumatology, 44, 1899–1903. https://doi.org/10.3899/jrheum.161112 Zigmond, A., & Snaith, R. (1983). The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 67, 361–370.

History Received November 5, 2018 Revision received March 11, 2019 Accepted March 12, 2019 Published online June 3, 2019 Acknowledgments We thank Putu Khorisantono, James Gwinnutt, Fatma Mehmet and Egli Ioannou for assistance with data collection.

Funding This paper represents independent research funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.

Conflict of Interest Trudie Chalder is the author of self-help books for chronic fatigue.

Faith Matham has received honoraria for speaking from Pfizer.

Authors disclose no other conflicts of interest.

ORCID Faith Matcham https://orcid.org/0000-0002-4055-904X Faith Matcham Department of Psychological Medicine Institute of Psychiatry, Psychology & Neuroscience 16 De Crespigny Park London, SE5 8AF United Kingdom faith.matcham@kcl.ac.uk

European Journal of Health Psychology (2019), 26(1), 25–29

News and Announcements Meeting Calendar July 2–5, 2019 16th European Congress of Psychology (ECP2019) Moscow, Russian Federation Contact: Russian Psychological Society, E-mail info@ecp2019.ru, and European Federation of Psychologists’ Associations, E-mail headoffice@ efpa.eu, Web https://ecp2019.ru/

September 16–17, 2019 Mental Health and Wellness 2019 London, UK Contact: Lexis Conferences, Ave. Roger Vandendriessche 18, Brussels, Belgium, E-mail mentalhealth@lexisevent.com, Web https://www.lexisconferences.com/mentalhealthwellness

July 10–11, 2019 BPS Division of Health Psychology Annual Conference Manchester, UK Contact: British Psychological Society, Preena Tailor & Claire Woodward, E-mail dhpconference@bps.org.uk, Web https://www.bps.org.uk/events/ division-health-psychology-annualconference-2019

September 19–20, 2019 9th International Conference on Health, Wellness and Society Berkeley, CA, USA Contact: Health, Wellness & Society Research Network, Web https:// healthandsociety.com/2019conference

August 8–11, 2019 APA Annual Convention Chicago, IL, USA Contact: American Psychological Association, Washington, DC, USA, E-mail convention@apa.org, Web https://convention.apa.org/ September 3–7, 2019 33rd Annual Conference of the European Health Psychology Society Dubrovnik, Croatia Contact: European Health Psychology Society, E-mail secretary@ehps.net, Web ehps.net/conferences

October 23–24, 2019 International Conference on Wellness, Resilience and Nursing Frankfurt, Germany Contact: Contact desk, E-mail henrybenjamin@memeetings.net, Web https://wellness-resilience.annualcongress.com/ November 7–9, 2019 Annual Conference on Advancing School Mental Health Austin, TX, USA Contact: University of Maryland, E-mail ncsmh@som.umaryland.edu, Web http://csmh.umaryland.edu/ Conferences/Annual-Conference-onAdvancing-School-Mental-Health/

European Journal of Health Psychology (2019), 26(1), 30 https://doi.org/10.1027/2512-8442/a000025

November 28–29, 2019 2nd Euro Depression Conference Amsterdam, The Netherlands Contact: Contact desk, E-mail henrybenjamin@memeetings.net, Web https://depressioncongress.neurologyconference.com/europe/ July 19–24, 2020 32th International Congress of Psychology (ICP) Prague, Czech Republic Contact: Congress Secretariat, Computer System Group a.s., 5. Kvetna 65, 140 21 Prague, Czech Republic, E-mail secretariat@icp2020.com, Web http://www.icp2020.com/ September 2021 17th European Congress of Psychology (ECP2021) Ljubljana, Slovenia Contact: Slovenian Psychological Association, Web https://www. slovenia-convention.com/europeancongress-psychology-2021-held-ljubljana/, and European Federation of Psychologists’ Associations, E-mail headoffice@efpa.eu.

Ó 2019 Hogrefe Publishing

Instructions to Authors The European Journal of Health Psychology strives to promote theory and practice in the analysis of psychological approaches to health and disease. Its aim is, therefore, to publish high quality empirical or experimental research as well as sound practiceoriented articles, current methodological developments, and comprehensive critical reviews of the scientific literature.

European Journal of Health Psychology publishes the following types of article: Original Articles, Review Articles, Short Reports, Commentaries, and Book Reviews.

Manuscript submission: All manuscripts should in the first instance be submitted electronically at http://www.editorialmanager. com/zgp. Detailed instructions to authors are provided at http:// www.hogrefe.com/periodicals/zgp/advice-for-authors/ Copyright Agreement: By submitting an article, the author confirms and guarantees on behalf of him-/herself and any coauthors that the manuscript has not been submitted or published elsewhere, and that he or she holds all copyright in and titles to the submitted contribution, including any figures, photographs, line drawings, plans, maps, sketches, and tables, and that the article and its contents do not infringe in any way on the rights of third parties. The author indemnifies and holds harmless the publisher from any third party claims. The author agrees, upon acceptance of the article for publication, to transfer to the publisher the exclusive right to reproduce and distribute the article and its contents, both physically and in nonphysical, electronic, or other form, in the journal to which it has been submitted and in other independent publications, with no limitations on the number of copies or on the form or the extent of distribution. These rights are transferred for the duration of copyright as defined by international law. Furthermore, the author

Ó 2019 Hogrefe Publishing

transfers to the publisher the following exclusive rights to the article and its contents: 1. The rights to produce advance copies, reprints, or offprints of the article, in full or in part, to undertake or allow translations into other languages, to distribute other forms or modified versions of the article, and to produce and distribute summaries or abstracts.

2. The rights to microfilm and microfiche editions or similar, to the use of the article and its contents in videotext, teletext, and similar systems, to recordings or reproduction using other media, digital or analog, including electronic, magnetic, and optical media, and in multimedia form, as well as for public broadcasting in radio, television, or other forms of broadcast.

3. The rights to store the article and its content in machine readable or electronic form on all media (such as computer disks, compact disks, magnetic tape), to store the article and its contents in online databases belonging to the publisher or third parties for viewing or downloading by third parties, and to present or reproduce the article or its contents on visual display screens, monitors, and similar devices, either directly or via data transmission.

4. The rights to reproduce and distribute the article and its contents by all other means, including photomechanical and similar processes (such as photocopying or facsimile), and as part of so-called document delivery services.

5. The right to transfer any or all rights mentioned in this agreement, as well as rights retained by the relevant copyright clearing centers, including royalty rights to third parties.

Online Rights for Journal Articles: Guidelines on authors’ rights to archive electronic versions of their manuscripts online are given in the ’’Guidelines on sharing and use of articles in Hogrefe journals’’ on the journal’s web page at http://www.hogrefe.com/j/zgp.

December 2017

European Journal of Health Psychology (2019), 26(1)

Assessment methods in health psychology “This book is an excellent overview of measurement issues that are central to health psychology.”

David French, PhD, Professor of Health Psychology, University of Manchester, UK

Yael Benyamini / Marie Johnston / Evangelos C. Karademas (Editors)

Assessment in Health Psychology (Series: Psychological Assessment – Science and Practice – Vol. 2) 2016, vi + 346 pp.

US $69.00 / € 49.95 ISBN 978-0-88937-452-2 Also available as eBook

Assessment in Health Psychology presents and discusses the best and most appropriate assessment methods and instruments for all specific areas that are central for health psychologists. It also describes the conceptual and methodological bases for assessment in health psychology, as well as the most important current issues and recent progress in methods.

A unique feature of this book, which brings together leading authorities on health psychology assessment, is its emphasis on the bidirectional link between theory and practice.

www.hogrefe.com

Assessment in Health Psychology is addressed to masters and doctoral students in health psychology, to all those who teach health psychology to researchers from other disciplines, including clinical psychology, health promotion, and public health, as well as to health policy makers and other healthcare practitioners.

This volume provides a thorough and authoritative record of the best available assessment tools and methods in health psychology, making it an invaluable resource both for students and academics as well as for practitioners in their daily work.

Yoga für Gesundheitsprofis und Klienten

Ingrid Kollak

Yoga in Vorsorge und Therapie Fachbuch mit Übungen für Atmung, Bewegung und Konzentration 2019. 264 S., 193 Abb., 9 Tab., € 34,95 / CHF 45.50 ISBN 978-3-456-85893-7 Auch als eBook erhältlich

Dieses Fachbuch eröffnet einen Zugang zu neuem Wissen und praktischer Befähigung, indem es Yoga aus Sicht aktueller Studien mit praktischer Erkenntnis verknüpft.

Verbesserte Bewegungsabläufe gegen Rückenschmerzen, bewusste Atemkontrolle zur Verminderung von Asthma Anfällen oder methodischer Spannungsabbau zur Senkung des Angstniveaus werden vermittelt und gleichzeitig die Wirkungsweisen durch wissenschaftliche Studien und Untersuchungen belegt. Der Nutzen des Yoga-Übens wird deutlich zum Erhalt der Gesundheit und zur Unter-

www.hogrefe.com

stützung von Therapie und Rehabilitation somatischer und psychischer Erkrankungen.

Die Autorin vertritt die Auffassung, dass Yoga die Menschen befähigt: von Tag zu Tag, von der Kindheit bis ins Alter und vor allem in Krisen- und Krankheitszeiten. Damit Yoga vielen Menschen nützlich sein kann, muss er individuell gestaltet werden und sich den wechselnden körperlichen und psychischen Situationen anpassen. Dazu zeigt dieses Buch zielgerichtete Yoga-Übungen vor klassischen Referenzpositionen.

A concise guide to the assessment and treatment of insomnia for busy professionals “This book provides an excellent concise introduction to insomnia and its treatment with cognitive behavioral therapy.

A great addition to any therapist’s bookshelf!”

Philip Gehrman, PhD, CBSM, Associate Professor, Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA

William K. Wohlgemuth / Ana Imia Fins

Insomnia (Series: Advances in Psychotherapy – Evidence-Based Practice – Volume 42) 2019, viii + 94 pp.

US $29.80 / € 24.95 ISBN 978-0-88937-415-7 Also available as eBook About 40% of the population experiences difficulty falling or staying asleep at some time in a given year, while 10% of people suffer chronic insomnia. This concise reference written by leading experts for busy clinicians provides practical and upto-date advice on current approaches to assessment, diagnosis, and treatment of insomnia. Professionals and students learn to correctly identify and diagnose insomnia and gain hands-on information on how to

www.hogrefe.com

carry out treatment with the best evidence base: cognitive behavioral therapy for insomnia (CBT-I). The American Academy of Sleep Medicine (AASM) and the American College of Physicians (ACP) both recognize CBT-I as the first-line treatment approach to insomnia. Appendices include useful resources for the assessment and treatment of insomnia, which readers can copy and use in their clinical practice.

“It is the benefits that our products can bring, both for individuals and for society, that spur us on.”

Dr. G.-Jürgen Hogrefe, Publisher

How to assess and treat persistent depressive disorders “Persistent Depressive Disorders does a masterful job of laying out the nature of and treatments for those depressions that do not remit, with an emphasis on the cognitive behavioral analytic system of psychotherapy (CBASP), the most efficacious and best tested of them all. It is a real tour-de-force.”

Steven D. Hollon, PhD, Gertrude Conaway Vanderbilt Professor of Psychology, Vanderbilt University, Nashville, TN

J. Kim Penberthy

Persistent Depressive Disorders (Series: Advances in Psychotherapy – Evidence-Based Practice – Volume 43) 2019, vi + 106 pp.

US $29.80 / € 24.95 ISBN 978-0-88937-505-5 Also available as eBook This compact guide is packed with the latest knowledge on the assessment and treatment of persistent depressive disorders (PDDs) – the new DSM-5 diagnosis that amalgamates the categories dysthymic disorder (DD), chronic major depression (MDD), and DD with major depressive episode (MDE).

Written by a leading expert, the book guides us through the complexities of assessing PDDs and the models for understanding how these difficult to identify and potentially life-threatening disorders develop and are maintained over long periods. It then outlines those therapies

www.hogrefe.com

that have the strongest evidence base. The author goes on to explore in detail the cognitive behavioral analysis system of psychotherapy (CBASP), a treatment specifically developed for PDDs. This compelling integrated approach incorporates components of learning, developmental, interpersonal, and cognitive theory with aspects of interpersonal mindfulness. We are led expertly through the therapeutic process using clinical vignettes and practical tips, with particular attention paid to identifying the assessment and therapy methods most valuable in CBASP. Printable tools in the appendices can be used in daily practice.