- Email: [email protected]
Depression scores predict adherence in a dietary weight loss intervention trial
Clinical Nutrition 30 (2011) 593e598
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Depression scores predict adherence in a dietary weight loss intervention trial S.M. Somerset*, L. Graham, K. Markwell School of Public Health, and Griffith Health Institute, Griffith University, Meadowbrook 4131, Brisbane, Queensland, Australia
a r t i c l e i n f o
s u m m a r y
Article history: Received 5 October 2010 Accepted 18 April 2011
Background & aims: Depression has a complex association with cardiometabolic risk, both directly as an independent factor and indirectly through mediating effects on other risk factors such as BMI, diet, physical activity, and smoking. Since changes to many cardiometabolic risk factors involve behaviour change, the rise in depression prevalence as a major global health issue may present further challenges to long-term behaviour change to reduce such risk. This study investigated associations between depression scores and participation in a community-based weight management intervention trial. Methods: A group of 64 overweight (BMI > 27), otherwise healthy adults, were recruited and randomised to follow either their usual diet, or an isocaloric diet in which saturated fat was replaced with monounsaturated fat (MUFA), to a target of 50% total fat, by adding macadamia nuts to the diet. Subjects were assessed for depressive symptoms at baseline and at ten weeks using the Beck Depression Inventory (BDI-II). Both control and intervention groups received advice on National Guidelines for Physical Activity and adhered to the same protocol for food diary completion and trial consultations. Anthropometric and clinical measurements (cholesterol, inflammatory mediators) also were taken at baseline and 10 weeks. Results: During the recruitment phase, pre-existing diagnosed major depression was one of a range of reasons for initial exclusion of volunteers from the trial. Amongst enrolled participants, there was a significant correlation (R ¼ 0.38, p < 0.05) between BDI-II scores at baseline and duration of participation in the trial. Subjects with a baseline BDI 10 (moderate to severe depression symptoms) were more likely to dropout of the trial before week 10 (p < 0.001). BDI-II scores in the intervention (MUFA) diet group decreased, but increased in the control group over the 10-week period. Univariate analysis of variance confirmed these observations (adjusted R2 ¼ 0.257, p ¼ 0.01). Body weight remained static over the 10-week period in the intervention group, corresponding to a relative increase in the control group (adjusted R2 ¼ 0.097, p ¼ 0.064). Conclusions: Depression symptoms have the potential to affect enrolment in and adherence to dietbased risk reduction interventions, and may consequently influence the generalisability of such trials. Depression scores may therefore be useful for characterising, screening and allocating subjects to appropriate treatment pathways. Ó 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Depression Adherence Metabolic syndrome Dropout Dietary intervention
1. Introduction The accumulation of excessive adiposity (overweight and obesity) has emerged as a global health issue, requiring urgent preventive and therapeutic intervention.1 The primary justification for the public health focus on overweight and obesity prevention is the association with increased risk of a spectrum of cardiometabolic disease associated with metabolic syndrome (MetS), including coronary heart disease, hypertension and type II diabetes.2 In addition, overweight and obesity are associated with increased risk of other health consequences such as cancer3 and musculoskeletal injury.4 * Corresponding author. Tel.: þ61 (0) 733821027. E-mail address: s.somerset@griffith.edu.au (S.M. Somerset).
There is substantial evidence that if populations and individuals are able to adopt appropriate behavioural change in terms of diet, physical activity and medication, then large reductions in cardiometabolic risk would ensue.1 A major challenge to risk reduction interventions is maximising uptake (initiation) and persistence (adherence) within these various treatment/prevention modalities.5 This translation from efficacy to effectiveness is particularly critical to obesity prevention and treatment.6 Behaviour-based weight loss interventions in general have very poor long-term outcomes. Diet and physical activity weight loss trials have substantial dropout rates, and even those subjects who persist with interventions often have difficulty maintaining long-term adherence to dietary protocols, with only 21% (or less) overweight/obese subjects successfully achieving long-term weight loss and maintenance.7,8 This phenomenon occurs in
0261-5614/$ e see front matter Ó 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2011.04.004
594
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
a range of contexts. For example, a recent study by Tsiros et al.9 recruited 47 subjects to participate in a randomised controlled trial testing the effects of cognitive behaviour therapy (CBT) on dietary outcomes in a group of overweight/obese adolescents. After a 20 week period, only 18 subjects (38%) remained despite the dietary intervention being very moderate (education on general aspects of a healthy diet and physical activity). In a more homogenous patient group, Luyster et al. found that both depression and anxiety symptoms compromised dietary adherence in heart failure patients with defibrillators.10 Conceptual frameworks based on Self Determination Theory11 and the Transtheoretical Model of behaviour change12 have been applied to understand the determinants of engagement in and adherence to weight loss intervention by obese subjects. However, such frameworks assume no underlying psychological dysfunction. This assumption seems prevalent in the design of weight loss intervention trials. For example, a recent study by Messier et al.13 applied a comprehensive psychosocial profiling of participants in a weight loss trial (body esteem, self esteem, perceived stress, self efficacy, dietary restraint), but did not report testing for depressive symptoms. In an environment of increasing depression prevalence, approaches involving self-determination and transtheoretical models may have only limited relevance in determining successful weight loss. On a global level, depression has been identified as a major public health issue with prevalence expanding rapidly.14,15 Various lines of evidence identify depression as having profound and complex associations with the cardiometabolic comorbidities of obesity and MetS. Firstly, depression acts directly on coronary risk16 and indirectly through its effects on determinants of coronary risk such as BMI, diet, physical activity, smoking, medication treatment
fidelity and various biochemical mediators.17e20 More specifically, there are strong and consistent relationships between depression and obesity, physical activity and energy intake in middle-aged women,21 and between depression and physical activity in elderly men.22 Depression is also associated with increased risk of developing type II diabetes.23 Conversely, obesity and CHD both correlate with an elevated risk of depression.24 Depression therefore has the potential to affect cardiometabolic risk through multiple pathways. Beyond these observations from cross-sectional data, there is some evidence that depression treatment protocols such as Cognitive Behaviour Therapy (CBT) can enhance intervention adherence. For example, Tsiros et al.9 showed improvements in diet and body composition in a group of overweight/obese adolescents when CBT was integrated into a healthy eating and physical activity intervention. Welschen et al.25 have proposed that CBT also may be beneficial in changing lifestyle in patients with diabetes. As a means to screening for early drop-outs and characterising trial generalisability, we sought to investigate how depression symptoms might affect participation in a community-based weight loss intervention which employed changes to diet and physical activity behaviours in overweight and obese subjects. 2. Methods The effect of depression scores on adherence to a weight loss trial was conducted as part of a trial comparing the effects of two diet/physical activity protocols on weight loss in a group of overweight subjects. The trial design is outlined in Fig. 1. This trial is registered with the Australia New Zealand Clinical Trial Registry (ACTRN12607000106437).
Fig. 1. Experimental design of randomised trial comparing a monounsaturated fat-enriched diet compared to usual (control) diet.
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
Brief advertisements were placed in several local newspapers over a four-week period calling for volunteers to participate in a University-based study on weight loss using dietary modification. A contact phone number was included in the ad, which was attended to during working hours and had an answering machine connected for after-hours calls. All calls received in response to the ads were logged, and a telephone consult was conducted either on the day of receipt, or at a negotiated time to determine subject suitability. Respondents were screened initially via telephone using BMI (>27), age (25e65 years), and a current prescribed medication audit. They were then interviewed face-to-face to ascertain medical history. Those with a self-reported pre-existing medical condition (including Major Depression) or whose medication audit indicated pre-existing medical condition were excluded from the trial. Volunteers were given a seven-day diet diary to complete. After a two-week period, those who were able to complete this diary competently were invited to participate. A total of 64 overweight (BMI > 25), otherwise healthy subjects were enrolled in this study, and randomised to follow their usual diet or a monounsaturated fat (MUFA) enriched version of their usual diet. Subjects were randomised consecutively as they entered the trial, in blocks of 10 subjects. For the intervention group, baseline energy and macronutrient intakes were calculated from two independent three-day diet diaries conducted two weeks apart (comprising 2 weekdays and 1 weekend day), and saturated fat was replaced with MUFA (to 50% total fat) by adding macadamia nuts to the diet. Both control and intervention groups received advice on national guidelines for physical activity and adhered to the same protocol for diet diary record keeping and trial consultations to control for Hawthorne effect.26 Patients were scheduled consultations at 0, 4 and 10 weeks, to supervise adherence to the diet and physical activity protocol as previously described.27e29 Prior to each consultation, subjects were given a three-day diet diary to complete, and a three-pass 24 h food recall interview was conducted during each consultation.30,31 At the 4-week consult, MUFA intake was assessed and the diet reviewed if MUFA intake fell outside the range of 45e55% total fat. Anthropometric and clinical measurements were taken at baseline and 10 weeks. For the purposes of assessing the impact of depression scores on trial participation, persistence with the trial to enable sampling of biochemical, anthropometric and dietary data at weeks 0, 4 and 10 (ie. Non-adherence corresponded to notification of the study team of severance from the trial) was used as an indicator of adherence. Data were analysed at 10 weeks. All diet data were analysed using diet analysis software based on the Australian food composition database AUSNUT. Data were entered into SPSS 17.0, cleaned and checked for normality. Statistical associations were tested using the Pearson correlation coefficient. Subjects were stratified according to BDI-II score being < or 10, and statistical differences were detected using independent sample t-test analysis. 3. Results The profiles of 129 volunteers responding to the newspaper ads, and reasons for exclusion from the trial are summarised in Table 1. Pre-diagnosed Major Depression accounted for only a small proportion of subject exclusion. Volunteers were predominantly female, comprising 84.4% and 89.2% of volunteers who were included and excluded, respectively. Baseline characteristics of the 55 subjects who submitted blood samples are summarised in Table 2. A total of 66.7% and 56.5% of male and female subjects were obese (BMI > 35) at baseline, respectively. Females had significantly (p < 0.05) higher cardiometabolic risk profile than males (lower mean HDL: 1.5 0.3 vs 1.2 0.1, higher mean fasting blood glucose: 5.2 0.5 vs 5.8 0.9 mM/L, respectively), in the study group.
595
Table 1 Proportions of reasons for non-participation or exclusion from entry into the trial. Reason for exclusion/non-participation
%
No response after initial contact Smoker Family/personal reasons Not interested after further explanation Age > 55 years Hypertension Glucose intolerance BMI > 40 Taking fish oil supplements Major depression Fatty liver disease Trying to conceive, extreme dietary practices, high recent weight loss, hormone replacement therapy, oral contraceptive use, time commitment too great, age < 25 years Nut allergy, arthritis, asthma, lactating, high blood cholesterol, BMI < 25, Hashimoto disease, work commitments
34.1 9.3 7.8 7.0 5.4 4.7 3.1 3.1 3.1 2.3 2.3 1.6 each
0.8 each
There was a significant correlation (R ¼ 0.38, p < 0.05) between BDI-II scores at baseline and duration of participation in the trial. Subjects with a baseline BDI 10 (moderate to severe depression symptoms) were more likely to dropout of the trial before 10 weeks (p < 0.001). Bivariate analysis of variance confirmed these observations (adjusted R2 ¼ 0.257, p ¼ 0.01). A statistical difference (p < 0.05) of mean time in trial up until 10 weeks post-baseline was observed with 28.72 16.1 and 10.4 16.1 days for BDI score groups <10 and 10, respectively. Mean weight changes (SD) at 10 weeks were 2.29 4.3, and þ0.2 1.2 kg for BDI score groups <10 and 10, respectively. The difference (independent sample t-test) between these groups was not significant (p > 0.05). 4. Discussion Results from the present study showed a significant association between depression symptoms and progression of subjects to the 10-week trial completion, thereby implicating depression as a determinant of trial resignation (dropout). The trial resignation phenomenon represents a significant burden on the development and delivery of effective weight loss intervention programs for a range of reasons. Firstly it imposes substantial up-front costs to research studies, since baseline data are collected and subjects are then lost to follow-up. In addition to the added cost, this compromises the statistical analysis of trial outcomes.32 This situation is replicated in the clinical setting, where major investment in baseline assessment and intervention delivery is spent on patients who dropout. In essence, the evidence base for effective weight loss intervention is restricted to those patients who persist with treatment. Table 2 Clinical profiles of trial subjects (n ¼ 55).
BMI Waist (cm) Total cholesterol (mmol/L) HDL (mmol/L) LDL (mmol/L) Total triglycerides (mmol/L) Glucose (mmol/L) CRP (mg/L) Lp(a) (mg/L) IL-6 (pg/ml) BDI-II score (baseline) BDI-II score (10 wks)
Mean
SD
Max
Min
34.33 103.0 5.3 1.42 3.2 1.8 5.29 7.0 415 3.1 9.5 8.3
4.4 11.72 0.8 0.31 0.74 1.3 0.57 3.5 80 2.4 6.2 7.3
43.5 194.5 7.1 2.6 4.9 5.7 7.6 20 1820 31.0 23.0 26.0
27.3 85 3.4 0.9 1.4 0.9 4.2 5.0 80 2.0 1.0 1.0
596
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
Secondly, trial resignation introduces a potential systematic sampling bias, since those subjects who remain to trial end are more likely to have lower depression scores. An additional issue that may arise in such trials is that changes in depression scores may occur between control and intervention groups, a phenomenon which has the potential to undermine randomised controlled trial (RCT) design.33e35 This highlights the utility of collecting psychological data to characterise subjects in dietary intervention trials. Methods used to recruit subjects can have a significant influence on the profile of volunteers. In the present study, subjects would be expected to comprise those who live in the local area, read the local newspaper (or know someone who does), are contemplating the need to make lifestyle changes and perhaps are unable to meet the costs of such services themselves. Consequently, the cohort for this study could not be construed as representative of the local population. However, they do represent a patient profile that might respond to such a recruitment protocol, a design which is not unusual in this type of study.36 Clinical characteristics of subjects shown in Table 2 confirm that the recruitment strategy was successful in accessing those at heightened coronary risk. Of particular note in terms of patient characteristics was the large majority of female volunteers. This gender disparity is not uncommon in community based intervention trials.37,38 In addition, female subjects had a significantly different (poorer) coronary risk profile (HDL, blood glucose). The discontinuum between the need for intervention in males and volunteering for such trials has been observed previously, for example in type II diabetes.39 Given the higher prevalence of coronary risk in adult men (cf women), the recruitment method used in the present study may only be of limited use in accessing at risk populations. Weight loss interventions which apply behaviour change to diet and physical activity habits have only modest long-term outcomes.40 There is ample evidence identifying adherence as a major issue,5 with far fewer longer-term trials being reported in
the literature.40 The potential for depression to play a major role via affecting determinants of adherence is compelling (see Fig. 2). The low proportion of volunteers excluded because of prediagnosed Major Depression implies that this recruitment protocol did not necessarily capture this patient profile. However, it is clear from the BDI-II scores that a proportion of subjects with depressogenic tendancy were recruited. The elevated depression symptoms observed in the sample size indicate an important proportion of seemingly undiagnosed depressogenic tendency, especially in relation to the infrequency of major depression as a reason for exclusion. Undiagnosed depression, especially as a comorbidity for factors associated with metabolic syndrome, is a well-documented phenomenon.41,42 Early intervention in major depression is critical, since once a Major Depressive Episode (MDE) has occurred, recurrence is high.15 An important challenge for health service providers is to identify risk prior to MDE. Routine screening for depression symptoms in General Practice has been promoted in the USA, but poses substantial cost to the health sector.43 The apparent comorbidity of obesity and depression may represent a more targeted approach to screening, since excess body weight is easily identified in clinical practice. The substantial attrition rates reported in many weight loss interventions represent relatively poor returns in relation to upfront investments in diagnosis and treatment. Depression scoring may deliver a cost-effective means to identify patients less likely to adhere to standard diet and physical activity interventions. Once such patients are identified, they can be diverted to appropriate (pre)interventions. There is an emerging evidence that CBT, a common behavioural treatment for depression, can affect adherence across a range of treatment modalities.9,44,45 The BDI was chosen as a depression score43 for the present study due to its widespread recognition and its well-established validity for clinical assessment. However, it is acknowledged that it measures a fairly narrow perspective on the range of the psychological factors which could potentially impact on adherence to an
Fig. 2. Conceptual model of potential interaction of depression risk and cardiometabolic risk via treatment adherence (adapted from Shay 200847).
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
intervention. In this context, tools such as the Depression, Anxiety and Stress (DAAS) score46 may be more informative and useful. Adherence is a major factor in the transition of interventions from efficacy to effectiveness (ie. from controlled clinical/research facilities to free-living community settings) across a comprehensive spectrum of treatment modalities. In the healthcare and health promotion context, the term adherence is generally defined as an ability to maintain specific behaviours related to treatment or prevention plans.47 As a multi-dimensional phenomenon, adherence encompasses an array of behaviours such as attendance at counselling session, completion of self reported behaviours, fidelity to prescribed treatments and achievements in relation to desired outcome variable changes.48 Specific definitions of adherence can vary according to treatment modality. For example, adherence in weight loss relates to the ability to maintain a weight change for at least 2 years.47 Physical activity is another treatment/prevention modality in which adherence plays a key role in risk reduction.22 Boyette et al.49 identified factors such as age, gender, ethnicity, occupation, educational level, socioeconomic status, smoking status, and past exercise participation as dominant antecedents to adherence. For physical activity, various criteria including maintenance of regular exercise for at least 6 months are commonly used as a benchmark.50,51 In relation to treatment fidelity to medication, a figure of 80% of the proportion of days covered by medication is often used.52 The present study focused on attendance at data collection points as the pragmatic indicator of trial participation, since this is a critical occasion for data collection and outcome measurement, and therefore has an important role to play in the costeffectiveness of intervention delivery. Depression can affect treatment adherence across all modalities where behaviour change is integral (eg. medication, physical activity, diet, addiction). Thus, in addition to direct influences of depression on each of the components of MetS (obesity, glucose intolerance, hypertension, dyslipidaemia), depression and/or its symptoms could render MetS more resistant to treatment. There are indeed established links between depression and MetS.53 The precise causal pathways between these links remain to be elucidated, but appear to be bidirectional and influenced by both biochemical and behavioural mediators.54 5. Conclusion Depression and its symptoms were a prevalent mitigating factor in this intervention trial, both at the recruitment stage and during the intervention phase. Emerging literature supports the need for integration between interventions addressing depression and those focussing on diet and physical activity. Pre-entry screening for depression symptoms and appropriate subsequent intervention has potential to maximise adherence to diet and physical activitybased risk reduction interventions. Better characterisation of the psychosocial profiles of subjects in such trials is an important step in generalising results to the broader community. Conflict of interest None declared. Acknowledgements This project is supported by Horticulture Australia (HAL 32633) and listed with the Australia New Zealand Clinical Trial Registry (ACTRN12607000106437). Ethical Clearance was obtained through the Griffith University Human Research Ethics Committee (PBH/09/ 05/HREC). KM is supported by a Griffith University Postgraduate Research Scholarship.
597
References 1. Bays H, Blonde L, Rosenson R. Adiposopathy: how do diet, exercise and weight loss drug therapies improve metabolic disease in overweight patients? Expert Rev Cardiovasc Ther 2006;4:871e95. 2. Sarti C, Gallagher J. The metabolic syndrome: prevalence, CHD risk, and treatment. J Diabetes Complications 2006;20:121e32. 3. WRCF/AICR. Food, nutrition and the prevention of cancer: a global perspective. Washington, DC: American Institute for Cancer Research; 1997. 4. Anandacoomarasamy A, Caterson I, Sambrook P, Fransen M, March L. The impact of obesity on the musculoskeletal system. Int J Obes (Lond) 2008;32:211e22. 5. Fabricatore AN, Wadden TA, Moore RH, Butryn ML, Gravallese EA, Erondu NE, et al. Attrition from randomized controlled trials of pharmacological weight loss agents: a systematic review and analysis. Obes Rev 2009;10:333e41. 6. Yaskin J, Toner RW, Goldfarb N. Obesity management interventions: a review of the evidence. Popul Health Manag 2009;12:305e16. 7. Ayyad C, Andersen T. Long-term efficacy of dietary treatment of obesity: a systematic review of studies published between 1931 and 1999. Obes Rev 2000;1:113e9. 8. Mant A, Lowinger J, Hall W, Whicker S, Ringland C, Stark H. Statins down under. Br J Clin Pharmacol 2007;63:125e7. 9. Tsiros MD, Sinn N, Brennan L, Coates AM, Walkley JW, Petkov J, et al. Cognitive behavioral therapy improves diet and body composition in overweight and obese adolescents. Am J Clin Nutr 2008;87:1134e40. 10. Luyster FS, Hughes JW, Gunstad J. Depression and anxiety symptoms are associated with reduced dietary adherence in heart failure patients treated with an implantable cardioverter defibrillator. J Cardiovasc Nurs 2009;24:10e7. 11. Williams GC, Grow VM, Freedman ZR, Ryan RM, Deci EL. Motivational predictors of weight loss and weight-loss maintenance. J Pers Soc Psychol 1996;70:115e26. 12. Wilson GT, Schlam TR. The transtheoretical model and motivational interviewing in the treatment of eating and weight disorders. Clin Psychol Rev 2004;24:361e78. 13. Messier V, Hayek J, Karelis AD, Messier L, Doucet E, Prud’homme D, et al. Anthropometric, metabolic, psychosocial and dietary factors associated with dropout in overweight and obese postmenopausal women engaged in a 6-month weight loss programme: a MONET study. Br J Nutr; 2009:1e6. 14. Tiong AC, Patel MS, Gardiner J, Ryan R, Linton KS, Walker KA, et al. Health issues in newly arrived African refugees attending general practice clinics in Melbourne. Med J Aust 2006;185:602e6. 15. Barrera AZ, Torres LD, Munoz RF. Prevention of depression: the state of the science at the beginning of the 21st century. Int Rev Psychiatry 2007;19:655e70. 16. Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. Int J Geriatr Psychiatry 2007;22:613e26. 17. Wagner JA, Tennen H. History of major depressive disorder and diabetes outcomes in diet- and tablet-treated post-menopausal women: a case control study. Diabet Med 2007;24:211e6. 18. Goldstein LB, Adams R, Alberts MJ, Appel LJ, Brass LM, Bushnell CD, et al. Primary prevention of ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council: cosponsored by the Atherosclerotic Peripheral Vascular Disease Interdisciplinary Working Group; Cardiovascular Nursing Council; Clinical Cardiology Council; Nutrition, Physical Activity, and Metabolism Council; and the Quality Of Care And Outcomes Research Interdisciplinary Working Group: the American Academy Of Neurology affirms the value of this guideline. Stroke 2006;37:1583e633. 19. Teychenne M, Ball K, Salmon J. Associations between physical activity and depressive symptoms in women. Int J Behav Nutr Phys Act 2008;5:27. 20. Rondina Rde C, Gorayeb R, Botelho C. Psychological characteristics associated with tobacco smoking behavior. J Bras Pneumol 2007;33:592e601. 21. Simon GE, Ludman EJ, Linde JA, Operskalski BH, Ichikawa L, Rohde P, et al. Association between obesity and depression in middle-aged women. Gen Hosp Psychiatry 2008;30:32e9. 22. Kamphuis MH, Geerlings MI, Tijhuis MA, Giampaoli S, Nissinen A, Grobbee DE, et al. Physical inactivity, depression, and risk of cardiovascular mortality. Med Sci Sports Exerc 2007;39:1693e9. 23. Cosgrove MP, Sargeant LA, Griffin SJ. Does depression increase the risk of developing type 2 diabetes? Occup Med (Lond) 2008;58:7e14. 24. Baune BT, Adrian I, Arolt V, Berger K. Associations between major depression, bipolar disorders, dysthymia and cardiovascular diseases in the general adult population. Psychother Psychosom 2006;75:319e26. 25. Welschen LM, van Oppen P, Dekker JM, Bouter LM, Stalman WA, Nijpels G. The effectiveness of adding cognitive behavioural therapy aimed at changing lifestyle to managed diabetes care for patients with type 2 diabetes: design of a randomised controlled trial. BMC Public Health 2007;7:74. 26. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P. The Hawthorne effect: a randomised, controlled trial. BMC Med Res Methodol 2007;7:30. 27. Colquhoun DM, Moores D, Somerset SM, Humphries JA. Comparison of the effects on lipoproteins and apolipoproteins of a diet high in monounsaturated
598
28.
29.
30.
31. 32.
33. 34.
35. 36.
37.
38.
39.
40.
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598 fatty acids, enriched with avocado, and a high-carbohydrate diet. Am J Clin Nutr 1992;56:671e7. Colquhoun DM, Humphries JA, Moores D, Somerset SM. Effects of a macadamia nut enriched diet on serum lipids and lipoproteins compared to a low fat diet. Food Aust 1996;48:216e22. Colquhoun DM, Somerset S, Glasziou P, Richards D, Weyers J. comparison of an olive oil enriched diet to a low fat diet intervention study using vascular endpoints e assessed by repeat quantitative angiography (OLIVE study). Aust J Nutr Dietet 1998;55:S24e29. Shuaibi AM, Sevenhuysen GP, House JD. Validation of a food choice map with a 3-day food record and serum values to assess folate and vitamin B-12 intake in college-aged women. J Am Diet Assoc 2008;108:2041e50. Parnell WR, Wilson NC, Russell DG. Methodology of the 1997 New Zealand national nutrition survey. N Z Med J 2001;114:123e6. Elobeid MA, Padilla MA, McVie T, Thomas O, Brock DW, Musser B, et al. Missing data in randomized clinical trials for weight loss: scope of the problem, state of the field, and performance of statistical methods. PLoS One 2009;4:e6624. Mullner M, Matthews H, Altman DG. Reporting on statistical methods to adjust for confounding: a cross-sectional survey. Ann Intern Med 2002;136:122e6. Rios LP, Odueyungbo A, Moitri MO, Rahman MO, Thabane LCIN. Quality of reporting of randomized controlled trials in general endocrinology literature. J Clin Endocrin Metab 2008;93:3810e6. Somerset SM. Reporting of dietary control frameworks in studies on simvastatin efficacy. Clin Nutr 2010;29:352e7. Yancey AK, Miles OL, McCarthy WJ, Sandoval G, Hill J, Leslie JJ, et al. Differential response to targeted recruitment strategies to fitness promotion research by African-American women of varying body mass index. Ethn Dis 2001;11:115e23. Burroughs AR, Visscher WA, Haney TL, Efland JR, Barefoot JC, Williams Jr RB, Siegler IC. Community recruitment process by race, gender, and SES gradient: lessons learned from the Community Health and Stress Evaluation (CHASE) Study experience. J Community Health 2003;28:421e37. Garrett SK, Thomas AP, Cicuttini F, Silagy C, Taylor HR, McNeil JJ. Communitybased recruitment strategies for a longitudinal interventional study: the VECAT experience. J Clin Epidemiol 2000;53:541e8. Davis RM, Hitch AD, Nichols M, Rizvi A, Salaam M, Mayer-Davis EJ. A collaborative approach to the recruitment and retention of minority patients with diabetes in rural community health centers. Contemp Clin Trials 2009;30:63e70. Franz MJ, VanWormer JJ, Crain AL, Boucher JL, Histon T, Caplan W, et al. Weight-loss outcomes: a systematic review and meta-analysis of weight-loss
41.
42.
43. 44.
45.
46. 47. 48.
49.
50.
51.
52.
53. 54.
clinical trials with a minimum 1-year follow-up. J Am Diet Assoc 2007;107:1755e67. Davidson KW, Korin MR. Depression and cardiovascular disease: selected findings, controversies, and clinical implications from. Cleve Clin J Med 2009;77(Suppl. 3):S20e6. Holt RI, Phillips DI, Jameson KA, Cooper C, Dennison EM, Peveler RC. The relationship between depression and diabetes mellitus: findings from the Hertfordshire Cohort study. Diabet Med 2009;26:641e8. McAlpine DD, Wilson AR. Screening for depression in primary care: what do we still need to know? Depress Anxiety 2004;19:137e45. Mefferd K, Nichols JF, Pakiz B, Rock CL. A cognitive behavioral therapy intervention to promote weight loss improves body composition and blood lipid profiles among overweight breast cancer survivors. Breast Cancer Res Treat 2007;104:145e52. Rodriguez-Hernandez H, Morales-Amaya UA, Rosales-Valdez R, RiveraHinojosa F, Rodriguez-Moran M, Guerrero-Romero F. Adding cognitive behavioural treatment to either low-carbohydrate or low-fat diets: differential short-term effects. Br J Nutr 2009;102:1847e53. Crawford J, Henry J. Normative data and latent structure in a large non-clinical sample. Br J Clin Psychol 2003;42:111e31. Shay LE. A concept analysis: adherence and weight loss. Nurs Forum 2008;43:42e52. Williamson DA, Anton SD, Han H, Champagne CM, Allen R, LeBlanc E, et al. Adherence is a multi-dimensional construct in the POUNDS LOST trial. J Behav Med 2010;33:35e46. Boyette LW, Lloyd A, Boyette JE, Watkins E, Furbush L, Dunbar SB, et al. Personal characteristics that influence exercise behavior of older adults. J Rehabil Res Dev 2002;39:95e103. Jack K, McLean SM, Moffett JK, Gardiner E. Barriers to treatment adherence in physiotherapy outpatient clinics: a systematic review. Man Ther 2010 Jun;15:220e8. Turk MW, Yang K, Hravnak M, Sereika SM, Ewing LJ, Burke LE. Randomized clinical trials of weight loss maintenance: a review. J Cardiovasc Nurs 2009;24:58e80. Schneeweiss S, Patrick AR, Maclure M, Dormuth CR, Glynn RJ. Adherence to statin therapy under drug cost sharing in patients with and without acute myocardial infarction: a population-based natural experiment. Circulation 2007;115:2128e35. Kyrou I, Tsigos C. Stress hormones: physiological stress and regulation of metabolism. Curr Opin Pharmacol 2009;9:787e93. Cohen BE, Panguluri P, Na B, Whooley MA. Psychological risk factors and the metabolic syndrome in patients with coronary heart disease: findings from the Heart and Soul Study. Psychiatry Res Jan 30 2010;175:133e7.
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Depression scores predict adherence in a dietary weight loss intervention trial S.M. Somerset*, L. Graham, K. Markwell School of Public Health, and Griffith Health Institute, Griffith University, Meadowbrook 4131, Brisbane, Queensland, Australia
a r t i c l e i n f o
s u m m a r y
Article history: Received 5 October 2010 Accepted 18 April 2011
Background & aims: Depression has a complex association with cardiometabolic risk, both directly as an independent factor and indirectly through mediating effects on other risk factors such as BMI, diet, physical activity, and smoking. Since changes to many cardiometabolic risk factors involve behaviour change, the rise in depression prevalence as a major global health issue may present further challenges to long-term behaviour change to reduce such risk. This study investigated associations between depression scores and participation in a community-based weight management intervention trial. Methods: A group of 64 overweight (BMI > 27), otherwise healthy adults, were recruited and randomised to follow either their usual diet, or an isocaloric diet in which saturated fat was replaced with monounsaturated fat (MUFA), to a target of 50% total fat, by adding macadamia nuts to the diet. Subjects were assessed for depressive symptoms at baseline and at ten weeks using the Beck Depression Inventory (BDI-II). Both control and intervention groups received advice on National Guidelines for Physical Activity and adhered to the same protocol for food diary completion and trial consultations. Anthropometric and clinical measurements (cholesterol, inflammatory mediators) also were taken at baseline and 10 weeks. Results: During the recruitment phase, pre-existing diagnosed major depression was one of a range of reasons for initial exclusion of volunteers from the trial. Amongst enrolled participants, there was a significant correlation (R ¼ 0.38, p < 0.05) between BDI-II scores at baseline and duration of participation in the trial. Subjects with a baseline BDI 10 (moderate to severe depression symptoms) were more likely to dropout of the trial before week 10 (p < 0.001). BDI-II scores in the intervention (MUFA) diet group decreased, but increased in the control group over the 10-week period. Univariate analysis of variance confirmed these observations (adjusted R2 ¼ 0.257, p ¼ 0.01). Body weight remained static over the 10-week period in the intervention group, corresponding to a relative increase in the control group (adjusted R2 ¼ 0.097, p ¼ 0.064). Conclusions: Depression symptoms have the potential to affect enrolment in and adherence to dietbased risk reduction interventions, and may consequently influence the generalisability of such trials. Depression scores may therefore be useful for characterising, screening and allocating subjects to appropriate treatment pathways. Ó 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Depression Adherence Metabolic syndrome Dropout Dietary intervention
1. Introduction The accumulation of excessive adiposity (overweight and obesity) has emerged as a global health issue, requiring urgent preventive and therapeutic intervention.1 The primary justification for the public health focus on overweight and obesity prevention is the association with increased risk of a spectrum of cardiometabolic disease associated with metabolic syndrome (MetS), including coronary heart disease, hypertension and type II diabetes.2 In addition, overweight and obesity are associated with increased risk of other health consequences such as cancer3 and musculoskeletal injury.4 * Corresponding author. Tel.: þ61 (0) 733821027. E-mail address: s.somerset@griffith.edu.au (S.M. Somerset).
There is substantial evidence that if populations and individuals are able to adopt appropriate behavioural change in terms of diet, physical activity and medication, then large reductions in cardiometabolic risk would ensue.1 A major challenge to risk reduction interventions is maximising uptake (initiation) and persistence (adherence) within these various treatment/prevention modalities.5 This translation from efficacy to effectiveness is particularly critical to obesity prevention and treatment.6 Behaviour-based weight loss interventions in general have very poor long-term outcomes. Diet and physical activity weight loss trials have substantial dropout rates, and even those subjects who persist with interventions often have difficulty maintaining long-term adherence to dietary protocols, with only 21% (or less) overweight/obese subjects successfully achieving long-term weight loss and maintenance.7,8 This phenomenon occurs in
0261-5614/$ e see front matter Ó 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2011.04.004
594
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
a range of contexts. For example, a recent study by Tsiros et al.9 recruited 47 subjects to participate in a randomised controlled trial testing the effects of cognitive behaviour therapy (CBT) on dietary outcomes in a group of overweight/obese adolescents. After a 20 week period, only 18 subjects (38%) remained despite the dietary intervention being very moderate (education on general aspects of a healthy diet and physical activity). In a more homogenous patient group, Luyster et al. found that both depression and anxiety symptoms compromised dietary adherence in heart failure patients with defibrillators.10 Conceptual frameworks based on Self Determination Theory11 and the Transtheoretical Model of behaviour change12 have been applied to understand the determinants of engagement in and adherence to weight loss intervention by obese subjects. However, such frameworks assume no underlying psychological dysfunction. This assumption seems prevalent in the design of weight loss intervention trials. For example, a recent study by Messier et al.13 applied a comprehensive psychosocial profiling of participants in a weight loss trial (body esteem, self esteem, perceived stress, self efficacy, dietary restraint), but did not report testing for depressive symptoms. In an environment of increasing depression prevalence, approaches involving self-determination and transtheoretical models may have only limited relevance in determining successful weight loss. On a global level, depression has been identified as a major public health issue with prevalence expanding rapidly.14,15 Various lines of evidence identify depression as having profound and complex associations with the cardiometabolic comorbidities of obesity and MetS. Firstly, depression acts directly on coronary risk16 and indirectly through its effects on determinants of coronary risk such as BMI, diet, physical activity, smoking, medication treatment
fidelity and various biochemical mediators.17e20 More specifically, there are strong and consistent relationships between depression and obesity, physical activity and energy intake in middle-aged women,21 and between depression and physical activity in elderly men.22 Depression is also associated with increased risk of developing type II diabetes.23 Conversely, obesity and CHD both correlate with an elevated risk of depression.24 Depression therefore has the potential to affect cardiometabolic risk through multiple pathways. Beyond these observations from cross-sectional data, there is some evidence that depression treatment protocols such as Cognitive Behaviour Therapy (CBT) can enhance intervention adherence. For example, Tsiros et al.9 showed improvements in diet and body composition in a group of overweight/obese adolescents when CBT was integrated into a healthy eating and physical activity intervention. Welschen et al.25 have proposed that CBT also may be beneficial in changing lifestyle in patients with diabetes. As a means to screening for early drop-outs and characterising trial generalisability, we sought to investigate how depression symptoms might affect participation in a community-based weight loss intervention which employed changes to diet and physical activity behaviours in overweight and obese subjects. 2. Methods The effect of depression scores on adherence to a weight loss trial was conducted as part of a trial comparing the effects of two diet/physical activity protocols on weight loss in a group of overweight subjects. The trial design is outlined in Fig. 1. This trial is registered with the Australia New Zealand Clinical Trial Registry (ACTRN12607000106437).
Fig. 1. Experimental design of randomised trial comparing a monounsaturated fat-enriched diet compared to usual (control) diet.
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
Brief advertisements were placed in several local newspapers over a four-week period calling for volunteers to participate in a University-based study on weight loss using dietary modification. A contact phone number was included in the ad, which was attended to during working hours and had an answering machine connected for after-hours calls. All calls received in response to the ads were logged, and a telephone consult was conducted either on the day of receipt, or at a negotiated time to determine subject suitability. Respondents were screened initially via telephone using BMI (>27), age (25e65 years), and a current prescribed medication audit. They were then interviewed face-to-face to ascertain medical history. Those with a self-reported pre-existing medical condition (including Major Depression) or whose medication audit indicated pre-existing medical condition were excluded from the trial. Volunteers were given a seven-day diet diary to complete. After a two-week period, those who were able to complete this diary competently were invited to participate. A total of 64 overweight (BMI > 25), otherwise healthy subjects were enrolled in this study, and randomised to follow their usual diet or a monounsaturated fat (MUFA) enriched version of their usual diet. Subjects were randomised consecutively as they entered the trial, in blocks of 10 subjects. For the intervention group, baseline energy and macronutrient intakes were calculated from two independent three-day diet diaries conducted two weeks apart (comprising 2 weekdays and 1 weekend day), and saturated fat was replaced with MUFA (to 50% total fat) by adding macadamia nuts to the diet. Both control and intervention groups received advice on national guidelines for physical activity and adhered to the same protocol for diet diary record keeping and trial consultations to control for Hawthorne effect.26 Patients were scheduled consultations at 0, 4 and 10 weeks, to supervise adherence to the diet and physical activity protocol as previously described.27e29 Prior to each consultation, subjects were given a three-day diet diary to complete, and a three-pass 24 h food recall interview was conducted during each consultation.30,31 At the 4-week consult, MUFA intake was assessed and the diet reviewed if MUFA intake fell outside the range of 45e55% total fat. Anthropometric and clinical measurements were taken at baseline and 10 weeks. For the purposes of assessing the impact of depression scores on trial participation, persistence with the trial to enable sampling of biochemical, anthropometric and dietary data at weeks 0, 4 and 10 (ie. Non-adherence corresponded to notification of the study team of severance from the trial) was used as an indicator of adherence. Data were analysed at 10 weeks. All diet data were analysed using diet analysis software based on the Australian food composition database AUSNUT. Data were entered into SPSS 17.0, cleaned and checked for normality. Statistical associations were tested using the Pearson correlation coefficient. Subjects were stratified according to BDI-II score being < or 10, and statistical differences were detected using independent sample t-test analysis. 3. Results The profiles of 129 volunteers responding to the newspaper ads, and reasons for exclusion from the trial are summarised in Table 1. Pre-diagnosed Major Depression accounted for only a small proportion of subject exclusion. Volunteers were predominantly female, comprising 84.4% and 89.2% of volunteers who were included and excluded, respectively. Baseline characteristics of the 55 subjects who submitted blood samples are summarised in Table 2. A total of 66.7% and 56.5% of male and female subjects were obese (BMI > 35) at baseline, respectively. Females had significantly (p < 0.05) higher cardiometabolic risk profile than males (lower mean HDL: 1.5 0.3 vs 1.2 0.1, higher mean fasting blood glucose: 5.2 0.5 vs 5.8 0.9 mM/L, respectively), in the study group.
595
Table 1 Proportions of reasons for non-participation or exclusion from entry into the trial. Reason for exclusion/non-participation
%
No response after initial contact Smoker Family/personal reasons Not interested after further explanation Age > 55 years Hypertension Glucose intolerance BMI > 40 Taking fish oil supplements Major depression Fatty liver disease Trying to conceive, extreme dietary practices, high recent weight loss, hormone replacement therapy, oral contraceptive use, time commitment too great, age < 25 years Nut allergy, arthritis, asthma, lactating, high blood cholesterol, BMI < 25, Hashimoto disease, work commitments
34.1 9.3 7.8 7.0 5.4 4.7 3.1 3.1 3.1 2.3 2.3 1.6 each
0.8 each
There was a significant correlation (R ¼ 0.38, p < 0.05) between BDI-II scores at baseline and duration of participation in the trial. Subjects with a baseline BDI 10 (moderate to severe depression symptoms) were more likely to dropout of the trial before 10 weeks (p < 0.001). Bivariate analysis of variance confirmed these observations (adjusted R2 ¼ 0.257, p ¼ 0.01). A statistical difference (p < 0.05) of mean time in trial up until 10 weeks post-baseline was observed with 28.72 16.1 and 10.4 16.1 days for BDI score groups <10 and 10, respectively. Mean weight changes (SD) at 10 weeks were 2.29 4.3, and þ0.2 1.2 kg for BDI score groups <10 and 10, respectively. The difference (independent sample t-test) between these groups was not significant (p > 0.05). 4. Discussion Results from the present study showed a significant association between depression symptoms and progression of subjects to the 10-week trial completion, thereby implicating depression as a determinant of trial resignation (dropout). The trial resignation phenomenon represents a significant burden on the development and delivery of effective weight loss intervention programs for a range of reasons. Firstly it imposes substantial up-front costs to research studies, since baseline data are collected and subjects are then lost to follow-up. In addition to the added cost, this compromises the statistical analysis of trial outcomes.32 This situation is replicated in the clinical setting, where major investment in baseline assessment and intervention delivery is spent on patients who dropout. In essence, the evidence base for effective weight loss intervention is restricted to those patients who persist with treatment. Table 2 Clinical profiles of trial subjects (n ¼ 55).
BMI Waist (cm) Total cholesterol (mmol/L) HDL (mmol/L) LDL (mmol/L) Total triglycerides (mmol/L) Glucose (mmol/L) CRP (mg/L) Lp(a) (mg/L) IL-6 (pg/ml) BDI-II score (baseline) BDI-II score (10 wks)
Mean
SD
Max
Min
34.33 103.0 5.3 1.42 3.2 1.8 5.29 7.0 415 3.1 9.5 8.3
4.4 11.72 0.8 0.31 0.74 1.3 0.57 3.5 80 2.4 6.2 7.3
43.5 194.5 7.1 2.6 4.9 5.7 7.6 20 1820 31.0 23.0 26.0
27.3 85 3.4 0.9 1.4 0.9 4.2 5.0 80 2.0 1.0 1.0
596
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
Secondly, trial resignation introduces a potential systematic sampling bias, since those subjects who remain to trial end are more likely to have lower depression scores. An additional issue that may arise in such trials is that changes in depression scores may occur between control and intervention groups, a phenomenon which has the potential to undermine randomised controlled trial (RCT) design.33e35 This highlights the utility of collecting psychological data to characterise subjects in dietary intervention trials. Methods used to recruit subjects can have a significant influence on the profile of volunteers. In the present study, subjects would be expected to comprise those who live in the local area, read the local newspaper (or know someone who does), are contemplating the need to make lifestyle changes and perhaps are unable to meet the costs of such services themselves. Consequently, the cohort for this study could not be construed as representative of the local population. However, they do represent a patient profile that might respond to such a recruitment protocol, a design which is not unusual in this type of study.36 Clinical characteristics of subjects shown in Table 2 confirm that the recruitment strategy was successful in accessing those at heightened coronary risk. Of particular note in terms of patient characteristics was the large majority of female volunteers. This gender disparity is not uncommon in community based intervention trials.37,38 In addition, female subjects had a significantly different (poorer) coronary risk profile (HDL, blood glucose). The discontinuum between the need for intervention in males and volunteering for such trials has been observed previously, for example in type II diabetes.39 Given the higher prevalence of coronary risk in adult men (cf women), the recruitment method used in the present study may only be of limited use in accessing at risk populations. Weight loss interventions which apply behaviour change to diet and physical activity habits have only modest long-term outcomes.40 There is ample evidence identifying adherence as a major issue,5 with far fewer longer-term trials being reported in
the literature.40 The potential for depression to play a major role via affecting determinants of adherence is compelling (see Fig. 2). The low proportion of volunteers excluded because of prediagnosed Major Depression implies that this recruitment protocol did not necessarily capture this patient profile. However, it is clear from the BDI-II scores that a proportion of subjects with depressogenic tendancy were recruited. The elevated depression symptoms observed in the sample size indicate an important proportion of seemingly undiagnosed depressogenic tendency, especially in relation to the infrequency of major depression as a reason for exclusion. Undiagnosed depression, especially as a comorbidity for factors associated with metabolic syndrome, is a well-documented phenomenon.41,42 Early intervention in major depression is critical, since once a Major Depressive Episode (MDE) has occurred, recurrence is high.15 An important challenge for health service providers is to identify risk prior to MDE. Routine screening for depression symptoms in General Practice has been promoted in the USA, but poses substantial cost to the health sector.43 The apparent comorbidity of obesity and depression may represent a more targeted approach to screening, since excess body weight is easily identified in clinical practice. The substantial attrition rates reported in many weight loss interventions represent relatively poor returns in relation to upfront investments in diagnosis and treatment. Depression scoring may deliver a cost-effective means to identify patients less likely to adhere to standard diet and physical activity interventions. Once such patients are identified, they can be diverted to appropriate (pre)interventions. There is an emerging evidence that CBT, a common behavioural treatment for depression, can affect adherence across a range of treatment modalities.9,44,45 The BDI was chosen as a depression score43 for the present study due to its widespread recognition and its well-established validity for clinical assessment. However, it is acknowledged that it measures a fairly narrow perspective on the range of the psychological factors which could potentially impact on adherence to an
Fig. 2. Conceptual model of potential interaction of depression risk and cardiometabolic risk via treatment adherence (adapted from Shay 200847).
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598
intervention. In this context, tools such as the Depression, Anxiety and Stress (DAAS) score46 may be more informative and useful. Adherence is a major factor in the transition of interventions from efficacy to effectiveness (ie. from controlled clinical/research facilities to free-living community settings) across a comprehensive spectrum of treatment modalities. In the healthcare and health promotion context, the term adherence is generally defined as an ability to maintain specific behaviours related to treatment or prevention plans.47 As a multi-dimensional phenomenon, adherence encompasses an array of behaviours such as attendance at counselling session, completion of self reported behaviours, fidelity to prescribed treatments and achievements in relation to desired outcome variable changes.48 Specific definitions of adherence can vary according to treatment modality. For example, adherence in weight loss relates to the ability to maintain a weight change for at least 2 years.47 Physical activity is another treatment/prevention modality in which adherence plays a key role in risk reduction.22 Boyette et al.49 identified factors such as age, gender, ethnicity, occupation, educational level, socioeconomic status, smoking status, and past exercise participation as dominant antecedents to adherence. For physical activity, various criteria including maintenance of regular exercise for at least 6 months are commonly used as a benchmark.50,51 In relation to treatment fidelity to medication, a figure of 80% of the proportion of days covered by medication is often used.52 The present study focused on attendance at data collection points as the pragmatic indicator of trial participation, since this is a critical occasion for data collection and outcome measurement, and therefore has an important role to play in the costeffectiveness of intervention delivery. Depression can affect treatment adherence across all modalities where behaviour change is integral (eg. medication, physical activity, diet, addiction). Thus, in addition to direct influences of depression on each of the components of MetS (obesity, glucose intolerance, hypertension, dyslipidaemia), depression and/or its symptoms could render MetS more resistant to treatment. There are indeed established links between depression and MetS.53 The precise causal pathways between these links remain to be elucidated, but appear to be bidirectional and influenced by both biochemical and behavioural mediators.54 5. Conclusion Depression and its symptoms were a prevalent mitigating factor in this intervention trial, both at the recruitment stage and during the intervention phase. Emerging literature supports the need for integration between interventions addressing depression and those focussing on diet and physical activity. Pre-entry screening for depression symptoms and appropriate subsequent intervention has potential to maximise adherence to diet and physical activitybased risk reduction interventions. Better characterisation of the psychosocial profiles of subjects in such trials is an important step in generalising results to the broader community. Conflict of interest None declared. Acknowledgements This project is supported by Horticulture Australia (HAL 32633) and listed with the Australia New Zealand Clinical Trial Registry (ACTRN12607000106437). Ethical Clearance was obtained through the Griffith University Human Research Ethics Committee (PBH/09/ 05/HREC). KM is supported by a Griffith University Postgraduate Research Scholarship.
597
References 1. Bays H, Blonde L, Rosenson R. Adiposopathy: how do diet, exercise and weight loss drug therapies improve metabolic disease in overweight patients? Expert Rev Cardiovasc Ther 2006;4:871e95. 2. Sarti C, Gallagher J. The metabolic syndrome: prevalence, CHD risk, and treatment. J Diabetes Complications 2006;20:121e32. 3. WRCF/AICR. Food, nutrition and the prevention of cancer: a global perspective. Washington, DC: American Institute for Cancer Research; 1997. 4. Anandacoomarasamy A, Caterson I, Sambrook P, Fransen M, March L. The impact of obesity on the musculoskeletal system. Int J Obes (Lond) 2008;32:211e22. 5. Fabricatore AN, Wadden TA, Moore RH, Butryn ML, Gravallese EA, Erondu NE, et al. Attrition from randomized controlled trials of pharmacological weight loss agents: a systematic review and analysis. Obes Rev 2009;10:333e41. 6. Yaskin J, Toner RW, Goldfarb N. Obesity management interventions: a review of the evidence. Popul Health Manag 2009;12:305e16. 7. Ayyad C, Andersen T. Long-term efficacy of dietary treatment of obesity: a systematic review of studies published between 1931 and 1999. Obes Rev 2000;1:113e9. 8. Mant A, Lowinger J, Hall W, Whicker S, Ringland C, Stark H. Statins down under. Br J Clin Pharmacol 2007;63:125e7. 9. Tsiros MD, Sinn N, Brennan L, Coates AM, Walkley JW, Petkov J, et al. Cognitive behavioral therapy improves diet and body composition in overweight and obese adolescents. Am J Clin Nutr 2008;87:1134e40. 10. Luyster FS, Hughes JW, Gunstad J. Depression and anxiety symptoms are associated with reduced dietary adherence in heart failure patients treated with an implantable cardioverter defibrillator. J Cardiovasc Nurs 2009;24:10e7. 11. Williams GC, Grow VM, Freedman ZR, Ryan RM, Deci EL. Motivational predictors of weight loss and weight-loss maintenance. J Pers Soc Psychol 1996;70:115e26. 12. Wilson GT, Schlam TR. The transtheoretical model and motivational interviewing in the treatment of eating and weight disorders. Clin Psychol Rev 2004;24:361e78. 13. Messier V, Hayek J, Karelis AD, Messier L, Doucet E, Prud’homme D, et al. Anthropometric, metabolic, psychosocial and dietary factors associated with dropout in overweight and obese postmenopausal women engaged in a 6-month weight loss programme: a MONET study. Br J Nutr; 2009:1e6. 14. Tiong AC, Patel MS, Gardiner J, Ryan R, Linton KS, Walker KA, et al. Health issues in newly arrived African refugees attending general practice clinics in Melbourne. Med J Aust 2006;185:602e6. 15. Barrera AZ, Torres LD, Munoz RF. Prevention of depression: the state of the science at the beginning of the 21st century. Int Rev Psychiatry 2007;19:655e70. 16. Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. Int J Geriatr Psychiatry 2007;22:613e26. 17. Wagner JA, Tennen H. History of major depressive disorder and diabetes outcomes in diet- and tablet-treated post-menopausal women: a case control study. Diabet Med 2007;24:211e6. 18. Goldstein LB, Adams R, Alberts MJ, Appel LJ, Brass LM, Bushnell CD, et al. Primary prevention of ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council: cosponsored by the Atherosclerotic Peripheral Vascular Disease Interdisciplinary Working Group; Cardiovascular Nursing Council; Clinical Cardiology Council; Nutrition, Physical Activity, and Metabolism Council; and the Quality Of Care And Outcomes Research Interdisciplinary Working Group: the American Academy Of Neurology affirms the value of this guideline. Stroke 2006;37:1583e633. 19. Teychenne M, Ball K, Salmon J. Associations between physical activity and depressive symptoms in women. Int J Behav Nutr Phys Act 2008;5:27. 20. Rondina Rde C, Gorayeb R, Botelho C. Psychological characteristics associated with tobacco smoking behavior. J Bras Pneumol 2007;33:592e601. 21. Simon GE, Ludman EJ, Linde JA, Operskalski BH, Ichikawa L, Rohde P, et al. Association between obesity and depression in middle-aged women. Gen Hosp Psychiatry 2008;30:32e9. 22. Kamphuis MH, Geerlings MI, Tijhuis MA, Giampaoli S, Nissinen A, Grobbee DE, et al. Physical inactivity, depression, and risk of cardiovascular mortality. Med Sci Sports Exerc 2007;39:1693e9. 23. Cosgrove MP, Sargeant LA, Griffin SJ. Does depression increase the risk of developing type 2 diabetes? Occup Med (Lond) 2008;58:7e14. 24. Baune BT, Adrian I, Arolt V, Berger K. Associations between major depression, bipolar disorders, dysthymia and cardiovascular diseases in the general adult population. Psychother Psychosom 2006;75:319e26. 25. Welschen LM, van Oppen P, Dekker JM, Bouter LM, Stalman WA, Nijpels G. The effectiveness of adding cognitive behavioural therapy aimed at changing lifestyle to managed diabetes care for patients with type 2 diabetes: design of a randomised controlled trial. BMC Public Health 2007;7:74. 26. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P. The Hawthorne effect: a randomised, controlled trial. BMC Med Res Methodol 2007;7:30. 27. Colquhoun DM, Moores D, Somerset SM, Humphries JA. Comparison of the effects on lipoproteins and apolipoproteins of a diet high in monounsaturated
598
28.
29.
30.
31. 32.
33. 34.
35. 36.
37.
38.
39.
40.
S.M. Somerset et al. / Clinical Nutrition 30 (2011) 593e598 fatty acids, enriched with avocado, and a high-carbohydrate diet. Am J Clin Nutr 1992;56:671e7. Colquhoun DM, Humphries JA, Moores D, Somerset SM. Effects of a macadamia nut enriched diet on serum lipids and lipoproteins compared to a low fat diet. Food Aust 1996;48:216e22. Colquhoun DM, Somerset S, Glasziou P, Richards D, Weyers J. comparison of an olive oil enriched diet to a low fat diet intervention study using vascular endpoints e assessed by repeat quantitative angiography (OLIVE study). Aust J Nutr Dietet 1998;55:S24e29. Shuaibi AM, Sevenhuysen GP, House JD. Validation of a food choice map with a 3-day food record and serum values to assess folate and vitamin B-12 intake in college-aged women. J Am Diet Assoc 2008;108:2041e50. Parnell WR, Wilson NC, Russell DG. Methodology of the 1997 New Zealand national nutrition survey. N Z Med J 2001;114:123e6. Elobeid MA, Padilla MA, McVie T, Thomas O, Brock DW, Musser B, et al. Missing data in randomized clinical trials for weight loss: scope of the problem, state of the field, and performance of statistical methods. PLoS One 2009;4:e6624. Mullner M, Matthews H, Altman DG. Reporting on statistical methods to adjust for confounding: a cross-sectional survey. Ann Intern Med 2002;136:122e6. Rios LP, Odueyungbo A, Moitri MO, Rahman MO, Thabane LCIN. Quality of reporting of randomized controlled trials in general endocrinology literature. J Clin Endocrin Metab 2008;93:3810e6. Somerset SM. Reporting of dietary control frameworks in studies on simvastatin efficacy. Clin Nutr 2010;29:352e7. Yancey AK, Miles OL, McCarthy WJ, Sandoval G, Hill J, Leslie JJ, et al. Differential response to targeted recruitment strategies to fitness promotion research by African-American women of varying body mass index. Ethn Dis 2001;11:115e23. Burroughs AR, Visscher WA, Haney TL, Efland JR, Barefoot JC, Williams Jr RB, Siegler IC. Community recruitment process by race, gender, and SES gradient: lessons learned from the Community Health and Stress Evaluation (CHASE) Study experience. J Community Health 2003;28:421e37. Garrett SK, Thomas AP, Cicuttini F, Silagy C, Taylor HR, McNeil JJ. Communitybased recruitment strategies for a longitudinal interventional study: the VECAT experience. J Clin Epidemiol 2000;53:541e8. Davis RM, Hitch AD, Nichols M, Rizvi A, Salaam M, Mayer-Davis EJ. A collaborative approach to the recruitment and retention of minority patients with diabetes in rural community health centers. Contemp Clin Trials 2009;30:63e70. Franz MJ, VanWormer JJ, Crain AL, Boucher JL, Histon T, Caplan W, et al. Weight-loss outcomes: a systematic review and meta-analysis of weight-loss
41.
42.
43. 44.
45.
46. 47. 48.
49.
50.
51.
52.
53. 54.
clinical trials with a minimum 1-year follow-up. J Am Diet Assoc 2007;107:1755e67. Davidson KW, Korin MR. Depression and cardiovascular disease: selected findings, controversies, and clinical implications from. Cleve Clin J Med 2009;77(Suppl. 3):S20e6. Holt RI, Phillips DI, Jameson KA, Cooper C, Dennison EM, Peveler RC. The relationship between depression and diabetes mellitus: findings from the Hertfordshire Cohort study. Diabet Med 2009;26:641e8. McAlpine DD, Wilson AR. Screening for depression in primary care: what do we still need to know? Depress Anxiety 2004;19:137e45. Mefferd K, Nichols JF, Pakiz B, Rock CL. A cognitive behavioral therapy intervention to promote weight loss improves body composition and blood lipid profiles among overweight breast cancer survivors. Breast Cancer Res Treat 2007;104:145e52. Rodriguez-Hernandez H, Morales-Amaya UA, Rosales-Valdez R, RiveraHinojosa F, Rodriguez-Moran M, Guerrero-Romero F. Adding cognitive behavioural treatment to either low-carbohydrate or low-fat diets: differential short-term effects. Br J Nutr 2009;102:1847e53. Crawford J, Henry J. Normative data and latent structure in a large non-clinical sample. Br J Clin Psychol 2003;42:111e31. Shay LE. A concept analysis: adherence and weight loss. Nurs Forum 2008;43:42e52. Williamson DA, Anton SD, Han H, Champagne CM, Allen R, LeBlanc E, et al. Adherence is a multi-dimensional construct in the POUNDS LOST trial. J Behav Med 2010;33:35e46. Boyette LW, Lloyd A, Boyette JE, Watkins E, Furbush L, Dunbar SB, et al. Personal characteristics that influence exercise behavior of older adults. J Rehabil Res Dev 2002;39:95e103. Jack K, McLean SM, Moffett JK, Gardiner E. Barriers to treatment adherence in physiotherapy outpatient clinics: a systematic review. Man Ther 2010 Jun;15:220e8. Turk MW, Yang K, Hravnak M, Sereika SM, Ewing LJ, Burke LE. Randomized clinical trials of weight loss maintenance: a review. J Cardiovasc Nurs 2009;24:58e80. Schneeweiss S, Patrick AR, Maclure M, Dormuth CR, Glynn RJ. Adherence to statin therapy under drug cost sharing in patients with and without acute myocardial infarction: a population-based natural experiment. Circulation 2007;115:2128e35. Kyrou I, Tsigos C. Stress hormones: physiological stress and regulation of metabolism. Curr Opin Pharmacol 2009;9:787e93. Cohen BE, Panguluri P, Na B, Whooley MA. Psychological risk factors and the metabolic syndrome in patients with coronary heart disease: findings from the Heart and Soul Study. Psychiatry Res Jan 30 2010;175:133e7.