Exploring the relationship between obesity and depression and weight-loss and depression in adults with obesity

The increasing prevalence of both obesity and depression is a significant public health issue. Worldwide, 10% of men and 14% of women are categorised as obese (Body Mass Index > 30). Similarly, worldwide prevalence rates of depression range between 10-15%. Both conditions are associated with an increased risk of serious chronic disease, psychological comorbidities and functional impairment; moreover, these health risks are likely exacerbated when these two conditions are comorbid. It is therefore concerning that prevalence rates of depression are elevated in populations with obesity when compared to non-obese populations. In light of the increased prevalence of depression in populations with obesity, and the associated health consequences, a comprehensive understanding of the relationship between obesity and depression is a critical step towards developing appropriate prevention and intervention strategies. While an empirically tested and accepted model of the relationship between obesity and depression would be ideal, the field of research has yet to progress to this point. Recent studies have consistently demonstrated a bi-directional relationship between obesity and depression. Research that identifies bio-psychosocial variables associated with the relationship between obesity and depression has been less consistent and published reviews on the subject have not used a systematic or replicable methodology to review the existing literature. To address this gap, and review the literature to date, a systematic review entitled, “A systematic review of variables associated with the relationship between obesity and depression” was conducted. Forty-six studies met the inclusion criteria for review and the following variables were identified as being consistently associated with the relationship between obesity and depression: severity of obesity, educational attainment, body image dissatisfaction, self-reported binge-eating behaviours, perceived physical health, perceived experience of weight-based stigmatisation, psychological characteristics and interpersonal effectiveness. While the review identified some consistent results, it was noted that a number of conclusions were tentative as the research reviewed was inconsistent in key aspects of its design and analyses. Given these variations it was suggested that future research should adopt the following recommendations: study a sample of people with obesity, rather than compare a sample with obesity with a non-obese sample; assess depression as a key construct with reliable and valid measures; apply multivariate statistical methods to assess multiple variables concurrently; and, use longitudinal research designs to identify mediators and moderators of the relationship. Adopting the first three recommendations listed above, the first empirical paper presented in this thesis tested whether selected bio-psychosocial variables were risk factors for depressive symptoms and/or depressive disorders in a sample of surgery-seeking adults with obesity. A battery of questionnaires and a structured clinician led interview was administered to 145 adults with obesity, aged between 18-65 years, seeking bariatric surgery. Depressive symptoms were measured using Beck’s Depression Inventory (II) and a diagnosis of a depressive disorder was obtained via the administration of the mood module of the Structured Clinical Interview for DSM-IV. Results indicated that gender, self-esteem, body image dissatisfaction, eating concern, perceived physical health and interpersonal relationships confer risk of elevated depressive symptoms in people with obesity, while self-esteem confers risk of a diagnosable depressive disorder. These results provide important information about risk factors for depressive symptoms and a depressive disorder in a population with obesity. Further longitudinal research is required to identify moderators and mediators of the relationship. Using a longitudinal design, the second empirical paper aimed to identify psychosocial predictors of improvement in depressive symptoms following weight-loss in a sample of adults with obesity who underwent gastric-band surgery. These variables were also tested as moderator and mediators of the relationship between weight-loss and depressive symptom in this same sample. Prior to undergoing gastric-band surgery, 99 adults with obesity aged between 18-65 years completed a battery of self-report questionnaires assessing depressive symptoms (BDI-II), perceived physical health, body image dissatisfaction, experience of weight-based stigmatisation and self-esteem. Participants also completed a short questionnaire assessing depressive symptoms (BDI-II) and the psychosocial variables measured at baseline at monthly intervals commencing one month post-surgery. Depressive symptoms and all psychosocial variables reduced rapidly and significantly between baseline (pre-surgery) and one month post-surgery and these improvements were maintained at 6 months follow-up. Weight-loss was significantly associated with improvements in depressive symptoms. A series of hierarchical regressions indicated that body image dissatisfaction and self-esteem significantly contributed to change in depressive symptoms at one and six-months post-surgery after controlling for weight loss. Additionally, moderation analyses indicated that weight-loss was predictive of change in depressive symptoms from baseline to six-months post-surgery for participants with low baseline body image dissatisfaction. No other significant moderating or mediating relationships were found. In sum, the findings of the current thesis suggest that the following variables confer risk of increased depressive symptoms in people with obesity: gender, self-esteem, body image dissatisfaction, eating concern, perceived physical health, self-reported binge-eating behaviours and interpersonal relationships. In a clinical setting, this information may help identify people with obesity at risk of comorbid depression. Additionally, the current findings indicate that in a population with obesity, body image dissatisfaction and self-esteem explain some of the improvement in depressive symptoms following a surgical weight-loss intervention after controlling for weight-loss. Moreover, it seems likely that body image dissatisfaction is a moderator of the relationship between weight-loss and depressive symptoms. Consequently, therapeutic interventions for people with obesity at risk of elevated depressive symptoms may benefit from a focus on improving body image dissatisfaction and self-esteem, which may also be useful prior to the implementation of weight-loss interventions, in order to maximise reductions in depressive symptoms. While more longitudinal research is required to identify and test additional causal relationships, the current findings advance the research field towards a model of the relationship between obesity and depression with the aim of improving prevention and treatment of these worldwide public health issues