Presurgical assessment of bariatric patients with the Patient Health Questionnaire (PHQ)—A screening of the prevalence of psychosocial comorbidity

Bariatric surgery has been on the rise over the past two decades. Increasing prevalence of obesity and the development of laparoscopic surgery have contributed to this increase, but above all bariatric surgery has proved to be the most effective therapy for morbid obesity. Despite an overall improvement, long-term postsurgical weight loss is fraught by some setbacks as approximately 20 to 30 % of patients regain weight following surgery [16, 22]. To some portion this is due to the lack of changes in lifestyle and eating behaviours, which not only comprise surgical outcome by weight regain and malnutrition, but also the deterioration of pre-existing mental health disorders have being reported [16, 23]. Surgical outcome can be impaired both by physiological postoperative changes and psychosocial factors [12, 22, 24]. Therefore, presurgical psychological evaluation of bariatric surgery candidates is recommended by the guidelines and frequently demanded by insurance companies [1, 25, 26]. However, no standards of how best to assess bariatric surgery candidates have currently been defined [27, 28]. In this study, we present our results on screening for mental health disorders of morbidly obese individuals seeking bariatric treatment using the PHQ. Originally designed to screen for eight different mental health disorders in primary health care, the PHQ has been validated in numerous studies and is a well-accepted test instrument [17, 18, 21]. It allows assignment of patients to one of four subgroups reflecting the severity of mental health disorders. Assessing the severity of symptoms has been shown to impact stronger than the specificity of diagnoses [22].

Employing the PHQ in a computerized version, almost 85 % of the participants were screened positive for a least one psychiatric syndrome, which is much in keeping with data reported in existing literature for the lifetime history of major mental disorders in the morbidly obese [22, 29, 30]. More than 50 % of the tested candidates exhibited three or more mental health disorders, which underlines the need for a differentiated psychological evaluation. In particular, obese patients seeking bariatric surgery are known to have markedly higher prevalence rates of psychopathologies than normal weight individuals [16]. Owing to the bio-psycho-social model proposed by Engel [31], factors triggering morbid obesity may be associated to biological, psychological, or social factors. There is ample reason to assume the apparent physical symptom “obesity” to result from a hazardously intertwined interaction between these factors. The hazardous nature of this interaction is also evident since the correction of psychic or social problems mostly fails to curtail the exacerbation of obesity once weight has climbed beyond 40 kg/m² [22]. The PHQ may qualify as a test tool since it not only screens for mental health disorders but also includes physical symptoms pertaining to mental health disorders such as pain. Nearly 60 % of the participants in this study exhibited a high somatic symptom burden, with pain being the most frequently presented complaint. It is likely that the high frequency of somatic complaints can be attributed to the physical consequences of morbid obesity, e.g. musculoskeletal pain or shortness of breath. Increased prevalence rates of somatoform disorders in overweight and obese individuals were also reported in various other studies [32, 33].

Anxiety disorders and depressive syndromes were the second most common syndromes found in our study population. A positive association between obesity and anxiety disorders was also described by some other authors [34, 35]. Obese individuals have a life-time prevalence of depression ranging from 29 to 51 % [36‐38] and obesity is associated with a higher risk of possible depression [39]. In a population-based study of 40,000 US adults, the risk of major depression in persons with obesity grade III was nearly five times that in normal weight individuals [40]. Furthermore, patients who remain depressed following bariatric surgery, tend to report poorer postsurgical quality of life and weight loss [41]. Andersen and colleagues found that following bariatric surgery, anxiety and depression were reduced and that these changes were closely associated with improvements in self-reported physical health [42].

Numerous studies have focused on the eating behaviour of individuals seeking bariatric surgery and the most common and likely relevant condition is binge eating disorder [5]. Twenty percent of our study sample exhibited a binge eating disorder, which is in the range of prevalences reported for bariatric surgery candidates [38, 43]. Binge eating could have detrimental effects on surgical outcome as it is associated with greater weight regain [44, 45]. Therefore, some authors recommend binge eating as a marker of poorer outcome and an appropriate target for post-surgical intervention [45].

An important scale of the PHQ screens for alcohol syndrome which is a contraindication for bariatric procedures due to increased sensitivity to alcohol postoperatively. Furthermore, alcohol consumption may compromise weight loss following bariatric surgery [1, 46]. Kudsi and colleagues systematically assessed 653 patients seeking weight-loss surgery and reported alcohol abuse in 4 % of their candidates which corresponds to the results in our PHQ assessed patient cohort [46]. Other authors found alcohol abuse or alcohol dependency in more than 30 % of bariatric surgery candidates [38].

The male to female ratio in our cohort was 1:1.9, which is also a standard phenomenon in studies on bariatric surgery candidates [30, 47]. Males more seldom tend to request surgical treatment for their obesity, while females tend to more regularly seek effective methods for treating their disease [47, 48]. Further, the median number of PHQ syndromes was significantly higher in women than in men. In particular, a higher somatic symptom burden and anxiety disorders were found significantly more often in women than in men. These results are in line with other studies, which found more psychopathologies among women [47, 49]. Gender effects on motivation for bariatric surgery and psychological co-morbidity therefore need to be considered in presurgical evaluation of bariatric candidates.

It is striking that 39 % of the participants were unemployed, which is considerably higher than the unemployment rate in the German population in general (8.6 % in 2010; source: German Federal Statistical Office). Different explanations of why obese individuals display a higher unemployment rate compared to non-obese counterparts are given in the literature. Firstly, obesity is a severely debilitating heath condition that negatively impacts productivity and employment and is a risk factor for a wide number of diseases [1, 50]. In our cohort, a high somatic symptom burden, depressive disorders and panic disorders were significantly more often among the unemployed. Additionally, obese individuals may have lower self-esteem and may experience discrimination at work [50, 51]. Furthermore, obesity may accelerate the departure from the labour market through early retirement [52]. In our study, 14.5 % of the participants were early retirees.

The PHQ was applied in a specially developed computer-based version. Completion of the electronic version took on average 8.5 minutes, which is somewhat shorter than completion time reported for the paper-pen version (approx. 10 minutes) [18]. Computerized tests are known to offer advantages over pencil-and-paper based assessment such as rapid completion time, high acceptance rates and immediate output of results [53‐55]. The electronic documentation of the output files might mitigate or minimize diagnostic errors [56]. As skipping of a displayed question was impossible for the participant, all questions were answered and no values were missing. This is a relevant advantage of the computerized tests compared to paper-pen versions. Furthermore, computer-based tests facilitate repetitive monitoring and detection of change processes.

There are some limitations of the present study that need to be addressed. As the study population was recruited from patient information meetings, it is subject to selection bias and may therefore not be representative of bariatric surgery candidates on the whole. Approximately 63 % of the individuals who attended the information meeting about bariatric surgery decided to participate in the testing; the others did not participate predominantly because they were not interested in bariatric surgery. These individuals were not assessed surgically or for mental health disorders and we are left with speculation on the health of these individuals. As all patients who underwent bariatric surgery in the further course performed the PHQ testing, the study results provide a comprehensive picture of prebariatric patients.

Psychiatric syndromes obtained from the PHQ relied on patient self-report and participants might answer with the intention not to jeopardise bariatric surgery. Furthermore, the PHQ does not screen for personality disorders, which may contribute to the bariatric outcome (e.g. borderline disorders) [11, 57]. Finally, the PHQ is so far not validated in morbidly obese patients and somatic complaints might have biased reports on somatoform or affective disorders. To overcome these shortcomings, all patients underwent a semi-structured interview administered by a psychosomatic specialist (co-author VP) experienced in the evaluation of bariatric patients prior to surgery. If a participant was diagnosed positive for one or more syndromes, the psychosomatic specialist determined if the severity of a mental health disorder was a contraindication for bariatric surgery. Then the most appropriate therapy plan was discussed, which included nutrition counselling, outpatient or in-patient psychotherapy.