Marek, Heinberg, Lavery, Rish, and Ashton (2016) offer a thorough review of psychological assessment instruments for bariatric surgery patients. Through their literature review, they highlight the association of pre-surgical psychological factors with weight gain post-op and recurrence of behavioral problems. Additionally, they note that bariatric patients have a higher prevalence for psychological disorders compared to the general population (Kalarchian et al., 2007; Mitchell, Selzer, et al., 2012, as cited in as cited in Marek, Heinberg, Lavery, Rish, & Ashtom, 2016). Thus, and consistent with their review, the National Institute of Health (NIH) has recommend psychological assessment for bariatric surgical candidates. The authors highlight the goals for such an evaluation are to “identify and treat preexisting psychopathology,” “identify patients who may need additional postoperative care,” and to “identify alternative treatment strategies” if a patient is deemed not appropriate for a selected procedure (Block & Sarwer, as cited in Marek et al., 2016, p. 1143).
The authors review the domains of a semi-structured interview for the assessment of bariatric surgery candidates and provide references for the clinical interview (see references for information on clinical interviews). They indicate that though many practitioners use common broadband assessments (i.e., MMPI-2 or BID-II), the instruments used tend to vary and often lack sound psychometric properties for use with this population. In general, they recommend that the psychological domains of internalizing psychopathology, eating-related behaviors, externalizing psychopathology, and thought disorder or poor cognitive functioning be assessed. The authors indicate that depression and anxiety are prevalent among this population, and further note that antidepressants may be inadequately absorbed after surgery (Roerig et al., 2012; as cited in Marek et al., 2016). If left untreated, alcohol and substance use are contraindicated with this surgical procedure. Marek et al. (2016) state that “pharmacokinetic changes following some bariatric surgery procedures further accelerate alcohol absorption, making postsurgical risk of alcohol misuse problematic” (p. 1144). Further, continued marijuana use may impact eating habits; while the effects of other substances are currently unknown. Finally, an unstable or untreated thought disorder is considered contraindicated for bariatric procedures. Here, concerns of weight gain from psychiatric medications, side effects from anesthesia (e.g., delirium), adherence and understanding of the procedure and aftercare, and deficits in neurocognitive domains are considered to be significant factors that could lead to problems post-surgery.
The authors offer a thorough review of several common assessment instruments used in bariatric surgery evaluations. For the broadband instruments, the Minnesota Multiphasic Personality Inventory, Second Edition (MMPI-2) is the most widely used, and the Minnesota Multiphasic Personality Inventory, Second Edition-Restructured Form (MMPI-2-RF) also shows good reliability, validity, and predictive utility. Marek et al. (2016) reported preference for the MMPI-2-RF with this population. The authors highlight concerns with MMPI-2 profile elevations related to underreporting as this response style may not only suppress clinical scales, but also may indicate an underreporting on other self-report measures. High scores on the hysteria, masculinity/femininity, and paranoia scales, along with elevations of Health Concerns and the Infrequency scale differentiated patients who lost less than 50 percent of their weight. The Personality Assessment Inventory (PAI) is less commonly used, but is suggested as a viable option. The Symptom Item Checklist-90-Revised (SCL-90-R) lacks validity scales and assessment of externalizing psychopathology; yet based on past research, bariatric patients that score higher on depression, anxiety, and hostility scales were more likely to be delayed for surgery. The Million Behavioral Medication Diagnostic (MBMD) has bariatric normative data and report options, yet there is limited psychometric data published. Research with the Millon Clinical Multiaxial Inventory-II (MCMI-II) suggested patients with elevation on scales of schizoid, schizotypal, and compulsiveness had less weight loss 6 months post-surgery. However, the authors note that the MBMD and MCMI-II lack adequate research supporting the use with this population. For the Basic Personality Inventory, low alienation scores were associated with successful weight loss.
Narrowband instruments can function as a supplement to gauge eating disorder behavior or other specific domains of concern. For depression and anxiety, the Beck Depression Inventory, Second Edition (BDI-II) is suggested to be an adequate screening measure, and per research findings (Hayden et al., 2012, as cited in Marek et al., 2016) a cutoff score of 13 should be utilized. The authors suggest additional discriminant validity is needed for use of the BDI-II with this population. The Patient Health Questinnaire-9 (PHQ-9) is also a useful screening tool and is a strong choice with his population per Marek et al. (2016). The authors suggest a recommended cutoff of 15 to indicate further screening for depression. The Mood Disorders Questionnaire (MDQ) has good sensitivity for assessing bipolar spectrum symptoms, with a recommended cutoff of less than 7. The Beck Anxiety Inventory has good reliability and validity for use with this population. The Center for Epidemiologic Studies Depression Scale and the Generalized Anxiety Disorder-7 lack psychometric data for use with bariatric assessments.
For substance abuse screening, the Alcohol Use Disorders Identification Test (AUDIT) has good sensitivity and specificity for use with bariatric populations. The Michigan Alcoholism Screening Test is useful but may be more so reflective of lifetime use rather than more recent drinking patterns; furthermore, psychometric properties have not been reported with bariatric samples. The Substance Abuse Subtle Screening Inventory-3 has shown to have low sensitivity in identifying alcohol dependence in some populations. There also is reportedly no data relative to bariatric samples.
Instruments to assess eating behaviors are useful in identifying persistent eating disorder pathology, which may contributed to less successful weight loss post-surgery. The authors recommend that an evaluation of eating behaviors be included in bariatric assessments, as well as the need to confirm reported eating behavior through a clinical interview. The Eating Disorders Examination Questionnaire is commonly used and has strong internal consistency and validity. The Questionnaire of Eating and Weight Patterns-Revised is the most commonly utilized measure in the literature. It assesses behavioral aspects of disordered eating, as well as weight history and body image. The Three-Factor Eating Questionnaire assesses restraint, hunger, and disinhibition. This instrument is used frequently and has shown to be able to distinguish between binge eating and non-binge eating. The Binge Eating Scale is also commonly used and is able to distinguish between minimal, moderate, and severe binge eating problems; however, this instrument should be used with caution due to a tendency to over diagnose. The Eating Disorders Inventory-III has been validated with obese populations, but not with bariatric populations. The 11 subscales provide assessment for drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distrust, introceptive awareness, maturity fears, asceticism, impulse regulation, and social insecurity. The Night Eating Questionnaire assesses severity of nocturnal ingestion, evening hyperphagia, morning anorexic, and mood/sleep problems. It has also been validated with weight loss surgery candidates and the authors recommend this as a component of the assessment, though more psychometric development is needed. To assess for loss of control related to binge eating, the Loss of Control Eating Scale has shown good psychometric properties. The concept of loss of control is noted to be predictive of psychopathology and distress rather than the amount of food consumed.
Overall, a broadband assessment appears necessary to assess and rule out existing psychopathology that either is contraindicated with weight loss surgery or to target treatment in order to maximize benefits post-surgery. Furthermore, though a clinical interview is necessary to diagnose disorders, the use of screening measures to support diagnoses or to suggest areas of intervention is recommended. Interestingly despite recommendation by the NIH for pre-surgical evaluation, only about two-thirds of bariatric surgery clinics reportedly adhere to this recommendation (Marek et al., 2016). The use of psychological testing helps provide normative data and additional evidence to support a diagnosis, aid in treatment planning, and assess behavioral tendencies (eating patterns, substance use). The use of a broadband measure that assesses response styles is also necessary. Interpretation of response styles can help guide decision making and diagnosis. Marek et al. (2016) further indicated that a portion of bariatric surgery patients minimize psychopathology, specifically impulse-control and sensation-seeking. Overall, Marek et al. (2016) suggest the assessment of eating, mood, and substance use is the foundation for bariatric assessments, with the overall goal to enhance the evaluation in order to inform treatment and decision making to best assist patients.
Finally, for additional information on templates for a structured clinical interview and recommendations to include in the interview, see the references below:
Sogg, S., & Mori, D. L (2004). The Boston Interview for Gastric Bypass: Determining the psychological suitability of surgical candidates. Obesity Surgery, 14, 370-380.
Sogg, S., & Mori, D. L. (2009). Psychosocial evaluation for bariatric surgery: The Boston Interview and opportunities for intervention. Obesity Surgery, 19, 369-377.
Dannie S. Harris, MA
WKPIC Doctoral Intern
