Of the over 33 million individuals eligible for bariatric surgery in the United States, less than 1% of eligible patients yearly complete surgery [1, 2]. Utilization of this life-altering intervention is inadequate. Perhaps more alarming is that among patients who express interest in this therapy and engage with bariatric centers, the dropout rate of eligible surgical candidates has been reported as high as 60% [3,4,5]. There is a lack of understanding how the healthcare system can better support patients to obtain treatment of this serious chronic disease.

A number of factors for this high attrition rate have been proposed such as inadequate or lack of insurance coverage and the high burden of pre-surgical requirements. With regard to insurance coverage, studies using quantitative methods have shown that increased cost-sharing decreases bariatric surgery uptake [6, 7]. Gasoyan et al. found that insurance type (private insurance [8], preferred provider organization plans [9], and fee-for-services-plans [9]) increases bariatric surgery utilization as well [8]. However, health insurance coverage is unlikely to be significant enough as an independent reason for attrition [5, 10]. Even in countries with universal healthcare, the attrition rate is approximately 50% [11]. For pre-surgical requirements, both the number and type of pre-surgical requirements (e.g., requiring months of medically supervised weight loss [9], be accompanied by a support person during clinic visits [3], cardiology evaluation [12]) have been shown to be significantly associated with attrition from bariatric surgery programs. However, the nuances of how these proposed factors as well as other not yet identified factors work together to cause dropout remains unknown. Additionally, prior research focuses on insurance variables or comorbidity burden which are largely non-modifiable.

In order to identify strategies to reduce this attrition, we need a better understanding of the factors that contribute to patients’ decisions to drop out of the bariatric surgery process. In this context, we utilize a qualitative approach to understand the experiences of patients who started the process leading to bariatric surgery but chose not to complete surgery. We learn from patients themselves to inform nuanced, multifaceted, theory-informed strategies to promote completion of bariatric surgery.

Materials and methods

Study design

We identified patients from three clinical sites who had dropped out of bariatric surgery programs. These clinical sites represented a suburban community center, a suburban academic center, and an urban academic center. The study protocol was approved by the University of Michigan Institutional Review Board (HUMID#00171265). Verbal informed consent was obtained from all participants of the study prior to their interview. The study is reported according to the Consolidated Criteria for Reporting Qualitative Studies (COREQ) guidelines [13] (Supplement 1).

Study population

Participants were recruited using purposive sampling. We reached out via phone to participants who dropped out of the bariatric surgery programs from December 2017 to January 2019. Attrition was defined as persons who initiated the process of bariatric surgery (attending an initial evaluation visit in clinic) but did not complete bariatric surgery. Patients dropped out of the process anywhere along the continuum from after the initial evaluation up to scheduling a surgical date (e.g., after the initial evaluation, after participating for months, after scheduling a surgery date). We reached out to 31 patients by phone and met our goal of 20 participants (64.5% response rate). We excluded patients who did not meet NIH criteria for surgery or were deemed not to be a surgical candidate by the multidisciplinary board. Our final group of participants was diverse with respect to gender (60% Female) and age (35% < 45 years; 50% 45–65 years; 5% > 65 years; 10% No Answer). Of our participants, 85% self-identified as “White,” 10% as “Black,” and 5% as “Mixed Race.” With regard to self-reported BMI, 10% had a BMI of < 35 kg/m2, 15% of 35–40 kg/m2, 60% of 40 kg/m2 or greater, and 15% did not answer their height and weight.

The Theoretical Domains Framework and interview guide

We used the Theoretical Domains Framework to develop our semi-structured interview guide and then to analyze patient responses in a systematic manner to identify behavioral change techniques. The Theoretical Domains Framework is a robust implementation science framework that applies organizational theory towards behavioral change [14,15,16]. The framework integrates 35 theoretical models of human behavior change into 14 domains. These 14 domains are (1) knowledge; (2) skills; (3) social or professional role and identity; (4) beliefs about capabilities; (5) optimism; (6) beliefs about consequences; (7) reinforcement; (8) intentions; (9) goals; (10) memory, attention, and decision processes; (11) environmental context; (12) social influences; (13) emotion; and (14) behavioral regulation [16]. Each of these domains includes constructs which in turn map to specific behavioral change techniques (Fig. 1). The TDF has been used extensively in healthcare to identify behavioral change practices to increase physical activity [17, 18], to improve smoking cessation rates [19], to increase cohesion between provider practice and guidelines [14, 20, 21], and to minimize trainee prescribing errors [22]. We chose this framework because this is an individual-facing framework that links to behavior change techniques. We use the consensus guidelines proposed by Atkins et al. in applying the Theoretical Domains Framework [23].

Fig. 1
figure 1

Behavior Wheel [21]

Full size image

Data collection

Interviews were conducted May 2020–March 2021 via phone due to COVID-19 social distancing measures. Reasons for non-participation were not elicited. Of the 31 patients we reached by phone, 20 patients (65%) agreed to participate. Independent interviews were conducted by 5 authors (G.F.C., K.L., L.D., Y.A., and S.M.J.), two surgical health services researchers, two health services research assistants, and a medical student. Interviews began with the question, “What first brought you to be interested in bariatric surgery?” The participants then guided the rest of the conversation. The interview guide was designed to probe further into topics the participants brought up themselves. All participants were also asked “Can you tell me about the process of getting ready for bariatric surgery?”, “Was there something about the bariatric surgery process that confused or frustrated you?”, and “What would convince you to get bariatric surgery in the future?” Demographic information was asked at the conclusion of the interview. The full interview guide with cross-walked linking questions is available in Appendix Table 6.

Interviews lasted 20 to 50 min and were digitally recorded, transcribed verbatim, and de-identified. Transcripts were not returned to participants for review. Field notes, observations about each interview, were included in analysis. One participant did not consent to be recorded but did allow for field notes from the interview to be used for the study.

Data analysis

Coding was approached through an iterative, inductive process. Three members of the team (G.F.C., K.L., and L.D.) independently reviewed transcripts to identify an initial set of codes. The codes were further refined with the senior author, an implementation science expert (D.A.Te.). Next, three transcripts were independently coded by two of the team members (G.F.C. and K.L.) blinded to each other’s work. The coders met iteratively to discuss discrepancies and refine the codebook. The rest of the interviews were coded by one of the two coders. Once all data were coded, the entire research team met to discuss patterns and “issues around which codes cluster,” [24] reach consensus, and map codes to TDF domains. To increase validity, we selected a research team representing diverse professional fields (anthropology, implementation science, nursing, public health, and surgery). All transcribed interviews were coded using NVivo 12.6.1 (QSR International, 2020), a computer-assisted qualitative data analysis software.

Results

Of study participants, 40% identified as men and 60% as women with an average age of 46.7 years (SD 10.7) and average BMI of 44.8 kg/m2 (SD 8.7). Participants in our study were predominantly non-Hispanic White (85%). Results for this study focus on three codes: “perceptions of bariatric surgery,” “reasons for not undergoing surgery,” and “factors for re-considering surgery.” Fourteen of the twenty participants stated they would re-consider surgery sometime in the future. There are still 18 other codes to review for analysis in future studies.

Key themes that emerged were (1) burden of pre-operative workup, (2) stigma against bariatric surgery, (3) fear of surgery, and (4) anticipated regret which organized into four TDF domains: environmental context and resources, social role and identity, emotion, and beliefs about consequences. Exemplar quotations are included below and in tables.

Burden of pre-operative workup: environmental context and resources

Participants cited the long pre-operative workup time and number of requirements as a major barrier to their completing bariatric surgery, mapping to the TDF domain of Environmental Context and Resources (Table 1). Some even believed their insurance company purposely prolonged the process to discourage them from getting bariatric surgery. One participant shared that she was so disappointed after being told she had to wait three months after quitting smoking and then a few more weeks between other appointments that:

I even considered the whole Mexico thing too…I can quit smoking for two weeks and, you know, get the surgery, and you have the surgery. You can’t smoke again after the surgery, so I don’t understand why the waiting. It just like brings you that more time to fail, you know what I mean? (Participant 11, White Female)

Table 1 Burden of pre-operative workup
Full size table

This participant was even willing to consider going to another country for surgery. Her response was emblematic of the discouragement many participants felt over the long wait times. While two participants cited inability to take off work as a reason why the visits were burdensome, overall participants described the burden as the number of and the time between pre-operative workup appointments and tests.

The longer the pre-operative wait time, the greater the loss of optimism for patients as well. When asked why participants decided to initiate the bariatric surgery process, almost all cited specific health concerns (e.g. diabetes, cardiovascular health) and a desire to lose weight after failures of non-operative methods. However, not all participants decided to proceed past the initial information session or one or two appointments. We found that participants who decided to proceed further cited a sense of hope that they could do something to improve their life. These participants referred to bariatric surgery as a “light at the end of the tunnel” (Participant 7, Black Female) or “some saving thing for my life” (Participant 13, White Male). Many expressed optimism that bariatric surgery would lead to personal physical health benefits of decreased obesity-related conditions as well as the emotional and social benefits of bariatric surgery such as being able to travel more easily or participate in social events. This optimism that initially propelled patients to attend subsequent appointments and evaluations was diminished by the long wait times.

There were certain aspects of the pre-operative workup that participants found informative and that sustained their interest in pursuing bariatric surgery. As one participant shared, “Well, one of the positive things was the openness of everything. There was nothing that they were trying to hide like in regards to like the surgery and the after-effects” (Participant 14, White Female). Most participants described the initial information session and group classes as beneficial to their understanding of their health and spoke positively about them.

Stigma against bariatric surgery: social role and identity

With regard to the TDF domain of Social Role and Identity, participants in this waiting time began to think more about what it would mean to be a person who had undergone bariatric surgery which was guided by stigma against bariatric surgery (Table 2). They began to focus on how to undergo bariatric surgery would be to admit loss of control over their life. Participants’ view of bariatric surgery changed from perceptions of it mostly as a helpful tool to achieve health to something people who were weak used as “a crutch” (Participant 13, White Male) rather than taking control of their own lives. Thus, many stated that they dropped out of the program because they would rather try to lose weight through diet and exercise which they viewed as activities more under their personal control and responsibility compared to bariatric surgery.

Table 2 Stigma of bariatric surgery
Full size table

Fear of surgery: emotion

Individuals discussed how their fear of surgical procedures which they had prior to starting the bariatric surgery process contributed to their decision not to pursue surgery, mapping to the TDF domain of Emotion (Table 3). Surgery was seen as invasive, potentially dangerous, and irreversible. One participant shared: “I want to place the emphasis on drastic because I feel like the procedure itself is, is a very drastic procedure…hey, like instead of trying to work at this on my own, I’m going to give up and sacrifice my body in the process.” (Participant 7, Black Female). Bariatric surgery was seen putting one’s body in danger.

Table 3 Fear of surgery
Full size table

Anticipated regret: beliefs about consequences

Lastly, participants stated they were afraid they would regret undergoing bariatric surgery, contributing to their decision to drop out of the process (Table 4). Because the pre-operative process, surgery itself, and post-operative recovery required substantial work to be put in by patients, many worried that they would regret undergoing bariatric surgery and this work would have been done in vain. Participants described this anticipated regret as related to three factors: possible weight regain, further physical disfigurement, and possible worsening of health.

Table 4 Anticipated Regret
Full size table

The first area of anticipated regret was undergoing weight loss and then regaining weight afterwards. The main marker of success after bariatric surgery for participants was weight loss. Thus, some participants felt that it would be disappointing if they had put in all of the work to go through with bariatric surgery and then regained the weight. For some participants, this was a theoretical concern while others cited individuals they met at group sessions or family members, friends, or acquaintances who had undergone bariatric surgery and then regained weight.

The second area of anticipated regret was future physical disfigurement. Some physical changes with bariatric surgery, both those that were visible outwardly and those that were inside the body, were considered undesirable for participants. One participant noted:

Is it worse to look obese, or worse to look like you were once obese and have all this skin hanging around. I mean, obviously skin hanging around, to me, seems more disgusting. (Participant 1, White Male)

As he described, some physical changes that might occur after bariatric surgery like loose skin further discouraged participants from undergoing bariatric surgery. Internal disfigurement was also a concern for some participants. One participant even described this as having a “Frankenstein-ish kind of body once you’re done…going under a scalpel and getting your body kind of chopped up to be a certain way” (Participant 7, Black Female). These images of a disfigured future self were a major barrier for some patients to undergo surgery.

Finally, some participants were concerned about complications with their physical health after bariatric surgery. This included anticipated regret getting bariatric surgery regarding general post-surgical concerns, dietary concerns, and others. Dietary concerns included needing to rely on supplements or liquids for the rest of their lives or gastrointestinal symptoms such as dumping syndrome. Other concerns included too many disruptions to metabolism and needing to postpone childbearing.

Factors for re-considering surgery

There were fourteen participants who stated that they would consider surgery in the future. Overall, half of participants cited worsening of weight gain or other obesity-related comorbidity as a potential driver to reconsider surgery. All participants’ reasons for reconsidering surgery are summarized in Table 5.

Table 5 Reasons to Reconsider Bariatric Surgery
Full size table

Discussion

Our study is the first to use a qualitative analysis method to understand the nuances in why patients decided to drop out of bariatric surgery programs. Because of the length of time and the number of requirements prior to surgery, patients became discouraged from continuing the process. The burden of pre-operative requirements led patients to lose their initial optimism about bariatric surgery improving their health. Their perceptions regarding being seen as weak for choosing bariatric surgery, fear of surgery itself, and possible regret over having had surgery also grew as more time passed between their initial evaluation and a future surgical date. By mapping these themes onto the Theoretical Domains Framework, we can begin to identify theory-based interventions.

Our data add to the existing literature on pre-operative attrition from bariatric surgery programs by illuminating the most important reasons for dropping out of bariatric surgery programs according to patients themselves. In a study by Sadhasivam et al.where patients’ charts were reviewed, approximately 20% of eligible patients’ assigned reason for not completing surgery was “patient decision.” [5] Our study is the first to begin to characterize the factors most important to patient decisionmaking. Our findings also align with prior quantitative work by Gasoyan, Alvarez, and Love et al.which found that certain pre-operative requirements were significantly associated with dropout from bariatric surgery programs, especially months of medically supervised weight loss [9], additional psychological evaluations [3], longer diet requirement [12], and advanced laboratory testing [12].

By using the Theoretical Domains Framework, we can map these important determinants of attrition from bariatric surgery from patients in our study to theory-based interventions [25]. We can use this framework to identify sources of behavior and intervention functions to target in the future. Given the importance of shorter wait times to completion of surgery and that the median wait time from 2006 to 2016 increased from 86 to 159 days [26], reducing wait times should be a priority for future interventions to decrease attrition. One way we can reduce pre-operative requirements is through advocating for reduced pre-operative requirements. Other proposed pragmatic strategies to decrease the time to surgery involve restructuring how care is delivered. For example, telehealth uptake due to the COVID-19 pandemic increased significantly among surgical practices [27] and could be leveraged to minimize the burden of pre-operative visits for patients who choose this option. Practices of different specialties can also coordinate visits and laboratory testing for bariatric surgical patients to be on the same day or within a few days of one another.

Patients’ anticipated regret warrants special consideration. While we understand the legitimacy of concerns about body and life after bariatric surgery that contributed to patients deciding not to pursue surgery, we were struck by how patients connected these concerns affecting their decision to the lengthy time from their initial enthusiasm for bariatric surgery. Thus, we believe our findings represent the burden of pre-operative requirements working against patients’ desire to undergo bariatric surgery rather than these concerns being of major concern at the beginning of their workup process. Moving forward, interventions will need to distinguish between whether patient concerns are due to long waiting time or represent fundamental beliefs that would lead to decisional regret after bariatric surgery. Prior work by Wee et al.have shown that individuals who underwent Roux-en-Y gastric bypass have low rates of decisional regret ranging from 2.2% at post-operative year 1 to 5.1% at post-operative year 4 [28], but decisional regret in future intervention studies will be vital.

Our study has several limitations. First, the majority of our patients self-identified as White. Thus, it is unknown whether these findings are equally relevant to our Black, Hispanic, and Asian American Pacific Islander patients. However, a strength in the diversity of our sample is that while most studies in bariatric surgery include predominantly women, ours shares the voices of a significant number of men. Additionally, there remains a lack of formal guidance on how to apply the TDF in healthcare settings [23]. Thus, research with and interventions designed from the TDF requires expert guidance from an implementation science expert. Lastly, all of our participants were recruited from Southeastern Michigan. Their experiences with Michigan bariatric surgery practices may not be generalizable to other settings.

This study addresses the barriers and facilitators for patients being able to achieve their goal of bariatric surgery. Using the Theoretical Domains Framework, we identified areas of greatest concern for patients that can be used for future intervention design. This is the first step in understanding how we best support our patients who express interest in bariatric surgery achieve their goals and live healthier lives.