Introduction

Obesity is a significant public health concern leading to an epidemic [1, 2]. Bariatric surgery has been suggested to provide long-term weight loss and improve obesity-related diseases, with optimal weight loss usually occurring 18 to 24 months after surgery [3, 4]. In order to achieve the best result of bariatric surgery, guidelines present some recommendations, including increasing the patient’s knowledge about surgery performance through training sessions, at least 30 min of daily physical activity, psychological evaluation, and nutritional and sports medicine consultant [5]. Therefore, postoperative follow-up and multidisciplinary evaluations are critical in patients who have undergone obesity and metabolic surgery [1, 4,5,6,7,8]. The typical health programs suggest four to six medical, psychological, and nutritional counseling visits during the first year after surgery [1, 9].

Multidisciplinary follow-up care can play an essential role in preventing adverse outcomes through early detection and treatment of medical complications [1, 4, 7, 10,11,12,13,14,15]. Besides the advantages of attending to postoperative follow-up visits, the adherence of some patients is not sufficiently high.

The attrition rates vary considerably from 3 to 63% according to the patients or operative procedure characteristics and the type of specialized examinations from the multidisciplinary program [4, 11, 15,16,17,18,19,20,21]. The literature shows that adherence to postoperative medical follow-up is related to nature, attributes, and the result of the operation [4, 6, 7, 15], gender [4, 17], age [4, 7, 11, 22], baseline body mass index (BMI) [4], marital status, occupation [10], health insurance coverage, and distance to the follow-up center [11]. Identifying significant factors could help health program developers adjust their strategies for the patient’s benefit. However, there are contradictory findings from different studies about the role of each of these factors [17, 18]. We have, therefore, explored the effect of the patient’s baseline status, demographical factors, comorbidities, type of surgery, and its attributes and complications on the adherence to the follow-up program in the first year and beyond the first year after surgery.

Methods

Study Design

We used the data from the Iranian National Obesity Surgery cohort database [23, 24]. In a retrospective cohort study design, we selected patients who had undergone bariatric surgery performed by a single surgeon from June 2009 to March 2019 and had at least 1 year of follow-up. Other inclusion criteria were age above 18 years old at the time of surgery, and those who underwent sleeve gastrectomy (SG), one-anastomosis gastric bypass (OAGB), or Roux-en-Y gastric bypass (RYGB) as the most common types of bariatric surgery. The last eligible follow-up time was 19 February 2020, based on the first COVID-19 case reported in Iran to avoid any interference from the changing adherence pattern following the COVID-19 epidemic.

According to the cohort protocol, patients should attend clinic visits at 10 days and 1, 3, 6, 9, 12, 18, and 24 months after their surgery. The visits will continue at 1-year intervals for up to 10 years. All patients were reminded of their visit 2 weeks before the scheduled visit via text messages. In case of non-attendance at any predefined session, the follow-up team called the patient, provided the necessary advice, and rescheduled their visit if possible. All the predefined follow-up visits after operative procedures, while the patients were alive, were regarded as eligible. Since patients did not attend precisely on predefined dates, delays up to 20% of the calculated time for each visit were allowed.

For patients and the follow-up team, the importance of the first year after surgery differed from the other follow-up period. As a result, non-compliance with postoperative visits was investigated in two periods: visits in the first year after surgery and visits in the second year and beyond until 10 years. We assumed that the contribution of each predefined session to the overall attendance at eligible sessions is not equal. We weighted each predefined session by the proportion of attendance of all cohort members scheduled for that session. We then calculated the proportion of attendance for each individual at each period. Cohort participants with a proportion of attendance above the median were compared with those below the median in the first-year post-operation and the period after that.

All possible factors affecting patients’ adherence were explored. Patient’s characteristics included gender, age, body mass index at the time of surgery, history of comorbidities including hypertension, diabetes, dyslipidemia, cardiovascular disease, history of stroke, low back pain, and knee pain, marital status, education, smoking history, city of residence, and type of hospital (private or public-educational hospitals). The variables related to the operative procedures included the waiting time until surgery, the type of surgery, the length of hospital stay, the presence of surgical complications, and the need for readmission. Unfortunately, we did not have direct information about patients’ economic status, such as occupation, income, and health insurance. We used educational status and the type of hospital instead.

Statistical analysis

We described continuous variables using means (standard deviation) and categorical variables using frequency (percentage). The multivariable logistic regression analysis was used to determine factors that affect adherence as the outcome. The variables with a p-value less than 0.2 in univariable analysis were entered in this model. This approach was repeated for each period. Data analysis was performed using STATA statistical software. All the test and confidence intervals were evaluated at the 95% confidence level.

Results

In this study, we followed 5245 (4159 females and 1086 males) patients who had undergone bariatric surgery for up to 10 years. The median follow-up time was 2 years. Cohort members were mostly (62%) in the 30–50 years age and (62.5%) 40–50 kg/m2 BMI categories. Frequencies of SG, OAGB, and RYGB were 1204 (23.0%), 2601 (49.5%), and 1440 (27.5%) respectively. The majority of patients had at least one comorbidity; 34.0% dyslipidemia, 19.4% hypertension, 16.9% type 2 diabetes, and 71.5% musculoskeletal diseases (Table 1, Table S1 in the Supplementary information).

Table 1 Demographic characteristics and initial evaluations of preoperative patients in three groups, all patients, first-year follow-up, and beyond the first-year follow-up
Full size table

The contributions of each predefined session to the overall attendance at 10 days, and 1, 3, 6, 9, and 12 months after surgery were 87.6%, 80.1%, 68.0%, 57.8%, 33.0%, and 44.5%, respectively (Fig. 1) which were used as a weight in the calculation of the proportion of attendance for each individual at each period.

Fig. 1
figure 1

Predetermined postoperative visits, number of permitted visits, number and percentage of adherence to each visit

Full size image

In the logistic regression of total follow-up time, we observed an increase in adherence by each unit increase in age at the time of surgery (OR = 1.02, 95% CI: 1.01–1.03), similar to the first-year follow-up period and BMI at the first visit (OR = 1.01, 95% CI: 1.00–1.02) similar to the first year and the beyond the first-year follow-up periods. Having a history of smoking significantly decreased patient adherence (OR = 0.78, 95% CI: 0.64–0.94). Staying 2–3 days or more than 3 days in the hospital decreased the adherence by about 40% compared with a 1-day stay (OR = 0.61, 95% CI: 0.51–0.73; and OR = 0.58, 95% CI: 0.40–0.85) (Table 2).

Table 2 Factors affecting patients’ adherence to scheduled visits in total patients and patients in visits under 1 year and above 1 year
Full size table

In the first-year follow-up, patients undergoing OAGB or RYGB tended to attend the follow-up visits about 33% more than patients undergoing sleeve surgery (OR = 1.34, 95% CI: 1.15–1.55; and OR = 1.33, 95% CI: 1.13–1.56). Smoker patients were less likely to adhere compared to non-smokers (OR = 0.82, 95% CI: 0.68–0.99). Staying 2–3 days or more than 3 days in the hospital decreased adherence by about 40% compared with a 1-day stay (OR = 0.64, 95% CI: 0.54–0.77; and OR = 0.57, 95% CI: 0.39–0.83) (Table 2).

In the follow-up period beyond the first year, patients without a college education were more likely to attend the visit sessions than those with a college education (OR = 1.37, 95% CI: 1.18–1.60). In this period, in contrast to the first year, staying more than 3 days in the hospital increased adherence by 64% compared with a 1-day stay (OR = 1.64, 95% CI: 1.10–2.48). Adherence to follow-up visits in patients undergoing OAGB or RYGB was 100% higher than in those undergoing sleeve surgery (OR = 2.17, 95% CI: 1.76–2.69; and OR = 2.06, 95% CI: 1.65–2.57). History of smoking and cardiovascular diseases reduced patients' adherence to follow-up visits (OR = 0.71, 95% CI: 0.56–0.90; and OR = 0.60, 95% CI: 0.41–0.86) (Table 2).

Discussion

We found that in the first year of follow-up, an increase in age at the time of surgery, BMI at the first visit, history of dyslipidemia, admission to a public rather than the private hospital, and OAGB or RYGB surgeries were factors associated with increased adherence. In contrast, an increased length of stay in the hospital and a history of smoking decreased the adherence in total follow-up time.

Increased adherence to follow-up visits in patients who underwent bariatric malabsorptive procedures (OAGB, RYGB) in contrast to the restrictive ones [9, 18, 20, 25] could be partly attributed to the patient’s awareness of the possibility of malabsorption and malnutrition complications in these types of surgeries. This is because surgeons usually explain the advantages and disadvantages of each type of surgery, warnings, and precautions in the pre-operative orientation classes.

We have separately examined the impact of the factors associated with adherence in the first year and beyond the first year because the patterns and factors affecting adherence in these two periods differed. Combining these two intervals could dilute the effects and lead to non-significant findings.

Our study is the first that has investigated determinants of patient adherence over a long follow-up time in two separate periods. In the available literature, the Larjani and Khorgami study included 2 years of follow-up after bariatric surgery, and others covered only the first postoperative year [4, 15]. Our study, for the first time, provides the opportunity to compare the factors associated with adherence in the first postoperative year with the period beyond the first year. Length of stay, history of cardiovascular disease, type of surgery, and readmission were factors associated with patients’ adherence which had different effects in intensity or pattern in these two periods.

We found that readmission increases the chance of adherence by more than 100% in the period beyond the first year, although it does not affect adherence in the first year. This is contrary to the findings by Goldenshluger et al., which reported a decrease in patients’ adherence after readmission in the first year (OR = 0.08, p-value = 0.046) [7]. The small sample size in the study probably explains his finding. The large sample size in our study provided a sufficient number of readmissions (133 (5.1%) and 119 (4.5%) of patients in adherence and non-adherence groups to provide more stable and valid estimates of the effect on adherence.

One of the issues in studying adherence is the definition of adherers and non-adherers. Studies have used different approaches. Goldenshluger’s study used visit number as a continuous variable and applied linear models [7]. However, this approach makes a normality assumption for the variable of visit number, which is not entirely accurate. Other studies have divided the participants into adherers and non-adherers and have made it a binary variable [7, 15, 26] using the same weights for all the sessions. However, the predefined visit sessions do not have the same weight, and this approach does not address it. We took into account the importance of different predefined sessions by weighting each predefined session by the proportion of attendance of all cohort members scheduled for that session.

One of the limitations in analyzing the post-bariatric surgery follow-ups is the multidisciplinary nature of the visits. The patients are usually referred to the clinic for several visits by different specialties on the same day. Therefore, adherence to a particular service in the multidisciplinary program is not measurable. Goldenshluger has tried to address this issue by reporting the results by specialty (surgeon team or nutrition specialist) [7]. However, the way of presentation is likely to mislead because of the simultaneous visits. Simultaneous routine multidisciplinary visits are one of the most important reasons for the difference in patients’ adherence in each session in the first year.

We could use current findings to help increase attendance at postoperative care by identifying factors that could potentially reduce the adherence in each patient. People who are likely to miss their visits should be targeted by varied and more intensive recall approaches. We may also plan simultaneous multidisciplinary visits to increase their adherence.