Abstract
Background
Effects of the COVID-19 pandemic on rates of early postoperative follow-up after bariatric surgery are poorly understood. Our study characterizes 30-day follow-up after bariatric surgery prior to COVID-19 (years 2015–2019) and during the pandemic of COVID-19 (year 2020) and evaluates general predictive factors of short-term follow-up.
Methods
Data was extracted from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) data registry from 2015 to 2020. Cohorts were divided into pre-pandemic and pandemic years and patients with and without 30-day follow-up. Multivariable logistic regression analysis was used to identify general factors independently predictive of 30-day follow-up. The primary aim was to evaluate the impact of the COVID-19 pandemic on short-term 30-day follow-up adherence. A secondary outcome was to characterize general short-term postoperative 30-day follow-up associated with elective bariatric surgery and identify independent predictors of 30-day follow-up among bariatric surgery patients using multivariable logistic regression analysis.
Results
A total of 834,646 patients were identified. Follow-up rates significantly increased in the COVID era in 2020 (p < 0.0001). Patients who achieved 30-day follow-up were older and had an increased burden of medical comorbidities, including non-insulin and insulin-dependent diabetes mellitus, hypertension, dyslipidemia, as well as increased BMI compared to patients lacking follow-up. The cohort with successful 30-day follow-up was more likely to receive gastric bypass and had increased rates of metabolic comorbidities. After adjusting for comorbidities, the greatest independent predictors of follow-up were the 2020 COVID-19 era year, Asian race, black race, and gastroesophageal reflux disease.
Conclusions
After adjusting for comorbidities, the 2020 COVID-19 era year was one of the greatest predictors of follow-up after bariatric surgery. Postoperative follow-up rates after elective bariatric surgery are excellent at > 95% and increased during the 2020 COVID-19 era year. Several independent predictors of follow-up were identified which may help in development of strategies aimed to mitigate lack of postoperative follow-up.
Graphical Abstract
Supplementary Information
The online version contains supplementary material available at 10.1007/s11695-022-06423-z.
Keywords: Bariatric surgery, Postoperative follow-up, Pandemic, COVID-19, 30-day, Early, Obesity, Predictors
Introduction
Bariatric surgery continues to be the most effective long-term treatment for severe obesity and its metabolic comorbidities [1]. Previous studies have demonstrated that in addition to the extensive medical workup and educational programs typically required prior to bariatric surgery, beneficial outcomes are linked to adherence to participation in perioperative follow-up [2]. Adherence to postoperative follow-up has been associated with increased weight loss and improvement in comorbidities after bariatric surgery [3, 4]. Perhaps even more critical is the capacity of early follow-up to allow for timely recognition of concerning symptoms and expedition of treatment, ultimately resulting in better outcomes [5]. However, the coronavirus disease (COVID-19) pandemic has disrupted access to routine healthcare, including postoperative follow-up and this impact on the elective bariatric landscape is not yet understood [6].
As the COVID-19 pandemic limited the feasibility of in-person follow-up, this prompted a rapid transition to the inclusion of numerous alternatives modalities of healthcare delivery (i.e., virtual care). These methods of healthcare were previously rarely used, as only 7% of physicians reported providing at least one virtual care visit in 2019 [7]. While a variety of different strategies have been employed to ensure appropriate follow-up, there is a paucity of data on how these or patient factors affected follow-up during the pandemic [8]. Successful short-term postoperative follow-up is thought to be valuable in optimizing patient outcomes; therefore, understanding follow-up rates during the COVID-19 pandemic may provide insight into how these varying strategies have truly influenced patient care and their interaction with the bariatric team.
Our primary aim was to evaluate the impact of the COVID-19 pandemic on short-term 30-day follow-up adherence to continue to optimize perioperative outcomes by identifying barriers to postoperative follow-up adherence. Our secondary aim was to characterize general short-term postoperative 30-day follow-up associated with elective bariatric surgery and identify independent predictors of 30-day follow-up among bariatric surgery patients using multivariable logistic regression analysis.
Materials and Methods
Data Source
All data was extracted from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database from 2015 to 2020 inclusive. As the largest clinical data registry of bariatric surgery patients, the MBSAQIP captures most bariatric procedures conducted in over 800 accredited North American bariatric centers. Meticulous and frequent review of practice and data integrity are performed at contributing centers. Data is collected prospectively by trained clinical reviewers and consists of a standardized set of pre-, intra-, and postoperative variables. Postoperative variables are captured up to 30 days following bariatric surgery.
Study Design, Variable Definitions, and Population
This retrospective cohort study included all patients who underwent sleeve gastrectomy (SG) or Roux-en-Y gastric bypass (RYGB) between 2015 and the 2020 COVID-19 era year.
We divided patients into 2 main comparative groups based on operative year: 2015–2019 and 2020. We operated under the assumption that prior to 2020, the effects of COVID-19 had not affected the MBSAQIP patient population.
Patients who underwent prior bariatric surgery, emergency surgery, conversion procedures, or revision procedures were excluded. We defined our outcome of interest as those patients who followed up within 30 days postoperatively (achieved 30-day follow-up) and those who did not (lacked 30-day follow-up).
Our primary objective was to evaluate the impact of the COVID-19 pandemic on short-term 30-day follow-up adherence.
As a secondary objective, we sought to characterize 30-day follow-up and identify independent predictors of 30-day follow-up among bariatric surgery patients using multivariable logistic regression analysis.
Clinical characteristics of patients with and without postoperative 30-day follow-up included a comprehensive list of pre-, intra-, and postoperative factors. Patient factors included age, sex, body mass index (BMI), and smoking status. Comorbidities investigated included diabetes (non-diabetic and diet controlled, non-insulin dependent, and insulin dependent), hypertension, gastroesophageal reflux disease (GERD), chronic obstructive pulmonary disease (COPD), dyslipidemia, renal insufficiency, history of deep vein thrombosis (DVT) or pulmonary embolism (PE), obstructive sleep apnea (OSA), and prior myocardial infarction (MI). Additional comorbid factors included chronic steroid use, dialysis dependence, anticoagulation status, prior cardiac surgery, and prior percutaneous coronary intervention (PCI). Operative factors assessed included type of procedure and surgical approach.
Trends in characteristics from 2015 to the 2020 COVID-19 era year related to postoperative follow-up included type of procedure, follow-up at 30 days, age, BMI, surgical approach, and comorbidities including diabetes (non-diabetic and diet controlled, non-insulin dependent, and insulin dependent), hypertension, and GERD.
Specific postoperative complications that were investigated included leak, urinary tract infection (UTI), cardiac complications, pneumonia, acute kidney injury (AKI), deep surgical site infections (SSI), wound disruption, sepsis, unplanned intubation, and cerebrovascular accident (CVA). Additionally, 30-day reoperation, intervention, readmission, and mortality rates were assessed.
Statistical Analysis
Categorical variables were expressed as absolute values and percentages and univariate analysis was performed using chi-squared tests. Continuous variables were expressed as weighted means ± standard deviations (SD), and analysis was performed using independent two sample t-tests.
To determine the influence of 30-day postoperative follow-up on postoperative complications and 30-day mortality, a non-parsimonious multivariable logistic regression model was developed using a hypothesis driven purposeful selection methodology. Univariate analysis of variables with a p-value < 0.10 or from variables previously deemed to be clinically relevant to our primary outcome was used to generate a preliminary main effects model. Significant variables in the multivariable model were then identified (Wald test p < 0.05), and linear assumption of continuous variables and multicollinearity were checked. The Brier score and the receiver operating characteristic (ROC) curve were used to assess goodness of fit, as the ROC is one tool that is used to evaluate the performance of statistical models [9]. All statistical analysis was performed using Stata 15 (StataCorp LP, College Station, TX).
Results
Basic Demographics and Univariate Analysis of Achieved 30-Day Follow-Up vs. Lacked 30-Day Follow-Up Cohorts
A total of 834,646 patients that underwent primary SG and RYGB were identified from 2015 to the 2020 COVID-19 era year using the MBSAQIP data registry. Of these, 797,080 (95.5%) were identified as having follow-up within 30 days following bariatric surgery. Overall, patients with 30-day follow-up were older (44.7 ± 12.0 years vs. 42.0 ± 11.8; p < 0.0001), had decreased smoking rates (7.8% vs. 9.6%; p < 0.0001), and had increased BMI (45.0 ± 7.8 vs. 44.8 ± 7.9; p < 0.0001) when compared to patients without 30-day postoperative follow-up (Table 1; Fig. 1). No difference in sex was found between those with and without follow-up within 30 days after bariatric surgery.
Table 1.
Basic demographics of patients with and without follow-up at 30 days following bariatric surgery
| Non-follow-up (n = 37,566) | Follow-up (n = 797,080) | p-value | |
|---|---|---|---|
| Bypass | 9447 (25.2) | 217,058 (27.2) | < 0.0001 |
| Age | < 0.0001 | ||
| < 18 | 118 (0.3) | 1342 (0.17) | |
| 18–30 | 5901 (15.7) | 86,954 (10.9) | |
| 30–40 | 10,503 (28.0) | 199,430 (25.0) | |
| 40–50 | 10,797 (28.7) | 233,243 (29.3) | |
| 50–60 | 7142 (19.0) | 179,582 (22.5) | |
| > 60 | 3105 (8.3) | 96,529 (12.1) | |
| Female | 30,144 (80.3) | 641,034 (80.4) | 0.682 |
| BMI | < 0.0001 | ||
| < 35 | 1950 (5.2) | 38,734 (4.9) | |
| 35–40 | 8479 (22.8) | 181,481 (23.0) | |
| 40–50 | 18,952 (51.0) | 398,368 (50.4) | |
| 50–60 | 6022 (16.2) | 133,716 (16.9) | |
| 60–70 | 1417 (3.8) | 29,912 (3.8) | |
| > 70 | 373 (1.0) | 7780 (0.98) | |
| Racial status | < 0.0001 | ||
| White | 24,628 (65.6) | 566,385 (71.1) | |
| Black | 7665 (20.4) | 146,409 (18.4) | |
| Other | 699 (1.8) | 10,478 (1.2) | |
| Not reported | 4574 (12.2) | 73,808 (9.3) | |
| Smoker | 3611 (9.6) | 62,438 (7.8) | < 0.0001 |
| Comorbidities | |||
| Diabetes | < 0.0001 | ||
| Non-diabetic and diet controlled | 29,705 (79.1) | 597,159 (74.9) | |
| Non-insulin-dependent | 5522 (14.7) | 137,815 (17.3) | |
| Insulin-dependent | 2339 (6.2) | 62,106 (7.8) | |
| HTN | 22,470 (59.8) | 423,145 (53.1) | < 0.0001 |
| GERD | 9600 (25.6) | 250,903 (31.5) | < 0.0001 |
| COPD | 427 (1.1) | 11,827 (1.5) | < 0.0001 |
| DLD | 6693 (17.8) | 185,905 (23.3) | < 0.0001 |
| Chronic steroids | 590 (1.6) | 15,020 (1.9) | < 0.0001 |
| Renal insufficiency | 184 (0.5) | 4911 (0.6) | 0.002 |
| Dialysis | 124 (0.33) | 2488 (0.31) | 0.543 |
| Prior DVT | 497 (1.3) | 13,603 (1.7) | < 0.0001 |
| Prior PE | 391 (1.0) | 10,098 (1.3) | < 0.0001 |
| Venous stasis | 202 (0.5) | 7102 (0.9) | < 0.0001 |
| Therapeutic anticoagulation | 840 (2.2) | 23,254 (2.9) | < 0.0001 |
| Sleep apnea | 116,963 (31.1) | 299,398 (37.6) | < 0.0001 |
| Prior MI | 330 (0.9) | 9625 (1.2) | < 0.0001 |
| Prior cardiac surgery | 318 (0.9) | 8200 (1.0) | 0.001 |
| Previous PCI/PTCA | 484 (1.3) | 14,449 (1.8) | < 0.0001 |
| Surgical approach | < 0.0001 | ||
| Conventional laparoscopic (thoracoscopic) | 33,302 (88.6) | 711,509 (89.3) | |
| Endoscopic | 3 (0.01) | 31 (0.00) | |
| Hand-assisted | 22 (0.06) | 400 (0.05) | |
| Laparoscopic assisted (thoracoscopic assisted) | 1063 (2.8) | 22,842 (2.9) | |
| N.O.T.E.S | 12 (0.03) | 293 (0.04) | |
| Open | 18 (0.05) | 190 (0.02) | |
| Robotic-assisted | 3033 (8.1) | 60,686 (7.6) | |
| Single incision | 113 (0.3) | 1129 (0.1) |
Bolded p-values are statistically significant
BMI body mass index, ASA American Society of Anesthesiologists, HTN hypertension, GERD gastroesophageal reflux disease, COPD chronic obstructive pulmonary disease, DLD dyslipidemia, DVT deep vein thrombosis, PE pulmonary embolism, MI myocardial infarction, PCI percutaneous coronary intervention
Fig. 1.
Summary of key predictors of 30-day postoperative follow-up including operative year, racial status, body mass index, sex, and smoking
Patients that achieved 30-day follow-up had an increased burden of medical comorbidities. These included non-insulin and insulin-dependent diabetes mellitus (17.3% vs. 14.7%; p < 0.0001 and 7.8% vs. 6.2%; p < 0.0001, respectively), hypertension (53.1% vs. 59.8%; p < 0.0001), dyslipidemia (23.3% vs. 17.8%; p < 0.0001) and sleep apnea (37.6% vs. 31.1%; p < 0.0001) (Table 1). Additional comorbidities significantly associated with achievement of 30-day follow-up were GERD (31.5% vs. 25.6%; p < 0.0001), COPD (1.5% vs. 1.1%; p < 0.0001), chronic steroid use (1.9% vs. 1.6%; p < 0.0001), and renal insufficiency (0.6% vs. 0.5%; p = 0.002). Prior medical events associated with follow-up included prior DVT (1.7% vs. 1.3%; p < 0.0001), prior PE (1.3% vs. 1.0%; p < 0.0001), and prior MI (1.2% vs. 0.9%; p < 0.0001).
Trends in Follow-Up and Related Characteristics in the Pre-pandemic and 2020 COVID-19 Era Year
Several key characteristics related to early follow-up postbariatric surgery were analyzed to determine trends from 2015 to 2020 (Table 2). In the pre-pandemic period (2015–2019), rates of follow-up fluctuated between 2015 and 2019 from 95.1 to 95.6% (Fig. 2). Trends in age and BMI decreased over this period, as well as the presence of insulin-dependent diabetes. Conversely, rates of non-diabetic and diet-controlled and non-insulin-dependent diabetes increased.
Table 2.
Trends in characteristics related to follow-up postbariatric surgery
| 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | p-value | |
|---|---|---|---|---|---|---|---|
| Bypass | 20,029 (29.8) | 23,832 (27.1) | 45,649 (26.7) | 47,112 (26.9) | 50,131 (28.3) | 39,753 (25.5) | < 0.0001 |
| Follow-up at 30 days | 64,254 (95.6) | 83,639 (95.1) | 162,922 (95.2) | 166,999 (95.3) | 168,637 (95.2) | 150,629 (96.7) | < 0.0001 |
| Surgical approach | < 0.0001 | ||||||
| Conventional laparoscopic | 60,382 (89.9) | 78,983 (89.8) | 152,027 (88.8) | 151,815 (86.6) | 145,833 (82.3 | 155,772 (99.96) | |
| Endoscopic | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) | 34 (0.02) | |
| Hand-assisted | 97 (0.14) | 79 (0.09) | 96 (0.06) | 75 (0.05) | 62 (0.04) | ||
| Laparoscopic assisted (thoracoscopic assisted) | 3432 (5.1) | 3315 (3.8) | 4756 (2.8) | 5646 (3.2) | 6756 (3.8) | 0 (0.00) | |
| N.O.T.E.S | 3 (0.00) | 37 (0.04) | 141 (0.08) | 78 (0.04) | 46 (0.03) | 0 (0.00) | |
| Open | 30 (0.04) | 24 (0.03) | 50 (0.03) | 39 (0.02) | 41 (0.02) | 24 (0.02) | |
| Robotic-assisted | 3219 (4.8) | 5292 (6.0) | 13,522 (7.9) | 17,379 (9.9) | 24,307 (13.7) | - | |
| Single incision | 42 (0.06) | 191 (0.22) | 572 (0.33) | 277 (0.16) | 160 (0.09) | 0 (0.00) | |
| Comorbidities | |||||||
| Diabetes | < 0.0001 | ||||||
| Non-diabetic and diet controlled | 49,726 (74.0) | 65,438 (74.4) | 127,755 (74.6) | 130,890 (74.7) | 132,790 (74.9) | 120,266 (77.2) | |
| Non-insulin dependent | 11,601 (17.3) | 15,234 (17.3) | 29,664 (17.3) | 30,407 (17.3) | 30,947 (17.5) | 25,484 (16.4) | |
| Insulin dependent | 5878 (8.8) | 7249 (8.2) | 13,745 (8.03) | 14,022 (8.00) | 13,471 (7.6) | 10,080 (6.5) | |
| HTN | 32,887 (48.9) | 42,107 (47.9) | 80,921 (47.3) | 81,974 (46.8) | 81,913 (46.2) | 69,229 (44.4) | < 0.0001 |
| GERD | 20,773 (30.9) | 27,654 (31.5) | 53,158 (31.1) | 53,692 (30.6) | 55,508 (31.3) | 49,719 (31.9) | < 0.0001 |
Bolded p-values are statistically significant
N.O.T.E.S Natural Orifice Transluminal Endoscopic Surgery, HTN hypertension, GERD gastroesophageal reflux disease
Fig. 2.
Trends in follow-up by postoperative year for bariatric surgery patients who underwent either Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) procedures between 2015 and 2020
Rates of follow-up peaked in the 2020 COVID-19 era year at 96.7%. Compared to the pre-pandemic era, bariatric surgical patients were of older age and lower BMI in the 2020 COVID-19 era year. Additionally, comorbidities including non-insulin and insulin-dependent diabetes mellitus and HTN significantly decreased while the diagnosis of GERD significantly increased during the pandemic period.
Bi-variate Analysis of Postoperative Complications in Patients Who Achieved 30-Day Follow-Up vs. Lacked 30-Day Follow-Up Cohorts
We compared complication rates between achieved 30-day follow-up and lacked 30-day follow-up cohorts to identify if postoperative follow-up at 30 days was associated with development of postoperative complications (Table 3). No significant difference was found between patients with or without postoperative follow-up with regard to leaks, cardiac complications, deep SSI, reoperation, pneumonia, AKI, wound disruption, sepsis, unplanned intubation, or CVA. At 30-day postoperation, complications including need for reoperation (1.3% vs. 1.1%; p < 0.0001), reintervention (1.2% vs. 0.8%; p < 0.0001), and readmission (3.8% vs. 3.0%; p < 0.0001) were all increased in patients with postoperative follow-up. There was no difference in the mortality rate between the two cohorts.
Table 3.
Postoperative complications of patients with and without follow-up at 30 days following bariatric surgery
| Complication | Non-follow-up (n = 34,285) | Follow-up at 30 days (n = 717,667) | p-value |
|---|---|---|---|
| Reoperation | 405 (1.1) | 10,467 (1.3) | < 0.0001 |
| Reintervention | 311 (0.8) | 9512 (1.2) | < 0.0001 |
| Readmission | 1126 (3.0) | 29,923 (3.8) | < 0.0001 |
| Leak | 8 (0.15) | 448 (0.3) | 0.153 |
| UTI | 86 (0.2) | 2728 (0.3) | < 0.0001 |
| Cardiac | 10 (0.03) | 205 (0.03) | 0.915 |
| Deep SSI | 16 (0.04) | 537 (0.07) | 0.295 |
| Wound disruption | 14 (0.04) | 464 (0.06) | 0.236 |
| Pneumonia | 72 (0.19) | 1686 (0.21) | 0.650 |
| Unplanned intubation | 48 (0.13) | 1072 (0.13) | 0.728 |
| CVA | 7 (0.02) | 112 (0.01) | 0.467 |
| AKI | 22 (0.06) | 578 (0.07) | 0.324 |
| Sepsis | 30 (0.09) | 739 (0.1) | 0.381 |
| Unplanned intubation | 65 (0.19) | 907 (0.13) | 0.728 |
| Mortality | 44 (0.12) | 747 (0.09) | 0.150 |
Bolded p-values are statistically significant
UTI urinary tract infection, SSI surgical site infection, CVA cerebrovascular accident, AKI acute kidney injury
Multivariable Logistic Regression Analysis for Predictors of 30-Day Postoperative Follow-Up
A multivariable logistic regression model was then developed to identify if clinical characteristics were independently predictive of 30-day postoperative follow-up after adjusting for covariates (Table 4). Receiver operating characteristic (ROC) and Brier scores (BS) for this model were 0.68 and 0.0, respectively.
Table 4.
Predictors of follow-up at 30 days following bariatric surgery
| Predictors of follow-up | Odds ratio | 95% confidence interval | p-value | |
|---|---|---|---|---|
| Older age (per 1 year) | 1.16 | 1.14–1.17 | < 0.0001 | |
| Higher BMI (per kg/m2) | 1.04 | 1.03–1.05 | < 0.0001 | |
| Male | 0.87 | 0.84–0.90 | < 0.0001 | |
| Racial status | ||||
| Black vs. white | 1.33 | 1.17–1.53 | < 0.0001 | |
| Asian vs. white | 1.43 | 1.18–1.73 | < 0.0001 | |
| Pacific Islander vs. white | 1.23 | 0.90–1.40 | 0.292 | |
| Other vs. white | 0.30 | 0.24–0.39 | < 0.0001 | |
| Unknown vs. white | 1.18 | 1.03–1.35 | 0.017 | |
| Smoker | 0.82 | 0.80–0.85 | < 0.0001 | |
| Diabetes | ||||
| Non-insulin dependent vs. non-diabetic/diet controlled | 1.09 | 1.06–1.12 | < 0.0001 | |
| Insulin dependent vs. non-diabetic/diet controlled | 1.06 | 1.01–1.11 | 0.020 | |
| Bypass | 1.02 | 0.99–1.04 | 0.243 | |
| OSA | 1.17 | 1.14–1.20 | < 0.0001 | |
| HTN | 1.09 | 1.02–1.07 | < 0.0001 | |
| GERD | 1.18 | 1.15–1.21 | < 0.0001 | |
| DLD | 1.09 | 1.06–1.12 | < 0.0001 | |
| History ofDVT | 1.08 | 0.98–1.19 | 0.116 | |
| Prior MI | 1.03 | 0.92–1.15 | 0.613 | |
| COPD | 0.95 | 0.86–1.05 | 0.343 | |
| Renal insufficiency | 1.08 | 0.92–1.26 | 0.365 | |
| Therapeutic anticoagulation | 0.98 | 0.91–1.06 | 0.636 | |
| Operative year | ||||
| 2016 vs. 2015 | 0.90 | 0.86–0.94 | < 0.0001 | |
| 2017 vs. 2015 | 0.92 | 0.88–0.96 | < 0.0001 | |
| 2018 vs. 2015 | 0.93 | 0.90–0.98 | 0.002 | |
| 2019 vs. 2015 | 0.92 | 0.88–0.96 | < 0.0001 | |
| 2020 vs. 2015 | 1.40 | 1.33–1.46 | < 0.0001 |
Bolded p-values are statistically significant
BMI body mass index, OSA obstructive sleep apnea, HTN hypertension, GERD gastroesophageal reflux disease, DLD dyslipidemia, DVT deep vein thrombosis, MI myocardial infarction, COPD chronic obstructive pulmonary disease
Our model evaluated predictors of 30-day postoperative follow-up (Table 4). A total of 12 characteristics were found to be independent positive predictors of postoperative follow-up, and eight were found to be independent negative predictors. The five variables that were found to be the strongest independent predictors of having postoperative follow-up from greatest to lowest effect, respectively, were the following: Asian race (OR 1.43, 95% CI 1.18–1.73, p < 0.0001), surgical year of 2020 (OR 1.40, 95% CI 1.33–1.46, p < 0.0001), black race (OR 1.33, 95% CI 1.17–1.53, p < 0.0001), GERD (OR 1.18, 95% CI 1.15–1.21, p < 0.0001), and sleep apnea (OR 1.17, 95% CI 1.14–1.20, p < 0.0001). A higher BMI, unknown race, and presence of comorbidities including diabetes, HTN, and dyslipidemia were also independent predictors of having postoperative follow-up. Racial status classified as other was the single strongest independent predictor of not having postoperative follow-up. Smoking, male sex, and later operative year up to 2019 were also independently negatively associated with 30-day postoperative follow-up.
Achievement of follow-up was positively correlated with readmission, reoperation, and reintervention (Tables S1–S3). The 2020 COVID-19 era year was a negative independent predictor of both readmission and reintervention (OR 0.84, 95% CI 0.80–0.88, p < 0.0001 and OR 0.64, 95% CI 0.59–0.69, p < 0.0001, respectively).
Discussion
Our study highlights that rates of short-term 30-day postoperative follow-up after elective bariatric surgery improved during the pandemic and that the 2020 COVID-19 era year was the single greatest independent predictor of achieving follow-up. Patients who achieved 30-day follow-up were older and had an increased burden of medical comorbidities compared to patients lacking follow-up. Multivariable logistic regression identified predictors of follow-up. These can be clinically valuable to inform strategies aimed at overcoming barriers to follow-up.
Rapid adoption of virtual healthcare has been well documented over the course of the COVID-19 pandemic, and there are benefits to patients, healthcare providers, and the overall healthcare system [10–12]. Benefits of telemedicine visits include decreased need for patient travel, high rates of patient satisfaction, and fiscal savings [8, 10]. Benefits to the clinician included increased flexibility in scheduling and work location and the ability to follow-up with patients more closely [12]. Finally, there is evidence to suggest that there is a lower overall expenditure from the healthcare system on patients who participate in virtual follow compared to those that only seek in person care [11]. Studies have found that this lower expenditure may be due to decreased hospital admissions, as well as allowing healthcare providers to see more patients in a clinic day [12]. Surveys have shown that both patients and providers support the continuation of virtual care as an option for accessing healthcare in the postpandemic period [13].
Although there are numerous benefits to the increased adoption of virtual healthcare, there are challenges that should also be considered. Care will need to be taken to not unintentionally exclude high-risk patient populations, as evidence suggests that racial minority groups may be offered access to telemedicine less frequently than white patients [14]. This population may also have more concerns about confidentiality when using telemedicine [15]. Additionally, although virtual care may benefit patients living in remote communities, it further disadvantages those who do not have access to Internet or phone services or devices such as smartphones or computers [14]. Therefore, we suggest offering all patients the opportunity to participate in virtual follow-up visits and provide clear documentation on strategies employed to maintain patient privacy and confidentiality.
Taken together, our results suggest that although the current rate of 30-day postoperative follow-up for bariatric surgical patients is high, rates of follow-up could be improved, especially in specific patient populations. By identifying characteristics positively or negatively associated with rates of early follow-up, targeted strategies can be implemented to encourage patients at higher risk for lack of follow-up to attend these appointments. This is a critical aspect to a successful bariatric procedure, as there is strong evidence supporting that adequate follow-up is associated with improved outcomes, not only for weight loss but also in reducing the burden of related comorbidities [2–4, 16]. Early follow-up is of particular importance given that some complications, such as dietary intolerances, wound infections, and VTE often present early in the postoperative period and are thought to be more easily managed with prompt intervention [17–19]. Recent guidelines provide recommendations for minimum frequency and duration of both short-term and long-term follow-up postbariatric surgery, including the American Society of Metabolic and Bariatric Surgery, who emphasizes the importance of these visits in the view of obesity as a chronic disease [20, 21].
Prior literature has attempted to describe rates of follow-up for surgical patients as well as provide insight into factors that may reduce the likelihood of participation in postoperative follow-up. Saunders et al. [22] conducted a retrospective review of 6620 general surgery patients, with an overall 84% follow-up rate and higher rates of follow-up in older patients of white racial status. Another retrospective study by Vidal et al. reported the rate of follow-up visits after bariatric surgery as 82.5% and identified work-related problems as the main factor associated with nonadherence [23]. Thereaux et al. reported predictive factors of 5-year follow-up rates in 16,620 patients following bariatric surgery [24]. They found that older age, female sex, and comorbidities such as type 2 diabetes and sleep apnea were predictive of follow-up at 5 years. Our findings reinforce current evidence with a large data set and provide novel predictive factors of early patient follow-up after bariatric surgery within the COVID-19 pandemic. As the largest bariatric study to characterize and evaluate predictors of follow-up, these findings are critical in developing targeted strategies to increase postoperative follow-up rates in patient populations at higher risk for loss to follow-up.
Our study has a few limitations consistent with the retrospective nature of our study design. As MBSAQIP data is limited to follow-up at 30 days, we are unable to comment on whether similar factors are also associated with rates of longer term follow-up. Therefore, we were unable to elaborate on important aspects of long-term follow-up, including nutritional deficiencies or metabolic outcomes. Additionally, MBSAQIP does not capture nuanced data, such as surgeon or center specific practices, or characteristics, such as education level or socioeconomic status, which make our results less actionable. Although the database includes smoking status, it does not describe if a patient previously smoked or the substance that they smoke. The database also does not capture factors such as patient income, geographical location, and proportion of different ethnicities, all of which may contribute to the likelihood of a patient attending follow-up visits. It is also possible that a confounder exists which we were unable to account for and subsequently may have influenced the results of our multivariable regression models. Qualitative data is not collected by the MBSAQIP data registry; therefore, we are unable to comment on the influence of factors such as patient attitudes and beliefs regarding medical care. The MBSAQIP only captured data from North American bariatric centers, and our results may not be as applicable to centers in other countries. Finally, given the nature of our study design, we are not able to identify explanations responsible for our findings.
Despite these limitations, our work has several novel implications. We importantly highlight that rates of postoperative follow-up are excellent and have not been adversely impacted by the COVID-19 era. Lastly, we identify a subset of at-risk groups which may have barriers to adherence of short-term postoperative follow-up. Given these at-risk groups, we advocate for healthcare providers to emphasize the importance of follow-up as well as consider incorporation of strategies to improve adherence into their practice such as virtual healthcare.
Conclusions
After adjusting for comorbidities, the 2020 COVID-19 era year was one of the greatest predictors of follow-up after bariatric surgery. Postoperative follow-up rates after elective bariatric surgery are excellent at > 95% and increased during the 2020 COVID-19 era year. Several independent predictors of follow-up were identified which may help in the development of strategies aimed to mitigate lack of postoperative follow-up.
Supplementary Information
Below is the link to the electronic supplementary material.
Declarations
Ethical Approval and Informed Consent
For this type of study, formal consent is not required. All procedures performed that contributed to the data registry were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Conflict of Interest
The authors declare no competing interests.
Footnotes
Key Points
Key finding #1 — Postoperative follow-up rates increased during the 2020 COVID-19 era year.
Key finding #2 — Metabolic comorbidities are predictive of successful follow-up.
Key finding #3 — The 2020 pandemic year was among the greatest independent predictors for achieving 30-day follow-up.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Hillary A. Wilson, Valentin Mocanu, Cheynne McLean, Daniel W. Birch, Shahzeer Karmali, and Noah J. Switzer have contributed equally to this work.
Contributor Information
Hillary A. Wilson, Email: haw@ualberta.ca
Valentin Mocanu, Email: vmocanu@ualberta.ca.
Cheynne McLean, Email: cheynne@ualberta.ca.
Daniel W. Birch, Email: dbirch@ualberta.ca
Shahzeer Karmali, Email: shahzeer@ualberta.ca.
Noah J. Switzer, Email: nswitzer@ualberta.ca
References
- 1.Smith MD, Patterson E, Wahed AS, Belle SH, Berk PD, Courcoulas AP, Dakin GF, Flum DR, Machado L, Mitchell JE, Pender J, Pomp A, Pories W, Ramanathan R, Schrope B, Staten M, Ude A, Wolfe BM. Thirty-day mortality after bariatric surgery: independently adjudicated causes of death in the longitudinal assessment of bariatric surgery. Obes Surg. 2011;21:1687–1692. doi: 10.1007/s11695-011-0497-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Keren D, Matter I, Rainis T, Lavy A. Getting the most from the sleeve: the importance of post-operative follow-up. Obes Surg. 2011;21:1887–1893. doi: 10.1007/s11695011-0481-3. [DOI] [PubMed] [Google Scholar]
- 3.Spaniolas K, Kasten KR, Celio A, Burruss MB, Pories WJ. Postoperative follow-up after bariatric surgery: effect on weight loss. Obes Surg. 2016;26:900–903. doi: 10.1007/s11695-016-2059-6. [DOI] [PubMed] [Google Scholar]
- 4.Lujan J, Tuero C, Landecho MF, Moncada R, Cienfuegos AJ, Rotellar F, Silva C, Lapuente F, Martínez P, Frühbeck G, Valenti V. Impact of routine and long-term follow-up on weight loss after bariatric surgery. Obes Surg. 2020;11:4293–4299. doi: 10.1007/s11695-020-04788-7. [DOI] [PubMed] [Google Scholar]
- 5.Mogoanta SS, Paitici S, Mogoanta CA. Postoperative follow-up and recovery after abdominal surgery. IntechOpen. 2021 doi: 10.5772/intechopen.97739. [DOI] [Google Scholar]
- 6.Rubin R. COVID-19’s crushing effects on medical practices, some of which might not survive. JAMA. 2020;324:321–323. doi: 10.1001/jama.2020.11254. [DOI] [PubMed] [Google Scholar]
- 7.Bhatia RS, Chu C, Pang A, Tadrous M, Stamenova V, Cram P. Virtual care use before and during the COVID-19 pandemic: a repeated cross-sectional study. CMAJ Open. 2021;9:E107–E114. doi: 10.9778/cmajo.20200311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Auge M, Dejardin O, Menahem B, Lee Bion A, Savey V, Launoy G, Bouvier V, Alves A. Analysis of the lack of follow-up of bariatric surgery patients: experience of a reference center. J Clin Med. 2022;11:6310. doi: 10.3390/jcm11216310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zou KH, O’Malley AJ, Mauri L. Receiver-operating characteristic analysis for evaluating diagnostic tests and predictive models. Circulation. 2007;115:654–657. doi: 10.1161/CIRCULATIONAHA.105.594929. [DOI] [PubMed] [Google Scholar]
- 10.Glazier RH, Green ME, Wu FC, Frymire E, Kopp A, Kiran T. Shifts in office and virtual primary care during the early COVID-19 pandemic in Ontario, Canada. CMAJ. 2021;193:E200–E210. doi: 10.1503/cmaj.202303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McGrail KM, Ahuja MA, Leaver CA. Virtual visits and patient-centered care: results of a patient survey and observational study. J Med Internet Res. 2017;19:e177. doi: 10.2196/jmir.7374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Williams AM, Bhatti UF, Alam HB, Nikolian VC. The role of telemedicine in postoperative care. Mhealth. 2018;4. 10.21037/mhealth.2018.04.03 [DOI] [PMC free article] [PubMed]
- 13.Nanda M, Sharma R. A review of patient satisfaction and experience with telemedicine: a virtual solution during and beyond COVID-19 pandemic. Telemed J E Health. 2021;12:1325–1331. doi: 10.1089/tmj.2020.0570. [DOI] [PubMed] [Google Scholar]
- 14.Anthony DL, Campos-Castillo C, Lim PS. Who isn’t using patient portals and why? Evidence and implications from a national sample of US adults. Health Aff. 2018;37:1948–1954. doi: 10.1377/hlthaff.2018.05117. [DOI] [PubMed] [Google Scholar]
- 15.George S, Hamilton A, Baker RS. How do low-income urban African Americans and Latinos feel about telemedicine? A diffusion of innovation analysis. Int J Telemed Appl. 2012;2012:715194. doi: 10.1155/2012/715194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schwoerer A, Kasten K, Celio A, Pories W, Spaniolas K. The effect of close postoperative follow-up on co-morbidity improvement after bariatric surgery. Surg Obes Relat Dis. 2017;13:1347–1352. doi: 10.1016/j.soard.2017.03.02. [DOI] [PubMed] [Google Scholar]
- 17.Lim R, Beekley A, Johnson DC, Davis KA. Early and late complications of bariatric operation. Trauma Surg Acute Care Open. 2018;9:e000219. doi: 10.1136/tsaco-2018-000219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Elrazek AE, Elbanna AE, Bilasy SE. Medical management of patients after bariatric surgery: principles and guidelines. World J Gastrointest Surg. 2014;6:220–228. doi: 10.4240/wjgs.v6.i11.220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Nijland LMG, Reiber BMM, Jacobs A, Hazebroek EJ, van Veen RN, Demirkiran A, de Castro SMM. The association between patient attendance to a perioperative group-based lifestyle program and weight loss after bariatric surgery. SOARD in press. 2022 doi: 10.1016/j.soard.2022.02.011. [DOI] [PubMed] [Google Scholar]
- 20.Public Education Committee. Life after bariatric surgery. American Society of Metabolic and Bariatric Surgery. 2021; Retrieved from: https://asmbs.org/patients/life-after-bariatric-surgery#b7
- 21.O’Kane M, Parretti HM, Hughes CA, Sharma M, Woodcock S, Puplampu T, Blakemore AI, Clare K, MacMillan I, Joyce J, Sethi S, Barth JH. Guidelines for the follow-up of patients undergoing bariatric surgery. Clin Obes. 2016;6:210–224. doi: 10.1111/cob.12145. [DOI] [PubMed] [Google Scholar]
- 22.Saunders RS, Fernandes-Taylor S, Rathouz PJ, Saha S, Wiseman JT, Havlena J, Matsumura J, Kent KC. Outpatient follow-up versus 30-day readmission among general and vascular surgery patients: a case for redesigning transitional care. Surgery. 2014;156:949–958. doi: 10.1016/j.surg.2014.06.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Vidal P, Ramón JM, Goday A, Parri A, Crous X, Trillo L, Pera M, Grande L. Lack of adherence to follow-up visits after bariatric surgery: reasons and outcome. Obes Surg. 2014;24:179–183. doi: 10.1007/s11695-013-1094-9. [DOI] [PubMed] [Google Scholar]
- 24.Thereaux, J Lesuffleur T, Païta M, Czernichow S, Basdevant A, Msika S, Millat B, Fagot-Campagna A. Long-term follow-up after bariatric surgery in a national cohort. Br J Surg. 2017;1362–1371. 10.1002/bjs.10557 [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.