Associations of Bariatric Surgery With Changes in Interpersonal Relationship Status: Results From 2 Swedish Cohort Studies (2024)

Study Cohorts

The SOS study is a prospective, matched intervention trial, and its design has been described previously.¹⁹ In brief, 2010 participants who chose bariatric surgery formed the surgery group, and a contemporaneously matched control group (n = 2037) was created using 18 matching variables (sex, age, weight, height, waist and hip circumferences, systolic blood pressure, serum cholesterol and triglyceride levels, smoking status, diabetes, menopausal status, 4 psychosocial variables [social support, life events, health perception, and psychosocial functioning], and 2 personality traits [monotony avoidance and psychasthenia]). In the surgery group, 265 participants (13.2%) underwent gastric bypass, 376 participants (18.7%) underwent gastric banding, and 1369 participants (68.1%) had vertical banded gastroplasty. The control group received the customary treatment for obesity at their primary health care centers.

Power calculation for the SOS study was based on the primary outcome, which was overall mortality. The inclusion criteria in both surgery and control groups were an age between 37 and 60 years and a body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) of 34 or more for men and 38 or more for women. Both study groups had identical follow-up with physical examinations and questionnaires at baseline and after 0.5, 1, 2, 3, 4, 6, 8, 10, 15, and 20 years. Participants were recruited between September 1, 1987, and January 31, 2001. In the current analysis, the median follow-up time was 10 years, with follow-up examinations occurring until July 1, 2015. This analysis includes 1958 patients who had gastric banding (n = 368 [18.8%]), vertical banded gastroplasty (n = 1331 [68.0%]), or gastric bypass (n = 259 [13.2%]) and 1912 matched, obese controls who had known relationship status at baseline and follow-up.

In addition, this study incorporated data from SOReg, a prospective, electronically captured register for bariatric surgery in which data are stored as part of clinical practice. In brief, SOReg was started in 2007 and is estimated to cover 98.5% of all bariatric procedures in Sweden.¹⁷ In this study, we used data from January 2007 through December 2012. In the current analysis, all the 29 234 SOReg participants older than 18 years who had undergone a gastric bypass operation, of whom 12 531 were married and 16 703 were unmarried at baseline, were included. For comparisons of changes in relationship status after bariatric surgery with a general population sample, we matched up to 10 comparators from the Swedish Total Population Registry¹⁸ to each included patient who had undergone gastric bypass, using birth year, sex, and place of residence (parish). Extended descriptions of the SOS, SOReg, and general population groups can be found in eAppendix 1 and 2 in the Supplement.

Main Outcome Measures and Follow-up

Relationship status was not a predefined end point in either the SOS study or the SOReg and general population cohort. For the current analysis, the main outcome measure in the SOS cohort was self-reported relationship status (questionnaires at baseline and 0.5, 1, 2, 3, 4, 6, 8, 10, 15, and 20 years after baseline data collection). Relationship status was defined by response to the questions: (1) Are you currently living with a spouse/partner? and (2) Have you gone through a divorce or separated since the previous questionnaire? A positive reply to the first question was interpreted as being married or in a cohabiting relationship. The main outcome measures in the SOReg and general population cohort were date of legal marriage or divorce, as recorded in the Swedish Total Population Registry.

In the analysis of divorce/separation, participants who were widowed were included in the analysis until the date of the death of their spouse and thereafter censored from the analysis. In the SOS cohort, if no separation was reported, the participant was censored at the most recent questionnaire. Participants were also censored at date of emigration or death. Participants in the SOS control group who underwent bariatric surgery and participants in the SOS surgery group who underwent surgical reversal to normal anatomy (eg, band removal) were censored from analysis at the time of surgery.

Covariates

Body mass index data were based on measured heights and weights; these data were not available for the general population. In the SOS cohort, education level was self-reported and, in the SOReg and general population cohort, data on the level of educational attainment were retrieved from the Swedish Register of Education. Information on previous divorce/separation and years living together/ married was based on self-reported data in the SOS cohort and on data from the Swedish Total Population Registry in the SOReg and general population cohort. Self-reported data on perceived family relations on a 4-grade scale (very good, good, not so good, and not good at all) were available in the SOS cohort but were analyzed here as a dichotomous variable by combining the 2 best and 2 worst responses. Information on participants becoming a widow/widower was based on self-reported data (in the SOS cohort) and on data from the Swedish Total Population Registry (in the SOReg and general population cohorts). Data on health care visits for substance abuse were retrieved from the National Patient Register (which includes inpatient data from 1969 to December 31, 2014, and hospital-based outpatient data from 2001 to December 31, 2014) and converted into a dichotomous variable, as previously described.¹² Data on psychotropic drug use before inclusion were retrieved from self-reported data (in the SOS cohort) and from the Prescribed Drug Register (in the SOReg and general population cohort; this register started on July 1, 2005, and data up till September 30, 2015, were used) and converted into a dichotomous variable, as previously described.¹²

Statistical Analyses

To analyze the association of bariatric surgery with relationship status, the study cohorts were divided according to whether the participants were married or in a relationship or not at baseline. Time to change in relationship status in the surgery and control groups was analyzed with Kaplan-Meier estimates of cumulative incidence and Cox proportional hazard models (expressed as hazard ratios [HR] with 95% CIs). Unadjusted univariate analysis comparing surgical participants with control group participants, as well as multivariate analysis with adjustment for preselected baseline factors known to influence relationship status (age, sex, BMI [in the SOS cohort only], education level, previous divorce/separation, time in current relationship, substance abuse, psychotropic medication, and perceived family relation quality [in the SOS cohort only]) were performed.

Statistical analyses were carried out using Stata, version 14 (StataCorp) and SAS, version 9.4 (SAS Institute Inc). Two-sided tests with a 5% α level were considered statistically significant.

Relationship Status and Characteristics at Baseline

In the SOS study, 2871 of the 3870 included participants (74.2%) were married or living in a relationship at baseline (Table). There were no major differences in characteristics between the surgery and control groups, with the exception that university education was more common and previous divorce/separation less common in the control group compared with the surgery group (Table).

In the SOReg cohort, 12 531 of the 29 234 included participants (42.9%) were married at baseline (Table). Use of psychotropic drugs and substance abuse were more common in the surgery group compared with the general population group, both among those who were married or unmarried at baseline. University education was more common in the general population group compared with the surgery group (Table).

Marriages and New Relationships After Surgery

Among the 999 SOS study participants who were single at baseline, the incidence of marriage/new relationship was higher after bariatric surgery (n = 152 events; incidence rate/1000 person-years, 38.0) compared with controls (n = 76 events; incident rate [IR]/1000 person-years, 19.1; HR = 1.95; 95% CI, 1.48-2.57; P < .001; Figure 1A), and this remained statistically significant after adjustment (adjusted HR [aHR] = 2.03; 95% CI, 1.52-2.71; P < .001; eTable 1 in the Supplement). After 4 years, the cumulative incidence of marriage/new relationship was 20.9% (n = 98 events) in the surgery group compared with 11.2% (n = 47 events) in the control group, and at 10 years, it was 34.8% (n = 146 events) in the surgery group compared with 19.4% (n = 70 events) in the control group.

The incidence of marriage in the SOReg and general population cohort was higher in those who had gastric bypass (n = 1895 events; IR/1000 person-years, 39.1) than in general population comparison group (n = 14612 events; IR/1000 person-years, 31.6; HR = 1.24; 95% CI, 1.18.-1.30; P < .001; Figure 1B; aHR = 1.35; 95% CI 1.28-1.42; P < .001; eTable 1 in the Supplement). After 4 years, the cumulative incidence of marriage was 14.6% (n = 1771 events) in SOReg cohort compared with 11.8% (n = 13661 events) in the general population.

In both study cohorts, the multivariate analyses show that younger age and previous divorce/separation were significantly associated with increased incidence of marriage/new relationship (in the SOS cohort) and marriage (in the SOReg and general population cohort) (eTable 1 in the Supplement). In the SOReg and general population cohort, a history of substance abuse and psychotropic medication use were associated with a decreased incidence of marriage (eTable 1 in the Supplement).

In the SOS surgery group, those whose 1-year weight loss reached or exceeded the median level of 25.0% of body weight had higher incidence of marriage/new relationship (HR = 1.94; 95% CI, 1.32-2.84; P < .001; aHR = 1.98; 95% CI, 1.33-2.93; P < .001) compared with those with weight loss of less than the median level (Figure 2A and eTable 2 in the Supplement). In the SOReg cohort, those whose 1-year weight loss reached or exceeded the median level of 31.8% of body weight had a higher incidence of marriage (HR = 1.35; 95% CI, 1.20-1.51; P < .001; aHR = 1.17; 95% CI, 1.04-1.32; P = .008) compared with those with weight loss of less than the median level (Figure 2B and eTable 2 in the Supplement).

Separations and Divorces After Surgery

Among the 2871 participants in the SOS cohort who were married or were living in a relationship at baseline, bariatric surgery was associated with increased incidence of divorce/separation (n = 235 events; IR/1000 person-years, 16.5) compared with controls (n = 142 events; IR/1000 person-years, 11.1; HR = 1.54; 95% CI, 1.25-1.90; P < .001; Figure 3A), and this remained statistically significant after multivariate adjustment (aHR = 1.28; 95% CI, 1.03-1.60; P = .03; eTable 1 in the Supplement). After 4 years, the cumulative incidence of divorce/separation was 9.4% (n = 128 events) in the surgery group compared with 5.5% (n = 71 events) in the control group, and after 10 years it was 17.1% (n = 211 events) in the surgery group compared with 11.6% (n = 126 events) in the control group.

The incidence of divorce in the SOReg and general population cohort was higher in those who underwent gastric bypass (n = 1380 events; IR/1000 person-years, 37.8) than in the general population comparators (n = 8221 events; IR/1000 person-years, 21.7; HR = 1.74; 95% CI, 1.64-1.84; P < .001; Figure 3B), and this remained statistically significant after multivariate adjustments (aHR = 1.41; 95% CI, 1.33-1.49; P < .001); eTable 1 in the Supplement). After 4 years, the cumulative incidence of divorce was 14.4% (n = 1289 events) in SOReg compared with 8.2% (n = 7657 events) in the general population.

In both study cohorts, the multivariate analyses show that younger age, previous divorce/separation, and shorter time in a relationship/marriage were significantly associated with increased incidence of divorce/separation (in the SOS study) or divorce (in the SOReg and general population cohort) (eTable 1 in the Supplement). In the SOS study, poor family relations before surgery were a strong predictor of divorce/separation (eTable 1 in the Supplement). In the SOReg and general population cohort, a history of substance abuse and psychotropic medication use were associated with an increased incidence of divorce (eTable 1 in the Supplement).

In the SOS surgery group, the association between weight loss and incidence of divorce/separation was not statistically significant (HR, 1.22; 95% CI, 0.91-1.63; P = 0.18; aHR, 1.31; 95% CI, 0.96–1.77; P = 0.08; Figure 4A; eTable 2 in the Supplement). However, in the SOReg cohort, those whose 1-year weight loss reached the median level of 31.4% body weight or more had higher incidence of divorce (HR = 1.58; 95% CI, 1.37-1.81; P < .001; aHR = 1.26; 95% CI, 1.09-1.44; P = .001) compared with those with weight loss of less than the median level (Figure 4B and eTable 2 in the Supplement).

Limitations

A limitation of this study is that the included cohorts were not randomized; the SOS study could not be randomized for ethical reasons, and the SOReg cohort is a register that captures patients who undergo bariatric surgery. In addition, our study only includes persons living in Sweden, and it is unknown if the results can be generalized to other countries and cultures.

Supplement.

eMethods 1 SOS study design and recruitment

eMethods 2 SOReg study design and matching

eTable 1 Multivariate analysis of relationship status in the SOS study and the SOReg/general population cohort.

eTable 2 Multivariate analysis of relationship status in the SOS study surgery group and the SOReg cohort.

1. Nguyen NT, Varela JE. Bariatric surgery for obesity and metabolic disorders: state of the art. Nat Rev Gastroenterol Hepatol. 2017;14(3):160-169. [PubMed] [Google Scholar]

2. Sjöström L.Review of the key results from the Swedish Obese Subjects (SOS) trial—a prospective controlled intervention study of bariatric surgery. J Intern Med. 2013;273(3):219-234. [PubMed] [Google Scholar]

3. Angrisani L, Santonicola A, Iovino P, Formisano G, Buchwald H, Scopinaro N. Bariatric surgery worldwide 2013. Obes Surg. 2015;25(10):1822-1832. [PubMed] [Google Scholar]

4. Coulman KD, MacKichan F, Blazeby JM, Owen-Smith A. Patient experiences of outcomes of bariatric surgery: a systematic review and qualitative synthesis. Obes Rev. 2017;18(5):547-559. [PMC free article] [PubMed] [Google Scholar]

5. Jumbe S, Bartlett C, Jumbe SL, Meyrick J. The effectiveness of bariatric surgery on long term psychosocial quality of life—a systematic review. Obes Res Clin Pract. 2016;10(3):225-242. [PubMed] [Google Scholar]

6. Raaijmakers LC, Pouwels S, Thomassen SE, Nienhuijs SW. Quality of life and bariatric surgery: a systematic review of short- and long-term results and comparison with community norms. Eur J Clin Nutr. 2017;71(4):441-449. [PubMed] [Google Scholar]

7. Karlsson J, Taft C, Rydén A, Sjöström L, Sullivan M. Ten-year trends in health-related quality of life after surgical and conventional treatment for severe obesity: the SOS intervention study. Int J Obes (Lond). 2007;31(8):1248-1261. [PubMed] [Google Scholar]

8. Svensson PA, Anveden Å, Romeo S, et al.. Alcohol consumption and alcohol problems after bariatric surgery in the Swedish obese subjects study. Obesity (Silver Spring). 2013;21(12):2444-2451. [PubMed] [Google Scholar]

9. Conason A, Teixeira J, Hsu CH, Puma L, Knafo D, Geliebter A. Substance use following bariatric weight loss surgery. JAMA Surg. 2013;148(2):145-150. [PubMed] [Google Scholar]

10. Backman O, Stockeld D, Rasmussen F, Näslund E, Marsk R. Alcohol and substance abuse, depression and suicide attempts after Roux-en-Y gastric bypass surgery. Br J Surg. 2016;103(10):1336-1342. [PubMed] [Google Scholar]

11. Arterburn DE, Courcoulas AP. Bariatric surgery for obesity and metabolic conditions in adults. BMJ. 2014;349:g3961. [PMC free article] [PubMed] [Google Scholar]

12. Neovius M, Bruze G, Jacobson P, et al.. Risk of suicide and non-fatal self-harm after bariatric surgery: results from two matched cohort studies. Lancet Diabetes Endocrinol. 2018;S2213-8587(17)30437-0. [PMC free article] [PubMed] [Google Scholar]

13. Neill JR, Marshall JR, Yale CE. Marital changes after intestinal bypass surgery. JAMA. 1978;240(5):447-458. [PubMed] [Google Scholar]

14. Rand CS, Kuldau JM, Robbins L. Surgery for obesity and marriage quality. JAMA. 1982;247(10):1419-1422. [PubMed] [Google Scholar]

15. Bocchieri LE, Meana M, Fisher BL. Perceived psychosocial outcomes of gastric bypass surgery: a qualitative study. Obes Surg. 2002;12(6):781-788. [PubMed] [Google Scholar]

16. Ferriby M, Pratt KJ, Balk E, Feister K, Noria S, Needleman B. Marriage and weight loss surgery: a narrative review of patient and spousal outcomes. Obes Surg. 2015;25(12):2436-2442. [PubMed] [Google Scholar]

17. Hedenbro JL, Näslund E, Boman L, et al.. Formation of the Scandinavian obesity surgery registry, SOReg. Obes Surg. 2015;25(10):1893-1900. [PubMed] [Google Scholar]

18. Ludvigsson JF, Almqvist C, Bonamy AK, et al.. Registers of the Swedish total population and their use in medical research. Eur J Epidemiol. 2016;31(2):125-136. [PubMed] [Google Scholar]

19. Sjöström L, Narbro K, Sjöström CD, et al.; Swedish Obese Subjects Study . Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med. 2007;357(8):741-752. [PubMed] [Google Scholar]

20. Sharman MJ, Venn AJ, Hensher M, et al.. Motivations for seeking bariatric surgery: the importance of health professionals and social networks. Bariatr Surg Pract Patient Care. 2016;11(3):104-109. [Google Scholar]

21. Munoz DJ, Lal M, Chen EY, et al.. Why patients seek bariatric surgery: a qualitative and quantitative analysis of patient motivation. Obes Surg. 2007;17(11):1487-1491. [PubMed] [Google Scholar]

22. Sogg S, Gorman MJ. Interpersonal changes and challenges after weight-loss surgery. Prim Psychiatry. 2008;15(8):61-66. [Google Scholar]

23. Aramburu Alegría C, Larsen B. Contextual care of the patient following weight-loss surgery: relational views and maintenance activities of couples. J Am Assoc Nurse Pract. 2017;29(1):17-25. [PubMed] [Google Scholar]

24. Moore DD, Cooper CE. Life after bariatric surgery: perceptions of male patients and their intimate relationships. J Marital Fam Ther. 2016;42(3):495-508. [PubMed] [Google Scholar]

25. Clark SM, Saules KK, Schuh LM, Stote J, Creel DB. Associations between relationship stability, relationship quality, and weight loss outcomes among bariatric surgery patients. Eat Behav. 2014;15(4):670-672. [PubMed] [Google Scholar]

26. Kinzl JF, Traweger C, Trefalt E, Biebl W. Psychosocial consequences of weight loss following gastric banding for morbid obesity. Obes Surg. 2003;13(1):105-110. [PubMed] [Google Scholar]

27. Kinzl JF, Trefalt E, Fiala M, Hotter A, Biebl W, Aigner F. Partnership, sexuality, and sexual disorders in morbidly obese women: consequences of weight loss after gastric banding. Obes Surg. 2001;11(4):455-458. [PubMed] [Google Scholar]

28. Pories ML, Hodgson J, Rose MA, Pender J, Sira N, Swanson M. Following bariatric surgery: an exploration of the couples’ experience. Obes Surg. 2016;26(1):54-60. [PubMed] [Google Scholar]

29. Slotman GJ.Gastric bypass: a family affair—41 families in which multiple members underwent bariatric surgery. Surg Obes Relat Dis. 2011;7(5):592-598. [PubMed] [Google Scholar]

30. Romo LK, Dailey RM. Weighty dynamics: exploring couples’ perceptions of post-weight-loss interaction. Health Commun. 2014;29(2):193-204. [PubMed] [Google Scholar]