- Email: [email protected]
The impact of history of exposure to abuse on outcomes after bariatric surgery: data from the Ontario Bariatric Registry
1 2 3 Surgery for Obesity and Related Diseases ] (2016) 00–00 4 5 6 Original article 7 8 Q3 9 10 11 Jennifer M. Hensel, M.D., M.Sc.a,*, Keren Grosman Kaplan, M.D.b,c, 12 13 Q1 Mehran Anvari, M.B., B.S., Ph.D.d, Valerie H. Taylor, M.D., Ph.D.a a 14 Department of Psychiatry, Women’s College Hospital & University of Toronto, Toronto, Ontario, Canada b Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada 15 c Department of Psychiatry, Emek Medical Center, Afula, Israel 16 d Department of Surgery, McMaster University, Hamilton, Ontario, Canada 17 Received January 1, 2016; accepted March 15, 2016 18 19 20 Abstract Background: Studies worldwide have reported an increased prevalence of abuse histories among 21 bariatric surgery candidates. The impact of abuse history on weight loss after surgery has not been 22 examined in Canada. 23 Objectives: Determine the prevalence of abuse and its impact on postoperative outcomes in 24 Ontario, Canada. 25 Setting: Data from the Ontario Bariatric Registry. 26 Methods: A retrospective cohort study of laparoscopic gastric bypass and sleeve gastrectomy surgeries from 2010 to 2014, for which any follow-up data were available (N ¼ 6016). Weight loss 27 outcomes at 3 months (n ¼ 5147), 6 months (n ¼ 4749), and 1 year (n ¼ 4024) were compared 28 between those with and without a self-reported history of any of emotional, physical or sexual abuse 29 and those with and without a history of sexual abuse specifically. Mixed repeated measures models 30 were adjusted for age, sex, type of surgery, and baseline body mass index. One-year postoperative 31 occurrence of revisions or repairs, hospitalization, and death were also examined. 32 Results: The prevalence of documented abuse was 21.5%. Emotional abuse was most common 33 (13.1%), followed by sexual abuse (10.6%), then physical abuse (8.9%). There was no significant 34 association between presence of abuse history and weight loss at any time point in repeated 35 measures analyses. 36 Conclusion: Abuse histories are common in bariatric surgery candidates in Ontario, but at a lower 37 prevalence than what has been reported elsewhere. History of abuse does not appear to affect weight loss out to 1 year postoperatively and may alert providers to offer additional support perioperatively 38 and postoperatively, particularly in the setting of psychiatric co-morbidity. (Surg Obes Relat Dis 39 2016;]:00–00.) r 2016 American Society for Metabolic and Bariatric Surgery. All rights reserved. 40 41 42 Keywords: Bariatric surgery; Weight loss; Abuse; Psychiatric; Suicide 43 44 45 46 A history of early life abuse is associated with higher mechanisms for this association include overweight as a defense 47 body mass index (BMI) in adulthood [1–4]. Hypothesized mechanism against future abuse, higher incidence of psychiatric 48 co-morbidity, lifestyle factors such as substance use and 49 sedentary behavior, and worse social determinants of health * 50 Correspondence. Jennifer M. Hensel, M.D., M.Sc., Women’s College [2,5–7]. Sexual abuse has been the most studied form of abuse Hospital, 76 Grenville St., Toronto, Ontario, M5 S1 B2 Canada. 51 in relation to obesity [1,8] and has the strongest association E-mail: [email protected] 52 53 http://dx.doi.org/10.1016/j.soard.2016.03.016 54 1550-7289/r 2016 American Society for Metabolic and Bariatric Surgery. All rights reserved. 55
The impact of history of exposure to abuse on outcomes after bariatric surgery: data from the Ontario Bariatric Registry
56 57 58 59 60 61 62 63 64 65 66
2
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
relative to other forms of abuse [4]. This has resulted in 2 widely held beliefs about the relationship between bariatric surgery and sexual abuse: (1) There are higher rates of sexual abuse in individuals seeking surgery than there are in the general population, and (2) those with a history of sexual abuse have poorer weight loss outcomes after surgery [6,7]. As a result, the majority of published research on abuse and bariatric surgery focuses on sexual abuse. A recent systematic review documents the prevalence of sexual abuse among bariatric surgery candidates between 15% and 32% [9]. Detection depends on the evaluation method, and underreporting is likely prevalent [9]. An analysis of 8 published studies examining weight loss after surgery concluded that initial weight loss may be slightly lower among people with sexual abuse histories, but over time it reaches the same level as the nonabused group [9]. Although these data support that a history of sexual abuse should not preclude surgery [10], it is noteworthy that among surgical candidates, a history of sexual abuse has been positively associated with more mental health co-morbidities [11], postoperative psychiatric hospitalization [12], and suicidal ideation [13]. There have been few studies to date on these latter outcomes, and there is much less published on the prevalence and impact of other forms of abuse that are more prevalent in the general population [14] and also highly associated with obesity [2,3]. Moreover, there is a dearth of published data on this topic in Canada. Consistent with studies elsewhere, one Canadian study found that 21.8% of individuals enrolled in a provincial obesity program reported a history of sexual abuse [15], but not all patients in this program received surgery. To our knowledge, no Canadian studies have examined prevalence of abuse among people undergoing bariatric surgery. Ontario, Canada’s most populous province, is the site of approximately half of the bariatric surgeries performed annually across Canada [16]. In 2009, the Ontario Bariatric Registry (OBR) was established as part of the Bariatric Services Strategy outlined by Ontario’s Ministry of Health and Long Term Care to address obesity and obesity-related illness, which accounts for an estimated 1%–3% of all health expenditures [16], and 4C$7 billion in direct and indirect costs annually [17]. This study used OBR data to examine the prevalence of abuse, including sexual, physical, and emotional abuse; its impact on weight loss at 3, 6, and 12 months postoperative; and its relationship to hospitalizations and death at 1 year.
Materials and methods
at participating centers in Ontario [18]. Clinical charts for all consenting patients are reviewed by a trained chart reviewer who completes a data collection form, which is then entered into the OBR. The OBR has research ethics approval from all participating sites, and no patient identifiers are collected. We conducted a secondary analysis of de-identified data from the OBR for individuals undergoing a nonrevision laparoscopic Roux-en-Y gastric bypass (LRYGB) or sleeve gastrectomy (SG) procedure between April 1, 2010 and March 31, 2014 (N ¼ 6315). SG and LRYGB procedures comprise 495% of surgeries recorded in the OBR within this time frame. Before surgery, all individuals undergo a detailed demographic profile and medical review including measurement of height and weight (baseline assessment) and a screening assessment for the presence of mental health conditions and abuse (psychological assessment). In the first postoperative year, data are entered in the OBR for 3-month, 6-month, and 1-year follow-up visits. At all visits, individuals are weighed by a clinician. Our cohort consisted of all individuals who were older than 18 years of age at the time of baseline assessment, completed both assessments, and had data available for at least 1 of the follow-up time points (n ¼ 6016, 95.3% of LRYGB and SG surgeries). Deidentified data for the cohort was provided to the study investigators as a password-protected Excel file. History of abuse and psychiatric illness Lifetime history of abuse was assessed during the psychological assessment, a clinical interview completed as part of the workup for surgery. This assessment was completed by a social worker (67%), psychologist (17%), psychiatrist (3%), nurse (2%), or “other” healthcare professional (11%). In more recent years of the OBR, there has been a trend toward more assessments being completed by social workers and psychologists and fewer by the other health disciplines. Surgery candidates are asked directly about a history of abuse, but no standardized assessment tools are used. If an individual screens positive for any abuse, he or she is asked specifically about physical, emotional, and sexual abuse. In the chart review used to capture data for the OBR, lifetime exposure to each type of abuse is recorded on the OBR data collection form as present or absent. In this study, we created indicators for exposure to any abuse, defined as any of the 3 types of abuse, and for each type of abuse separately. A referent group was composed of those who had no exposure to any of the 3 types of abuse. Past or present psychiatric illness identified during the clinical interview is also collected as a binary variable.
Design and data sources Data for this study were retrieved from the OBR. The OBR is a multisite, observational registry containing the data for all consenting patients who receive bariatric surgery
Data analysis We examined baseline demographic and weight variables for those with any type of abuse (physical, emotional or sexual) and
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
Q4
those with sexual abuse (with or without another type of abuse), both compared with the referent group of individuals with no abuse reported using independent t tests (continuous variables) or χ2 statistics (categorical variables). A separate analysis for sexual abuse was conducted to provide comparison with other studies that have focused exclusively on this type of abuse in the bariatric surgery population. Baseline variables included (1) age, (2) sex, (3) BMI, and (4) presence of past or present psychiatric illness. We categorized individuals based on their baseline BMI into BMI 460 or r60. This cutoff was selected to differentiate those who would be more likely to be offered SG over LRYGB [19]. We examined outcomes at 3 months, 6 months, and 1 year postoperative using data from Similarly, we compared (1) change in BMI, (2) percent of total weight loss (%TWL), and (3) percent of excess weight loss (%EWL) at 3 months, 6 months, and 1 year postoperative. Percent TWL and %EWL were calculated using the following formulas: % TWLt ¼
baseline weightweightt 100 baseline weight
% EWLt ¼
baseline BMIBMIt baseline BMI25
100
where t ¼ the time point of measurement (e.g., 3 mo). In the first postoperative year, we also examined (1) any revision or repair, (2) all cause hospitalization, (3) hospitalization for complications of surgery, (4) hospitalization for any other reason, (5) psychiatric hospitalizations, (6) all cause death, and (7) death by suicide. Because of their rarity, these outcomes were aggregated across the entire year, and as such, if any time point was missing for an individual, that time point was recorded as no event. This may result in underestimation of the outcome but allows inclusion of individuals who may have had an outcome at one time point and no data available at the other time points. We conducted a general linear mixed model analysis with time as a repeated measure and %EWL as the outcome. Sex, type of surgery (LRYGB or SG), and abuse history were fixed
3
factors, and age and baseline BMI were entered as co-variates. Percent EWL varies significantly by baseline BMI [20], which should be controlled for in any statistical models using this outcome. There were no associations between missing data at any time point and history of abuse. We used maximum likelihood estimation to calculate marginal means at each time point for those with and without a history of abuse. We evaluated statistically significant main effects of time and history of abuse, as well as the interaction between these variables. The analysis was performed for the entire cohort and separately for the subset with either a history of sexual abuse or no history of abuse. All analyses were conducted with the Statistical Package for the Social Sciences, version 23 (IBM Corp., Armonk, NY, USA). Results A history of physical, sexual or emotional abuse was documented in 1293 individuals (21.5%) (Table 1). Emotional abuse was the most common in 13.1%, sexual abuse was reported by 10.6%, and physical abuse by 8.9%. All 3 types of abuse were documented in 2.8% of the cohort. There was a linear trend in rates of documented abuse over time, rising from 16.0% in 2010–2011 to 30.9% in 2013– 2014 for any type of abuse, and 9.5% in 2010–2011 to 15.3% in 2013–2014 for sexual abuse. Table 1 summarizes baseline variables across groups. Individuals with abuse were more significantly more likely to be female and have a psychiatric co-morbidity than those with no history of abuse, but these groups did not differ significantly in baseline BMI. Relative to the no abuse group, there was a slightly higher proportion in both abuse categories who had a baseline BMI 460 and underwent SG. A summary of the outcomes is reported in Table 2. Small bivariate differences were found in weight loss at each time point between those with any abuse and no abuse. Revision or repair and hospitalizations at 1 year did not differ significantly
Table 1 Baseline variables by abuse exposure history Baseline variable
No. of patients (%) Age, mean (SD) Female gender, n (%) Baseline BMI, mean (SD) BMI 460, n (%) Any mental health co-morbidity, n (%) Surgery LRYGB, n (%) SG, n (%)
Exposure to abuse None reported
Any type of abuse
Sexual abuse
4,723 (78.5) 44.2 (10.4) 3829 (81.1) 49.4 (7.9) 472 (10.0) 2025 (42.9)
1293 (21.5) 44.9 (10.1)* 1178 (91.1)† 49.8 (8.1) 155 (12.0)* 890 (68.8)†
640 (10.6) 44.9 (9.9) 605 (94.5)† 49.9 (8.2) 82 (12.8)* 451 (70.5)†
4204 (89.0) 519 (11.0)
1097 (84.8) 196 (15.2)†
547 (85.5) 93 (14.5)‡
BMI ¼ body mass index; LRYGB ¼ laparoscopic Roux-en-Y gastric bypass; SD ¼ standard deviation; SG ¼ sleeve gastrectomy. The group with no abuse reported is the referent category for all statistical comparisons. * P o .05. ‡ P o .01. † P o .001.
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 T1250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 T2265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
4
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
Table 2 Weight loss, revisions, and hospitalization outcomes at 3 months, 6 months, and 1 year postoperative by abuse exposure Outcome
No abuse reported
Any type of abuse
Sexual abuse
3 mo, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) 6 mo, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) 1 yr, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) Overall first postoperative yr , n Any revision or repair, n (%)‡,§,¶ Hospitalization for any reason, n (%)§,¶ Hospitalization for complications, n (%)‖ Hospitalization for other reason, n (%)‖ Psychiatric hospitalization, n (%)¶ All cause death, n (%)¶ Suicide, n (%)¶
4033 –9.7 (3.1) 19.6 (5.3) 41.5 (13.2) 3697 –13.6 (3.9) 27.6 (6.3) 58.5 (16.1) 3125 –16.7 (5.1) 33.6 (8.2) 71.1 (19.5) 4723 33 (.7) 683 (14.5) 512 (75.0) 208 (30.5) 5 (.1) 6 (.1) 0 (.0)
1106 –9.5 (3.2) 19.0 (5.5)* 40.2 (13.8)* 1046 –13.5 (3.8) 27.1 (6.4)† 57.1 (16.4)† 893 –16.5 (4.9) 33.1 (8.3) 69.6 (20.1)† 1293 8 (.6) 193 (14.9) 141 (73.1) 65 (33.7) 2 (.2) 4 (.3) 0 (.0)
548 –9.6 (3.4) 19.2 (5.8) 40.6 (14.6) 496 –13.8 (4.1) 27.5 (6.6) 58.0 (16.9) 437 –16.8 (4.9) 33.8 (8.3) 71.1 (20.4) 640 7 (1.1) 101 (15.8) 76 (75.2) 31 (30.7) 2 (.3) 3 (.5) 0 (.0)
ΔBMI ¼ change in body mass index from baseline; %EWL ¼ % of excess weight loss; SD ¼ standard deviation; %TWL ¼ % of total weight loss. The group with no abuse reported is the referent for all statistical comparisons. * P o .01. † P o .05. ‡ Any revision or repair documented at any time point in first postoperative year. § If data were missing for any time point in the first postoperative year, it was coded as no revision or repair and no hospitalization, as applicable. ¶ Calculated as proportion of entire abuse category. ‖ Calculated as proportion of all hospitalizations within abuse category.
F1
between groups. Psychiatric hospitalizations occurred in only 7 individuals, and death in 10. There were no suicides. In the mixed models for %EWL, there was no significant main effect of exposure to abuse (F[1, 5037.4] ¼ 1.99, P ¼ .16 and F[1, 4506.1] ¼ .77, P ¼ .38, for any abuse and sexual abuse, respectively), and no significant interaction between history of abuse and time (F[2, 4268.9] ¼ .33, P ¼ .72 and F[2, 3737.6] ¼ .73, P ¼ .48, for any abuse and sexual abuse, respectively). In both models, there was a significant main effect of time (F[2, 4268.9] ¼ 5954.2, P o .001 and F[2, 3737.5] ¼ 3334.6, P o .001, for any abuse and sexual abuse, respectively) (Fig. 1A and B). Discussion We examined the prevalence of emotional, sexual, and physical abuse among a 4-year cohort of individuals undergoing bariatric surgery over 4 years in Ontario, Canada, and its impact on weight loss up to 1-year postoperatively. Overall, 4 1 in 5 individuals (21.5%) had a documented abuse history; however, the prevalence rose substantially over time with almost 1 in 3 individuals having a history of abuse documented in 2013–2014. Emotional abuse was the most common, followed by sexual and then physical abuse. Those with a history of abuse were slightly more likely to have a baseline BMI 460 and
undergo SG compared with LRYGB. In the adjusted repeated measures analysis, there was no impact of abuse history on %EWL over time. Rates of revision or repair and postoperative hospitalization between those with and without a history of abuse did not differ. These first-of-a-kind results from Canada find a lower prevalence of documented sexual abuse compared with what has been reported in most other studies of bariatric surgery candidates [9]. Our rate is similar to a national population-based study that found a rate of childhood sexual abuse equal to 10% among survey respondents [14]. Even in the last year of our study data, with the highest rate of sexual abuse, it was below what was found in a systematic review of the published literature [9], and 45% below what was reported among a sample of individuals attending an obesity program elsewhere in Canada [15]. In this latter study the assessment of abuse was done independent of any clinical care, and only a proportion of those individuals go on to have surgery [15]. This difference in methodology and the outcomes observed may suggest 2 possibilities: (1) when obesity co-occurs with an abuse history, there may be a lower likelihood of receiving surgery, which could be patient or provider driven; and/or (2) disclosure may be inhibited when abuse history is assessed as part of clinical management because of fear of not being offered surgery or mistrust of the medical team.
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
5
Fig. 1. Marginal means for percentage of excess weight loss (%EWL) at 3 months, 6 months, and 1 year, comparing (A) those with exposure to any type of abuse versus no abuse and (B) those with exposure to sexual abuse versus no abuse. BMI ¼ body mass index; LRYGB ¼ laparoscopic Roux-en-Y gastric bypass; SG ¼ sleeve gastrectomy.
A. Marginal means evaluated at age = 44.8 and baseline BMI = 49.5 Time point
3 mo
6 mo
1 yr
n (% of cohort)
5139 (85.4)
4743 (78.8)
4018 (66.8)
B. Marginal means evaluated at age = 44.7 and baseline BMI = 49.5 Time point
3 mo
6 mo
1 yr
n (% of cohort)
4581 (85.4)
4193 (78.2)
3562 (66.4)
Error bars are 95% confidence intervals. Marginal means were calculated with mixed models with time as a repeated measure. Models are adjusted for age, sex, type of surgery (LRYGB or SG), and baseline BMI. The rise in prevalence of documented abuse over time may reflect a growing recognition, supported by an emerging literature, that a history of abuse is less likely to affect weight loss outcomes than was once believed [6,7,9]. Alternatively, screening processes have improved and abuse is more often being assessed or disclosed, hopefully with concurrent supports in place to address issues that may arise as people undergo surgery. Although underreporting may still be present, our data do support the growing literature that a history of abuse, including sexual abuse, does not affect early and later weight loss outcomes. Similar to what has been previously reported elsewhere [6,11], we found that individuals with abuse histories were more likely to have preoperative psychiatric co-morbidity, as would be expected given the increased susceptibility to psychiatric problems with early life adversity [14]. Despite this higher rate of psychiatric co-morbidity, weight loss outcomes did not differ. Furthermore, those with abuse histories were no more likely to be hospitalized in the first
postoperative year. We also found that people with abuse histories more often received the less-invasive SG procedure. This discrepancy in the type of surgery received was partially accounted for by a higher number of these individuals with a baseline BMI 460, an indication for this procedure; however, this did not entirely explain the discrepancy. We did not examine other indications for this procedure, such as the presence of certain medical co-morbidities. A limitation of this study is that we examined those who underwent surgery only and did not assess whether or not a history of abuse was a barrier to receiving bariatric surgery. History of abuse was assessed by a variety of healthcare providers using an unstandardized clinical assessment. We observed that the rates of abuse recorded among social worker, psychologist, and psychiatrist assessors were similar, with significantly lower rates recorded among nurses and “other” healthcare professionals. These latter disciplines, possibly with less mental health training, who do a minority of the psychological assessments, may not
452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
6
507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
screen for abuse as thoroughly. Conversely, patients with known mental health histories may be more likely to be assessed by a trained mental health professional, in which case the variability in the rates of abuse would reflect true differences. It is also possible that some surgery candidates may choose not to disclose a history of abuse. For example, studies examining rates of disclosure of psychiatric symptoms have documented that individuals will minimize or deny active symptoms to receive bariatric surgery [21,22]. The rising rate of abuse over time supports that disclosure may be increasing with improved screening and fewer perceived barriers, but our rates were still lower than what has been reported elsewhere [9,15]. A proportion of data were missing for all time points, and outcome data collected were based on clinical documents. This implies that some hospitalizations and deaths may not have been identified, and rates may be underestimates. We have only reported outcomes out to 1 year postoperative, and the study of longer-term outcomes is needed in Canada. For example, a recent Ontario population-level study examined patients presenting for medical care for self-harm attempts up to 3 years after bariatric surgery and found that these increased 1.54 times compared with a referent 3-year period before surgery [23]. Conclusion The findings in this study, based on first-of-a-kind data from Canada, do not support the hypothesis that history of abuse is associated with worse outcomes. With a heightened risk for obesity and obesity-related illness among survivors of abuse, this treatment can be an effective and life-altering intervention for these individuals. There is, however, a high burden of psychiatric co-morbidity in this population, the recognition and management of which will likely facilitate the optimization of surgical outcomes among survivors of abuse. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. Acknowledgments Data used in the study were obtained from the Ontario Bariatric Registry [18]. The registry is managed by Ruth Breau and the team at the Centre for Surgical Invention and Innovation, with analysis and data management provided by the Population Health Research Institute. References [1] Irish L, Kobayashi I, Delahanty DL. Long-term physical health consequences of childhood sexual abuse: a meta-analytic review. J Pediatr Psychol 2010;35(5):450–61. [2] Min MO, Minnes S, Kim H, Singer LT. Pathways linking childhood maltreatment and adult physical health. Child Abuse Negl 2013;37 (6):361–73.
[3] Hemmingsson E, Johansson K, Reynisdottir S. Effects of childhood abuse on adult obesity: a systematic review and meta-analysis. Obes Rev 2014;15(11):882–93. [4] Danese A, Tan M. Childhood maltreatment and obesity: systematic review and meta-analysis. Mol Psychiatry 2014;19(5):544–54. [5] Buser A, Dymek-Valentine M, Hilburger J, Alverdy J. Outcome following gastric bypass surgery: impact of past sexual abuse. Obes Surg 2004;14(2):170–4. [6] Mahony D. Assessing sexual abuse/attack histories with bariatric surgery patients. J Child Sex Abus 2010;19(4):469–84. [7] Buser AT, Lam CS, Poplawski SC. A long-term cross-sectional study on gastric bypass surgery: impact of self-reported past sexual abuse. Obes Surg 2009;19(4):422–6. [8] Wilson DR. Health consequences of childhood sexual abuse. Perspect Psychiatr Care 2010;46(1):56–64. [9] Steinig J, Wagner B, Shang E, Dolemeyer R, Kersting A. Sexual abuse in bariatric surgery candidates: impact on weight loss after surgery: a systematic review. Obes Rev 2012;13(10):892–901. [10] Adams ST, Salhab M, Hussain ZI, Miller GV, Leveson SH. Roux-en-Y gastric bypass for morbid obesity: what are the preoperative predictors of weight loss? Postgrad Med J 2013;89(1053):411–6;quiz 5, 6. [11] Wildes JE, Kalarchian MA, Marcus MD, Levine MD, Courcoulas AP. Childhood maltreatment and psychiatric morbidity in bariatric surgery candidates. Obes Surg 2008;18(3):306–13. [12] Clark MM, Hanna BK, Mai JL, et al. Sexual abuse survivors and psychiatric hospitalization after bariatric surgery. Obes Surg 2007;17 (4):465–9. [13] Chen EY, Fettich KC, Tierney M, et al. Factors associated with suicide ideation in severely obese bariatric surgery-seeking individuals. Suicide Life Threat Behav 2012;42(5):541–9. [14] Afifi TO, MacMillan HL, Boyle M, Taillieu T, Cheung K, Sareen J. Child abuse and mental disorders in Canada. CMAJ 2014;186(9):E324–32. [15] Gabert DL, Majumdar SR, Sharma AM, et al. Prevalence and predictors of self-reported sexual abuse in severely obese patients in a population-based bariatric program. J Obes 2013;2013:374050. [16] Bariatric Surgery in Canada—Report May 2014 [monograph on the Internet]. Canada: Canadian Institute of Health Information; c2015 [cited 2015 May 28]. Available from: https://secure.cihi.ca/free_ products/Bariatric_Surgery_in_Canada_EN.pdf. [17] Janssen I. The public health burden of obesity in Canada. Can J Diabetes. 2013;37(2):90–6. [18] Anvari M, Sharma A, Yusuf S, et al. Bariatric Registry. Registry data produced and distributed by the Population Health Research Institute and the Centre for Surgical Invention and Innovation, supported by the Ministry of Health and Long Term Care. Ontario, Canada: Ministry of Health and Long Term Care; 2015. [19] Karmali S, Schauer P, Birch D, Sharma AM, Sherman V. Laparoscopic sleeve gastrectomy: an innovative new tool in the battle against the obesity epidemic in Canada. Can J Surg 2010;53(2):126–32. [20] van de Laar A, de Caluwe L, Dillemans B. Relative outcome measures for bariatric surgery. Evidence against excess weight loss and excess body mass index loss from a series of laparoscopic Rouxen-Y gastric bypass patients. Obes Surg 2011;21(6):763–7. [21] Fabricatore AN, Sarwer DB, Wadden TA, Combs CJ, Krasucki JL. Impression management or real change? Reports of depressive symptoms before and after the preoperative psychological evaluation for bariatric surgery. Obes Surg 2007;17(9):1213–9. [22] Ambwani S, Boeka AG, Brown JD, et al. Socially desirable responding by bariatric surgery candidates during psychological assessment. Surg Obes Relat Dis 2013;9(2):300–5. [23] Bhatti JA, Nathens AB, Thiruchelvam D, Grantcharov T, Goldstein BI, Redelmeier DA. Self-harm emergencies after bariatric surgery: a population-based cohort study. JAMA Surg 2015:1–7.
562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616
The impact of history of exposure to abuse on outcomes after bariatric surgery: data from the Ontario Bariatric Registry
56 57 58 59 60 61 62 63 64 65 66
2
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
relative to other forms of abuse [4]. This has resulted in 2 widely held beliefs about the relationship between bariatric surgery and sexual abuse: (1) There are higher rates of sexual abuse in individuals seeking surgery than there are in the general population, and (2) those with a history of sexual abuse have poorer weight loss outcomes after surgery [6,7]. As a result, the majority of published research on abuse and bariatric surgery focuses on sexual abuse. A recent systematic review documents the prevalence of sexual abuse among bariatric surgery candidates between 15% and 32% [9]. Detection depends on the evaluation method, and underreporting is likely prevalent [9]. An analysis of 8 published studies examining weight loss after surgery concluded that initial weight loss may be slightly lower among people with sexual abuse histories, but over time it reaches the same level as the nonabused group [9]. Although these data support that a history of sexual abuse should not preclude surgery [10], it is noteworthy that among surgical candidates, a history of sexual abuse has been positively associated with more mental health co-morbidities [11], postoperative psychiatric hospitalization [12], and suicidal ideation [13]. There have been few studies to date on these latter outcomes, and there is much less published on the prevalence and impact of other forms of abuse that are more prevalent in the general population [14] and also highly associated with obesity [2,3]. Moreover, there is a dearth of published data on this topic in Canada. Consistent with studies elsewhere, one Canadian study found that 21.8% of individuals enrolled in a provincial obesity program reported a history of sexual abuse [15], but not all patients in this program received surgery. To our knowledge, no Canadian studies have examined prevalence of abuse among people undergoing bariatric surgery. Ontario, Canada’s most populous province, is the site of approximately half of the bariatric surgeries performed annually across Canada [16]. In 2009, the Ontario Bariatric Registry (OBR) was established as part of the Bariatric Services Strategy outlined by Ontario’s Ministry of Health and Long Term Care to address obesity and obesity-related illness, which accounts for an estimated 1%–3% of all health expenditures [16], and 4C$7 billion in direct and indirect costs annually [17]. This study used OBR data to examine the prevalence of abuse, including sexual, physical, and emotional abuse; its impact on weight loss at 3, 6, and 12 months postoperative; and its relationship to hospitalizations and death at 1 year.
Materials and methods
at participating centers in Ontario [18]. Clinical charts for all consenting patients are reviewed by a trained chart reviewer who completes a data collection form, which is then entered into the OBR. The OBR has research ethics approval from all participating sites, and no patient identifiers are collected. We conducted a secondary analysis of de-identified data from the OBR for individuals undergoing a nonrevision laparoscopic Roux-en-Y gastric bypass (LRYGB) or sleeve gastrectomy (SG) procedure between April 1, 2010 and March 31, 2014 (N ¼ 6315). SG and LRYGB procedures comprise 495% of surgeries recorded in the OBR within this time frame. Before surgery, all individuals undergo a detailed demographic profile and medical review including measurement of height and weight (baseline assessment) and a screening assessment for the presence of mental health conditions and abuse (psychological assessment). In the first postoperative year, data are entered in the OBR for 3-month, 6-month, and 1-year follow-up visits. At all visits, individuals are weighed by a clinician. Our cohort consisted of all individuals who were older than 18 years of age at the time of baseline assessment, completed both assessments, and had data available for at least 1 of the follow-up time points (n ¼ 6016, 95.3% of LRYGB and SG surgeries). Deidentified data for the cohort was provided to the study investigators as a password-protected Excel file. History of abuse and psychiatric illness Lifetime history of abuse was assessed during the psychological assessment, a clinical interview completed as part of the workup for surgery. This assessment was completed by a social worker (67%), psychologist (17%), psychiatrist (3%), nurse (2%), or “other” healthcare professional (11%). In more recent years of the OBR, there has been a trend toward more assessments being completed by social workers and psychologists and fewer by the other health disciplines. Surgery candidates are asked directly about a history of abuse, but no standardized assessment tools are used. If an individual screens positive for any abuse, he or she is asked specifically about physical, emotional, and sexual abuse. In the chart review used to capture data for the OBR, lifetime exposure to each type of abuse is recorded on the OBR data collection form as present or absent. In this study, we created indicators for exposure to any abuse, defined as any of the 3 types of abuse, and for each type of abuse separately. A referent group was composed of those who had no exposure to any of the 3 types of abuse. Past or present psychiatric illness identified during the clinical interview is also collected as a binary variable.
Design and data sources Data for this study were retrieved from the OBR. The OBR is a multisite, observational registry containing the data for all consenting patients who receive bariatric surgery
Data analysis We examined baseline demographic and weight variables for those with any type of abuse (physical, emotional or sexual) and
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
Q4
those with sexual abuse (with or without another type of abuse), both compared with the referent group of individuals with no abuse reported using independent t tests (continuous variables) or χ2 statistics (categorical variables). A separate analysis for sexual abuse was conducted to provide comparison with other studies that have focused exclusively on this type of abuse in the bariatric surgery population. Baseline variables included (1) age, (2) sex, (3) BMI, and (4) presence of past or present psychiatric illness. We categorized individuals based on their baseline BMI into BMI 460 or r60. This cutoff was selected to differentiate those who would be more likely to be offered SG over LRYGB [19]. We examined outcomes at 3 months, 6 months, and 1 year postoperative using data from Similarly, we compared (1) change in BMI, (2) percent of total weight loss (%TWL), and (3) percent of excess weight loss (%EWL) at 3 months, 6 months, and 1 year postoperative. Percent TWL and %EWL were calculated using the following formulas: % TWLt ¼
baseline weightweightt 100 baseline weight
% EWLt ¼
baseline BMIBMIt baseline BMI25
100
where t ¼ the time point of measurement (e.g., 3 mo). In the first postoperative year, we also examined (1) any revision or repair, (2) all cause hospitalization, (3) hospitalization for complications of surgery, (4) hospitalization for any other reason, (5) psychiatric hospitalizations, (6) all cause death, and (7) death by suicide. Because of their rarity, these outcomes were aggregated across the entire year, and as such, if any time point was missing for an individual, that time point was recorded as no event. This may result in underestimation of the outcome but allows inclusion of individuals who may have had an outcome at one time point and no data available at the other time points. We conducted a general linear mixed model analysis with time as a repeated measure and %EWL as the outcome. Sex, type of surgery (LRYGB or SG), and abuse history were fixed
3
factors, and age and baseline BMI were entered as co-variates. Percent EWL varies significantly by baseline BMI [20], which should be controlled for in any statistical models using this outcome. There were no associations between missing data at any time point and history of abuse. We used maximum likelihood estimation to calculate marginal means at each time point for those with and without a history of abuse. We evaluated statistically significant main effects of time and history of abuse, as well as the interaction between these variables. The analysis was performed for the entire cohort and separately for the subset with either a history of sexual abuse or no history of abuse. All analyses were conducted with the Statistical Package for the Social Sciences, version 23 (IBM Corp., Armonk, NY, USA). Results A history of physical, sexual or emotional abuse was documented in 1293 individuals (21.5%) (Table 1). Emotional abuse was the most common in 13.1%, sexual abuse was reported by 10.6%, and physical abuse by 8.9%. All 3 types of abuse were documented in 2.8% of the cohort. There was a linear trend in rates of documented abuse over time, rising from 16.0% in 2010–2011 to 30.9% in 2013– 2014 for any type of abuse, and 9.5% in 2010–2011 to 15.3% in 2013–2014 for sexual abuse. Table 1 summarizes baseline variables across groups. Individuals with abuse were more significantly more likely to be female and have a psychiatric co-morbidity than those with no history of abuse, but these groups did not differ significantly in baseline BMI. Relative to the no abuse group, there was a slightly higher proportion in both abuse categories who had a baseline BMI 460 and underwent SG. A summary of the outcomes is reported in Table 2. Small bivariate differences were found in weight loss at each time point between those with any abuse and no abuse. Revision or repair and hospitalizations at 1 year did not differ significantly
Table 1 Baseline variables by abuse exposure history Baseline variable
No. of patients (%) Age, mean (SD) Female gender, n (%) Baseline BMI, mean (SD) BMI 460, n (%) Any mental health co-morbidity, n (%) Surgery LRYGB, n (%) SG, n (%)
Exposure to abuse None reported
Any type of abuse
Sexual abuse
4,723 (78.5) 44.2 (10.4) 3829 (81.1) 49.4 (7.9) 472 (10.0) 2025 (42.9)
1293 (21.5) 44.9 (10.1)* 1178 (91.1)† 49.8 (8.1) 155 (12.0)* 890 (68.8)†
640 (10.6) 44.9 (9.9) 605 (94.5)† 49.9 (8.2) 82 (12.8)* 451 (70.5)†
4204 (89.0) 519 (11.0)
1097 (84.8) 196 (15.2)†
547 (85.5) 93 (14.5)‡
BMI ¼ body mass index; LRYGB ¼ laparoscopic Roux-en-Y gastric bypass; SD ¼ standard deviation; SG ¼ sleeve gastrectomy. The group with no abuse reported is the referent category for all statistical comparisons. * P o .05. ‡ P o .01. † P o .001.
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 T1250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 T2265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
4
287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
Table 2 Weight loss, revisions, and hospitalization outcomes at 3 months, 6 months, and 1 year postoperative by abuse exposure Outcome
No abuse reported
Any type of abuse
Sexual abuse
3 mo, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) 6 mo, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) 1 yr, n ΔBMI, mean (SD) %TWL, mean (SD) %EWL, mean (SD) Overall first postoperative yr , n Any revision or repair, n (%)‡,§,¶ Hospitalization for any reason, n (%)§,¶ Hospitalization for complications, n (%)‖ Hospitalization for other reason, n (%)‖ Psychiatric hospitalization, n (%)¶ All cause death, n (%)¶ Suicide, n (%)¶
4033 –9.7 (3.1) 19.6 (5.3) 41.5 (13.2) 3697 –13.6 (3.9) 27.6 (6.3) 58.5 (16.1) 3125 –16.7 (5.1) 33.6 (8.2) 71.1 (19.5) 4723 33 (.7) 683 (14.5) 512 (75.0) 208 (30.5) 5 (.1) 6 (.1) 0 (.0)
1106 –9.5 (3.2) 19.0 (5.5)* 40.2 (13.8)* 1046 –13.5 (3.8) 27.1 (6.4)† 57.1 (16.4)† 893 –16.5 (4.9) 33.1 (8.3) 69.6 (20.1)† 1293 8 (.6) 193 (14.9) 141 (73.1) 65 (33.7) 2 (.2) 4 (.3) 0 (.0)
548 –9.6 (3.4) 19.2 (5.8) 40.6 (14.6) 496 –13.8 (4.1) 27.5 (6.6) 58.0 (16.9) 437 –16.8 (4.9) 33.8 (8.3) 71.1 (20.4) 640 7 (1.1) 101 (15.8) 76 (75.2) 31 (30.7) 2 (.3) 3 (.5) 0 (.0)
ΔBMI ¼ change in body mass index from baseline; %EWL ¼ % of excess weight loss; SD ¼ standard deviation; %TWL ¼ % of total weight loss. The group with no abuse reported is the referent for all statistical comparisons. * P o .01. † P o .05. ‡ Any revision or repair documented at any time point in first postoperative year. § If data were missing for any time point in the first postoperative year, it was coded as no revision or repair and no hospitalization, as applicable. ¶ Calculated as proportion of entire abuse category. ‖ Calculated as proportion of all hospitalizations within abuse category.
F1
between groups. Psychiatric hospitalizations occurred in only 7 individuals, and death in 10. There were no suicides. In the mixed models for %EWL, there was no significant main effect of exposure to abuse (F[1, 5037.4] ¼ 1.99, P ¼ .16 and F[1, 4506.1] ¼ .77, P ¼ .38, for any abuse and sexual abuse, respectively), and no significant interaction between history of abuse and time (F[2, 4268.9] ¼ .33, P ¼ .72 and F[2, 3737.6] ¼ .73, P ¼ .48, for any abuse and sexual abuse, respectively). In both models, there was a significant main effect of time (F[2, 4268.9] ¼ 5954.2, P o .001 and F[2, 3737.5] ¼ 3334.6, P o .001, for any abuse and sexual abuse, respectively) (Fig. 1A and B). Discussion We examined the prevalence of emotional, sexual, and physical abuse among a 4-year cohort of individuals undergoing bariatric surgery over 4 years in Ontario, Canada, and its impact on weight loss up to 1-year postoperatively. Overall, 4 1 in 5 individuals (21.5%) had a documented abuse history; however, the prevalence rose substantially over time with almost 1 in 3 individuals having a history of abuse documented in 2013–2014. Emotional abuse was the most common, followed by sexual and then physical abuse. Those with a history of abuse were slightly more likely to have a baseline BMI 460 and
undergo SG compared with LRYGB. In the adjusted repeated measures analysis, there was no impact of abuse history on %EWL over time. Rates of revision or repair and postoperative hospitalization between those with and without a history of abuse did not differ. These first-of-a-kind results from Canada find a lower prevalence of documented sexual abuse compared with what has been reported in most other studies of bariatric surgery candidates [9]. Our rate is similar to a national population-based study that found a rate of childhood sexual abuse equal to 10% among survey respondents [14]. Even in the last year of our study data, with the highest rate of sexual abuse, it was below what was found in a systematic review of the published literature [9], and 45% below what was reported among a sample of individuals attending an obesity program elsewhere in Canada [15]. In this latter study the assessment of abuse was done independent of any clinical care, and only a proportion of those individuals go on to have surgery [15]. This difference in methodology and the outcomes observed may suggest 2 possibilities: (1) when obesity co-occurs with an abuse history, there may be a lower likelihood of receiving surgery, which could be patient or provider driven; and/or (2) disclosure may be inhibited when abuse history is assessed as part of clinical management because of fear of not being offered surgery or mistrust of the medical team.
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
5
Fig. 1. Marginal means for percentage of excess weight loss (%EWL) at 3 months, 6 months, and 1 year, comparing (A) those with exposure to any type of abuse versus no abuse and (B) those with exposure to sexual abuse versus no abuse. BMI ¼ body mass index; LRYGB ¼ laparoscopic Roux-en-Y gastric bypass; SG ¼ sleeve gastrectomy.
A. Marginal means evaluated at age = 44.8 and baseline BMI = 49.5 Time point
3 mo
6 mo
1 yr
n (% of cohort)
5139 (85.4)
4743 (78.8)
4018 (66.8)
B. Marginal means evaluated at age = 44.7 and baseline BMI = 49.5 Time point
3 mo
6 mo
1 yr
n (% of cohort)
4581 (85.4)
4193 (78.2)
3562 (66.4)
Error bars are 95% confidence intervals. Marginal means were calculated with mixed models with time as a repeated measure. Models are adjusted for age, sex, type of surgery (LRYGB or SG), and baseline BMI. The rise in prevalence of documented abuse over time may reflect a growing recognition, supported by an emerging literature, that a history of abuse is less likely to affect weight loss outcomes than was once believed [6,7,9]. Alternatively, screening processes have improved and abuse is more often being assessed or disclosed, hopefully with concurrent supports in place to address issues that may arise as people undergo surgery. Although underreporting may still be present, our data do support the growing literature that a history of abuse, including sexual abuse, does not affect early and later weight loss outcomes. Similar to what has been previously reported elsewhere [6,11], we found that individuals with abuse histories were more likely to have preoperative psychiatric co-morbidity, as would be expected given the increased susceptibility to psychiatric problems with early life adversity [14]. Despite this higher rate of psychiatric co-morbidity, weight loss outcomes did not differ. Furthermore, those with abuse histories were no more likely to be hospitalized in the first
postoperative year. We also found that people with abuse histories more often received the less-invasive SG procedure. This discrepancy in the type of surgery received was partially accounted for by a higher number of these individuals with a baseline BMI 460, an indication for this procedure; however, this did not entirely explain the discrepancy. We did not examine other indications for this procedure, such as the presence of certain medical co-morbidities. A limitation of this study is that we examined those who underwent surgery only and did not assess whether or not a history of abuse was a barrier to receiving bariatric surgery. History of abuse was assessed by a variety of healthcare providers using an unstandardized clinical assessment. We observed that the rates of abuse recorded among social worker, psychologist, and psychiatrist assessors were similar, with significantly lower rates recorded among nurses and “other” healthcare professionals. These latter disciplines, possibly with less mental health training, who do a minority of the psychological assessments, may not
452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
6
507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
J. M. Hensel et al. / Surgery for Obesity and Related Diseases ] (2016) 00–00
screen for abuse as thoroughly. Conversely, patients with known mental health histories may be more likely to be assessed by a trained mental health professional, in which case the variability in the rates of abuse would reflect true differences. It is also possible that some surgery candidates may choose not to disclose a history of abuse. For example, studies examining rates of disclosure of psychiatric symptoms have documented that individuals will minimize or deny active symptoms to receive bariatric surgery [21,22]. The rising rate of abuse over time supports that disclosure may be increasing with improved screening and fewer perceived barriers, but our rates were still lower than what has been reported elsewhere [9,15]. A proportion of data were missing for all time points, and outcome data collected were based on clinical documents. This implies that some hospitalizations and deaths may not have been identified, and rates may be underestimates. We have only reported outcomes out to 1 year postoperative, and the study of longer-term outcomes is needed in Canada. For example, a recent Ontario population-level study examined patients presenting for medical care for self-harm attempts up to 3 years after bariatric surgery and found that these increased 1.54 times compared with a referent 3-year period before surgery [23]. Conclusion The findings in this study, based on first-of-a-kind data from Canada, do not support the hypothesis that history of abuse is associated with worse outcomes. With a heightened risk for obesity and obesity-related illness among survivors of abuse, this treatment can be an effective and life-altering intervention for these individuals. There is, however, a high burden of psychiatric co-morbidity in this population, the recognition and management of which will likely facilitate the optimization of surgical outcomes among survivors of abuse. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. Acknowledgments Data used in the study were obtained from the Ontario Bariatric Registry [18]. The registry is managed by Ruth Breau and the team at the Centre for Surgical Invention and Innovation, with analysis and data management provided by the Population Health Research Institute. References [1] Irish L, Kobayashi I, Delahanty DL. Long-term physical health consequences of childhood sexual abuse: a meta-analytic review. J Pediatr Psychol 2010;35(5):450–61. [2] Min MO, Minnes S, Kim H, Singer LT. Pathways linking childhood maltreatment and adult physical health. Child Abuse Negl 2013;37 (6):361–73.
[3] Hemmingsson E, Johansson K, Reynisdottir S. Effects of childhood abuse on adult obesity: a systematic review and meta-analysis. Obes Rev 2014;15(11):882–93. [4] Danese A, Tan M. Childhood maltreatment and obesity: systematic review and meta-analysis. Mol Psychiatry 2014;19(5):544–54. [5] Buser A, Dymek-Valentine M, Hilburger J, Alverdy J. Outcome following gastric bypass surgery: impact of past sexual abuse. Obes Surg 2004;14(2):170–4. [6] Mahony D. Assessing sexual abuse/attack histories with bariatric surgery patients. J Child Sex Abus 2010;19(4):469–84. [7] Buser AT, Lam CS, Poplawski SC. A long-term cross-sectional study on gastric bypass surgery: impact of self-reported past sexual abuse. Obes Surg 2009;19(4):422–6. [8] Wilson DR. Health consequences of childhood sexual abuse. Perspect Psychiatr Care 2010;46(1):56–64. [9] Steinig J, Wagner B, Shang E, Dolemeyer R, Kersting A. Sexual abuse in bariatric surgery candidates: impact on weight loss after surgery: a systematic review. Obes Rev 2012;13(10):892–901. [10] Adams ST, Salhab M, Hussain ZI, Miller GV, Leveson SH. Roux-en-Y gastric bypass for morbid obesity: what are the preoperative predictors of weight loss? Postgrad Med J 2013;89(1053):411–6;quiz 5, 6. [11] Wildes JE, Kalarchian MA, Marcus MD, Levine MD, Courcoulas AP. Childhood maltreatment and psychiatric morbidity in bariatric surgery candidates. Obes Surg 2008;18(3):306–13. [12] Clark MM, Hanna BK, Mai JL, et al. Sexual abuse survivors and psychiatric hospitalization after bariatric surgery. Obes Surg 2007;17 (4):465–9. [13] Chen EY, Fettich KC, Tierney M, et al. Factors associated with suicide ideation in severely obese bariatric surgery-seeking individuals. Suicide Life Threat Behav 2012;42(5):541–9. [14] Afifi TO, MacMillan HL, Boyle M, Taillieu T, Cheung K, Sareen J. Child abuse and mental disorders in Canada. CMAJ 2014;186(9):E324–32. [15] Gabert DL, Majumdar SR, Sharma AM, et al. Prevalence and predictors of self-reported sexual abuse in severely obese patients in a population-based bariatric program. J Obes 2013;2013:374050. [16] Bariatric Surgery in Canada—Report May 2014 [monograph on the Internet]. Canada: Canadian Institute of Health Information; c2015 [cited 2015 May 28]. Available from: https://secure.cihi.ca/free_ products/Bariatric_Surgery_in_Canada_EN.pdf. [17] Janssen I. The public health burden of obesity in Canada. Can J Diabetes. 2013;37(2):90–6. [18] Anvari M, Sharma A, Yusuf S, et al. Bariatric Registry. Registry data produced and distributed by the Population Health Research Institute and the Centre for Surgical Invention and Innovation, supported by the Ministry of Health and Long Term Care. Ontario, Canada: Ministry of Health and Long Term Care; 2015. [19] Karmali S, Schauer P, Birch D, Sharma AM, Sherman V. Laparoscopic sleeve gastrectomy: an innovative new tool in the battle against the obesity epidemic in Canada. Can J Surg 2010;53(2):126–32. [20] van de Laar A, de Caluwe L, Dillemans B. Relative outcome measures for bariatric surgery. Evidence against excess weight loss and excess body mass index loss from a series of laparoscopic Rouxen-Y gastric bypass patients. Obes Surg 2011;21(6):763–7. [21] Fabricatore AN, Sarwer DB, Wadden TA, Combs CJ, Krasucki JL. Impression management or real change? Reports of depressive symptoms before and after the preoperative psychological evaluation for bariatric surgery. Obes Surg 2007;17(9):1213–9. [22] Ambwani S, Boeka AG, Brown JD, et al. Socially desirable responding by bariatric surgery candidates during psychological assessment. Surg Obes Relat Dis 2013;9(2):300–5. [23] Bhatti JA, Nathens AB, Thiruchelvam D, Grantcharov T, Goldstein BI, Redelmeier DA. Self-harm emergencies after bariatric surgery: a population-based cohort study. JAMA Surg 2015:1–7.
562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616