Perioperative pregnancy interval, contraceptive counseling experiences, and contraceptive use in women undergoing bariatric surgery

Accepted Manuscript Perioperative pregnancy interval, contraceptive counseling experiences and contraceptive use in women undergoing bariatric surgery Biftu M. Mengesha, MD MAS, Jonathan T. Carter, MD, Christine E. Dehlendorf, MD MAS, Ms. Amanda J. Rodriguez, MPH, Jody E. Steinauer, MD MAS PII:

S0002-9378(18)30286-2

DOI:

10.1016/j.ajog.2018.04.008

Reference:

YMOB 12132

To appear in:

American Journal of Obstetrics and Gynecology

Received Date: 15 December 2017 Revised Date:

2 April 2018

Accepted Date: 3 April 2018

Please cite this article as: Mengesha BM, Carter JT, Dehlendorf CE, Rodriguez AJ, Steinauer JE, Perioperative pregnancy interval, contraceptive counseling experiences and contraceptive use in women undergoing bariatric surgery, American Journal of Obstetrics and Gynecology (2018), doi: 10.1016/ j.ajog.2018.04.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1 Perioperative pregnancy interval, contraceptive counseling experiences and contraceptive use in

2

women undergoing bariatric surgery.

3

Biftu M. MENGESHA MD MAS

4

University of California, San Francisco Department of Obstetrics, Gynecology and Reproductive

5

Sciences, Division of Zuckerberg San Francisco General

6

Jonathan T. CARTER MD

7

University of California, San Francisco, Department of Surgery, Division of General Surgery

8

Christine E. DEHLENDORF MD MAS

9

University of California, San Francisco Department of Family & Community Medicine, Department of

M AN U

SC

RI PT

1

Obstetrics, Gynecology and Reproductive Sciences, Department of Epidemiology & Biostatistics

11

Ms. Amanda J. RODRIGUEZ MPH

12

University of California, San Francisco Department of Obstetrics, Gynecology and Reproductive

13

Sciences, Division of Zuckerberg San Francisco General

14

Jody E. STEINAUER MD MAS

15

University of California, San Francisco Department of Obstetrics, Gynecology and Reproductive

16

Sciences, Division of Zuckerberg San Francisco General

17

Disclosure statement: The authors report no conflict of interest.

18

Funding: The Society of Family Planning Research Fund (SFPRF), award number SFPRF 16-18.

19

SFPRF did not have any role in the study design; collection, analysis and interpretation of data; the

20

writing of the report or the decision to submit this article for

21

publication. The views and opinions expressed are those of the authors, and do not necessarily represent

22

the views and opinions of SFPRF.

AC C

EP

TE D

10

ACCEPTED MANUSCRIPT 2 Paper presentation: These findings were presented at the 34th annual ObesityWeek Conference in

24

Washington D.C., October 29, 2017 – November 2, 2017.

25

Corresponding author: Biftu M. Mengesha MD MAS

26

1001 Potrero Ave Ward 6D13

27

San Francisco, CA 94110

28

Abstract word count: 301

29

Main text word count: 2608

30

Condensation: The needs of reproductive-aged women for counseling about pregnancy and

31

contraception after bariatric surgery are not being met.

32

Title short version: Perioperative counseling experiences and contraceptive use in women undergoing

33

bariatric surgery

34

Implications and Contributions

35

A. To assess the prevalence of perioperative pregnancy interval and contraceptive counseling in women

36

undergoing bariatric surgery, as well as evaluate the use of contraception within the first year

37

postoperatively.

38

B. Approximately 74% of participants had perioperative discussions about recommended pregnancy

39

interval or contraceptive use postoperatively, and 66% of women used a contraceptive method in the

40

first postoperative year. More than half of the participants felt it was “very important” to have these

41

discussions perioperatively.

42

C. This study shows that there is a notable minority of women undergoing bariatric surgery who do not

43

have perioperative discussions with a provider about these reproductive health topics, and women highly

44

desire having these conversations before surgery.

45

AC C

EP

TE D

M AN U

SC

RI PT

23

ACCEPTED MANUSCRIPT 3 Abstract

47

BACKGROUND: Reproductive-aged women represent about half of those undergoing bariatric surgery

48

in the United States (US). Obstetric and bariatric professional societies recommend that women avoid

49

pregnancy for 12-18 months postoperatively due to concern for increased pregnancy risks, and that

50

providers should counsel women about these recommendations and their contraceptive options.

51

However, knowledge about women’s experience with perioperative counseling and postoperative

52

contraceptive use is limited.

53

OBJECTIVES: 1) Determine prevalence of perioperative contraceptive and pregnancy interval

54

discussions among women who have recently undergone bariatric surgery, 2) Describe postoperative

55

contraceptive use within the first year of surgery in this population.

56

STUDY DESIGN: We performed a cross-sectional study of US women, aged 18-45 years and recruited

57

through Facebook, who underwent bariatric surgery within the last 24 months.

58

RESULTS: We enrolled 363 geographically-diverse women. Three-quarters recalled perioperative

59

pregnancy or contraceptive discussions, the majority with a bariatric provider. Half felt it was “very

60

important” to discuss these issues perioperatively, and 41% of those who reported discussions wished

61

they had had more. Of the 66% of women who reported using contraception in the first 12 months

62

postoperatively; 27% used oral contraceptives and 26% used an intrauterine device. One-third of

63

contraceptive users who had undergone Roux-en-Y gastric bypass, a combined restrictive-malabsorptive

64

procedure, were using oral contraceptives. Perioperative contraceptive or pregnancy discussions were

65

independently associated with increased postoperative contraceptive use (OR 2.5, 95% CI 1.5-4.3,

66

P<0.001).

67

CONCLUSIONS: A substantial proportion of women who had undergone bariatric surgery reported

68

having had no perioperative pregnancy or contraception counseling, and many women who had felt the

AC C

EP

TE D

M AN U

SC

RI PT

46

ACCEPTED MANUSCRIPT 4 discussions were insufficient. Those who had had perioperative discussions were more likely to use

70

contraception postoperatively. Reproductive-aged women should be routinely counseled perioperatively

71

about pregnancy and contraception in the context of their reproductive desires, so they can make

72

informed decisions about perioperative pregnancy prevention and contraceptive method use.

RI PT

69

73

Keywords: bariatric surgery; contraception; family planning; obesity; patient-centered care; pregnancy;

75

social media recruitment

SC

74

76

M AN U

77 78 79 80

84 85 86 87 88 89 90 91

EP

83

AC C

82

TE D

81

ACCEPTED MANUSCRIPT 5 1. Introduction

93

Over 30% of reproductive-aged women in the United States (US) are obese.1 Bariatric surgery has

94

provided effective treatment for obesity with long-term success in weight loss and improvement or

95

resolution of medical comorbidities.2 Over 190,000 bariatric procedures were performed in 2014 in the

96

US, and the number of procedures performed each year continues to rise.3 Reproductive-aged women

97

represent nearly 50% of patients undergoing bariatric surgery in the US,4 with the most common

98

procedures performed being the sleeve gastrectomy, a restrictive procedure where most of the greater

99

curvature of the stomach is resected, and Roux-en-Y gastric bypass (RYGB), a combined restrictive-

SC

malabsorptive procedure.3

M AN U

100

RI PT

92

101

Because of concerns for increased maternal and fetal risks in an early postoperative pregnancy, obstetric

103

and bariatric professional societies recommend that women avoid pregnancy for 12-18 months after

104

bariatric surgery and use an effective form of contraception during that time.5 Further, the Center for

105

Disease Control US Medical Eligibility Criteria (MEC) lists a history of bariatric surgery within 2 years

106

as a condition associated with increased risk in pregnancy.6 In addition, because of concerns regarding

107

decreased efficacy due to malabsorption, the MEC rates oral contraceptives as a category 3 for women

108

who have had malabsorptive bariatric surgery.

EP

AC C

109

TE D

102

110

Knowledge about how well these recommendations are communicated from medical providers to

111

patients is limited. Previous literature has mainly evaluated contraceptive use at individual centers in the

112

US7,8 with the exception of one recently published analysis of a prospective multicenter cohort.9 This

113

study found that while pregnancy in the 18 months after surgery was relatively rare, contraceptive use

114

was inconsistent. To our knowledge, women’s counseling experiences have been described in a single

ACCEPTED MANUSCRIPT 6 study of 35 women,10 with no studies assessing perioperative counseling experiences on a broader scale.

116

We sought to determine the prevalence of perioperative contraceptive and pregnancy interval

117

discussions and postoperative contraceptive use from a national sample of reproductive-aged women,

118

with the goal to inform quality improvement efforts ensuring women’s reproductive health needs are

119

met in the context of bariatric surgery.

RI PT

115

120 2. Materials and Methods

122

This was a cross-sectional survey of reproductive-aged women from June 2016 to December 2016.

123

Ethical approval was obtained prior to beginning the study from the University of California, San

124

Francisco Institutional Review Board.

125

M AN U

SC

121

Women were recruited using targeted advertising on Facebook. Advertisements were shown to women

127

between the ages of 18-45 years who lived in the US and had demonstrated interest in bariatric surgery

128

or weight loss on social media. The advertisement contained a link to our secure, web-based survey. We

129

recruited both women who underwent bariatric surgery within the last 24 months as well as those who

130

have not undergone surgery yet but were actively engaged in the bariatric surgery preoperative process.

131

This analysis was restricted to postoperative women only. Inclusion criteria were women ages 18 to 45

132

years who had undergone bariatric surgery within the last 24 months. Those who had a hysterectomy or

133

tubal ligation preoperatively or were sexually active exclusively with women were excluded as we

134

wanted to identify those women who would potentially be at risk for pregnancy and to whom issues of

135

contraception and pregnancy would hold more relevance. Of note, we did not exclude women who were

136

not sexually active in the preoperative period as sexual activity is a dynamic status and we desired to

137

evaluate whether this had been discussed in the perioperative period in relation to their reproductive

AC C

EP

TE D

126

ACCEPTED MANUSCRIPT 7 counseling. Once found to be eligible through a series of screening questions, participants were allowed

139

to continue to the survey, which collected demographic, obstetric and surgical information as well as

140

pregnancy intention, contraceptive use and experiences with perioperative contraceptive counseling and

141

counseling about pregnancy interval recommendations.

142

RI PT

138

Our primary aim was to determine the prevalence of women’s receipt of perioperative contraceptive and

144

pregnancy interval information. We calculated a descriptive sample size estimating that 50% of women

145

would report having had received this information perioperatively, assuming a 7.5% margin of error and

146

using a 95% confidence interval. These parameters led to a sample size target of 171 women. In order

147

to ensure adequate recruitment of postoperative women, we used an inflation factor of 50% and thus

148

aimed to recruit a total of 342 women. Our secondary aim was to determine the prevalence of

149

postoperative contraceptive use within the first 12 months postoperatively. We again calculated a sample

150

size estimating that 60% of women using contraception postoperatively based on prior literature7,9,10,

151

using a 7.5% margin of error, a 95% confidence interval and 50% inflation factor. This resulted in a

152

desired sample size of 326 women.

M AN U

TE D

EP

153

SC

143

We performed statistical analysis using Stata version 14 (College Station, TX, USA). We first calculated

155

summary statistics to describe the demographic, medical and surgical characteristics of the study

156

population. We used nonparametric tests for continuous data that were not normally distributed and chi-

157

squared tests for categorical variables in order to compare these factors between those women who

158

received perioperative information and those who did not. Bivariate logistic regression was performed to

159

evaluate correlates of the receipt of contraceptive or pregnancy interval information as well as

160

contraceptive use within the first 12 months postoperatively. Multivariate logistic regression was used to

AC C

154

ACCEPTED MANUSCRIPT 8 examine factors associated with either perioperative pregnancy interval or contraceptive discussions as

162

well as to examine the association between having perioperative discussions and postoperative

163

contraceptive use. We tested having either a pregnancy interval or contraception discussion together as a

164

combined variable since we wanted to capture anyone who had their reproductive needs acknowledged

165

in any way perioperatively. We determined covariates in the multivariate regression model that we

166

considered to be potential confounders in the relationship between perioperative counseling and

167

postoperative contraceptive use on conceptual grounds, and also included any variables significantly

168

associated with contraceptive use in bivariate analysis. These covariates included age, race (categorized

169

as white, black and other), ethnicity (categorized as Hispanic or Latino or not Hispanic or Latino),

170

parity, bariatric center type (academic, private or community, and county hospital), bariatric surgery

171

type and if the participant was sexually active during their process of obtaining bariatric surgery. A p

172

value of <0.05 was considered to be statistically significant.

M AN U

SC

RI PT

161

TE D

173 3. Results

175

Approximately 1,100 women attempted to take the survey. A total of 363 were women who had bariatric

176

surgery in the past 24 months were eligible for study inclusion and submitted complete surveys.

177

Demographic, obstetric and surgical characteristics are in Table 1. The study population was

178

geographically diverse and women were a median of 12 months postoperative (range 1 to 24 months).

179

Most were privately insured, had surgery at a private or community hospital, and had undergone a

180

RYGB or sleeve gastrectomy. Preoperative infertility was reported by 140 respondents (39%), and 93

181

(25%) of all respondents reported that a desire to improve their fertility was one motivation for having

182

bariatric surgery.

183

AC C

EP

174

ACCEPTED MANUSCRIPT 9 Approximately 74% (n=268) reported having had a perioperative discussion about either contraception

185

or recommendations for surgery-to-pregnancy interval as it related to their bariatric surgery (Table 1).

186

Factors associated with having these perioperative discussions in bivariate analysis were younger age,

187

being sexually active during the bariatric surgery process and having a preoperative diagnosis of

188

infertility. In multivariate analysis, factors that were significantly associated with increased odds of

189

receiving either pregnancy interval or contraceptive information were being between the ages of 26-35

190

years (OR 2.2, 95% CI 1.4 - 3.5), black race (OR 2.0, 95% CI 1.0 – 3.9) and being sexually active

191

during the bariatric surgery process (OR 2.1, 95% CI 1.1 – 3.8). Having undergone gastric banding

192

surgery was associated with decreased odds of receiving perioperative pregnancy or contraceptive

193

information (OR 0.22, 95% CI 0.07-0.7) (Table 1). When asked in an open-ended question about what

194

was discussed regarding pregnancy after bariatric surgery, some respondents (n=10) reported a range of

195

recommended surgery-to-pregnancy intervals, from 10 months to 3 years. In a similar open-ended

196

question about contraception, respondents reported a variety of recommendations including using two

197

forms of contraception, required use of a method for 12 months postoperatively, using long-acting

198

reversible methods or needing to provide proof of contraception before surgery.

SC

M AN U

TE D

EP

199

RI PT

184

Respondents reported having pregnancy or contraception discussions at various time points throughout

201

their perioperative process. The majority reported discussions at their surgery orientation or immediately

202

before surgery (Table 2), and ten (5%) had discussions only in the postoperative period. A significant

203

proportion (n=123, 46%) had discussions about either contraception of surgery-to-pregnancy interval at

204

more than one time point during the perioperative period. Bariatric surgeons were cited most commonly

205

as the provider with whom they had had discussions (92% and 84%, respectively). Approximately 16%

206

of respondents were referred to another provider, such as a gynecologist or a primary care provider, for

AC C

200

ACCEPTED MANUSCRIPT 10 discussion about either contraception or pregnancy (Table 2). Just more than half felt that it was “very

208

important” and approximately one-quarter “somewhat important” to have these discussions

209

perioperatively. Nearly half (42%) of respondents who reported perioperative discussions wished they

210

had had more discussion about pregnancy and contraception. Respondents reported wishing their

211

reproductive plans had been assessed early in their bariatric process, they had received more detailed

212

information about maternal and fetal risks of postoperative pregnancy and about appropriate

213

contraceptive methods based on their surgical procedure, and they had had the opportunity to speak with

214

a women’s health specialist at the time of their bariatric clinic visits.

SC

RI PT

207

M AN U

215

With respect to contraceptive use, 241 (66%) reporting using a contraceptive method at some point in

217

the first 12 months after surgery: 27% used oral contraceptives and 26% used an intrauterine device

218

(Figure 2). Thirty-two percent of respondents used condoms as well. There were no significant

219

differences in the proportion of specific contraceptive methods used based on type of surgical procedure.

220

Restricting to only respondents who were 12 months or more postoperative, 69% (n=124) reported

221

contraceptive use during the first 12 months. Being sexually active during the bariatric surgery process

222

was associated with increased contraception use postoperatively in bivariate analyses. Nulliparity,

223

having surgery at an academic center and undergoing gastric banding were associated with decreased

224

postoperative contraception use in bivariate analysis (Table 3). There was no difference in postoperative

225

contraceptive use between respondents who pursued surgery for fertility reasons and those who did not

226

(68% versus 67% respectively, p=0.7). In multivariate analysis, sexual activity was associated with

227

increased postoperative contraceptive use, and having undergone gastric banding and surgery at an

228

academic center were associated with decreased use. Of note, 34% of contraceptive users who had a

229

RYGB, a combined restrictive-malabsorptive procedure, used oral contraceptives postoperatively. Of

AC C

EP

TE D

216

ACCEPTED MANUSCRIPT 11 230

those that had RYGB and used oral contraceptives postoperatively, 14 (35%) reported that a provider

231

discussed potential contraceptive efficacy issues with them, however what was exactly discussed is

232

unknown.

RI PT

233

Respondents who did not use a contraceptive method cited not being sexually active as the most

235

common reason (33%), followed by 18% citing that they didn’t want to use birth control and 17% who

236

did not think they could become pregnant. Respondents who had perioperative pregnancy or

237

contraceptive discussions had 2.5 times higher odds (95% CI 1.5-4.3, p=0.001) of using a contraceptive

238

method in the first 12 months postoperatively in multivariate analysis. No interactions were found in our

239

model between age and bariatric surgery type as well as race and bariatric center type. We asked

240

respondents whether they had a pregnancy or contraceptive discussion only once during their

241

perioperative process or at multiple times throughout in order to evaluate whether multiple

242

conversations had any effect on contraceptive use. There was no difference in postoperative

243

contraceptive use for respondents who had a discussion only once compared to those who had

244

discussions multiple times perioperatively (adjusted OR 0.96, p=0.9).

245

EP

TE D

M AN U

SC

234

4. Comment

247

Our findings suggest that women’s needs may be unmet for perioperative pregnancy interval and

248

contraceptive counseling. A substantial proportion of women reported having had no counseling, and

249

many women felt discussions were insufficient, despite recommendations by obstetric and bariatric

250

professional societies that women use contraception to prevent pregnancy for 12 to 18 months after

251

bariatric surgery. Our study adds substantially to our knowledge about women’s perioperative

252

counseling experiences and contraceptive use, and our finding of 66% of women using contraception

AC C

246

ACCEPTED MANUSCRIPT 12 253

within the first year postoperatively is consistent with the previous multicenter study which reported

254

60% of their cohort using a contraceptive method.9

255 Barriers to meeting women’s needs for perioperative counseling about contraception and pregnancy in

257

the context of bariatric surgery include provider knowledge and attitudes. Many studies have found that

258

while the vast majority of bariatric surgery providers recommend delaying pregnancy after surgery,

259

believe the conversations are important and that it is their responsibility to initiate them, the majority do

260

not feel comfortable, do not initiate the discussions and have inadequate contraceptive knowledge.11,12,13

261

What remains unclear is the optimal time to have perioperative discussions about contraception and

262

future pregnancy or when providers typically have these conversations with patients. Our study found

263

that women had these conversations at varied times throughout the perioperative period.

M AN U

SC

RI PT

256

264

Of note, we found that a substantial number of women who underwent RYGB were using oral

266

contraceptives, a method with possibly decreased efficacy as indicated by the US MEC. Concerns about

267

efficacy of oral methods after malabsorptive procedures came from a case series of 40 patients who

268

underwent biliopancreatic diversion, an infrequently performed procedure with higher malabsorptive

269

effects than RYGB14, where two oral contraceptive failures were reported in nine users over a period of

270

one year.15 While we do not have definitive evidence to support or refute findings of decreased efficacy

271

and current procedures have fewer malabsorptive effects, some experts recommend avoiding oral

272

contraceptives after all malabsorptive bariatric procedures, particularly in patients with long-term

273

diarrhea or vomiting.16 Therefore, the relatively high rate of use of these methods raises the question of

274

whether these patients had been adequately counseled about these concerns.

275

AC C

EP

TE D

265

ACCEPTED MANUSCRIPT 13 To our knowledge, this is the largest study to date evaluating women’s perioperative reproductive health

277

experiences in the US. Limitations include that the results are based on self-report and may be affected

278

by recall bias. However, evaluation of patient experiences requires self-report. We also included women

279

at varying points in their postoperative course, and as such our results may not be fully representative of

280

the complete postoperative experience. Facebook users may also be different from the general

281

population; for example, those that responded to our advertisements may have more interest in

282

reproductive health, which might limit generalizability. However, our study population was similar to

283

the general bariatric surgery population, which makes our findings potentially more generalizable.17,18

284

Additionally, women with obesity may have multiple medical comorbidities or risk factors that might

285

preclude the use of certain contraceptive methods. We did not directly assess the potential influence of

286

medical comorbidities on postoperative contraceptive use, however in our data there was no difference

287

in contraceptive use in women with and without hypertension or diabetes. Finally, our sample size

288

yielded relatively large confidence intervals around our odds ratios, which could result in failing to

289

detect meaningful associations.

SC

M AN U

TE D

290

RI PT

276

Women undergoing bariatric surgery would benefit from improved perioperative pregnancy and

292

contraception counseling from bariatric, gynecologic and primary care providers alike, ideally in a

293

multidisciplinary fashion. This counseling should be tailored to women’s reproductive desires, as some

294

women may be pursuing bariatric surgery for improvement in fertility and subsequent pregnancy. More

295

research is needed to understand women’s desires around pregnancy in the context of bariatric surgery,

296

as well as their preferences for counseling, including when and from whom they would prefer to receive

297

this counseling. Additionally, more research is needed to clarify the conflicting evidence about the risks

298

associated with birth spacing after surgery, as some studies suggest no risk associated with pregnancy

AC C

EP

291

ACCEPTED MANUSCRIPT 14 299

soon after surgery.19,20,21,22 This information will allow providers to deliver high-quality, evidence-based

300

patient-centered counseling in order to empower women to make informed decisions about pregnancy

301

and fertility after bariatric surgery, aligned with their desires and preferences.

RI PT

302 303 304

SC

305 306

M AN U

307 308 309 310

314 315 316 317 318 319 320 321

EP

313

AC C

312

TE D

311

ACCEPTED MANUSCRIPT 15 322 323

5. References 1. Ogden CL, Carroll MD, Fryar CD, Flegal KM. Prevalence of obesity among adults and youth: United States, 2011-2014. NCHS data brief, no 219. Hyattsville, MD: National Center for Health

325

Statistics. 2015.

327 328

2. Buchwald H, Avidor Y, Braunwald E et al. Bariatric surgery: a systematic review and metaanalysis. JAMA 2004; 292: 1724-37

3. Ponce J, Nguyen NT, Hutter M, Sudan R, Morton JM. American Society for Metabolic and

SC

326

RI PT

324

Bariatric Surgery estimation of bariatric procedures in the United States, 2011-2014. Surg Obes

330

Relat Dis 2015; 11: 1190-200.

334 335 336 337 338 339 340 341 342 343 344

5. Beard JH, Bell RL, Duffy AJ. Reproductive considerations and pregnancy after bariatric surgery: current evidence and recommendations. Obes Surg 2008; 18: 1023-7.

TE D

333

systematic review. JAMA. 2008;300: 2286–96

6. Curtis KM, Tepper NK, Jatlaoui TC et al. CDC. U.S. medical eligibility criteria for contraceptive use. MMWR Recomm Rep 2016; 65(No. RR-3). 7. Mody SK, Hacker MR, Dodge LE, Thornton K, Schneider B, Haider S, Contraceptive

EP

332

4. Maggard MA, Yermilov I, Li Z, et al. Pregnancy and fertility following bariatric surgery: a

Counseling for Women Who Undergo Bariatric Surgery. J Womens Health, 2011; 20: 1785-8. 8. Mengesha B, Griffin L, Nagle A, Kiley J. Assessment of contraceptive needs in women

AC C

331

M AN U

329

undergoing bariatric surgery. Contraception 2016; 94: 74-7. 9. Menke MN, King WC, White GE et al. Contraception and Conception After Bariatric Surgery. Obstet Gynecol 2017;130:979-87. 10. Casas R, Bourieily G, Vithiananthan S, Tong I. Contraceptive use in women undergoing bariatric surgery. Obes Res Clin Pract 2014; 8: e608-13.

ACCEPTED MANUSCRIPT 16

346 347 348

11. Chor J, Chico P, Ayloo S, Roston A, Kominiarek MA. Reproductive health counseling and practices: a cross-sectional survey of bariatric surgeons. Surg Obes Relat Dis 2015; 11: 187-92. 12. Jatlaoui TC, Cordes S, Goedken P, Jamieson DJ, Cwiak C. Family planning knowledge, attitudes and practices among bariatric healthcare providers. Contraception 2016; 93: 455-62.

RI PT

345

13. Graham YN, Mansour D, Small PK et al. A survey of bariatric surgical and reproductive health

350

professionals’ knowledge and provision of contraception to reproductive-aged bariatric surgical

351

patients. Obes Surg 2016; 26: 1918-23.

353 354 355 356

14. Tucker ON, Szomstein S, Rosenthal RJ. Nutritional consequences of weight-loss surgery. Med Clin North Am 2009; 91: 499-514.

M AN U

352

SC

349

15. Gerrits EG, Ceulemans R, van Hee R, Hendrickx L, Totte E. Contraceptive treatment after biliopancreatic diversion needs consensus. Obes Surg 2003; 13: 378-82. 16. Paulen ME, Zapata LB, Cansino C, Curtis KM, Jamieson DJ. Contraceptive use among women with a history of bariatric surgery: a systematic review. Contraception 2010; 82: 86-94.

358

17. Pratt GM, Learn CA, Hughes GD, Clark BL, Warthen M, Pories W. Demographics and

TE D

357

outcomes at American Society for Metabolic and Bariatric Surgery Centers of Excellence. Surg

360

Endosc 2009; 23: 795-9.

361

EP

359

18. DeMaria EJ, Pate V, Warthen M, Winegar DA. Baseline data from American Society for Metabolic and Bariatric Surgery-designated Bariatric Surgery Centers of Excellence using the

363

Bariatric Outcomes Longitudinal Database. Surg Obes Relat Dis 2010; 6: 347-55.

364 365 366 367

AC C

362

19. Adams TD, Hammoud AO, Davidson LE et al. Maternal and neonatal outcomes for pregnancies before and after gastric bypass surgery. Int J Obes (Lond) 2015; 39: 686-94. 20. Parker MH, Berghella V, Nijjar JB. Bariatric surgery and associated adverse pregnancy outcomes among obese women. J Matern Fetal Neonatal Med 2015; 29:1747-50.

ACCEPTED MANUSCRIPT 17 368

21. Wax JR, Cartin A, Wolff R, Lepich S, Pinette MG. Pregnancy following gastric bypass for

369

morbid obesity: effect of surgery-to-conception interval on maternal and neonatal outcomes.

370

Obes Surg 2008; 18:1517-21.

372

22. Sheiner E, Levy A, Silverberg D, et al. Pregnancy after bariatric surgery is not associated with

RI PT

371

adverse perinatal outcome. Am J Obstet Gynecol 2004;190:1335–40.

373

SC

374 375

M AN U

376 377 378 379

383 384 385 386 387 388 389 390

EP

382

AC C

381

TE D

380

ACCEPTED MANUSCRIPT 18 391 392

Table 1: Demographic characteristics of study population and predictors of perioperative pregnancy and contraception discussion

393 Discussed

Did not

Unadjusted

women

either

discuss

Odds ratio

Odds

pregnancy or

pregnancy or

(95% CI)

ratiob

contraception

contraceptive

N (%)

information N (%)

268 (74) 88a (24)

Age, Median (IQR)

36 (31-

35 (30-39)

40) Age, categorized,

18 – 25 years

21(5)

16 (6)

N (%)

Race, N (%)

125 (47)

5(6)

22 (25)

0.92 (0.9-0.9)

1.5 (0.6-3.7)

2.2 (1.4-3.5)

<0.001

0.9 (0.8-

<0.002

0.9) 0.4

2.0 (0.6-

0.6

6.3) 0.001

3.2 (1.8-

<0.001

5.8)

191 (53)

127 (47)

61 (69)

ref

-

ref

-

Non-Hispanic

303 (86)

221 (85)

76 (87)

ref

-

ref

-

Hispanic

51 (14)

39 (15)

11 (13)

1.2 (0.6-2.5)

0.6

1.3 (0.6-

0.7

White

Other

Sexually active, N (%)

2.9)

258 (71)

187 (70)

67 (76)

ref

-

ref

-

86 (24)

66 (25)

17 (19)

1.4 (0.8-2.5)

0.3

2.0 (1.0-

0.05

AC C

Black

19 (5)

279 (77)

3.9) 15 (5)

4 (5)

1.3 (0.4-4.2)

0.6

1.4 (0.4-

0.6

4.6) 214 (80)

61 (70)

1.8 (1.0-3.0)

0.04

2.1 (1.13.8)

Incomec, N (%)

p

36 – 45 years

EP

Ethnicityc, N (%)

151 (42)

TE D

26 – 35 years

39 (33-43)

Adjusted

(95% CI)

M AN U

N = 363

p

RI PT

All

SC

Characteristic

<250 FPLd

100 (46)

71 (43)

27 (53)

ref

-

>250% FPL

119 (54)

93 (57)

24 (47)

1.5 (0.8-2.9)

0.2

0.02

ACCEPTED MANUSCRIPT 19 Education, N (%)

College and

320 (88)

232 (87)

82 (93)

ref

-

43 (12)

36 (13)

6 (7)

1.3 (0.6-2.5)

0.5

Private

263 (73)

193 (72)

66 (76)

ref

-

Public

91 (25)

71 (26)

19 (22)

1.3 (0.7-2.3)

0.4

None

7 (2)

4 (1)

2 (2)

0.7 (0.1-3.8)

0.7

Living

Rural

102(28)

71 (26)

31 (35)

ref

-

environment, N

Urban or

261 (72)

197 (73)

57 (65)

1.5 (0.9-2.5)

(%)

Suburban 141 (39)

105 (39)

32 (36)

beyond High school or

Infertility preoperatively, N (%)

Have used contraception in past,

140 (39)

119 (44)

0.001

0.3

54 (20)

19 (22)

ref

-

26 (30)

1.2 (0.6 –

0.6

South

118 (33)

TE D

States where

89 (33)

1.2 (0.8-

0.3

2.4)

Midwest

73 (20)

54 (20)

18 (20)

1.0 (0.5-2.2)

0.9

Northeast

63 (17)

48 (18)

15 (17)

1.1 (0.5-2.5)

0.8

EP

0.8

1.8)

0.6 (0.2-1.7)

73 (20)

International

35 (10)

22 (9)

10 (11)

0.8 (0.3-1.9)

0.6

Private or

257 (71)

192 (72)

60 (68)

ref

-

ref

-

45 (17)

18 (20)

0.8 (0.4-1.5)

0.4

0.7 (0.3-

0.2

AC C

type, N (%)

2.9 (1.6-5.2)

0.9 (0.51.8)

84 (95)

West

Bariatric center

21 (24)

0.6

247 (92)

Region of United

(%)

1.1 (0.6-1.8)

0.1

337 (93)

N (%)

surgery done, N

SC

Nulliparous, N (%)

M AN U

Insurancec, N (%)

RI PT

lower

Community Academic

County

64 (18)

42 (11)

1.3) 31 (11)

10 (12)

0.9 (0.4-2.1)

0.9

0.9 (0.4-

0.9

2.1) Surgery type, N

Roux-en-Y

118 (32)

90 (34)

26 (30)

ref

-

ref

-

(%)

Sleeve

214 (59)

157 (59)

53 (60)

0.9 (0.5-1.5)

0.6

0.7 (0.4-

0.3

gastrectomy

1.3)

ACCEPTED MANUSCRIPT 20 Gastric

21 (6)

12 (4)

8 (9)

0.4 (0.2-1.2)

0.1

banding

medical

Other

10 (3)

9 (3)

1 (1)

2.0 (0.2-17.2)

0.5

-

-

Diabetes

88 (28)

64 (27)

24 (34)

0.7 (0.4-1.3)

0.2

-

-

Hypertension

166 (53)

123 (51)

43 (61)

0.7 (0.4-1.2)

0.2

-

-

Depression

140 (45)

112 (47)

27 (38)

1.4 (0.8-2.5)

0.2

-

-

c

comorbidities , N (%) a

Seven respondents did not remember whether they had received pregnancy or contraceptive information perioperatively

395

b

Adjusted for age, race, ethnicity, parity, sexual activity, preoperative infertility, surgery type and bariatric center type

396

c

397

d

Missing data for listed medical comorbidities, ethnicity, and income

M AN U

FPL = federal poverty level

398 399 400 401 402

408 409 410 411 412 413

EP

407

AC C

406

TE D

403

405

SC

394

404

414 415 416 417

0.01

0.7)

RI PT

Preoperative

0.2 (0.07-

Table 2: Characteristics of perioperative pregnancy and contraceptive discussions

ACCEPTED MANUSCRIPT 21 418 Received pregnancy

Received

information (n=187)

contraception

N (%)

information

RI PT

Perioperative discussion characteristics

(n=214) N (%)

115 (62)

Before surgery

173 (93)

After surgery

75 (40)

Bariatric providerb

179 (84)

51 (27)

62 (29)

50 (27)

66 (31)

29 (16)

34 (16)

Gynecologist

18 (62)

26 (76)

Primary care provider

4 (14)

8 (24)

Other

7 (24)

0 (0)

Gynecologist discusseda

Primary care provider

420

b

421

c

423

Participants were given the option to select more than one choice Includes physician, nurse or physician’s assistant

AC C

422

EP

a

TE D

Referral for discussion given

419

72 (34)

173 (92)

What type of provider

Type of provider referred toc

196 (92)

M AN U

discusseda

129 (60)

SC

At surgery orientation Perioperative time point when

Denominator is the number of respondents referred for discussion

424 425 426 427

Table 3: Factors associated with contraceptive use anytime within the first 12 months postoperatively

ACCEPTED MANUSCRIPT 22 428 Characteristic

Using contraception

Unadjusted OR

in first 12 months

(95% CI)

Adjusted ORa

p

p

(95% CI)

RI PT

postoperatively N = 241 Had perioperative

194 (80)

2.1 (1.3 – 3.3)

36 (31-39)

0.99 (0.95-1.02)

0.002

contraception or pregnancy

Age, median (IQR)

14 (6)

26 – 35 years

103 (43)

36 – 45 years

124 (51)

Ethnicityb, N (%)

0.4

M AN U

Age, categorized, N (%) 18 – 25 years

0.001

SC

interval discussion, N (%)

2.5 (1.5 – 4.3)

1.1 (0.4-2.8)

0.9

1.6 (0.6-4.5)

0.4

1.2 (0.7-1.8)

0.5

1.3 (0.8-2.2)

0.2

ref

-

ref

ref

ref

-

ref

0.9

199(85)

Hispanic

34 (15)

1.0 (0.6-1.9)

0.9

1.0 (0.5-1.9)

White

178 (74)

ref

-

ref

Black

51 (21)

0.6 (0.4-1.0)

0.1

0.7 (0.4-1.3)

0.3

12 (5)

0.8 (0.3-2.0)

0.6

0.7 (0.3-1.9)

0.5

196 (81)

2.0 (1.2-3.4)

0.005

2.0 (1.1-3.4)

0.01

Other

AC C

Sexually active, N (%)

EP

Race, N (%)

TE D

Non-Hispanic

Incomeb, N (%) <250FPL

67(55)

ref

-

>250% FPL

83 (55)

1.1 (0.6-2.0)

0.7

College and beyond

213 (88)

ref

-

High school or lower

28 (12)

1.0 (0.5-2.1)

0.9

Education, N (%)

ACCEPTED MANUSCRIPT 23 Insurance, N (%) 182 (75)

ref

-

Public

55 (23)

0.7 (0.4-1.1)

0.1

None

4 (2)

0.6 (0.1-2.7)

0.5

Rural

68 (28)

ref

-

Urban or Suburban

173 (72)

0.9 (0.6-1.6)

0.9

Nulliparous, N (%)

85 (35)

0.6 (0.4-1.0)

0.05

Infertility preoperatively, N (%)

93 (39)

1.0 (0.7-1.3)

surgery done, N (%) 51 (21)

South

78 (33)

Midwest

49 (21)

Northeast

44 (18)

International

0.9

ref

-

0.8 (0.4-1.6)

0.6

0.9 (0.4-1.7)

0.7

0.9 (0.5-2.1)

0.9

18 (7)

0.5 (0.2-1.0)

0.06

Private/Community

181 (75)

ref

-

ref

Academic

35 (14)

0.5 (0.3-0.9)

0.02

0.5 (0.3-0.9)

0.03

0.6 (0.3-1.2)

0.2

0.6 (0.3-1.1)

0.1

County

25 (11)

AC C

Surgery type, N (%)

EP

Bariatric center type, N (%)

TE D

West

0.09

M AN U

Region of United States where

0.7 (0.4-1.1)

SC

Living environment, N (%)

RI PT

Private

Roux-en-Y

81 (34)

ref

-

ref

Sleeve gastrectomy

143 (59)

0.9 (0.6-1.5)

0.7

0.9 (0.5-1.5)

0.7

Gastric banding

9 (4)

0.3 (0.1-0.9)

0.03

0.3 (0.1-0.9)

0.03

Other

8 (3)

1.4 (0.3-7.1)

0.7

1.1 (0.2-6.3)

0.2

57 (27)

0.9 (0.5-1.4)

0.6

Preoperative medical comorbiditiesb, N (%) Diabetes

ACCEPTED MANUSCRIPT 24

429

a

430

b

Hypertension

112 (53)

1.0 (0.6-1.7)

0.9

Depression

101 (48)

1.5 (0.9-2.5)

0.08

Multivariable model includes age, race, ethnicity, sexual activity, nulliparity, bariatric center type and surgery type.

AC C

EP

TE D

M AN U

SC

RI PT

Missing data for listed medical comorbidities, ethnicity and income.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT