Current national data indicate that 10.3 million U.S. women use female sterilization for birth control, making it the leading contraceptive method among married U.S. women, ages 15-44 years, and the second most common contraceptive of all U.S. women.1,2 Half of all sterilizations are performed postpartum.3 We reported previously that 29% of women delivering liveborn singleton neonates at University Hospital (San Antonio, TX) requested postpartum tubal ligation.4 Sixty-nine percent of women requesting postpartum tubal ligation subsequently received the procedure, a rate that is slightly higher than in other reported populations.5,6 We, like other investigators, found that insurance coverage exceptions, immigration status, and Medicaid sterilization-consent paperwork presented potential barriers to obtaining postpartum tubal ligation.5–10 Cesarean delivery was most predictive of whether a woman received a successful postpartum tubal ligation, a finding seen in other cohorts.5,6
Previous authors, in a small longitudinal study (n=34), found that women who did not receive a desired postpartum tubal ligation were more likely than were those not requesting permanent sterilization (control group) to be discharged from the obstetric service without contraception,7 and 18% became pregnant within 1 year postpartum.8 After conducting a MEDLINE and PubMed search from January 1990 to June 2010 using various combinations of the keywords “postpartum,” “postnatal,” “pregnancy,” “contraception,” “contraceptive,” “tubal ligation,” “female sterilization,” and “contraception decision making,” we believe that the cohort described by Gilliam and colleagues is the only study to evaluate women who did not receive a desired postpartum tubal ligation to examine future obstetric and contraceptive outcomes.7,8
Therefore, the objective of this study was to expand on the data that tracks outcomes of women who do not receive requested postpartum tubal ligations, by evaluating three groups of women from their index obstetric delivery to 1 year postpartum: 1) women who successfully received a requested postpartum tubal ligation (the postpartum tubal ligation [PPTL] YES group), 2) women who requested but did not receive postpartum tubal ligation (the PPTL NO group), and 3) women who did not request postpartum tubal ligation (the control group) to determine the primary outcome—pregnancy within 1 year postpartum among women in the PPTL NO group compared with those in the control group—and secondary outcomes, which included contraception requested at hospital discharge, attendance at a postpartum or other gynecologic examination, contraceptive use between the time of delivery and the time of the postpartum examination, and request for contraception at the postpartum examination.
MATERIALS AND METHODS
This study was approved by the Institutional Review Board at the University of Texas Health Sciences Center San Antonio. This cohort has been described previously,4 but, in brief, the authors identified all women who delivered liveborn singleton neonates at the University of Texas Health Sciences Center San Antonio between December 1, 2007, and May 31, 2008. This time frame was selected because it was approximately 1 year after commencement of the Child Health Insurance Perinate Program, which provides insurance coverage for prenatal care, obstetric delivery, and two postpartum visits to women who qualify economically for Medicaid insurance but who do not have the proper immigration documentation. Also during the time of the study, all outpatient and inpatient hospital records were electronic, which facilitated data collection.
Women were excluded from the analysis if they were 20 years old or younger because they are not eligible for publicly funded sterilization procedures. We also excluded women with multiple births and fetal demise because we hypothesized that their choices of postpartum contraception might be inherently different from those of women delivering liveborn singleton neonates.
A standardized code sheet was used to abstract the data. From the patient's prenatal emergency room visits and obstetric admission records, the authors determined basic demographic information and U.S. resident status (U.S. resident with a social security number compared with undocumented U.S. resident). In all of the University of Texas Health Sciences Center San Antonio prenatal records, there is a specific area that documents whether a woman requests postpartum tubal ligation antenatally, when the request is made, and when and if the proper consent forms have been completed. If women were transferred in from another facility, we abstracted the charts and records that we obtained during their admission to our center.
When women present to University of Texas Health Sciences Center San Antonio for obstetric admission, the clerks request identification. Patients are categorized as an undocumented U.S. resident if they do not have a social security number and their residence is in the United States. Admission personnel also ask patients their country of birth and residence to determine whether the patient is an undocumented U.S. resident or a foreign-born person living in the United States under various conditions such as a temporary work visa.
We abstracted whether the woman requested, received, or both requested and received postpartum tubal ligation during her obstetric admission. If postpartum tubal ligation was not requested (control group) or not obtained (PPTL NO group), we recorded which contraceptive the woman requested, if any, at the time of hospital discharge. The patient's choice of postpartum contraception is listed on the standard postpartum note of the inpatient electronic medical record. During the patient's postpartum admission, the attending and resident physician staff members of the obstetric service discuss methods of postpartum contraception with the patient. Patients also receive counseling regarding postpartum contraception routinely during their prenatal care visits.
We also examined the patients' electronic records to determine whether any of the following events occurred within 1 year of the index delivery: any positive pregnancy test, attendance at a postpartum visit within 8 weeks of the index delivery, any attendance for gynecologic care within 9 weeks to 1 year postpartum, report of contraceptive use between the index delivery and the postpartum visit, and any request for contraception at the postpartum visit. We did not contact patients directly to remind them of postpartum visits or inquire about intent of subsequent pregnancies or the outcome of these pregnancies.
A χ2 or Fisher's exact analysis was used for univariable comparisons of categorical variables and analysis of variance was used for univariable comparisons of continuous variables among multiple groups, with a Dunnett's test for comparison of the groups to the referent control group. Variables that were significant in the univariable analyses were used as predictors of pregnancy within 1 year of the index delivery, using multivariable logistic regression analyses. P<.05 was considered significant, and relative risk with a 95% confidence interval (CI) was used. SPSS was used for the analysis.
Between December 1, 2007, and May 31, 2008, 1,986 women delivered singleton liveborn neonates at the University of Texas Health Sciences Center San Antonio. We excluded 259 women: 245 due to missing data on postpartum contraception or intent to have postpartum tubal ligation, nine due to multiple gestations, and five due to intrauterine fetal demise. Also excluded were 267 women who were 20 years old or younger, leaving 1,460 eligible women. Of the 1,460, 429 (29.4%) requested postpartum tubal ligation and 1,031 (70.6%) did not (control group). Of the 429 women requesting postpartum tubal ligation, 296 (69%) received the procedure (PPTL YES) and 133 (31%) did not (PPTL NO). One-year follow-up data, as abstracted from clinic, emergency room, and hospital records, were available for 254 women in the PPTL YES group, 92 women in the PPTL NO group, and 581 women in the control group. A summary of the participants is illustrated in Figure 1.
Table 1 illustrates baseline demographic characteristics of the three cohorts. Further demographic description has been previously published.4 As previously reported, the gestational age, weight, and sex of the neonate was not predictive of whether a woman went on to receive a requested postpartum tubal ligation.4 Additionally, we found previously that women who successfully received postpartum tubal ligation were more likely (P<.05) to undergo cesarean delivery, be U.S. residents, have either public or private insurance during the pregnancy, and attend any prenatal care visits (as opposed to no prenatal care).4 Our previous data found that one variable—cesarean delivery—was most predictive of receiving postpartum tubal ligation (adjusted odds ratio [OR] 2.75, 95% CI 1.89-3.87).4
Table 2 details the immigration status of the three cohorts, demonstrating that women in the PPTL NO group were significantly more likely to be undocumented U.S. residents. Although the majority of our patients were Hispanic, the control cohort included significantly fewer Hispanic women than did the cohorts requesting postpartum tubal ligation.
Table 3 details the contraception requested by women in the control group compared with those in the PPTL NO group at the time of obstetric-admission discharge. Most women did not request contraception at their hospital discharge, the most common reason being that they wanted to wait until their postpartum examination to discuss this choice.
Table 4 shows the postpartum disposition of women at highest risk of subsequent pregnancy: the control and PPTL NO groups. Women in the PPTL YES group were significantly less likely to attend their postpartum visits than were those in the control group (38/296 [12.8%] compared with 209/1,031 [20.3%]; P=.004, OR 0.58, 95% CI 0.40-0.84) but had similarly low postpartum visit attendance as did women in the PPTL NO group (38/296 [12.8%] compared with 25/133 [18.8%]; P=.11, OR 0.64, 95% CI 0.37-1.10).
Women in the PPTL NO group were more likely to report any contraceptive use between the time of delivery and the time of the postpartum visit (mainly in the form of depot medroxyprogesterone acetate [data not shown]) than were those in the control group. Most women requested some form of prescription birth control at their postpartum visit (mainly in the form of oral contraceptive pills and depot medroxyprogesterone acetate [data not shown]). Women who did not receive a requested postpartum tubal ligation had significantly higher pregnancy rates within 1 year postpartum than did those in the control group, our primary outcome.
Multiple logistic regression analysis was used to determine independent predictors of pregnancy within 1 year of the index delivery, using the covariates of age, undocumented resident status, number of living children, gravidity, and delivery route. Pregnancy among women in the PPTL NO group was predicted by undocumented-resident status (adjusted OR 1.57, 95% CI 1.23-1.85) and having vaginal delivery for the index delivery (adjusted OR 1.14, 95% CI 1.03-1.26).
Our study found that women who did not receive a requested postpartum tubal ligation had a high rate of pregnancy within 1 year postpartum (46.7%), which was significantly higher than the rate for those in the control group (22.3%). Our data support that several possibilities exist to explain these differences, including general characteristics of the patients in the cohorts or, more likely, the socioeconomic influences of the health care and insurance system.
Previous studies have demonstrated that women who receive a requested postpartum tubal ligation generally are older and have had more children than women who do not receive a desired postpartum tubal ligation,6 whereas our cohort demonstrated the opposite: women in the PPTL NO group were older and had more living children than those in the PPTL YES group.4 Although women in the PPTL NO group were significantly older than were those in the control group, our population as a whole was young (mean age 28-31 years). Women in the PPTL NO group had significantly more pregnancies and more living children than did those in the control group, demonstrating potentially increased fecundability and fecundity. It is also possible that women who did not receive a requested postpartum tubal ligation subsequently changed their minds and conceived by choice. The retrospective nature of our data collection cannot exclude this possibility. Although the U.S. Collaborative Review of Sterilization reports that women who receive postpartum tubal ligation have similar rates of poststerilization regret as those who receive interval sterilization, we cannot exclude the possibility that women in our cohort who requested postpartum tubal ligation might have had different considerations than women who subsequently received an interval procedure.11 Regardless, the rate of subsequent short-interval pregnancies, defined as conceptions occurring so that the interval between live births is 18 months or less,12 among women who requested but did not receive postpartum tubal ligation was high.
More likely explanations of the increased pregnancy rate among women in the PPTL NO group are based on socioeconomic issues. Women in the PPTL NO and control groups were equally likely to interact with our clinic postpartum, at either a postpartum or general gynecologic visit. Women in these groups were equally likely to request contraception at hospital discharge and at the postpartum visit. In fact, the women in the PPTL NO group, although the sample size was small, were more likely to report using any birth control between hospital discharge and the postpartum examination.
Women who did not receive a requested postpartum tubal ligation were more likely to be undocumented U.S. residents. All undocumented residents in our cohort were of Hispanic ethnicity. Being an undocumented resident is a marker of altered access to health care. Although the Child Health Insurance Perinate Program has likely made prenatal care available to many women, it does not cover the cost of any postpartum contraception, including postpartum tubal ligation, nor does it cover the cost of any postpartum testing, including cervical cancer screening or testing for sexually transmitted infections.
We previously reported that the driving variable for obtaining postpartum tubal ligation in this cohort was cesarean delivery.4 Currently in Texas, women receiving insurance through the Child Health Insurance Perinate Program who undergo cesarean delivery may pay $1,000 for postpartum tubal ligation, and those who have vaginal delivery are required to pay $1,800 for postpartum tubal ligation (these costs were $200 for postpartum tubal ligation after cesarean delivery and $1,000 after vaginal delivery during the time of this study). Women with Child Health Insurance Perinate Program insurance, like women with Medicaid insurance, must also sign a Title XIX Medicaid sterilization consent form 30-180 days before the procedure.
We previously reported that the most common reason for not receiving postpartum tubal ligation was lack of funding, which occurred exclusively among patients with Child Health Insurance Perinate Program insurance.4 Approximately 85% of our patients who are documented U.S. residents used Medicaid as their health care payor, an insurance that provides for postpartum contraception. As U.S. residents, many of these patients may also qualify for the Texas Medicaid Women's Health Program (also known as the family planning waiver), which provides continued coverage for contraception beyond the 6-week postpartum period.
It is likely, although we cannot prove this with certainty from our data, that the increased pregnancy rates among women who did not receive postpartum tubal ligation resulted from a combination of increased fecundity, increased fecundability, and altered access to postpartum contraceptive services owing to immigration or insurance coverage status or both. Further qualitative interviews regarding pregnancy intention, contraceptive usage, and access to health care with women in the control and PPTL NO cohorts would strengthen this study. However, our data show that women in the PPTL NO group, who previously had decided that they desired permanent sterilization, had significantly higher pregnancy rates in the following year than did women who did not request permanent sterilization.
We did not evaluate women who received postpartum tubal ligation for the primary outcome of this study, pregnancy. Although we realize that pregnancies do occur after sterilization,13 we assumed that this event would be rare enough to not contribute to the overall analysis and interpretation of this study.
Attendance at the postpartum visit among all patients was low at 12.8-20.3%. We decided to count postpartum visits within 8 weeks of delivery, because the Child Health Insurance Perinate Program provides coverage for up to two postpartum visits within 60 days of the index delivery. Our postpartum attendance rates were consistent with other studies examining postpartum visit attendance.14 Postpartum visit attendance is known to be lower among women with public as opposed to private insurance, and among women in minority groups.15 We expected that postpartum visit attendance among women receiving permanent sterilization (PPTL YES) would be lower than among women at risk of subsequent pregnancy, as contraception is a major focus of the postpartum visit.16
Our data differ from those of Gilliam and coworkers in that women who did not receive postpartum tubal ligation chose similar contraceptive methods at hospital discharge as did women in the control group.7 The majority of women in the control and PPTL NO groups requested no birth control at hospital discharge. Among those requesting any contraceptive, women in the control group most commonly requested oral contraceptive pills and women in the PPTL NO group were slightly more likely (P=.12) to request intrauterine devices. This may reflect the higher proportion of women in the PPTL NO cohort who are undocumented immigrants, primarily from Mexico, a country that has higher intrauterine device-use rates than does the United States.17 Intrauterine devices were not inserted immediately after vaginal or cesarean delivery. Monitoring these women to the postpartum visit, our data demonstrate that women in the PPTL NO group were more likely to report using any contraceptive between delivery and the postpartum visit. However, this did not translate into longer-term (1 year) avoidance of pregnancy.
Previous authors have concluded that asking women, particularly those using emergency Medicaid, to pay out of pocket for sterilization services does not reduce overall hospital obstetric charges.5 Approximately half of pregnancies in the United States are unintended or mistimed, resulting in 3.1 million births.18 Half of these unintended pregnancies occur during a month in which the woman reports some contraceptive use.18 Contraception cost-effectiveness varies, but contraception in general is significantly more cost-effective than unintended pregnancy and its consequences.19,20
Our study has several limitations, most importantly the retrospective nature of data collection. We excluded 245 women initially because we could not determine from their antenatal medical records their choice of postpartum birth control. Because these women were not included in the analyses, we do not know their intentions regarding postpartum tubal ligation nor their postpartum outcomes. Not all women who delivered had a follow-up visit in our health system, and thus we could be underestimating the number of pregnancies. However, the University of Texas Health Sciences Center San Antonio, which is linked to most of the public health departments in San Antonio, clearly captures much of the health service to indigent patients and undocumented residents. We cannot exclude, however, women returning to their home country, moving out of state, or returning to other private clinics in the area.
The strength of our study is that it expands on previous work in understanding what happens to women who do not receive requested postpartum tubal ligations.7,8 In addition, the cohort in Gilliam and colleagues' studies was predominantly non-Hispanic African American, whereas our patients were primarily Hispanic. Our study may underestimate the number of undocumented U.S. residents with unfulfilled sterilization requests because although we discuss postpartum contraception with all our obstetric patients, we inform women with Child Health Insurance Perinate Program insurance about the additional costs of postpartum tubal ligation. It is possible that some women with this insurance do not request postpartum tubal ligation because they knew they could not afford the cost of the surgery. However, as policy makers consider the details of public funding of prenatal care and contraceptive services, our data will help anticipate the possible unintended consequences of various health care-funding policies.
In conclusion, women who request postpartum tubal ligation but do not obtain one are similar to women who do not request postpartum tubal ligation in their postpartum contraceptive choices and in their attendance at postpartum visits and gynecologic visits but have a higher likelihood of becoming pregnant within 1 year of the index delivery.
1. Chandra A, Martinez GM, Mosher WD, Abma JC, Jones J. Fertility, family planning, and reproductive health of U.S. women: data from the 2002 National Survey of Family Growth. Vital Health Stat 23 2005;(25):1–160.
2. Mosher WD, Martinez GM, Chandra A, Abma JC, Wilson SJ. Use of contraception and use of family planning services in the United States, 1982-2002. Advance data from vital and health statistics, no 350. Hyattsville (MD): National Center for Health Statistics; 2004. p. 1–36.
3. Benefits and risks of sterilization. ACOG Practice Bulletin No. 46. American College of Obstetricians and Gynecologists. Obstet Gynecol 2003;102:647–58.
4. Thurman AR, Harvey D, Shain RN. Unfulfilled postpartum sterilization requests. J Reprod Med 2009;54:467–72.
5. Rodriguez MI, Edelman A, Wallace N, Jensen JT. Denying postpartum sterilization to women with Emergency Medicaid does not reduce hospital charges. Contraception 2008;78:232–6.
6. Zite N, Wuellner S, Gilliam M. Failure to obtain desired postpartum sterilization: risk and predictors. Obstet Gynecol 2005;105:794–9.
7. Gilliam ML, Zite N, Wuellner S, Gunning P. Postpartum contraceptive use among women who did not undergo desired postpartum sterilization. Contraception 2005;72:240.
8. Gilliam M, Davis SD, Berlin A, Zite NB. A qualitative study of barriers to postpartum sterilization and women's attitudes toward unfulfilled sterilization requests. Contraception 2008;77:44–9.
9. Zite N, Wuellner S, Gilliam M. Barriers to obtaining a desired postpartum tubal sterilization. Contraception 2006;73:404–7.
10. Zite NB, Philipson SJ, Wallace LS. Consent to Sterilization section of the Medicaid-Title XIX form: is it understandable? Contraception 2007;75:256–60.
11. Hillis SD, Marchbanks PA, Tylor LR, Peterson HB. Poststerilization regret: findings from the United States Collaborative Review of Sterilization. Obstet Gynecol 1999;93:889–95.
13. Peterson HB, Xia Z, Hughes JM, Wilcox LS, Tylor LR, Trussell J. The risk of pregnancy after tubal sterilization: findings from the U.S. Collaborative Review of Sterilization. Am J Obstet Gynecol 1996;174:1161–8.
14. Clark HD, Graham ID, Karovitch A, Keely EJ. Do postal reminders increase postpartum screening of diabetes mellitus in women with gestational diabetes mellitus? A randomized controlled trial. Am J Obstet Gynecol 2009;200:634.e1–7.
15. Matijasevich A, Santos IS, Silveira MF, Domingues MR, Barros AJ, Marco PL, et al. Inequities in maternal postnatal visits among public and private patients: 2004 Pelotas cohort study. BMC Public Health 2009;9:335.
16. Katz VL. Obstetrics: normal and problem pregnancies. 5th ed. Philadelphia (PA): Churchill Livingstone; 2007.
17. Hubacher D. The checkered history and bright future of intrauterine contraception in the United States. Perspect Sex Reprod Health 2002;34:98–103.
18. Finer LB, Henshaw SK. Disparities in rates of unintended pregnancy in the United States, 1994 and 2001. Perspect Sex Reprod Health 2006;38:90–6.
19. Trussell J. The cost of unintended pregnancy in the United States. Contraception 2007;75:168–70.
20. Trussell J, Lalla AM, Doan QV, Reyes E, Pinto L, Gricar J. Cost effectiveness of contraceptives in the United States. Contraception 2009;79:5–14.