Abstract
Introduction:
Sterilization is more common in rural areas than in urban areas. We assessed whether desire for sterilization reversal also differs by rural/urban areas and by aspects of the sterilization care visit—specifically, method of payment and type of facility.
Methods:
We used data from 469 female 2015–2019 National Survey of Family Growth respondents who received a tubal ligation in the last 5 years. Multivariate logistic regression tested associations between wanting a sterilization reversal and residence, method of payment, and facility type, controlling for sociodemographic factors.
Results:
Overall, roughly one-quarter of respondents living in rural and urban areas wanted a reversal. Receiving a tubal ligation at a public facility was associated with lower odds of desire for reversal (adjusted odds ratio [aOR] = 0.48, p = 0.04, confidence intervals [CIs]: 0.24–0.97). Additionally, using Medicaid/receiving a tubal ligation for free was associated with marginally higher odds of desire for reversal (aOR = 1.92, p = 0.070, CIs: 0.95–3.91). Consistent with prior research, we found higher odds in the desire for reversal among those who received a tubal ligation for reasons other than ending childbearing and for non-Hispanic Black respondents, and lower odds for those with a history of unwanted pregnancy.
Conclusions:
The desire for sterilization reversal is similarly high among rural and urban women. However, aspects of the care visit did matter. To prevent sterilization procedures, people later want reversed; improvements in counseling are warranted across sites. At the same time, attention is needed to address structural barriers that limit contraceptive options.
Introduction
Tubal ligation is a frequently used method of contraception, with 18.1% of contraceptive method users aged 15–49 relying on sterilization in 2017–2019. 1 Population rates of sterilization differ by demographic characteristics 1,2 and are particularly high among rural women. 3 Data from 2015 to 2019 show that 31% of rural, sexually active women relied on sterilization compared with 19% of those in urban areas. 3 High rates of sterilization are not inherently concerning; sterilization is a highly effective method of pregnancy prevention 4,5 and is preferred by many. 5 However, sterilization is permanent and requires surgical intervention if users later decide they no longer want it. In fact, research finds that a substantial proportion of people—approximately 25%—who have had the procedure later report wanting it reversed. 5 –8
As the family planning field increasingly seeks to implement patient-centered contraceptive care, 9 understanding the use of preferred contraceptive methods—including sterilization—is important to monitor. 10 Wanting a sterilization reversed is associated with worse mental health 11 and may be a sign of limited contraceptive autonomy. 12,13 Furthermore, high levels of desire for sterilization reversal are especially troublesome given the history of forced and coerced sterilization in the United States, especially among some of the most marginalized populations. 14 –16 To better meet contraceptive preferences, it is critical to identify which populations are at particular risk of desire for reversal as well as identify the (potentially malleable) factors associated with this outcome.
In this article, we first provide updated data on differences in the desire for a tubal sterilization reversal between people living in rural and urban areas. Those living in rural areas experience significant barriers to accessing sexual and reproductive health care. 17 –21 Despite established rural/urban differences in the prevalence of sterilization, 3,22 limited research has examined differences in the desire for reversal. This prior research—which finds no significant differences across those in rural and urban areas—has relied on older data 23 and limited samples. 22,24 Since then, the rural/urban gap in sterilization use has grown. 25 Additionally, rural communities have faced ongoing challenges, including almost two decades of hospital closures, the emigration of reproductive health providers, and more limited resources. 17,26 –28 Those sites that remain often have lower rates of adequate staff, resources, training, and contraceptive methods than clinics in urban areas. 18 –20,29 –32 Furthermore, rural patients face additional barriers, such as low appointment availability and long travel times. 19,21,33 These factors may push rural patients, in particular, to receive sterilization, a procedure that they might not otherwise want if care were more accessible.
We additionally examine whether two aspects of the sterilization care visit itself—method of payment and facility type—are linked to the desire for reversal. Patients with Medicaid insurance are federally mandated to fill out consent forms and adhere to a 30-day waiting period prior to accessing sterilization. 34 Several studies have documented how these practices, while implemented to prevent forced or coerced sterilization, may also inhibit patients’ access to desired sterilization. 35 –38 It is possible that those who successfully navigate these administrative burdens were particularly motivated and thus may be less likely to later want their procedure reversed. Facility type may also play a role. Publicly funded facilities—including those funded by Title X—tend to have a wider range of contraceptive methods available and provide financial support to help cover the cost of care. 39,40 As a result, clients may not resort to sterilization when Long-acting reversible contraceptives are available. To the extent this is true, we may expect clients at publicly funded clinics to have a lower likelihood of a desire for reversal. No research we are aware of has examined these relationships. However, understanding the role that aspects of sterilization care delivery may play in the later desire for reversal can help inform and strengthen patient-centered family planning practice and policy. 28
Data, Measures, and Methods
Data and sample
To address our research aims, we analyzed pooled data from the 2015–2017 and 2017–2019 female respondent files a of the National Survey of Family Growth (NSFG). The NSFG, conducted by the National Center for Health Statistics, is a cross-sectional survey of U.S. residents aged 15–49 years and is the leading source of nationally representative data on pregnancy, births, marriage and cohabitation, contraceptive use, and general and reproductive health. 32 The NSFG oversamples people ages 15–19 years and non-Hispanic Black and Hispanic people. b Our primary analytic sample was limited to respondents who received a tubal sterilization in the past 5 years (N = 469), as these respondents were asked a series of questions related to their sterilization care experiences, including method of payment, type of facility, and reason for receiving sterilization. c
Measures
“Wanting a reversal” of tubal sterilization was the primary dependent variable examined in our analyses. Respondents who reported having a sterilization were asked, “As things look to you now, if your tubal sterilization could be reversed safely, would you want to have it reversed?” Respondents answered on a 4-point Likert scale ranging from “definitely yes” to “definitely no.” Consistent with prior research, those who answered “probably yes” or “definitely yes” were categorized as wanting a reversal. 8,23,41,42 Additionally, one respondent who reported having had their sterilization reversed was categorized as wanting a reversal (this makes up 0.2% of those who wanted a reversal).
We examine the role of several independent variables to address our research aims. A binary variable is used to measure rural/urban status. Those living outside a metropolitan statistical area (MSA) were coded as rural, 33 while those residing within an MSA—either the principal city of an MSA or other MSA—were coded as urban. We also included two binary variables related to aspects of sterilization service delivery: the payment method used for the procedure (private insurance/a copay/some other method versus Medicaid/free care) and the type of facility where the sterilization procedure was performed (public facility versus private facility).
We also include a series of measures in our analysis to control for sociodemographic measures linked to sterilization and the desire for reversal, including a continuous measure of age at interview (centered on the mean and an age-squared term to capture a possible nonlinear relationship), a three-category measure of parity (no children or one child, two children, three or more children), a four category race and ethnicity measure (non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic “other” or multiple races), and a binary measure of educational attainment (some college or more versus a high school diploma or less). We also include two reproductive health history measures: a binary measure of the respondents’ history of unwanted pregnancy (yes or no), d as unintended pregnancies are independently associated with the desire for reversal, 11,23,41 and a binary measure indicating the main reason for receiving the procedure (having had all the children the respondent wanted versus all other reasons, including medical reasons, respondents’ partner had all the children he wanted, problems with other methods of contraception, and “some other reason not mentioned”), which is also associated with the desire for reversal. 11,23
Analytic methods
This article includes descriptive, bivariate, and multivariate analyses of our sample of respondents whose sterilization was in the past 5 years. First, we conducted a univariate analysis to describe the characteristics of the sample across all dependent and independent measures. We next conducted bivariate logistic regression analyses to assess whether our key predictors of interest—rural/urban areas and aspects of sterilization service delivery—are independently associated with the desire for reversal. We then moved forward with multivariate logistic regression models to test whether these associations persisted when controlling for sociodemographic and reproductive health history controls. e We ran two-stage models, first testing the association of rural/urban area without including aspects of sterilization service delivery but controlling for covariates and then a model incorporating aspects of sterilization service delivery. Across all regressions, we performed Wald tests of the two categorical covariates (race/ethnicity and parity) to assess the significance of its overall association with the desire for reversal. All analyses were performed in Stata 16.1 and using the “svy” command to weight the data, accounting for the NSFG’s complex sample design. 34,35
Results
Sample characteristics
Table 1 shows weighted descriptive characteristics of respondents who had a sterilization in the past 5 years, overall and separately by rural and urban status (N = 469). Twenty-eight percent of all respondents reported wanting a sterilization reversal, and this did not differ significantly by rural or urban area. Turning to our main independent variables, roughly half of the sample (50.1%) used public insurance to pay for their procedure or had it subsidized, and just over a quarter received their sterilization at a public healthcare facility (27.3%). These percentages also did not vary significantly by rural or urban area.
Characteristics of Respondents Who Received a Tubal Ligation in the Past 5 Years, 2015–2019
“Other” insurance reflects the response “some other way” to the NSFG’s variables series PAYRSTER, which asks “Looking at Card 16, please tell me all of the ways in which the bill for this operation was paid.”
“Other” reasons for sterilization include the respondents’ partners’ desire to end childbearing, problems with other contraceptive methods, and some other reasons not listed.
NSFG, National Survey of Family Growth; NH, non-Hispanic.
Regression analyses
Table 2 displays results from bivariate and multivariate logistic regressions testing associations between desire for sterilization reversal and rural/urban area, aspects of sterilization service delivery, and other demographic and reproductive health history characteristics. Bivariate results indicate that rural residents are not more likely than urban residents to want a reversal (OR = 1.18, p = 0.674). The type of facility where the sterilization procedure was performed was also not significantly associated with the desire for a reversal. However, bivariate results do suggest that those who used public insurance or whose procedure was free had more than two times higher odds of wanting a reversal (OR = 2.27, p = 0.024).
Results from Bivariate and Multivariate Logistic Regressions Testing the Associations of Respondent Characteristics with Wanting a Tubal Ligation Reversal, 2015–2019
“Other” reasons include medical reasons, the respondents’ partners’ desire to end childbearing, problems with other contraceptive methods, and some other reasons not listed.
“Other” insurance reflects the response “some other way” to the NSFG’s variables series PAYRSTER, which asks “Looking at Card 16, please tell me all of the ways in which the bill for this operation was paid.”
aOR, adjusted odds ratio; CIs, confidence intervals; NSFG, National Survey of Family Growth; NH, non-Hispanic; OR, odds ratio.
We found significant disparities in the desire for reversal by sociodemographic and reproductive health history characteristics, without accounting for other characteristics. The odds of wanting a reversal were higher among non-Hispanic Black respondents (OR = 2.92, p = 0.026) and lower among those with at least some college education (OR = 0.52, p = 0.032) compared with their counterparts. Age also had a significant, linear association with the desire for reversal; we found that the odds of desire decreased as respondents’ age increased (OR = 0.88, p = 0.000), and the squared term was not significant (OR = 1.00, p = 0.544). Additionally, those who experienced at least one unwanted pregnancy had lower odds of wanting a reversal (OR = 0.48, p = 0.050) compared with those who never had, and participants who received a sterilization for reasons other than being done with childbearing had higher odds (OR = 3.56, p = 0.001) of wanting a reversal than those whose main reason for reversal was that they were finished having children.
Table 2 also displays results from our multivariate models. Model 1 considers rural and urban areas, controlling for sociodemographic and reproductive health history characteristics. Consistent with the bivariate analysis, there were no differences by rural and urban areas of residence, though associations with the control variables persisted. Model 2 adds aspects of sterilization service delivery. Again, rural/urban area was not significantly associated with the desire for reversal. Compared with the bivariate analyses, receiving a publicly subsidized procedure was still positively associated with a desire for reversal but only marginally significant (aOR = 1.94, p = 0.070). However, unlike in the bivariate findings, receiving the procedure at a public facility was associated with lower odds of desiring reversal (aOR = 0.48, p = 0.043). Associations between sociodemographic and reproductive health history characteristics and desire for reversal were similar to Model 1, with the exception that the coefficients for higher levels of education and higher parity no longer reached significance.
Discussion
Identifying factors linked to the desire for sterilization reversal can help health care providers determine how to best support patients’ reproductive health needs. To this end, this article provided updated information on rates of desire for a sterilization reversal, paying attention to differences by where respondents live and two aspects of sterilization service delivery—payment type and facility type.
Given the higher rates of sterilization and more limited access to care, we expected that respondents in rural areas might be more likely to want a sterilization reversal than those in urban areas, but this was not the case. Consistent with previous research, 8,11 we found that roughly one-quarter of sterilization users (who received their procedure in the past 5 years) in both rural and urban areas report wanting a reversal. This is true despite the notably higher rates of sterilization in rural communities. It is possible that those who live in rural areas may remain content with sterilization, knowing they have more limited access to alternative contraceptive methods. 20,30 –32 More research should explore differences in rural and urban women’s motivations for sterilization, with particular attention to their reproductive health care environments.
Notably, aspects of the sterilization care visit were associated with a desire for reversal. We found that those who received their sterilizations at public facilities were less likely than those who received their procedure at a private facility to later want a reversal when taking other characteristics into account. Publicly funded clinics, particularly Title X-funded facilities, tend to have a wider range of contraceptive methods available for patients. 18,40 It is possible that the wider accessibility of methods means that those who received sterilization care there are more actively motivated to choose sterilization over other methods and, as such, less likely to later want it reversed. While Title X and other (public and private) reproductive health care providers offer patient-centered contraceptive care at similar rates, 43 more studies should explore what clinic-level factors (including the nature of sterilization counseling) are linked to the desire for reversal.
In unadjusted analyses, people who paid for their sterilization with Medicaid or had their procedure subsidized were more than two times as likely to want a reversal than those who paid with private insurance. 44,45 However, this effect was attenuated and no longer significant when accounting for other characteristics. This finding may suggest that federally mandated consent and waiting periods linked to Medicaid use to pay for sterilization may not be protective against the desire for reversal, as these policies were intended. Recent interviews with providers and postpartum sterilization patients highlight some limitations of the Medicaid waiting period while also noting the need for interventions to support noncoercive, patient-centered sterilization care. 46 –49 In fact, some medical providers and researchers have questioned whether these practices are still necessary or even ethical, 50,51 and the Coalition to Expand Contraceptive Access recently published guidelines for altering these practices. 28
While some prior research found no association between a history of unintended pregnancy and the desire for a reversal, 41 our study suggests that having had an unwanted pregnancy substantially lowered the odds of desire for a reversal. Measurement differences may play a role. In contrast to some research that measures unintended pregnancy (a combination of both mistimed and unwanted pregnancies), 41,52 we focused specifically on the role of unwanted pregnancy. Still, in our sample, one in five of those who had an unwanted pregnancy still want a reversal. Increased access to a wide range of preferred contraceptive methods may help prevent unwanted pregnancies and sterilizations that patients later want reversed. 53
We also confirm prior work finding that those whose primary reason for sterilization (as reported at the time of interview) was for nonchildbearing reasons (including medical reasons, having a partner who wanted to end childbearing, or problems with other contraceptive methods) is linked to increased desire for reversal. 11 Other articles have suggested that receiving the procedure for medical reasons specifically may reflect the influence or potentially coercive role that providers have over sterilization decisions. 23,54 More information is needed to understand how both medical and other non-contraceptive-related reasons for sterilization are navigated in practice given the high rates of desire for reversal among these respondents. This may require a deeper look at the alternatives to sterilization available to people with reproductive medical conditions or the consenting practices needed in such situations. Decisions to receive sterilization are multifaceted and complex, 55 –58 and this complexity should be accounted for in sterilization counseling.
Consistent with prior research, 23,42,54 we document racial disparities in the desire for sterilization reversal, with non-Hispanic Black respondents experiencing the highest rates of desire for a reversal. Such findings have motivated calls for policies that prevent racially biased coercion, 59 more research to understand unique pathways to sterilization among distinct racial and ethnic identities, 54 and interventions that increase access to the full range of contraceptive methods for medically underserved populations. 23 Also consistent with prior research, 8,17 those with a high school diploma or less were significantly more likely to want a reversal, before accounting for sterilization care characteristics. Given that research has rated the federal Medicaid sterilization consent form poorly in terms of readability, 60 particular attention to person-centered sterilization counseling and consenting practices for those with lower levels of education is warranted. The field may need to explore alternative (and standardized) approaches that support patient autonomy and decision making, particularly for Black women, and across patients with all types of insurance coverage.
There are some limitations to our study. Our sample size is relatively small, especially for rural residents. Unfortunately, we could not boost our sample size by incorporating additional waves of NSFG data given a skip pattern error in the 2011–2013 and 2013–2015 waves of data that excluded single women from the question regarding the desire for reversal. 61 As future waves of the NSFG (and appropriate linked weights) become available, it will be important to continue these lines of inquiry. Additionally, there were limited measures in the NSFG that allowed us to understand how characteristics of sterilization service delivery at the time of procedure influence the current desire for reversal. Prospective studies with more detail on visit and provider characteristics would help better explore these longitudinal links. Furthermore, we assessed the desire for reversal using a Likert scale; while our measure was highly correlated with the desire for more children, it is difficult to interpret the experiences of those who reported they would “probably” want a reversal. Further research—quantitative and qualitative—should explore the role that interpersonal and cultural factors, 45 as well as structural measures not available in the NSFG, play in sterilization rates for rural and urban women across the United States.
Conclusions
The desire for sterilization reversal remains high, particularly among women who received their sterilization at a private facility, identify as non-Hispanic Black, have lower levels of education, did not have a prior unwanted pregnancy, and whose main reason for tubal ligation (as reported at the time of the interview) was one other than being done with childbearing. Research suggests counseling should be accessible across patients’ literacy levels and primary languages, and should aim to increase participants’ understanding of the permanency of this procedure, 62 which can increase patients’ knowledge and certainty about sterilization. 63 More research is warranted to understand the role of factors we did not measure, including different counseling approaches, on the later desire for sterilization reversal. At the same time, attention is needed to address the structural oppressions and barriers to contraception that may motivate people—for example, those who do not have access to clinics with wide-ranging contraceptive options—to receive a sterilization that they will later want reversed.
Footnotes
Authors’ Contributions
E.P.: Conceptualization (lead); methodology; formal analysis; and writing—original draft (lead). E.W.: Conceptualization (supporting); methodology; supervision; and writing—review and editing (lead). J.S.M.: Writing—original draft (supporting) and writing—review and editing (equal). K.W.: Writing—review and editing (equal).
Author Disclosure Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Funding Information
This publication was made possible by Grant Number 5 FPRPA006076-02–00 from the Office of Population Affairs (OPA), U.S. Department of Health and Human Services (HHS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of OPA or HHS.
