Informed choice and female sterilization in South Asia and Latin - - PDF document

Page 1 of 25 Informed choice and female sterilization in South Asia and Latin America Apoorva Jadhav, University of Michigan Emily Vala-Haynes, Western Oregon University Corresponding author: Apoorva Jadhav Population Studies Centre 426 Thompson St, Rm 3471 Ann Arbor, MI 48104 USA

Page 2 of 25 SUMMARY Globally, female sterilization is one of the most popular contraceptive methods despite concerns about quality of care for women who report being sterilized. We quantify informed choice among sterilized women using Demographic and Health Survey (DHS) data from 2000 to 2012 from countries in South Asia and Latin America. We use three responses to measure informed choice and knowledge about whether women were informed by a health worker or provider that: sterilization is permanent, potential side effects, other methods of contraception. We create an ascending composite Method Information Index with scores ranging from 0 (women received no information) to 3 (women received information across all three indicators). Using ordinal logistic regression analysis we find that women younger than 25 and older than 35 at the time of sterilization, and those at high parities had lower odds of a high score on the index, while the opposite is true for women sterilized in the private sector in Latin America. Educated women in India had higher odds of a high score on the index, while the same is true for educated and wealthy women in Colombia. Our findings indicate that not enough health care providers spend time informing women about different aspects of sterilization, and that there are specific groups

• f women that are most affected. There is an urgent need to improve quality of care within health

systems providing sterilization for this very important and effective type of contraception.

Page 3 of 25 INTRODUCTION Female sterilization is one of the most common methods of contraception in the developing world,

• wing to its long-term effectiveness and low cost (Hosseni, Torabi, & Bagi, 2014; Joshi, Khadilkar, &

Patel, 2015). According to United Nations estimates, about 1 out of 5 women worldwide are sterilized, with a large portion in Asia – 28% of women in China and close to 36% in India – and Latin America – 47.4% in the Dominican Republic, 35% in Colombia, and 24% in Nicaragua (United Nations, 2013). It is thus a highly important option for contraception, provided that the decision to undergo sterilization is voluntary and informed. Yet, doubts remain regarding the extent to which this method is promoted through coercion and lack of informed choice regarding the procedure (Hardee et al., 2014; WHO, 2014). There have been allegations and proven cases of coercion associated with female sterilization

• ver the years in countries around the world- Namibia (Mallet & Kalambi, 2008), South Africa (Essack &

Strode, 2012), United States (Salvo, Powers, & Cooney, 1992; Stern, 2005), Peru (Miranda & Yamin, 2004), and India (Koenig, Foo, & Joshi, 2000; Singh, Ogollah, Ram, & Pallikadavath, 2012). Yet, it is among the most used methods among women in the developing world, so it is important to ensure that women are making an informed choice to use this method. This paper assesses the recent evidence from countries in the developing world that women are making informed choices about sterilization. In a climate of discussion around the extremes of forced and coerced female sterilization, there remains a lack of empirical research on the topic of informed choice—reports tend to be limited to specific human rights violations, health policy, or analyses of limited samples. In this study, we bridge that gap and provide a comprehensive picture of sterilization using the best and most current quantitative data to examine informed choice and female sterilization in 7 countries in Latin America and South Asia. Our research is motivated by underlying mechanisms of coercion (Hardee et al., 2014; WHO, 2014) and highlights specific groups of women who are not given sufficient information to make informed decisions about this permanent method. Coerced sterilization, informed choice and contraceptive counselling Coercion in the context of sterilization is defined as “actions or factors that compromise individual autonomy, agency or liberty in relation to contraceptive use or reproductive decision making through force, violence, intimidation or manipulation” (Hardee et al., 2014). It has been especially problematic within family planning programs in many countries for decades, in some cases where population control strategies are synonymous with female sterilization (Basu, 1984; Singh et al., 2012). Moreover, incentive payments in India, Bangladesh, Nepal and the Philippines have raised concerns that women in those

Page 4 of 25 countries are unduly pressured into being sterilized (Hardee et al., 2014). Throughout the world, women who are poor, indigenous, and/or rural residents have historically been more vulnerable to coercive sterilization (Hardee et al., 2014; Miranda & Yamin, 2004; Pieper Mooney, 2010). Questions thus remain about how to reconcile promotion of female sterilization as a family planning method with this problematic history. As programs move forward, an important place to begin is the interaction between a woman and her health care provider and, specifically, whether that provider ensures the woman receives comprehensive information to make an informed choice about whether or not to be sterilized. A woman’s ability to freely and responsibly choose a contraceptive method depends upon several components, primarily being the information supplied to her by a medical provider. In Brazil, female sterilization is a popular method particularly for women who give birth in a private hospital and who have health insurance (Amaral & Potter, 2015), indicating the importance of discussion about family planning options with a health provider. In Ethiopia, home visits by family planning workers are associated with an increase in long acting reversible and permanent contraceptives, including female sterilization (Teferra & Wondifraw, 2015). While the importance of proper information about family planning methods and choice is an essential factor in deciding to choose sterilization, it is unclear how many women are actually adequately counselled. A review in India found that women were not informed about other methods and spacing options; limited information was passed on from health workers to women about side effects and how to deal with these side effects, and quality of care was poor during and following the operation (Koenig et al., 2000). A case control study in Bangladesh found that sites in which female field workers paid regular visits and emphasized a variety of contraceptive methods, and where awareness of clinics to treat complications arising from sterilization was acute, were more successful in increasing contraceptive use compared to control areas where there were no field workers (Koenig, Rob, Khan, Chakraborty, & Fauveau, 1992). According to this study, these factors even outweighed the importance of socioeconomic status and education on increases in family planning, which is of critical importance, given policy focus on education in increasing contraceptive use. Studies ranging from Pakistan (Saeed, Fakhar, Rahim, & Tabassum, 2008), South Africa (Credé et al., 2012) to the United States (Lee, Parisi, Akers, Borrerro, & Schwarz, 2011) have shown that there is an increase in contraceptive uptake among women who were provided counselling and education about modern family planning methods, pointing to the importance of counselling and speaking to a health provider in informed choice. National family planning programs and the rise of female sterilization

Page 5 of 25 Many countries emphasize female sterilization as part of their family planning programs. There is no single factor that can explain the rise of this method in the two diverse regions of South Asia and Latin

• America. Rather, the push toward increasing accessibility of female sterilization and in some cases

explicitly promoting it, are rooted in a variety of historical and cultural elements. From pragmatic consideration of the procedure’s effectiveness and pressure to meet targets in India, to government efforts meant to systematically reduce fertility within indigenous, disabled, and HIV positive women in Latin America, context varies and country-specific examination is imperative. We briefly review some of these programs in the countries we focus on in this paper. South Asia In Asia – particularly India and Nepal – governments have put considerable effort into the promotion of female sterilization in family planning programs. India began a state-sponsored campaign to lower fertility in the 1950s, which succeeded in steadily reducing the country’s TFR from approximately 6.5 children per woman to less than 3 children per woman in 2010 (Visaria, Jejeebhoy, & Merrick, 1999). However, in a 1992 report the Indian government acknowledged significant problems in the program— namely, the use of numerical targets that incentivized health workers to coerce couples into accepting sterilization in order to avoid negative career consequences. This led not only to misreporting and data manipulation, but also to a concerning disregard for the health and agency of women (Visaria et al., 1999). Indeed, since the government began to heavily promote sterilization as a preferred method of family planning for Indian women, there have been reports of coercion and poor quality of care for

• women. The 12 deaths following a mass sterilization camp in 2014 brought malpractice to light and

illuminated shoddy procedures that leave marginalized women the most vulnerable to coercion (Barry and Raj, 2014). In Nepal, women have access to family planning services through a public sector healthcare system, as well as several NGOs. While those services were only offered in clinics during the early years

• f the country’s family planning outreach, they are now offered in mobile outreach camps that target

women in rural areas (Thapa & Friedman, 1998). Similar to India, mobilizing the delivery of female sterilization has been integral to the reduction of fertility rates in Nepal. While Thapa and Friedman (1998) found no evidence supporting increased risk for coercion or complications in these camps, they also indicate that the majority of camps neither offer nor provide other methods of family planning. A more recent study found that there were no efforts to promote temporary methods of family planning, and most were not available in mobile clinics (EngenderHealth, 2003). In interviews, providers said that

Page 6 of 25 women did not return for follow-up visits and discontinuation rates were high—ultimately arguing for higher quality contraceptive counselling (EngenderHealth, 2003). Latin America In contrast to South Asia, Latin American programs more specific population groups: indigenous women, women living with HIV, and women with disabilities (WHO, 2013). Latin America, a mixture of religious influences, international aid, and national family planning programs resulted in high rates of female sterilization adoption and several cases of coercion, with the most egregious being that in Peru under the Fujimori regime (Boesten, 2007). In Peru, the commitment to expansion of family planning services for all women began in the 1980s with the creation of the National Family Planning Program. Aided by donations and a supportive presidential administration, newly available contraceptives succeeded in reaching poor and rural women—the same demographic groups targeted in South Asia—who had previously lacked access (Gribble, Sharma, & Menotti, 2007). Female sterilization was not officially legalized until 1995; however, widespread reports of coerced and forced sterilization of indigenous women emerged during this period (Miranda & Yamin, 2004). The Peruvian government has hedged several times in its willingness to acknowledge these events, but in 2015 it created a registry for women who were forcibly sterilized and is gathering further evidence in order to properly determine the extent to which it was practiced (Lizarzaburu, 2015). Since the 1960s, Colombia has enjoyed steady progress in reducing fertility with well-run family planning programs. Sterilization is the most common family planning method in Colombia, and has grown in popularity over the past three decades. With that growth has come a decline of the temporary methods traditionally provided by private sector pharmacies, and the public sector now accounts for the majority of family planning service delivery (Bertrand, 2011). However, doubts regarding the voluntary nature of sterilization were raised in the mid-1980s after Profamilia, the leading private sector provider, began mobilizing medical teams and equipment to reach hospitals in remote areas of the country (Rizo & Roper, 1986). In addition, there have been reports of forced sterilizations, some of which targeted girls with learning disabilities (IWHR, 2013). Private clinics were the first to offer sterilization in the late 1970s in Nicaragua. Currently, most family planning—including female sterilization, which is the most popular method—is accessed through public Ministry of Health units (Santiso-Galvez, Ward, & Bertrand, 2015). Still, concerning reports of forced sterilization of HIV-positive women have surfaced (Kendall & Albert, 2015). These countries

Page 7 of 25 demonstrate how governments in concert with international aid provide family planning to women, with increasing demand and emphasis for female sterilization. Conversations about sterilization and family planning in general have been difficult in conservative and religious countries in Latin America- particularly in Bolivia and Honduras. Due to pronatalist governments and pressure from the Roman Catholic Church, Bolivia had a relatively late start in terms of effective family planning programs. Bolivia has a large indigenous population that has proven difficult to reach; moreover, that population resisted family planning in general for many years (Terborgh et al., 1995). Although sterilization does not dominate method mix in Bolivia as in some other Latin American countries, there are some reports of forced sterilization of indigenous Quechua women in that country by Peace Corps volunteers during the 1960s (Geidel, 2010). These reports are unsubstantiated, but given the prevalence of coerced sterilization among indigenous women, as solidly evidenced in neighbouring Peru, we keep Bolivia in the analysis. Family planning programs in Honduras, funded by USAID, had minimal success until the 1990s due to political and religious barriers, as well as implementation challenges. Quality clinical services were not available in rural areas, and Ministry of Health clinics provided low quality services with limited patient education (Martin, Buttari, Macias, & Cobb, 1993). Female sterilization is currently one of the most widely used family methods in the country (Coa & Ochoa, 2009). There have clearly been allegations and proven cases of coercion associated with female sterilization in several countries around the world. Yet, female sterilization remains an integral part of any family planning program. It is therefore important to ensure that women are making an informed decision to use this method. In this study, we explore decision-making surrounding sterilization for women by assessing recent evidence from the developing world. Specifically, we determine whether women were informed about: sterilization is a permanent method of contraception, potential side effects of sterilization, other available methods of contraception. This is the first study to our knowledge to examine this issue closely and to this extent in conjunction with female sterilization. METHODS Sample and Data We use the most recent waves of Demographic and Health Surveys (DHS) for our analyses. The DHS are nationally representative datasets collected by ICF International in partnership with Ministries of Health

Page 8 of 25 from respective countries using a stratified random sampling approach. We use the following country- specific inclusion criteria for women between the ages of 15 and 49: countries where female sterilization is reported as one of the main methods of modern contraception, where women who reported being sterilized also answered the three questions regarding informed choice, and countries with historical emphasis on female sterilization as part of national family planning programs or where coercive sterilization was practiced. These include countries in South Asia: India (2005), Nepal (2001, 2006); and Latin America: Peru (2000, 2009, 2012), Colombia (2000, 2005), Bolivia (2003, 2008) Nicaragua (2001) and Honduras (2005, 2011). We pool these cross-sectional surveys for each country and present analyses at the country level with year-level fixed effects. We restrict our sample to women who were sterilized within 5 years of the survey to ensure limited recall bias. Our total sample size of sterilized women varies by country: India (9,577), Nepal (1,046), Bolivia (727), Colombia (8,797), Honduras (2,672), Nicaragua (989), and Peru (4,027). We used publically available secondary data from the DHS, thus did not require ethical approval from our institutions. Dependent variable We identify key variables of informed choice in specific countries based on questions that conform to the 2014 interagency statement on Eliminating Forced, Coercive and Otherwise Involuntary Sterilization from the World Health Organization and six other United Nations programs (WHO, 2014), which specifies that counselling on sterilization should include three important points: a) the procedure is considered permanent; b) there are potential side effects of sterilization; and c) alternative temporary methods of contraception (WHO, 2014). The answers to three questions on informed consent are the basis of our three dependent variables among women who report being sterilized, coded as “0” if the respondent was not informed and “1” if respondent was informed. The first examines whether the respondent was made aware by a health or family planning worker that sterilization is a permanent method: “Before your sterilization operation, were you told that you would not be able to have any more children because of the operation?” The second question asks whether the respondent was informed about side effects associated with female sterilization: “Were you ever told by a health or family planning worker about side effects or problems you might have?” Finally, the respondent was asked whether she had been informed of other available contraceptive methods: “Were you told about

• ther methods of family planning which you could use?” We then created a composite index to capture

informed choice based on responses to these three questions, in line with the Method Information Index (MII) adapted from the FP2020 2014-15 Report (Track20, 2015). Our index thus consists of 4 levels

Page 9 of 25

• f informed choice prior to sterilization in ascending order: 0 (not informed across any of the 3

indicators), 1 (informed on one indicator), 2 (informed on two indicators), and 3 (informed across all three indicators). Independent variables We use two main sets of independent variables in the analysis to determine which respondent characteristics are associated with increased or decreased levels of informed consent. The first set consists of sterilization-related variables, specifically, age at sterilization (<25, 25-29, 30-34 and 35+), parity at sterilization (2, 3, 4, 5+ children), and source of sterilization (public sector, private sector,

• ther). In a review of 19 studies analysing the relationship between sterilization and regret, Curtis et al.

(2016) concluded that younger age at sterilization was significantly associated with regretting that decision at a later age. There is also some evidence that program systematically target women at higher parity for sterilization (Donaldson 2002). Source of sterilization and its relationship to informed choice is important to consider given the history of coercion within public family planning programs, as well as to direct policy recommendations (Sullivan et al. 2006). The second set of independent variables includes demographic characteristics to serve as

• controls. These include respondent’s wealth quintile, type of residence (rural vs. urban), and education

(none, primary, secondary, higher). Evidence from Latin America shows that sterilization is more common among wealthy women, but some reports suggest that although poor women are less likely to be sterilized, those who had undergone the procedure were less informed and in many cases coerced (da Costa Leite, Gupta, & do Nascimento Rodrigues, 2004; Miranda & Yamin, 2004). Sterilization camps in India specifically target poor women, many of whom live in rural areas with little opportunity for education (Barry & Raj, 2014). Thus, while wealthy women may be the majority of those who opt for sterilization, it is important to highlight the different decision-making processes—or lack thereof— among women of different wealth and education levels. Analysis Our analytic strategy is threefold. We first present demographic (education, wealth quintile, type of residence) and sterilization-specific (age and parity at sterilization, source of sterilization) characteristics for women who are sterilized by country. Next, we present the proportions of women in each country by the 3 indicators that constitute the MII as well as the composite index. Finally, we use an ordinal logistic regression analysis to determine the odds of informed choice after controlling for key

Page 10 of 25 background and sterilization-related variables. For countries with multiple survey years, we include year-level fixed effects. We use person-level weights for descriptive and regression analysis using Stata 13 software. RESULTS Descriptive Results Of modern contraceptive users in each country, women in Latin America and South Asia use female sterilization as their contraceptive method compared to other modern methods (Figure 1). India is a striking example, with about 77% of women sterilized compared to 23% using all other modern methods (pill, IUD, injectables, condoms, etc.). Close to half of all modern contraceptive users in Nepal (43.0%), Colombia (42.4%), and Nicaragua (42.3%) report female sterilization, followed closely by Honduras (38.3%). Peru (19.2%) and Bolivia (18.9%) report the lowest proportions, but still constitute close to one- fifth of modern contraceptive methods in those countries. Table 1 shows demographic variables for sterilized women with t-tests comparing sterilized women to

• ther modern method users and contraceptive non-users. Three main patterns are apparent. First, most

women in the sample who are sterilized have at least a primary education, with the exception of Nepal and India, where 78% and 47%, respectively, report no formal education. T-tests indicate that these characteristics are significantly different for women who are sterilized from other modern method users and contraceptive non-users in all countries save for Nicaragua. Second, there does not seem to be a clear relationship between wealth and sterilization, with women in the higher wealth quintiles sterilized as often as those in the second or middle quintiles. However, some countries— Bolivia, Nicaragua, and Peru—had notably small percentages of women in the lowest quintiles. These characteristics are significantly different from other modern method users and contraceptive non-users in India, Colombia, Honduras, Nicaragua, and Peru, and significantly different from contraceptive non-users in Nepal and

• Bolivia. Finally, sterilization is not necessarily a rural phenomenon, and within-region differences exist as
• well. More than 80% of sterilized women in Nepal live in rural areas, as well as 72% of sterilized women

in India. In all other countries, however, fewer than 50% of sterilized women live in rural areas. These characteristics are significantly different from other modern method users and contraceptive non-users in all countries save for Bolivia. Table 2 describes sterilization-related variables for the sample. Most women tend to undergo sterilization after age 25– with the notable exception of India where almost 43% are sterilized before

Page 11 of 25 that age. In Latin America, women tended to be older at the time of sterilization, with the majority in Bolivia, Colombia, and Peru sterilized at age 35 or higher. Women in India and Colombia report sterilization at low parities of 2 or fewer children, while women in Nepal, Honduras, and Peru are sterilized after having 3 children. In Bolivia and Nicaragua, women are at high parities of 5 or more children at the time of sterilization. Across the sample, a bulk of sterilization occurs in the public sector- governmental hospitals, rural sub-centres, mobile clinics, and camps – compared to the private sector, which is not surprising given that the public sectors operate on national family planning schedules and are the most affordable option. Of the three measures of informed consent—informed that sterilization is permanent, informed about potential side effects, informed about other methods of family planning—variation again follows country rather than regional patterns (Table 3). The first panel shows that of the three indicators, a higher proportion of women were informed that sterilization is permanent across all countries, with the proportions highest in Peru and Bolivia at around 95%, and lowest in India and Nepal at 66% and 78%

• respectively. While this indicator performs well, the other two do not. About 60% of women in Peru and

Bolivia are informed about side effects and other available methods of family planning; the same is true for close to half of women in Colombia, Honduras, and Nicaragua. The proportion of informed choice for these two indicators is lowest in India at 31% for information about side effects, and an even lower 24% for information about other available methods of family planning. The second panel shows results by the MII. About a quarter of women in India and 14% in Nepal reportedly did not receive any information related to sterilization, in contrast to less than 1% of women reporting the same in Latin America. Across all countries, most women received information on at least 1 of the indicators, with 40% in India, and about 33% in Colombia, Honduras, and Nicaragua. Notably, a majority of women in Bolivia (45%) and Peru (43%) reported receiving information across all 3

• indicators. The highest mean score on the MII (ranging from 0 to 4) was 2.2 in Bolivia and Peru, while

the lowest was in India at 1.3. Regression Results The construction of the MII sheds light on overall patterns of informed choice, and forms the basis of

• ur regression analysis (Table 4). For women who were sterilized in India before age 25 - which refers to

a majority of sterilization cases in the country – the odds of informed choice (a high score on the MII) are significantly lower compared to women who were sterilized at age 25-29. This is also true, though

Page 12 of 25 conversely, for women who were sterilized at older ages (35+) in Honduras, Nicaragua, and Peru. Women who were sterilized at higher parities compared to those at 2 or fewer children were also significantly less likely to be well informed about sterilization indicators in India, Colombia, and Peru across all parities. It appears that source of sterilization is a critical factor in the dissemination of information, with women sterilized in the private sector in Colombia, Honduras, and Nicaragua reporting higher odds of informed choice compared to women in those countries who were sterilized in the public sector, largely in the realm of government hospitals and clinics. It is not surprising that certain demographic characteristics are protective when it comes to who receives information and who does not. Women who are educated compared to those who receive no education have significantly higher odds of being informed in India, Bolivia, Colombia, and Honduras. This is particularly true at secondary education levels and beyond. Interestingly, wealth does not seem to matter in this scenario, save for Colombia, where women at higher wealth quintiles are significantly and progressively at a higher odds of receiving information compared to the poorest women in that

• country. There is some indication that women in Nicaragua in the fourth highest wealth quintile actually

had lower odds of receiving information compared to the poorest women, and warrants further

• investigation. In a testament potentially to national family planning programs, it seems that rural

women in India have a lower odds of informed choice, while the opposite is true in Colombia, where rural women actually had higher odds of informed choice regarding sterilization. DISCUSSION Very few studies to date quantitatively examine the relationship between female sterilization and informed choice. A study in India found that most sterilized women were not informed about other methods of contraception nor were they informed about possible post-sterilization health problems by providers, and the authors question the intention of the providers and governmental family planning program and quality of services provided (Pradhan & Ram, 2009). It has been shown that once women receive full information on various methods available to them by healthcare providers, they are able to make a decision away from permanent methods to reversible methods for spacing or stopping (Baveja et al., 2000), signalling that the high proportions of women opting for sterilization may actually reflect lack of knowledge rather than a preference for that method. While we are unable to determine

Page 13 of 25 whether coercion took place, our study is the first to quantitatively assess the current state of informed choice across countries where female sterilization is a common modern method of contraception. Our analysis shows the emergence important trends, with some variation across countries. The most important sterilization-specific characteristic across all indicators was that of parity at the time of

• sterilization. This is particularly true for India and Colombia. Women at the extremes of the age

distribution at the time of sterilization - younger than age 25 (India) and older than age 35 (Honduras, Nicaragua, Peru) were less likely to score high on the MII. The private sector performs better than public sector in Colombia, Nicaragua and Honduras in counselling women across the various indicators of informed choice. Age at sterilization: The relationship between age at sterilization and low levels of informed choice is worrisome, and points to lack of good quality of care for women perceived as important targets of family planning/sterilization programs. Women who are sterilized at very young ages are shown to have regret at older ages (Singh et al., 2012), thus appropriate counselling and introduction to other methods

• f contraception is critical. A comprehensive review of literature showed that the younger a woman is

when undergoing sterilization, the more likely she is to regret the decision at a later age, and potentially seek to reverse the procedure (Curtis, Mohllajee, & Peterson, 2006). This assumes agency in the decision to get sterilized, but may speak to the importance of a formal age of consent for sterilization across

• countries. This study identifies one other group at risk: women over the age of 35 in Honduras,

Nicaragua, and Peru. This is a result that previous studies focused on sterilization regret understandably did not address, and highlights women who may have previously been overlooked due to assumptions surrounding women at older ages. Parity at sterilization: National family planning programs set different targets and goals for countries depending on level of fertility. These targets have been shown to be achieved by emphasizing permanent methods of contraception, like female sterilization in India, rather than temporary methods for spacing (James, 2011). The consequence is that women may reach a certain desired threshold for number of children, and then choose to undergo sterilization. This is a viable and safe option, except that most of these women may not know of, or receive proper counselling on, the consequences of

• sterilization. Some research has found that family planning programs actually target women at high

parities for sterilization (Donaldson, 2002); thus, questions of informed choice are of profound

• importance. If women at all parities were targeted for sterilization, one would not see the type of

gradient associated with informed choice as is evidenced in India, Colombia, and Peru.

Page 14 of 25 Source of sterilization: Given that national family planning programs are meant to operate within the public healthcare system, most sterilizations in our sample occur in this sector. Sterilization in hospitals affiliated with the government are sometimes free, and may demonstrate a nexus between the healthcare providers and local politicians who see this as a tool for re-election (Caetano & Potter, 2004). It is no surprise then, to deal with large volumes of women passing through for sterilization, quality of counselling may be lacking in many dimensions. We find that this is especially the case in Latin American countries, and warrants further investigation. National policies that were in effect for a large part of the 20th century may favour method skew towards of sterilization, with evidence of this in all countries that appear in our analysis (Boesten, 2007; EngenderHealth, 2003; IWHR, 2013; Sullivan, Bertrand, Rice, & Shelton, 2006). Subsequently, provider bias is something that can come into play in the public and private sectors, and may be an explanation as to why different aspects of consent are not talked about in either setting- in an effort to push sterilization over other methods. A study that examined informed choice found that choice trumped provider bias in cases where an injectable (Norplant) was the preferred method of the provider (Baveja et al., 2000), thus female sterilization may be another potential example of such, although that qualitative aspect of provider bias is out of the purview of this analysis. LIMITATIONS There are several limitations to this study. First and perhaps most serious is recall bias among respondents, who at older ages may not remember the conversations they had with health workers prior to undergoing a sterilization procedure. While the surgical nature of the procedure makes it highly unlikely that a woman would make an error in recalling that she was sterilized, her memory of the counselling process is more problematic. To account for recall bias, we restricted our analysis to women who report being sterilized 5 or fewer years preceding the survey. Nevertheless, how the questions are understood by respondents, as well as the counselling process itself, cannot be fully analysed based on these data and render us unable to make more specific policy recommendations. Secondly, with these data we are unable to thoroughly examine the decision-making process that each woman goes through prior to a sterilization procedure. Women may feel constrained in their decision due to socio-cultural and economic factors that render them incapable of caring for their ideal number of children. Moreover, we cannot evaluate the impact of conversations they have had with other sterilized women, which inevitably play into their decision (Bongaarts & Watkins, 1996). Finally, historically, coercive sterilization programs targeted indigenous and ethnic minority women. However, a variable for ethnicity was not

Page 15 of 25 available in the majority of the surveys for this study, so we are unable to assess whether these populations are less likely to be fully informed about sterilization. In nearly every country in this study, indigenous or lower caste women are more likely to be poor and live in rural areas. We therefore rely on proxy variables such as the wealth index and rural residence to identify these populations that are particularly vulnerable to coercive sterilization. We conducted analyses with interaction terms, but they were not significant. Ultimately, we are unable to quantify the extent of informed choice for other vulnerable populations such as HIV-positive, disabled, and transgendered individuals. As data from developing countries become more comprehensive on these topics, we recommend that future research closely consider the sterilization experiences of these populations. CONCLUSIONS AND POLICY IMPLICATIONS Our findings point to an urgent need to improve quality of counselling within health systems that provide sterilizations for women in South Asia and Latin America. Our study is the first to comprehensively review the state of informed consent regarding sterilization around the world using the best available comparable global data. While we are unable to assess coercion due to lack of data on financial incentive for the method, we were able to conclusively determine that there is not enough information provided to women in order for them to make an informed choice about this procedure. This addresses an essential service gap within health systems and among sterilization providers. Sufficient time and willingness to counsel a woman and ensure she understands the implications of the decision to undergo sterilization will go a long way in ensuring her agency and reproductive rights. Acknowledgements The authors would like to thank participants at the International Family Planning Conference in Bali, Indonesia, (2016) for their invaluable comments and suggestions. Conflicts of interest None Figure Caption: Figure 1: Share of female sterilization and other modern methods for modern contraceptive users by country (2000-2012)

Page 16 of 25 REFERENCES Amaral, E. F., & Potter, J. E. (2015). Determinants of Female Sterilization in Brazil, 2001–2007. children, 1(1,681), 46,556. Barry, E., & Raj, S. (2014, November 11). 12 Women Die After Botched Government Sterilizations in

• India. The New York Times. Retrieved from

http://www.nytimes.com/2014/11/12/world/asia/botched-government-sterilizations-india.html Basu, A. M. (1984). Ignorance of Family Planning Methods in India: An Important Constraint on Use. Studies in Family Planning, 15(3), 136-142. doi:10.2307/1965464 Baveja, R., Buckshee, K., Das, K., Das, S. K., Hazra, M. N., Gopalan, S., . . . Saxena, B. N. (2000). Evaluating contraceptive choice through the method-mix approach: An Indian Council of Medical Research (ICMR) Task Force study. Contraception, 61(2), 113-119. doi:http://dx.doi.org/10.1016/S0010- 7824(00)00089-5 Bertrand, J. (2011). USAID Graduation from Family Planning Assistance: Implications for Latin America. Retrieved from https://www.populationinstitute.org/external/files/reports/FINAL_LAC_Report.pdf Boesten, J. (2007). Free Choice or Poverty Alleviation? Population Politics in Peru under Alberto Fujimori. European Review of Latin American and Caribbean Studies, 82, 3. Bongaarts, J., & Watkins, S. C. (1996). Social Interactions and Contemporary Fertility Transitions. Population and Development Review, 22(4), 639-682. doi:10.2307/2137804 Caetano, A. J., & Potter, J. E. (2004). Politics and Female Sterilization in Northeast Brazil. Population and Development Review, 30(1), 79-108. doi:10.1111/j.1728-4457.2004.00004.x Coa, R., & Ochoa, L. H. (2009). Bolivia Encuesta Nacional de Demografía y Salud 2008. Retrieved from Calverton, Maryland, USA: http://dhsprogram.com/pubs/pdf/FR228/FR228.pdf Credé, S., Hoke, T., Constant, D., Green, M. S., Moodley, J., & Harries, J. (2012). Factors impacting knowledge and use of long acting and permanent contraceptive methods by postpartum HIV positive and negative women in Cape Town, South Africa: a cross-sectional study. BMC Public Health, 12(1), 197. doi:10.1186/1471-2458-12-197 Curtis, K. M., Mohllajee, A. P., & Peterson, H. B. (2006). Regret following female sterilization at a young age: a systematic review. Contraception, 73(2), 205-210. da Costa Leite, I., Gupta, N., & do Nascimento Rodrigues, R. (2004). Female sterilization in Latin America: cross-national perspectives. J Biosoc Sci, 36(6), 683-698. Donaldson, P. J. (2002). The elimination of contraceptive acceptor targets and the evolution of population policy in India. Population Studies, 56(1), 97-110. doi:10.1080/00324720213793

• EngenderHealth. (2003). Contraceptive Use and Discontinuation Patterns in Nepal: Norplant, IUCD, Pill,

and Injectables Retrieved from https://www.engenderhealth.org/files/pubs/family- planning/nepal_report_on_contraceptive_use.pdf Essack, Z., & Strode, A. (2012). “I feel like half a woman all the time”: The impacts of coerced and forced sterilisations on HIV-positive women in South Africa. Agenda, 26(2), 24-34. doi:10.1080/10130950.2012.708583 Geidel, M. (2010). "Sowing death in our women's wombs": modernization and indigenous nationalism in the 1960s Peace Corps and Jorge Sanjines' "Yawar Mallku". Am Q, 62(3), 763-786. Gribble, J. N., Sharma, S., & Menotti, E. P. (2007). Family planning policies and their impacts on the poor: Peru's experience. International family planning perspectives, 33(4), 176-181. Hardee, K., Harris, S., Rodriguez, M., Kumar, J., Bakamjian, L., Newman, K., & Brown, W. (2014). Achieving the goal of the London Summit on Family Planning by adhering to voluntary, rights- based family planning: what can we learn from past experiences with coercion? Int Perspect Sex Reprod Health, 40(4), 206-214. doi:10.1363/4020614

Page 17 of 25 Hosseni, H., Torabi, F., & Bagi, B. (2014). Demand for Long-Acting and Permanent Contraceptive Methods among Kurdish Women in Mahabad, Iran. Journal of Biosocial Science, 46(06), 772-785. doi:doi:10.1017/S0021932013000710

• IWHR. (2013). From Forced Sterilization to Forced Psychiatry: Report on the violations of the human

rights of women with disabilities and transgender persons in Colombia. Retrieved from Geneva, Switzerland: http://www.law.cuny.edu/academics/clinics/iwhr/publications/IWHR-Report- Colombia-2013.pdf James, K. S. (2011). India’s Demographic Change: Opportunities and Challenges. Science, 333(6042), 576-

• 580. doi:10.1126/science.1207969

Joshi, R., Khadilkar, S., & Patel, M. (2015). Global trends in use of long-acting reversible and permanent methods of contraception: Seeking a balance. International Journal of Gynecology & Obstetrics, 131, Supplement 1, S60-S63. doi:http://dx.doi.org/10.1016/j.ijgo.2015.04.024 Kendall, T., & Albert, C. (2015). Experiences of coercion to sterilize and forced sterilization among women living with HIV in Latin America. Journal of the International AIDS Society, 18(1), 19462. doi:10.7448/IAS.18.1.19462 Koenig, M. A., Foo, G. H. C., & Joshi, K. (2000). Quality of Care Within the Indian Family Welfare Programme: A Review of Recent Evidence. Studies in Family Planning, 31(1), 1-18. doi:10.1111/j.1728-4465.2000.00001.x Koenig, M. A., Rob, U., Khan, M. A., Chakraborty, J., & Fauveau, V. (1992). Contraceptive Use in Matlab, Bangladesh in 1990: Levels, Trends, and Explanations. Studies in Family Planning, 23(6), 352-364. doi:10.2307/1966893 Lee, J. K., Parisi, S. M., Akers, A. Y., Borrerro, S., & Schwarz, E. B. (2011). The Impact of Contraceptive Counseling in Primary Care on Contraceptive Use. Journal of General Internal Medicine, 26(7), 731-736. doi:10.1007/s11606-011-1647-3 Lizarzaburu, J. (2015, December 2, 2015). Forced sterilisation haunts Peruvian women decades on. BBC

• News. Retrieved from http://www.bbc.com/news/world-latin-america-34855804

Mallet, J., & Kalambi, V. (2008). Coerced and forced sterilization of HIV-positive women in Namibia. HIV AIDS Policy Law Rev, 13(2-3), 77-78. Martin, R., Buttari, J., Macias, H., & Cobb, L. (1993). AIDs family planning program in Honduras. Miranda, J. J., & Yamin, A. E. (2004). Reproductive health without rights in Peru. Lancet, 363(9402), 68-

• 69. doi:10.1016/s0140-6736(03)15175-6

Pieper Mooney, J. E. (2010). Re-visiting Histories of Modernization, Progress, and (Unequal) Citizenship Rights: Coerced Sterilization in Peru and in the United States. History Compass, 8(9), 1036-1054. doi:10.1111/j.1478-0542.2010.00717.x Pradhan, M. R., & Ram, U. (2009). Female sterilization and ethical issues: The Indian experience. Social Change, 39(3), 365-387. doi:10.1177/004908570903900303 Rizo, A., & Roper, L. (1986). The Role of Sterilization in Colombia's Family Planning Program: A National

• Debate. International Family Planning Perspectives, 44-48.

Saeed, G. A., Fakhar, S., Rahim, F., & Tabassum, S. (2008). Change in trend of contraceptive uptake — effect of educational leaflets and counseling. Contraception, 77(5), 377-381. doi:http://dx.doi.org/10.1016/j.contraception.2008.01.011 Salvo, J. J., Powers, M. G., & Cooney, R. S. (1992). Contraceptive Use and Sterilization Among Puerto Rican Women. Family Planning Perspectives, 24(5), 219-223. doi:10.2307/2135873 Santiso-Galvez, R., Ward, V., & Bertrand, J. (2015). Family Planning In Nicaragua: The achievements of 50

• years. Retrieved from Chapel Hill, NC:

http://www.cpc.unc.edu/measure/resources/publications/sr-15-118f/

Page 18 of 25 Singh, A., Ogollah, R., Ram, F., & Pallikadavath, S. (2012). Sterilization regret among married women in India: implications for the Indian national family planning program. International perspectives on sexual and reproductive health, 187-195. Stern, A. M. (2005). STERILIZED in the Name of Public Health. American Journal of Public Health, 95(7), 1128-1138. doi:10.2105/AJPH.2004.041608 Sullivan, T., Bertrand, J., Rice, J., & Shelton, J. (2006). Skewed Contraceptive Method Mix: Why it Happens, Why it Matters. Journal of Biosocial Science, 38(04), 501-521. doi:doi:10.1017/S0021932005026647 Teferra, A. S., & Wondifraw, A. A. (2015). Determinants of long acting contraceptive use among reproductive age women in Ethiopia: Evidence from EDHS. Science, 3(1), 143-149. Terborgh, A., Rosen, J. E., Gálvez, R. S., Terceros, W., Bertrand, J. T., & Bull, S. E. (1995). Family planning among indigenous populations in Latin America. International Family Planning Perspectives, 143- 166. Thapa, S., & Friedman, M. (1998). Female sterilization in Nepal: a comparison of two types of service

• delivery. International Family Planning Perspectives, 78-83.
• Track20. (2015). Data Sources and Methodology for FP2020 Core Indicators: 2014-15 Report. Track 20:

Monitoring Progress in Family Planning. Retrieved from http://www.track20.org/pages/data/data_sources_and_methodology United Nations. (2013). World Contraceptive Patterns. from Department of Economic and Social Affairs, Population Division http://www.un.org/en/development/desa/population/publications/pdf/family/worldContracep tivePatternsWallChart2013.pdf Visaria, L., Jejeebhoy, S., & Merrick, T. (1999). From Family Planning to Reproductive Health: Challenges Facing India. International Family Planning Perspectives, 25, S44-S49. doi:10.2307/2991871 WHO (Producer). (2014, October 20, 2015). Eliminating forced, coercive and otherwise involuntary

• sterilization. An interagency statement OHCHR, UN Women, UNAIDS, UNDP, UNFPA, UNICEF and
• WHO. Retrieved from

http://apps.who.int/iris/bitstream/10665/112848/1/9789241507325_eng.pdf

Page 19 of 25 Table 1: Proportion of sterilized women in the sample by demographic characteristics and country (2000-2012) India Nepal Bolivia Colombia Honduras Nicaragua Peru Education None 0.467 0.781 0.065 0.024 0.058 0.145 0.035 (0.008) (0.022) (0.012) (0.002) (0.005) (0.011) (0.004) Primary 0.162 0.123 0.462 0.323 0.620 0.435 0.271 (0.005) (0.013) (0.025) (0.007) (0.012) (0.019) (0.011) Secondary 0.331 0.090 0.313 0.490 0.255 0.316 0.396 (0.007) (0.014) (0.024) (0.007) (0.011) (0.019) (0.013) Higher 0.040 0.007 0.160 0.163 0.068 0.105 0.298 (0.003) (0.003) (0.019) (0.006) (0.007) (0.013) (0.013) t-test compared to other modern method users 43.5* 14.6* 5.2* 26.2* 4.9* 0.7 2.9* t-test compared to contraceptive non-users 29.6* 18.2* 2.7* 20.0* 10.1* 1.9 5.0* Wealth quintile Lowest 0.195 0.150 0.098 0.173 0.126 0.108 0.054 (0.007) (0.018) (0.013) (0.006) (0.007) (0.011) (0.005) Second 0.217 0.230 0.150 0.226 0.172 0.200 0.152 (0.006) (0.016) (0.017) (0.006) (0.009) (0.015) (0.009) Middle 0.222 0.251 0.205 0.224 0.224 0.216 0.233

Page 20 of 25 (0.006) (0.023) (0.022) (0.006) (0.011) (0.017) (0.014) Fourth 0.212 0.211 0.231 0.201 0.238 0.231 0.249 (0.006) (0.023) (0.022) (0.006) (0.011) (0.019) (0.014) Highest 0.153 0.157 0.317 0.176 0.240 0.245 0.312 (0.006) (0.023) (0.026) (0.006) (0.013) (0.021) (0.017) t-test compared to other modern method users 48.3* 1.8 0.9 14.4*

• 4.2*
• 4.2*
• 15.4*

t-test compared to contraceptive non-users 13.2*

• 5.9*
• 3.4*

6.8*

• 4.6*
• 2.8*
• 13.2*

Rural 0.717 0.884 0.249 0.243 0.429 0.352 0.205 (0.009) (0.022) (0.022) (0.007) (0.015) (0.018) (0.010) t-test compared to other modern method users

• 27.6*
• 4.4*
• 0.9
• 5.9*

4.1* 2.8* 9.4* t-test compared to contraceptive non-users

• 6.2*
• 0.2*

2.2

• 3.6*

4.1* 2.6* 9.7* N 9,577 1,046 727 8,797 2,672 989 4,027

• Note. Robust std. errors in parentheses.

t-tests are significant at p<0.001.

Page 21 of 25 Table 2: Proportion of sterilized women in the sample by sterilization-related characteristics and country (2000-2012) India Nepal Bolivia Colombia Honduras Nicaragua Peru Age at sterilization <25 0.426 0.293 0.043 0.176 0.156 0.156 0.045 (0.007) (0.018) (0.008) (0.005) (0.009) (0.015) (0.005) 25-29 0.345 0.396 0.207 0.265 0.292 0.292 0.180 (0.006) (0.016) (0.020) (0.006) (0.010) (0.018) (0.010) 30-34 0.166 0.191 0.297 0.270 0.301 0.313 0.327 (0.005) (0.018) (0.022) (0.006) (0.011) (0.018) (0.012) 35+ 0.064 0.120 0.453 0.290 0.252 0.240 0.448 (0.003) (0.012) (0.024) (0.006) (0.010) (0.016) (0.013) Parity at sterilization 2 or fewer 0.368 0.205 0.122 0.420 0.198 0.171 0.197 (0.008) (0.029) (0.019) (0.007) (0.010) (0.015) (0.011) 3 0.279 0.299 0.236 0.309 0.317 0.280 0.348 (0.006) (0.018) (0.021) (0.006) (0.010) (0.018) (0.012) 4 0.157 0.209 0.168 0.142 0.201 0.195 0.193 (0.005) (0.016) (0.018) (0.005) (0.009) (0.014) (0.010) 5 or higher 0.196 0.287 0.475 0.129 0.285 0.354 0.263

Page 22 of 25 (0.006) (0.019) (0.028) (0.005) (0.010) (0.018) (0.011) Source of sterilization Public 0.804 0.954 0.706 0.760 0.642 0.643 0.860 (0.007) (0.009) (0.024) (0.006) (0.012) (0.018) (0.010) Private 0.195 0.028 0.294 0.240 0.345 0.333 0.129 (0.007) (0.006) (0.024) (0.006) (0.012) (0.018) (0.009) Other 0.001 0.018 0.000 0.000 0.013 0.024 0.011 (0.000) (0.005) (0.000) (0.000) (0.003) (0.006) (0.003) N 9,577 1,046 727 8,797 2,672 989 4,027

• Note. Robust std. errors in parentheses.

Page 23 of 25 Table 3: Proportion of sterilized women by informed choice indicators and method information index by country (2000-2012) India Nepal Bolivia Colombia Honduras Nicaragua Peru 3 indicators that constitute the Method Information Index for informed choice Informed that sterilization is permanent 0.660 0.781 0.935 0.885 0.874 0.896 0.955 (0.008) (0.019) (0.013) (0.004) (0.008) (0.012) (0.004) Informed about side effects related to sterilization 0.313 0.480 0.626 0.434 0.452 0.481 0.605 (0.008) (0.019) (0.025) (0.007) (0.012) (0.019) (0.012) Informed about other available methods

• f FP

0.238 0.317 0.642 0.487 0.463 0.468 0.602 (0.007) (0.016) (0.024) (0.007) (0.012) (0.018) (0.012) Method Information Index 0 (did not receive any information) 0.264 0.144 0.037 0.069 0.082 0.061 0.023 (0.007) (0.013) (0.011) (0.003) (0.007) (0.010) (0.003) 1 (received information on 1 indicator) 0.403 0.344 0.174 0.329 0.341 0.329 0.216 (0.008) (0.016) (0.020) (0.006) (0.011) (0.019) (0.010) 2 (received information on 2 indicators) 0.191 0.303 0.339 0.329 0.285 0.313 0.337 (0.006) (0.014) (0.022) (0.007) (0.010) (0.019) (0.012) 3 (received information on 3 indicators) 0.142 0.209 0.450 0.273 0.292 0.296 0.424 (0.006) (0.018) (0.025) (0.006) (0.011) (0.016) (0.012) Mean (SD) 1.3 (1.0) 1.6 (1.0) 2.2 (0.8) 1.7 (0.9) 1.8 (0.9) 1.9 (0.9) 2.2 (0.8) N 9,577 1,046 727 8,797 2,672 989 4,027

• Note. Robust std. errors in parentheses.

Page 24 of 25 Table 4: Ordinal logistic regression results for Method Information Index by country (2000-2012) Sterilization-related variables India Nepal Bolivia Colombia Honduras Nicaragua Peru Sterilization-related variables Age at sterilization (ref: 25-29) <25 0.85** 1.21 0.45 1.15 0.90 1.12 0.60 (0.05) (0.34) (0.21) (0.09) (0.13) (0.24) (0.22) 30-34 1.02 1.20 0.82 0.95 0.80* 0.70 0.79 (0.07) (0.22) (0.24) (0.06) (0.09) (0.13) (0.14) 35+ 0.88 1.27 0.60 0.92 0.74** 0.62* 0.59** (0.10) (0.28) (0.16) (0.06) (0.09) (0.13) (0.10) Parity at sterilization (ref: 2 children) 3 children 0.88* 1.09 1.08 0.71*** 0.85 0.63* 0.69* (0.05) (0.19) (0.36) (0.04) (0.11) (0.14) (0.12) 4 children 0.76*** 0.93 1.09 0.77*** 0.94 0.80 0.67* (0.06) (0.25) (0.38) (0.06) (0.13) (0.21) (0.14) 5+ children 0.79** 0.87 1.01 0.73*** 1.02 1.07 0.57** (0.06) (0.22) (0.34) (0.07) (0.15) (0.34) (0.11) Source of sterilization (ref: Public Sector) Private Sector 0.99 1.45 0.78 2.56*** 1.71*** 1.96*** 0.78 (0.06) (0.53) (0.18) (0.16) (0.16) (0.31) (0.14) Other 1.05 0.99 2.11* 1.20 0.47 (1.57) (0.61) (0.68) (0.52) (0.39) Demographic variables Education (ref: none) Primary 1.27*** 1.49 1.63 1.54* 0.98 0.87 1.27 (0.09) (0.34) (0.54) (0.28) (0.19) (0.17) (0.51) Secondary 1.28*** 1.33 1.09 1.66** 1.37 1.37 1.23 (0.08) (0.32) (0.44) (0.31) (0.32) (0.42) (0.51) Higher 1.62*** 0.99 3.08* 1.72** 1.92* 0.88 1.74 (0.23) (1.46) (1.47) (0.34) (0.57) (0.29) (0.76) Wealth quintile (ref: Lowest) Second 0.97 1.17 0.56 1.27** 1.36* 0.94 1.08 (0.08) (0.34) (0.20) (0.11) (0.17) (0.18) (0.23) Middle 1.01 1.51 0.63 1.56*** 1.13 0.74 0.89 (0.09) (0.39) (0.23) (0.16) (0.16) (0.17) (0.22) Fourth 1.00 1.74 0.60 1.50*** 1.05 0.51* 1.09 (0.10) (0.52) (0.25) (0.17) (0.17) (0.16) (0.29) Highest 1.21 1.13 0.75 1.78*** 0.98 0.60 1.16 (0.14) (0.36) (0.33) (0.21) (0.19) (0.20) (0.33) Rural residence (ref: 0.83* 1.11 0.96 1.26** 0.92 0.84 0.90

Page 25 of 25

• Note. Robust std. errors in parentheses.

*** p<0.001, ** p<0.01, * p<0.05 Urban) (0.07) (0.21) (0.27) (0.10) (0.11) (0.17) (0.15) Year of interview 1.02 1.02 1.01 0.92*** 1.00 (0.03) (0.05) (0.01) (0.01) (0.03) N 9,577 1,046 727 8,797 2,672 989 4,027