dissertating, diversions, and data.

On April 10, 2014 by ali heller


I’ve been spending some time (read: all of my time) trying to synthesize my findings into something cohesive, comprehensible, and cogent. Hundreds of pages (months, and, admittedly, tears) later, I have a half-written but already intractable behemoth that only a handful of people will ever read (aka: my dissertation) which feels to me rather disjointed and indecipherable. So, in an attempt to pull myself out of my self-made morass of theoretical ramblings, I decided to post a “lite” version of some of my findings – a half digested flavor-without-the-calories facsimile.

For those of you who haven’t been following my blog (I’m sure there is a friend-of-the-family somewhere out there who is giving this post an obligatory once-over), the following observations are based from one year of in-depth ethnographic research at four treatment centers for obstetric fistula in Niger (The Danja Fistula Center located about 750 km east of the capital, and three fistula centers in Niamey, including: Dimol, Centre National de Reference des Fistules Obstétricales – CNRFO, and Hôpital National de Lamordé).

I interviewed 100 women at least once (but many between two and five times). I also conducted in-depth interviews with 12 women previously healed of fistula; 6 family members and 2 husbands of women with fistula; 19 practitioners, specialists, individuals working with fistula-related non-profits, social workers, midwives, and village surgeons. Additionally, I conducted 7 post-surgical follow-up interviews in women’s homes. And, as if that weren’t already far too many hours of scratchy (and sometimes frustratingly unintelligible) voice-recordings to transcribe, I conducted 8 focus groups with women with fistula, covering topics such as concealment, management, treatment seeking, the clinical experience, sexuality, and gender expectations.

Some of this stuff I’ve said before. Some of this stuff I’ve been adamant about. But this time, there’s a little bit of data to back me up. Yep, even graphs and tables. And, for the pièce de résistance, there are even some statistics. That’s right, a cultural anthropologist did a little math. With numbers. So, without further self-congratulation, here are a few thoughts (but not all of the thoughts):

Over the past decade, fistula has been effectively branded as a highly stigmatizing illness affecting young girls and causing social abandonment. CNN called it “a fate worse than death”, while the New York Times labels its victims as “lepers of the 21st century”. In popular accounts of fistula, African (and frequently Muslim) men, as well as the assumed pathological culture from whence they came, are pinned as the clear victimizers – perpetrators of child marriage and barbaric demands of their wives, daughters, and sisters. In this archetypical narrative, the victim of fistula is young, powerless, pitiable, and in need of intervention from Western donors. Fistula surgery is then portrayed as the “quick and easy” fix, at once inexpensive and highly successful.

Many popular and academic treatments of fistula follow a well-rehearsed formula, an array of conspiring elements: fistula sufferers are young girls forced into “child” marriages; due to their youth, potential malnutrition, and small stature, these girls suffer through a complicated labor for up to a week without any medical intervention; following the onset of their postpartum incontinence, they are abandoned by their kin and exiled from their communities; they find redemption in a life-changing surgery that restores their continence, enabling their return to the fold of society.

When I began the project, I expected to find heavily stigmatized and largely abandoned young women who were finding corporeal and social redemption through surgical interventions. However, as I became more immersed in women’s stories, the fistula framework began to crumble, revealing a much less cohesive narrative – a patchwork of sufferers often less pitiable and interventions frequently less effective. I found that the fistula narrative fails to accurately reflect the vast diversity of women, both demographically, and in their experiences living with fistula and seeking care. My research began to deconstruct and de-center the three main tropes that constitute the fistula narrative: the despoiled child bride, the social pariah, and the quick and easy cure. I argue that these findings have important programmatic implications for fistula interventions.

The Archetypical Sufferer

The young girl who contracts fistula during her first pregnancy following a “forced” marriage in her pubescent (or pre-pubescent) years is made highly visible in the fistula narrative. She is a passive receiver of injurious decisions made by her (male) kin or husband regarding her life and her body. While young girls who suffer dramatic consequences from fistula do exist, they are not typical – they are not even the majority. Indeed, there is no “typical” sufferer of fistula. The average age at onset of fistula is mid to early 20s, the majority of women with fistula are not divorced, many women contract fistula after several pregnancies, and many have living children.

Of the 100 women I interviewed, I found that women contracted fistula at an average age of 23.5, but spanning from 13 years old to 52 years old. Although women with fistula are thought to be married unrepresentatively young, I found that the average age of marriage was 15.5, spanning from 10 to 24 years while the average age at first birth was 17.6, spanning from 13 to 27. Considering that the average age of marriage in Niger is 15.5, and the average age at first birth is 18.2, women with fistula are not demographically unrepresentative.




While the narrative situates fistula as a disease of young, first-time mothers, who are left childless and infertile due to the injury, the data do not. While just over half (51%) of women contracted fistula during their first pregnancies, the remainder of women who contracted fistula due to childbirth did so during their 2nd to 12th pregnancy. Additionally, although 51% of women contracted fistula during their first pregnancy, only 34% of women had experienced only one pregnancy, 37% had 2 to 5 pregnancies, while 25% had 6 to 10 pregnancies, meaning that many women continued to have pregnancies after the onset of their fistulas (contradicting popular assumptions which thoroughly de-sex women with fistula). While 55% of women had no living children, the remainder had between 1 and 6 living children.




Additionally, “forced”, “child”, and “early” marriage are often conflated. Not all “early” marriages are forced, nor are all “forced” marriages early.[1] Often young girls initiate their own marriage plans. While 55% of the sample was married at 15 years of age or younger (constituting what many may label “early marriage”), only 26% considered their marriages “forced”. Of these 26 women, 18 (69%) were married at the age of 15 or younger, while only 8 (31%) were married at the age of 16 or older.[2] Despite the apparent over-representation of young women within the group considering their marriages forced, the difference in the age of marriage between the group of women who considered their marriages forced and those women who did not did not reach statistical significance (p=0.096).[3] Indeed, within the larger sample, there is no statistically significant dependence between the age of first marriage and if the woman considered her marriage “forced” (p=0.256).[4]

Also, early marriage does not necessarily equate with early pregnancy. In my sample, there is a statistically significant relationship between the age of marriage and the length of time between marriage and first pregnancy. Women in my sample who married under the age of 16 were more likely to wait before getting pregnant than women who married at age 16 or older (p=0.0005).[5] The average amount of time between marriage and first pregnancy among the women with “early marriages” (or women who married under the age of 16) was 2.7 years while the average amount of time between marriage and first pregnancy among women who married after the age of 16 was 1.5 years.

Archetypical Suffering

Fistula is thought to render women as social non-entities: broken, valueless, and ultimately cast-aside. However, women’s experiences revealed complex networks of care and support in contrast to narratives of rejection and stigma. Women with fistula are not invariably defined by their disease or solely constituted by their pain. I found that most remain entangled in complex networks of obligations and attachments that defy their supposed relegation to the margins of society. Many attempt to manage relatively normal lives while living with fistula, continuing to have pregnancies and children and often concealing their fistulas from their communities. Although women with fistula are assumed to be wholly abandoned, and almost automatically divorced, the largest portion -38% of women in my sample- remained married, while only 23% of women were divorced and 36% were in a liminal state of marital separation. Nearly all of the women who considered themselves married continued to live with loving and supportive partners, and some even continued to engage in satisfying sexual lives.


I administered a standardized survey of 18 questions to all 100 women in my sample (typically before concluding their in-depth interview) in order to measure both external (etic) and internal (emic) stigma. The first set of questions focused on external stigma, investigating quotidian treatment by others (past mistreatment and avoidance behaviors such as the refusal to share a meal or verbal insult). The second set of questions focused on the internal experience of stigma, inquiring about daily internal struggles (measured through proxies such as shame, embarrassment, fear of judgment, and fear of mistreatment). Women were asked after each question about its frequency. For example, for each question, I asked: “Because of your fistula, how often did the following events happen?” Response options were “never” (scored as zero points), “once or twice” (1 point), “several times” (2 points), and “most of the time” (3 points).

The highest reports of stigma were for negative self-perception (1.78, average per question response out of 3) and verbal abuse (0.74). Low levels of stigma were reported for: social isolation (0.37) and fear of contagion (0.28). The seven questions measuring “negative self-perception” were categorized as measurements of internal (emic) stigma, while the eleven remaining questions measuring “verbal abuse”, “social isolation”, and “fear of contagion” were combined to measure external (etic) stigma. Scores for both internal and external were then normalized. While this is a crude instrument, offering far less insight than in-depth interviews, the ability to compare between women is quite elucidating.


Results from the standardized stigma survey, adapted from the HASI-P (Holzemer et al. 2007), indicating the stigma category and average score (between 0-3) for all 18 questions (organized in ascending order from lowest score to highest). N=100 women with fistula.


While 66% of women reported high or very high rates of internal stigma, and 15% reported no or low internal stigma, only 7% of women reported high or very high rates of external stigma, while 76% reported no or low external stigma. Although popular discussions of the social ramifications of stigma focus heavily on external mistreatment and tangible social consequences, women’s lived experiences minimize the prevalence of outward mistreatment and avoidance behaviors while underscoring the personal emotional and psychological burden of the illness.

The “Quick and Easy” Fix

While fistula surgery is framed as the “silver-bullet”, I found that surgical intervention was often unsuccessful and the social ramifications of surgical treatment were sometimes quite destructive. Even women with “simple fistula” in Niamey centers waited long periods of time at centers before receiving surgeries, or in some cases, even being seen by a clinician.

Of my sample, women with fistula in Niamey stayed an average of 5 months at clinics, and as long as 28 months without having received an operation. For many women, long absence results in weakened social networks and marital friction as women become increasingly peripheral to social life back home.

While fistula surgeries are reported to be successful between 75-95% of the time, only 36% of the women who actually received surgery in my sample left the clinic dry, while the remaining 64% remained incontinent.[6] (29% of the sample did not receive surgery during the research period either because they were not looking for surgeries or because of long wait times and infrequent surgeries). At least five women in my sample left their respective hospitals dry, only to experience repair breakdown within a few weeks of their return home.[7]

Of the 22 women who attained continence, 16 women (73%) were “new” cases, having previously undergone zero or one surgery before the research period, while 6 women (27%) underwent two or more surgeries prior to the research period. Women who attained continence were predominantly “new” cases. Of the 39 women who did not attain continence, 18 women (46%) were “new” cases, while 21 women (54%) were “old” cases.[8] The success rate for first time operations is widely understood to be higher than following surgeries.

Of the 37 healed women in my sample (including those who were not healed during the course of the research period), 51% were healed after the first surgery and the number of successful surgeries after the first continued to decline. Although women had as many as 11 failed surgeries, not a single woman was healed after her 7th surgery. Many women undergo one failed surgery after the next (sometimes as many as a dozen), spending years at the center. Indeed, rather than representing a young cohort of women, recently crippled by the injury, 44% of my sample had undergone three or more previous failed surgeries at the time of the interview, representing a cohort of chronic sufferers.


Of the 30 women who were seeking but did not receive surgical interventions, many had very large, complex, and potentially “incurable” fistula requiring interventions which frequently surpassed the expertise and abilities of available surgeons.

Of these 39 women not included, 31 had a surgical history (6 had never received surgery and 2 were cured by a catheter rather than surgical intervention). Of these 31 women, 10 (32%) were continent (dry) while 21 (68%) were not (wet).[9] When these 31 women who received surgery prior to (but not during) the research period are included in the total surgical success rate (along with the 61 who received surgery during the research period), the surgical success rate does not change significantly: 32 out of 92 women (35%) attained continence, while 60 women (65%) did not.

Slide1 copy

Culture as Culprit:

While what prevents a woman experiencing obstetric complications from contracting fistula is her access to quality emergency obstetric care, fistula campaigns often situates it largely as the consequence of culture, oppressive patriarchy, and neglectful families.

At various points throughout the fistula narrative, “culture” (or sometimes religion) is indicted. Often, culture is said to first intervene when the woman is a mere child, encouraging her premature sale into conjugal life; then during her labor, when often it is patriarchal cruelty or cultural norms encouraging birthing at home or alone that keep the woman from appropriate medical care; and finally after the contraction of her fistula, when cultural understandings of corporeal purity and reproductive success result in her degeneration into a social non-entity. These cultural failings are then juxtaposed with the potency and efficacy of fistula surgery, seen to mend the physical and social self all in one technical performance.

However, most women birthed in centers where they received poor care. Although the assumption is that women with fistula have waited a dangerously long time at home, this is not always the case. While some women I interviewed did spend days in labor at home without reaching out for biomedical care, many others reported heading to their local health clinic as soon as their labor began. They tried to deliver at health centers, but experienced poor medical treatment, including referral delays, refusal of services, physical and verbal abuse, inappropriate interventions, and botched Caesarean sections and episiotomies

In fact, 30% of women in my sample sought care within hours of their labor beginning while 88% eventually sought biomedical intervention. The 97 women in the sample who contracted fistula due to childbirth labored for an average of 3.0 days (±1.8). The average length of labor for women who birthed almost exclusively at health centers (defined as women who either went to clinics immediately after their labors began or stayed at home only “hours” before seeking care) was 2.1 days (±1.7), while the length of labor for women who birthed exclusively at home (never seeking care at health centers during the labor) was 2.6 days (±1.9). The length of labors for women who birthed almost exclusively in biomedical health centers is not significantly different from the length of labors of women who birthed entirely at home (p=0.45).[10] Although there may be several confounding variables, this does suggest that the quality of care women received at health centers was poor, demonstrated not only by poor outcomes but by prolonged delays to access care.

Interestingly, both of these groups of women had significantly shorter labors when compared to the remaining women in the sample (women who neither birthed entirely at home nor entirely at clinics) who labored for an average of 3.4 days (±1.8) (p=0.008).[11] This suggests that the women who labored entirely at home had slightly shorter labors than the average woman (2.6 days compared to 3.0 days for the full sample or 3.4 days for the women who birthed neither entirely at home nor entirely at health centers) and may have eventually sought care had their labors continued to be obstructed. Additionally, these findings suggest that women who first attempted to birth at home before seeking biomedical care suffered from all three phases of delay (see Thaddeus and Maine 1994 who discuss three barriers to maternal health care, first – delays in the decision to seek care, second – transportation delays to a center, and third – delays to access quality care once at a health center) – first experiencing prolonged labor at home before reaching the clinic where they were also made to wait.


The length of Labor Responsible for Fistula for the 12 women who did not seek biomedical care. The average labor of 2.6 days (±1.9)


The length of Labor Responsible for Fistula for the 29 women who immediately sought biomedical care. The average labor lasted 2.1 days (±1.7).


The length of labor responsible for fistula of the full sample (N=97). The average labor was 3.0 days (±1.8).

Patently, access to care is not the same as access to quality care. Fistula can be caused by a poorly trained clinician keeping a woman at a center for far too long, refusing to refer her to a higher level of care, referring her horizontally rather than vertically, or performing forceful and inappropriate interventions in the face of an obstetric complication. Much like my findings, a controversial study conducted by an epidemiologist working for Niger’s Network for the Eradication of Fistula under the ministry of health showed that contrary to popular belief, most women with fistula were not birthing at home, but at state-run health centers. This decenters the popular prevention approach whereby ‘changing the culture of birthing’ and ‘raising awareness’ about the importance of birthing at a health center is sufficient towards lowering incidence of fistula. Poor health outcomes at centers lead to prolonged 1st phase delays and increased medical suspicion. This is evidenced by large numbers of women who believed that their fistulas were caused by a biomedical healthcare provider. In my sample, 29 women (29%) expressed a belief that their fistulas were iatrogenic.

Generally, the state of health care in Niger is poor. Women with fistula reported much higher rates of child mortality than the national average, which is estimated at 124.5 children out of 1,000 – meaning that 12% (just over one out of ten) of children will die before reaching the age of five. Women with fistula have particularly high infant mortality rates, as the majority of reported deaths took place during (or just before or after) delivery (as child mortality rates are typically only calculated by live births, the high rate of perinatal deaths of babies among women with fistula might partially account for the large discrepancy). Still, the child mortality rates for women with fistula in my sample is disturbingly high – averaging 76% (meaning that just over three children out of every four pregnancies die). The rate is highest for women with only one pregnancy, at 91% (meaning that nine out of ten babies from first pregnancies die), and lowest for women who’ve had between 4 and 6 pregnancies (67%).



But why does it matter if the face of fistula is 15 or 51? Why does it matter is she is said to be divorced, abandoned, or rejected? What’s the difference if success rates are 39% or 93%? I suppose we anthropologists don’t spend a particularly long time ruminating on those questions. Mostly, we just reply – Well, because it’s truer. Still, there are some real policy implications to getting the story wrong (as well more insidious consequences on the way we westerners think about and approach development, humanitarian intervention, African men and women, and “African culture”).

One consequence of the fistula narrative is if women are assumed to be unequivocally rejected and wholly abandoned, then fistula centers are positioned as sites of rebirth and redemption for women (whose prospects back home are horrifying). Thus, the prolonged periods of time women are asked to wait at centers (due to long wait times for surgeries, poor clinical communication, and lengthy reinsertion courses, as well as a general disregard for women’s time) are not critically examined. Administrators and clinicians often justify these prolonged periods of time at centers by the assumption that women have nothing left at home and thus prefer to be at the clinic rather than returning to families that have rejected them. For the vast majority of women, this isn’t the case. In fact, it is often the prolonged absences from home (rather than the fistula itself) which cause social problems and marital rifts for women.

Additionally, the belief that fistula is highly curable leads to a near exclusive focus on surgical treatment, and thus a lack of focus on living with fistula as a chronic condition in fistula treatment efforts. Women determined “incurable” or “inoperable” are often rendered invisible in the penumbra of surgical efficacy’s glow.

These are just two examples among many of how getting the story wrong has very real and very harmful consequences for women with fistula. Benign misunderstandings, misrepresentations, and miscalculations harbor the malevolent.

[1] Although there is no real consensus over what constitutes “early” or “child” marriage, the age
of 16 is most often used as a cut-off.
[2] Specifically, 4 women were married at the age of 16, 2 women at 17, 1 woman at 19, and 1 at 20.
[3] Calculated with Chi-square test for significance, comparing the age of marriage (15 or younger
versus 16 or older) between women who considered their marriages forced and those who did not.
[4] These data were obtained through a Chi-Squared test for significance, comparing the age of
marriage (15 or younger versus 16 or older) to marriages perceived by women as “forced”.
[5] These data were obtained through a Chi-Squared test for significance, comparing the age of
marriage (15 or younger versus 16 or older) to the amount of time following marriage before
pregnancy (categorized as two years or less versus more than two years).
[6] Without a way to verify the state of a woman’s fistula itself, and believing that surgical
closure was not a particularly meaningful metric for women, I defined surgical success as a
binary: wet or dry. My definition of success followed the approach of Maulet et al. (2013)
whereby “Continence was considered as the sole factor for repair success and recorded as a
binary variable”. However, with the understanding that some degree of incontinence may be
normal following labors and aging, women were counted as “wet” only if their leakage was both
persistent and subjectively reported as problematic. Using this system of delimitation, five
women who reported very occasional urine leakage were still categorized as “dry”.
[7] Some of these women did return to their respective fistula clinics hoping to undergo an
additional fistula surgery during my research period, however, others did not, and I was only
able to follow up with them by phone.
[8] Due to the limitations of this study, and my inability to ascertain thorough clinical information
about each woman, I was not able to delimitate “simple” cases from “complex” cases.
[9] This sample may be biased as women with unsuccessful previous surgeries were probably more likely
to return to fistula centers than women who were previously healed (who had fewer reasons to
return, but did return for trainings, to collect money, for C-Sections, or other health
[10] This was calculated by comparing the length of time of deliveries of the women in the two groups
using a 2-Tailed T-test, p=0.45.
[11] This was calculated by comparing the length of time of deliveries of women who birthed either
entirely at home or at a clinic against the remaining women in the sample using a 2-Tailed T-
test, p=0.008.
*Special thanks to Anita Hannig, who through our collaborative effort, helped me to hone many of these ideas


3 Responses to “dissertating, diversions, and data.”

  • James

    Nice work, Ali! Especially the numbers.

    As you brought up, the shifting of the narrative with quantitative data really does lead to different policy prescriptions. Regardless of who reads your dissertation, I think this will be potentially quite important papers in the literature (he said, as someone who has not read it…) – it makes me wonder as well how universal the fistula experience is, and what differences would be found looking in other low-resource settings (ssAfrica and elsewhere).

    Keep it up! More words, fewer tears!

  • Nancy

    Thank you for this post and sharing your findings which validate many of my own observations.

  • Meri

    Great research. I think it adds nuance to the picture of fistula in Africa rather than completely overturns it. For example, yes, only 26% felt like their marriage was forced, but most of the others married young probably didn’t see other options and it is always better to choose your husband early than have someone forced on you. Also, though the age of marriage and first birth for women with fistula is not significantly different from their peers, it might help if the girls/women were older and more developed.

    As James mentioned, it does suggest some different policies might be better. Even in the U.S., successful fistula surgery is hard to come by (though making people wait at centers for months would never happen and as you point out, is also a problem in Africa.) Better nutrition, better health care will clearly make a difference–they take a longer time to implement.

    Thank you for your work. Do the world a favor and please submit your thesis and publish it if at all possible. Then you can refer to it when advocating for a more nuanced handling of fistula in Africa.

    Again, thank you!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>