Challenging the Conventional Thinking

This past week I successfully finished collecting data for my study's second and third sub-objectives, which involved work in communities outside the municipal district. I had originally had a smaller number of objectives for work in the communities but grant reviewers wanted to see more work in villages and among community members outside of health facilities. I had mixed feelings about revising my project to include these expanded community visits for a couple of reasons. Primarily because the barriers to care in communities are well known and well studied, and secondly, because it was really the facilities in these areas I wanted to look at, not necessarily the community dynamics. I also think this was one of the reasons another funding agency just was not having any of my grant. I submitted to them twice and didn't make it past the first round of reviews either time. They were looking for the conventional narrative, the line of thinking in anthropology that suggests we should be examining maternal death through the lens of the community and the subjectivity of those most at risk, the women themselves. I sought to break from this convention because this dialogue has sustained the Safe Motherhood movement since the 1980s and, unfortunately, in countries failing to make progress on significantly reducing maternal mortality, the problems that were challenging in the 1980s are still present. No point beating a dead goat (we don't have any horses around here) by continuing to repeatedly "discover" that transportation is a problem, education is a problem, bad roads are a problem, etc. And guess what I heard in all of these community meetings? You guessed it: still not enough education; transportation is still a challenge; people are still poor; the roads are still bad. 

When I started on this line of thinking-examining maternal deaths from inside the hospital, the social and institutional dynamics- I didn't think it was particularly radical, new maybe but that's what we all hope for our dissertations, right? By focusing on the health care providers and facilities, I was hoping to work most closely with those who also directly have the power to influence a woman's recovery or death. Sure, delays before she arrives at the hospital factor into it all, but, as I said above, these are well known. Where the ethnography has often stopped is at the doors of the hospital. If we do see inside a health facility in an ethnography, in so many cases, it is as the anthropologist accompanies a woman from the community and bears important witness to the ways in which (often minority) women are treated or discriminated against through overt or micro aggressions in the biomedical care setting. What we don't see as much-one chapter here or there- is the perspective of the health care providers themselves. How their personal lives, national or international policies, institutional rules and procedures, or social interactions with other providers and administrators deeply color the care they are able to provide. In my mind, this picture is crucial for filling in the blanks and moving forward. My work has also ended up shedding light on the ways in which the documents meant to represent a hospital (or dispensary or health center) and the daily practices often diverge. It is all well and good for a person in an office in Dar es Salaam to adamantly tell me what people are "supposed to do," but the real work, and the power of anthropology, is in discovering why it is that health care providers deviate from the protocols and guidelines. 

This focus on health care facilities has been largely well-received here in Tanzania, in theory, but I have encountered times when I am 100% convinced that people don't think the problem lies within their own house and therefore proceed to make judgments about what kind of data I should be collecting and what is relevant or not to my study of maternal death. Like the time when two top regional health administrators denied my request to attend a regional meeting on the basis that they thought it had nothing to do with my work and was just "internal business." Unbeknownst to them, I had already been privy to the major discussion of the meeting due to my relationships with other people. I was only interested in hearing how they were proposing to redistribute funds to help the regional hospital sustain its care for pregnant women in a more financially stable way. I wasn't necessarily interested in other dirty laundry, so to speak. Nevertheless, these administrators closed the proverbial gates squarely in my face with a suggestion that I head back to the ward, presumably where they thought I belonged. Again, I was struck by the idea that many people still need to broaden their conceptions of the causes of maternal death. These administrative issues undeniably affect maternal health outcomes because they affect supply availability and distribution, staff salaries and extra duty pay, district referral procedures, and communication; how could they not see the meeting was directly relevant to my research? It reminded me of my very first day on the maternity ward when one of the nurses basically told me I was barking up the wrong tree if I thought the nurses and doctors had anything to do with women's deaths or serious morbidity; malpractice was not in the question, she said. By the way, it is perfectly within the administrators' rights to deny me access to the meeting, I was mostly upset by what seemed like a disappointing lack of understanding of the goals of my research, even after having been at the hospital for a year.

So here I am, on the other side of the community portion of my study. I am glad I did it, it strengthens the overall study, but nothing earth-shattering was discovered. Each village had some specific issues but, overall, the threads were the same- not enough medications, need providers with more expertise, transportation is a major cost when trying to reach health services. What I may not have anticipated was that this portion of the study has provided further ways to connect health care service provision with broader political policies and reforms. From this portion of the study though, I am going to have to challenge everyone in the next maternal death audit meeting if they try to keep saying there was "delayed decision making in the family" in nearly every case, as in the past. Most of these community members try their best to reach health care services and the decision is often made early enough; it's other challenges down the line that often derail their attempts to arrive in a timely fashion. Relatives arrive at the dispensary with a woman in labor in the middle of the night. One relative stays with the woman while the other runs to the health care provider's house because there's no security guard or anyone else there at night. The nurse says she's tired and refuses to come out of her house, telling the family to go to the other provider's house. That one says the same thing and so the local "traditional" birth attendant helps the woman give birth on the porch of the dispensary. Where does this scenario fit in the three delays model? Is this a delay before reaching the hospital? At the hospital? Where does the responsibility lie then for carrying out some disciplinary action against the provider who refused to help the woman and her relatives? Will the family or community be empowered enough to even know that they can complain about this? If so, to whom do they complain, and will that person listen? There is also a blatant lack of connection and communication between facilities and communities, which, in my professional opinion, further exacerbates many of the challenges currently present and leads to bad will towards providers on the part of community members. 

In the village in which I was working this past week, the medical attendants were telling me that they no longer even bother to call the district ambulance because they know from experience that it will most likely not arrive. Instead, they choose to advise family members to find a motorcycle or private car to transport them in the hopes that it will be faster than waiting for the district ambulance to travel the short 29km from its supposed parking spot. This means families sell cows or tracts of land in order to pay a business man nearly $100 USD to take them and their sick relative to the nearest health center, which doesn't even do operations. Some choose to use motorcycles as their means of transport but this has led to the community members being familiar with the ways in which to transport dead bodies on the back of motorcycles. It's still prohibitively expensive for most people to even use a motorcycle and by the time the money is collected, the patient's condition can have significantly worsened to the point where they die on the way to the next level of care. It's tragic and macabre that they can tell me their strategies for returning these dead bodies back to the village and it's due not to the road quality, but to poor organization and communication at the district health level and due to a lack of some kind of price regulation for transportation for sick people. It's perfectly within the means of the village government's authority to make a rule saying it is illegal to increase the price of transportation for sick people. If you would use 20,000 TSH worth of fuel to go and come back, then you want a profit, so maybe you charge 40,000 TSH or 50,000 TSH. There is no reason to charge 100,000 TSH or 150,000 TSH for a one way trip.   

The next month will be my last here in Sumbawanga before I move to Dar for two months before the end of my project. The remaining time will be for finishing interviews with nurses, doctors, and administrators, as well as those providers who have been working in the region since the 1980s or before. I am also currently trying my darnedest to collect data for a social network analysis but that is going slowly and can most accurately be described as herding cats. We shall see how it turns out, I think it will be extremely interesting if I can collect high quality data for it. Wish me luck!