Obstetric fistulas in Uganda: scoping review using a determinant of health approach to provide a framework for health policy improvement

Obstetric fistulas are associated with severe inequity on multiple levels. Analysis using the determinant of health approach provides targets for intervention on those levels.

It has been called a “global social injustice” [1], the “result of the egregious failure of health systems” [2] and even a “human rights tragedy” [3]. While describing the burden of obstetric fistulas (OF) the use of superlatives is not feared. As a complication of obstructed labour, an OF is an abnormal connection between the vagina and the bladder or the rectum. This leaves the woman incontinent for urine, faeces, or even both. Besides the physical consequences of the constant leakage of bodily fluids, the associated foul smell and psychological stigmatization have enormous social and economic consequences, as well as devastating emotional and psychological consequences, that include shame, loss of self-esteem, depression, rejection, and loss of marital and sexual rights [4‐6].

The World Health Organization (WHO) estimates that worldwide between 50.000–100.000 women develop OF each year and 2 million women live with untreated fistula in Asia and sub-Sahara Africa (SSA), but the exact number remains difficult to measure [7, 8]. Within SSA, Uganda has the highest known prevalence of OF with an estimated lifetime prevalence of 19,2 per 1000 women of reproductive age [9]. The data of the last two Demographic and Health Surveys (DHS) show that 1 to 2% of Ugandan women have symptoms of OF, of whom only 62% have sought treatment [10, 11]. Those not seeking treatment give as reason being too embarrassed to seek treatment, while others did not know where to find treatment [10]. Furthermore, it is believed that even if all affected women would seek treatment, at the current rate that fistulas are surgically being managed, it will take at least 55 years to treat all existing patients, let alone to treat the new cases that develop every year [2].

According to WHO most cases of fistula can be avoided, firstly by delaying the age of first pregnancy and secondly by access to obstetric care [7]. Fistulas have been practically eliminated in many countries, yet in low- and middle-income countries “the most dispossessed, outcast, powerless group of women in the world” [12] are still at risk of this condition [13]. The uneven global and national distribution suggests a complex picture of determinants contributing to fistula development.

Altogether, the severity of the condition, its consequences, as well as the uneven distribution of OF explain the superlatives that are used to describe its global burden. This scoping review aims to give a critical overview of the determinants of health that contribute to the development of OF in Uganda, to create a better understanding of the process and to give a framework for improvement of health policy.

A scoping review of literature was performed in October 2018, using the PubMed/MEDLINE database as well as data from Ugandan DHS (UDHS), the Ministry of Health, the World Bank, and the Ministry of Finance. The key words for the PubMed search were synonyms for ‘obstetric fistula’ and ‘obstructed labour’ (Table 1) in combination with ‘Uganda’. No delimiters were used. The search results were screened by both authors independently. All studies reporting determinants of OF development in Uganda were included. Any disagreements were resolved through discussion between authors and reading of full text. Using the snowballing technique (within PubMed and Google Scholar) one more article was included, not exclusively about Uganda, but describing a pooled analysis of data from SSA, hence not showing up in the original search. The flow diagram of the search can be found in Fig. 1.

Any factors of influence for OF development were extracted from the selected publications. Publications describing determinants of obstructed labour in Uganda were also included, in assumption that these determinants indirectly play part in OF development. The results were analysed and organised using the different levels of the model for the main determinants of health by Dahlgren and Whitehead [14]. This model aims to unravel a wide range of factors that threaten, promote or protect health, and is consequently very suitable to use when formulating interventions within the setting of policy-making [14]. Dahlgren and Whitehead argue that individual lifestyles and attitudes, social and community networks, living and working conditions, as well as overall structural conditions shape a specific health divide resulting in health inequity. The model depicts four levels for policy intervention, and one level that is ‘fixed’ or uncontrollable, namely age, sex and genetics (Fig. 2) [14]. The epidemiological distribution of OF suggests considerable inequity in the presence of determinants of health at different levels of society, thus making this model highly relevant to approach the problem of OF.

Using the model, Ugandan statistics were further analysed to identify possible (indirect) determinants of OF. More literature was sought to explore these factors. The authors reached full consensus about the final presentation of the results.

The literature search yielded a variety of publications covering determinants of OF in Uganda. These determinants will be reviewed following the layers of the model, though most determinants overlap the different levels.

This scoping review of literature has rendered several starting points for improvement of health policy. Roughly, the process towards OF development can be divided in two: on the one hand development of obstructed labour, on the other hand the management thereof. Many determinants influence both aspects. For example, a low economic status might result in early marriage and childbearing, associated with obstructed labour, but can also increase delays in getting appropriate care due to lack of funds. Similarly, an increase in female education might result in better nutrition, but it also increases awareness of the necessity of skilled birth attendants during childbirth [58]. Consequently, aiming to improve the socioeconomic status of women in remote areas will theoretically impact both the prevention of obstructed labour, as well as the timely access to care in the case of obstructed labour.

The reviewed research on OF in Uganda covers many aspects of the model presented by Dahlgren and Whitehead, but not all proposed determinants are covered [14]. When further unravelling the determinants that are included in this review, the list extends far beyond what has been mentioned. For example, possible factors of influence could be the distribution of health facilities, the monitoring of labour and interpretation of findings, birth preparedness, fear of caesarean sections, or inadequate post-delivery care. One of the strengths of this review is the inclusion of a broad variety of studies to reflect the diverse set of determinants. However, performing a literature study to map existing determinants does imply the expectation that many determinants have been studied separately. Yet, as this review shows, there are many factors involved that are often intertwined with, dependant of, or reinforced by other factors. It is unlikely that all those complex factors can be studied independently, and it is thus unlikely that this review is fully comprehensive.

Furthermore, a limitation of this study is that for most included epidemiological publications the data of the UHDS of 2012 was analysed, not the more recent UHDS of 2016 published in 2018. Nevertheless, though the overall environment may have changed in Uganda, it is unlikely that the determinants of OF are different between 2012 and 2016. The relative weight or relevance of certain determinants might have shifted, but that is not the scope of this review.

Many reports of risk factors for OF are based on patient characteristics in facility-based studies, leaving limited options to compare the results of this study to risk factors in other countries [59]. There are only a few studies that report DHS data in relation to OF of which a pooled analysis of SSA countries was included in this review [9, 26, 60, 61]. Just like in Uganda, both in Ethiopia as well as in Pakistan, the geographical region and place of residence was a determinant for OF [60, 61]. This suggests that within different countries there is a similar conglomerate of risks factors for OF in specific regions. Also, low educational status, young age at first marriage, and low socio-economic status were significant determinants of OF, similar to the findings in Uganda [60, 61]. Some factors, like the effect of post-primary education or having problems obtaining permission to seek health care, were not found as significant in the pooled analysis of DHS data, but did show to be significant in different national reports [9, 26, 60, 61]. This shows that because of the many different cultural and context-specific factors that are involved in the process of OF development, pooling international data can result in missing the unique factors for a certain country. Similarly, some correlations can seem very strong in a pooled analysis, but might actually be less relevant for a specific country. This underlines the importance of studying the determinants of health within a specific context.

In line with this review, it would be interesting to further investigate the overlap of determinants of OF with determinants of other health conditions. For example, willingness to cervical cancer screening has been linked to a higher level of education, as well as formal forms of employment, and might even show more overlap with OF if further studied [62‐64]. This would mean that programs to address these socio-economical determinants could be of interest to several different stakeholders and might show to be very cost-effective. Nevertheless, the focus on socio-economical determinants of health should not surpass focus on health care services. During the decades of the Millennium Development Goals, government interventions like the recruitment of midwifes and the distribution of Emergency Obstetric Care equipment to health facilities, have resulted in impressive gains in maternal mortality and morbidity [65]. Also, though prevention is key, treating those that already have fistula remains of great importance especially considering the enormous impact of OF on the affected women. Fortunately, after the launch of the global campaign to end fistula in 2003, much progress has been made in improving the treatment of OF patients, but the best strategy for OF prevention still remains a difficult topic [66‐69].

Although the Determinants of Health model is a tool to create a framework for health policy improvement, it does not make predictions about the single most efficient or cost-effective policy. On the contrary, the identified determinants underline the necessity of a combination of a variety of programs, not only focussed on health-system strengthening but also on diverse population based strategies, like gender equality and fighting malnurtrition [68].

Early child bearing, childhood malnutrition, low socioeconomic status and empowerment of women, low community awareness of what causes OF and overall lack of access to emergency obstetric care paint the picture of OF in Uganda. Poverty and low education link these determinants, creating a web that is disproportionately hard to escape from for the poorest women. The Ministry of Health should aim to improve health care services of family planning, emergency obstetric care, skilled birth attendance, and obstetric fistula treatment, but even more so cooperate with other sectorial ministries to tackle the determinants that shape the generative environment for OF and limit the access to care. Because all in all, the picture of OF is principally a picture tainted by inequity.