Emergency obstetric care access dynamics in Africa’s cities: Analysis of women’s self-reported care-seeking pathways in Kampala city, Uganda

Birabwa Catherine, Beňová Lenka, van Olmen Josefien, Semaan Aline, Waiswa Peter, Banke-Thomas Aduragbemi
doi: https://doi.org/10.1101/2024.05.15.24307462

Abstract

The rapid urbanization, particularly in Africa has posed several challenges that affect provision and accessibility of healthcare. The complex mix of providers, socio-economic inequalities, and inadequate infrastructure have been found to limit clients’ ability to reach and utilize routine and emergency health services. A growing body of literature shows poor health outcomes in African cities, including high institutional maternal mortality, which reflect poorly performing health systems. Understanding care-seeking pathways is necessary for improving health service delivery and ultimately improving health outcomes. We describe typologies and attributes of care-seeking pathways, using self-reported data from a cross-sectional survey of 433 women (15-49 years) who had obstetric complications, from nine health facilities in Kampala city. Participants’ average age was 26 (SD=6) years, and 55% (237/433) lived in the city suburbs. We identified four broad pathway sequences based on number and location of steps: pathways with one step, directly to a facility that could provide required care (42%, 183/433); two steps, mostly including direct referrals from basic and comprehensive obstetric facilities (40%, 171/433); three steps, including potentially delayed referral trajectories as women first return home (14%, 62/433); and ≥4 steps (4%, 17/433). Comprehensive obstetric facilities referred out 43% (79/184) of women who initially sought care in these facilities. Peripheral facilities referred 65% of women directly to the National Referral Hospital. Majority (60%, 34/57) of referred women first returned home before going to the final care facility. Our findings suggest that care pathways of women with obstetric complications in Kampala mostly comprise of at least two formal providers. This implies that efforts to strengthen urban health systems for maternal health should adopt broad or integrated approaches; and calls for improved inter-facility communication, streamlined referral processes and emergency transport availability. Future studies should investigate quality and experiences of care along the pathways.

Key findings

  • The pathways to care were multiple and varied by complication. Majority included at least two formal providers, giving insights into the dynamics surrounding obstetric referrals in the city such as time spent in a facility before referral and travel time between facilities.

  • The issues identified from the facility referral pathways were: referrals from comprehensive obstetric facilities, bypassing of referral hierarchies by peripheral facilities, and low utilization of ambulances by peripheral facilities.

  • Additionally, a notable percentage of referral pathway trajectories were interrupted and delays created therein, by women having to go back home after first seeking care from a formal provider. This was largely to obtain funds and other requirements.

  • Majority of women seeking care for complications in Kampala city reside in neighboring suburbs.

Key implications

  • Policy makers and program stakeholders need to develop strategies and design services/systems that align with the care-seeking behaviors of clients in urban areas, including a substantial reliance on peripheral facilities, high contribution of referrals and the complexity of delays as clients move across different types or levels of providers.

  • Policy makers should strengthen and expand coverage of comprehensive obstetric care to reduce unnecessary referrals or bypassing. Policy initiatives to facilitate access through private hospitals could contribute to improved access.

  • Subnational stakeholders in urban areas, with support from national stakeholders, need to develop effective strategies and plans such as networks of care or care pathways for organizing and monitoring service delivery in metropolitan areas, with proper accountability for outcomes and resources.

Introduction

Urbanization is a major ongoing trend worldwide, with approximately 55% of the world’s population living in urban areas in 2018.1 Urbanization holds potential for socio-economic development but is associated with several challenges such as urban poverty and increasing inequality, which disproportionately affect low- and middle-income countries.2 The urbanization process in Africa is more rapid than that in other continents, and is characterized by inequality, informal settlement expansion, and inadequate infrastructure, including healthcare.3,4 These features have implications for urban health systems and could exacerbate barriers to improving health outcomes in urban settings.5,6 Studies in African cities show poor maternal and perinatal health outcomes such as high institutional maternal mortality and inequitable coverage of maternal health services.7,8 These partly reflect poorly performing urban health systems, in the context of a rapidly growing population with dynamic daily and seasonal migration patterns amidst constrained healthcare infrastructure.

Improving health outcomes is contingent on provision of quality health services, which also influences population access to healthcare.9 Urban areas generally have a high availability and coverage of health services compared to rural areas, even for maternal health.7,10 However, accessibility of quality care is challenged by complexities of urban health systems and their environment. For example, the high number and mix of providers result in failure or delay to seek healthcare and bypassing of health facilities because people do not know appropriate facilities to choose.11 Concurrently, dominance of fee-based private providers implies that essential care is unaffordable for majority urban poor.12 Furthermore, poor housing structures (slums/informal settlements) and inadequate road infrastructure also affect timely access.6 Long travel times, poor road infrastructure, and cost of both transport and care affect access to and use of maternal health services in urban settings.13,14 Understanding care-seeking pathways in urban settings is necessary for improving health service delivery and ultimately improving health outcomes.

Over 60% of global maternal deaths occur in sub-Saharan Africa (SSA), largely due to treatable obstetric complications such as postpartum hemorrhage and hypertension in pregnancy.15,16 As such, timely access to appropriate emergency obstetric care (EmOC) is critical for maternal and perinatal survival.17 EmOC includes nine interventions recommended for managing common obstetric complications. Facilities with capacity to provide seven of the nine interventions are designated as basic EmOC (BEmOC) facilities. These interventions include administration of parenteral uterotonics, antibiotics and anticonvulsants/antihypertensives, manually removing retained placenta, removing retained products of conception, performing assisted vaginal delivery and basic neonatal resuscitation. In addition to these, facilities conducting caesarean deliveries and providing blood transfusion are designated as comprehensive EmOC (CEmOC) facilities.18 Furthermore, quality EmOC includes appropriate referral that is, making timely decisions to refer, using appropriate transportation, and appropriate information exchange between health facilities.19

Studies on healthcare seeking in urban settings in Africa show varied patterns such as bypassing of primary health facilities for routine care and preference of public health facilities for maternal health services.13,20 Regarding EmOC particularly, studies have focused on the determinants of access,21 with a few studies describing pathways (series of steps women take along their care-seeking journeys) to care.22–24 Studies in urban settings in SSA elaborate travel dynamics of women in emergency situations, highlighting different patterns and determinants.22,25,26 However, these insights emerge from a few cities in the region, yet women’s care-seeking processes might differ in other cities due to differences in resource availability, governance structures and population dynamics, which influence organization of and timely access to health services.27 Additionally, existing studies largely use data from medical records rather than women’s first-hand accounts and provide less information on intermediate steps between place of origin and where women receive care. Insights from women’s lived experiences provide an opportunity to integrate women’s voices in service delivery planning/strategies and improve health system responsiveness.

Uganda has made substantial progress in reducing maternal mortality,28 however, poor maternal and perinatal health outcomes in the country’s capital city (Kampala) remain a challenge.29 Persistent poor health outcomes amidst healthcare deprivations and a struggling healthcare system in Kampala present important research areas with potential relevance for other cities in Africa.29,30 In 2020, the Ugandan government prioritized urban areas with a strategic focus on improving the organization of health services and health system performance.31 Existing evidence in Kampala related to pathways to EmOC describes conditions for which women are referred between health facilities, and sources of obstetric referrals.32–34 However, this information is limited to a few public health facilities and there is insufficient evidence on the steps women actually take to reach care. This study’s objective was to examine and sequentially map self-reported care-seeking pathways among women who had obstetric complications in selected EmOC facilities in Kampala city, Uganda.

Methods

Study design and setting

This cross-sectional study was conducted in EmOC facilities in Kampala city, Uganda’s oldest and largest urban area, part of the Greater Kampala Metropolitan Area together with Wakiso, Mukono and Mpigi districts. Kampala has a resident population of 1.7 million which swells to 3.5 million during the day.33 Kampala exhibits economic disparities with poverty rates as high as 25% in some parishes.35 Kampala’s health system, comprised of a mix of 1,497 public and private health facilities in 2020,33 operates through different levels ranging from health centres (HCs) to National Referral Hospitals (NRH) (Figure 1). Uncomplicated deliveries can occur at any level, but obstetric complications are managed in CEmOC facilities (HC IV and above).36,37 Although privately-owned facilities are the majority (98%, n=1471), only 2% (n=30) are designated CEmOC facilities, and women predominantly seek obstetric care from the 26 (2%) public facilities which are expected to provide free care.13 We purposefully selected nine EmOC facilities, seven public and two private (Figure 1). We aimed to include typical providers of obstetric care across Kampala’s administrative divisions.36,37 Eight of the selected facilities are among those conducting the highest number of deliveries in Kampala.

Figure. 1
Figure. 1 Health service delivery structure in Uganda for obstetric care and facilities included in the study.

Source: developed for the study based on Uganda’s health service delivery standards33,37

Study participants

Women who had obstetric complications were approached for inclusion in this study. We included all direct obstetric complications-obstetric hemorrhage (antepartum and postpartum), pre-eclampsia/eclampsia, obstructed/prolonged labor, pregnancy-related sepsis, complications of abortion and ectopic pregnancy. Women aged 15-49 years who provided written consent to participate in the study were included. Eligible women who declined to or could not participate were not included.

Sampling and recruitment

Using a 50% prevalence for pathways to EmOC within which women are referred from one facility to another34 and a 10% adjustment for non-response, we calculated a sample size of 428 women. We recruited participants from different units/wards within the facilities, including maternity/labor, postnatal, high-dependency unit, gynecology, and emergency departments. We reviewed women’s medical files to identify those with an obstetric complication and confirmed this through consultation with unit/ward in-charge (a midwife or senior nursing officer) or doctors on duty. Participants were selected consecutively throughout the study period. We aimed to select at least a half of the sample size from the NRH due to referral patterns in Kampala33. To ensure variation across the six complications at the NRH, recruitment of obstructed labor cases was stopped as their admissions superseded other complications. In all, we approached 440 women, of which seven declined to participate.

Study variables

Our primary outcome was self-reported care-seeking pathways, retrospectively captured by asking women to detail steps taken from deciding to seek help for symptoms or going to a health facility for childbirth or routine care, up to their final care facility. We utilized a sequential approach informed by Haenssgen and Ariana38 to capture each step’s location, activities, and duration. Locations included different health facilities, informal providers (like traditional birth attendants) and places of self-care/social support (home for the participant or another person). We captured any reported activity done or help received at each step. These included formal care (such as assessment, referral, delivery), self/informal care (including preparing childbirth items), and ignoring. We captured self-reported time spent in each step and between steps, initially categorized as minutes, hours, days, and weeks, and then quantified. Additional data on transportation, enablers and barriers faced were collected. Descriptions of pathway attributes and other study variables are provided in Table 1.

View this table:
Table 1: Definition of pathway attributes and other study variables

Data collection

Data were collected using an interviewer-administered questionnaire developed for the study, using KoboCollect. An excerpt of the questions used to generate the pathway data is provided in (Appendix Figure. 1). Nine research assistants, trained and supervised by the lead author, collected the data. These included midwives and graduates with prior experience in facility-based research. We reviewed women’s medical records for information on diagnosis on admission, pregnancy outcome and gestation age. Data were collected between July and September 2022.

Data analysis

Data were cleaned and analyzed in Stata v14 (Stata Corporation, College Station, Texas, US). Descriptive statistics including frequencies and percentages were used to summarize categorical characteristics of participants and pathway attributes; the mean and standard deviation (SD) were used to summarize age across the sample, while the median and inter-quartile range (IQR) were calculated for duration (within/between steps). Data were presented in tables and figures.

We analyzed all reported pathways to identify unique sequences, and then generated a typology of common pathways using the number of steps and their location. At each step, we mapped engagement with different sources of care/help along the pathways by counting each location. For each common pathway sequence, we summarized relevant attributes, considering key obstetric care practices and the urban context. During analysis of pathways involving facility referral, we distinguished CEmOC from other health facilities. We described individual pathways of selected women (using pseudonyms) to illustrate common pathway sequences. Next, we explored pathways by complication, analyzing the distribution of common pathways, travel and referral patterns, and how pathways started. We considered two possible ways in which women’s care-seeking started: 1) symptoms (a woman experienced something which prompted her to seek care – woman sensed something could be wrong); and 2) labor (A woman went to the facility for routine childbirth care and in the process of or after birth developed the complication – woman was not aware something could be wrong with the pregnancy when she first sought care). Instances where a woman’s journey started going to a health facility for routine care such as ANC and the complication detected during this time (signs), were summarized with symptoms.

Ethics

Ethical approval was obtained from the ethics committees of Makerere University School of Public Health (SPH-2021-169), the Uganda National Council for Science and Technology (HS1952ES), the Institute of Tropical Medicine Antwerp (1529/21) and the University of Antwerp Hospital (2021-1743). Written consent to participate in the study was obtained from each participant before the interview. No directly identifying information such as individual names and national identification numbers were collected. Each participant and study site were given a unique identifier.

Results

Participant characteristics

Among the 433 participants, 137 (32%) had obstructed labor, 105 (24%) had pre-eclampsia/eclampsia, 87 (20%) had hemorrhage, 54 (13%) had ectopic pregnancy complications, 50 (12%) had abortion-related complications and 26 (6%) had pregnancy-related sepsis. Women in the sample were 26 years on average (SD=6). The majority were married/cohabiting (79%), had secondary education (53%), and lived outside Kampala city (55%) particularly in Wakiso district (Table 2). More than 70% attended ANC, with a maximum of nine visits. Among all participants, 5% experienced a stillbirth. Place of delivery was more commonly decided by women alone (34%), while decisions related to transportation (33%) and spending money for pregnancy/childbirth care (39%) were more frequently made by women’s partners. Further, 32% of participants first sought permission to spend money to seek care, and this was mostly from their partners (89%, 117/131).

View this table:
Table 2 Characteristics of the sample of women who experienced obstetric complications in Kampala (n=433, 9 facilities)

Pathways to care among women who had obstetric complications

We identified 51 unique pathways among the 433 participants (Appendix Figure 2). Across all pathways, the initial points-of-contact were as follows: CEmOC facilities (43%, 184/433), peripheral facilities (25%, 110/433), NRH (14%, 60/433), BEmOC facility (12%, 50/433), self-care/social support (5%, 20/433) and informal care (2%, 9/433). Of the 184 participants who first sought care from a CEmOC facility, 43% (n=79) were referred out. We grouped the pathways into four sequences (A, B, C, D) based on the total number (ranging from one to six) and location of steps taken (Figure 2). Sequence A, with women who had one step on the pathway (direct to the final facility of care), was the most common (42%, 183/433). The most frequent complication in this sequence was obstructed labor (36%, 65/183). Sequence B represents pathways with two steps including the final facility of care (40%, 171/433). The first step was a CEmOC facility (seq. B1) in 40% (69/171), a peripheral facility (seq. B2) in 39% (66/171), a BEmOC facility (seq. B3) in 17% (29/163), and a place of self-care/social support in 4% (7/171, not shown). The most frequent complication in sequence B was pre-eclampsia (32%, 55/171). Sequence C contains pathways with three steps including the final facility of care and accounted for 14% (62/433) of the sample. The most frequent complication in this sequence was pre-eclampsia (29%, 18/62). Sequence D represents pathways with four or more steps including the final facility of care, representing 4% (17/433) of the sample, with obstructed labor being the most common complication (41%, 7/17). Figure 2 also shows distribution of places women visited at each location (step) of the care pathways. Women who did not go directly to final facility (n=250) more commonly first sought care from peripheral (44%, n=109) and CEmOC (32%, n=79) facilities. At subsequent steps, places of self-care/social were the most accessed (e.g., 58% at step 2; 46/79).

Figure 2.
Figure 2. Common pathways to emergency obstetric care among women from selected facilities in Kampala city, n=433

Continuous arrows between the place of origin and ‘location 1’ in the figure show the major pathway sequences and the dashed arrows show sub-sequences.

Embedded Image = home

Description of selected attributes of the common pathway sequences to EmOC

Women with Sequence A (direct to final care facility) travelled mostly (54%) by motorcycle, with majority going to HC IVs and hospitals (Table 3). Having someone to help women prepare was an important facilitator for care-seeking in this group. Among participants with Sequence B pathways, particularly those involving health facilities (sub-Sequences B1-B3), transport availability facilitated care access for at least 20% of participants. The median time spent in the facility before referral was 4 hours (IQR=1-33) in CEmOC, 5 hours (IQR=1-14) in BEmOC and 2 hours (IQR=1-7) in peripheral facilities. Fifty-five percent of women referred by CEmOC facilities and 20% of women referred by BEmOC or peripheral facilities used an ambulance for transport to the final facility. Among participants with Sequence C, pathways of 24% of women included two formal health facilities before the final facility. Also, places of self-care/social support accounted for 60% as the second step after women first seeking care from a health facility. All 17 women with sequence D had unique pathway sequences, with six including more than one formal health facility before the final facility. Additional details on each sequence are provided in (Appendix Table 1).

View this table:
Table 3: Key attributes of the major pathways among women who experienced obstetric complications in selected facilities in Kampala (n=433†)

Individual participant pathways

Illustrations of women’s individual journeys in each sequence described in Table 3 are shown in Figure 3. Starting from the place of origin, the figures show how women with different complication, age and obstetric history profiles reached the final facility, and their outcomes. Figures 3.1 and 3.5 illustrate the variation in pathways women with the same complication can take to care.

Figure 3.
Figure 3. Individual pathways illustrating the pathway sequences

All women’s names are pseudonyms

Description of pathways by obstetric complication

Pathways among women with sepsis (73%, 19/26), abortion (100%, 50/50), and ectopic pregnancy complications (98%, 53/54) mostly started with symptoms recognized by the woman or signs detected by a health worker (Appendix Figure 3). Among women with hemorrhage (63%, 55/87) and obstructed labor (91%, 125/137), pathways mostly started with women seeking care expecting routine childbirth (labor). Pathways of women with abortion complications mostly had one step (70%, 35/50), while over 50% pathways of women with all other complications had two or more steps. Additionally, women more frequently started their journey to the final facility of care in the morning across the six complications (168/428, 39%); others travelled in the afternoon (121/428, 28%), evening (68/428, 16%) and night (71/428, 17%). Over 50% of women across all complications, had pathways where the first facility visited was a designated CEmOC facility, except for ectopic pregnancy complications, for which the majority of women first sought care from a BEmOC or peripheral facility (52%, 28/54).

Discussion

Our study presents access dynamics for EmOC in African cities through examining care-seeking pathways in Kampala city. We found four common pathway sequences based on the number and nature of steps taken by women with complications: 1) direct pathways (42%), 2) pathways with two steps (40%), 3) pathways with three steps (14%), and 4) pathways with four or more steps (4%). Our results further showed that 40% of women who first sought care from CEmOC facilities were referred out. Among women who first sought care in a BEmOC or peripheral facility, 65% were referred directly to the NRH, thus bypassing expected levels of referral hierarchy. There was lower utilization of ambulances for transportation during referral from BEmOC or peripheral facilities (20%) compared to referrals out of CEmOC facilities (55%). Also, there were delays within referral sections of women’s pathways with the majority (60%) returning home before going to the final facility.

We found that four in 10 women took a direct path to care. This finding could be attributed to perceived quality of care and physical accessibility,40 which in our study influenced women’s health facility choice, in addition to familiarity with the health facility. Additionally, “early referral” advice given to women with high-risk pregnancies during ANC as is recommended in Uganda,36 could explain direct care-seeking. Our study showed that women’s social networks played a key role in enabling direct care-seeking, suggesting the importance of involving women’s network members in interventions addressing delays in care-seeking. A special focus on vulnerable women, lacking social support/capital for childbirth in urban areas is necessary.

Among pathways with two steps, our results showed that 40% of women initially seeking care from CEmOC facilities (predominantly HCIVs) were referred out, notably so among women who had pre-eclampsia (70%). This is inconsistent with the expected functionality of CEmOC facilities which should manage nearly all complications. Moreover, we found that 65% of women initially seeking care from BEmOC or peripheral facilities were referred directly to the NRH, showing non-adherence to referral guidelines.36 These suboptimal referral practices could be attributed to health system constraints such as a lack of skilled/critical human resources, commodities, and inadequate capacity of some CEmOC facilities to manage critically sick patients; or the inability of women to afford care in private hospitals and weak enforcement of referral guidelines.41,42 Furthermore, we found that the length of time women spent before being referred out ranged from 2 hours (in peripheral facilities) to 5 hours (in BEmOC facilities). While this may include pre-referral treatment or waiting for transportation, it could also indicate failure of providers to make timely decisions and unnecessarily delaying referrals. Capacity strengthening of health facilities in line with their designated levels of functionality and better care coordination are essential, and could reduce delays from unnecessary referral and bypassing, and also crowding at receiving facilities.

We found that fewer women referred out of BEmOC or peripheral facilities (20%) used an ambulance compared to women referred out of CEmOC facilities (55%). This indicates that peripheral facilities, which are expected to refer all major obstetric complications, have limited availability or accessibility to ambulance services. Kampala has a centralized coordination system for ambulances that targets public and private health facilities.33,43 Failure of BEmOC and peripheral facilities to use ambulances can arise when multiple requests are made, and priority is given to other types of facilities or patients. Additionally, some peripheral facilities may not be aware of available emergency transportation services and may have limited support to access them. These and other factors related to functionality of ambulances and administrative procedures may explain the apparent low use of ambulances by peripheral facilities. This may translate as additional costs incurred by women as they shoulder the burden of arranging and paying for transportation, and exacerbates inequalities in access to EmOC. Improving coverage of affordable ambulance services at peripheral facilities and in communities is crucial to address this issue.

We further found that 20% of women visited more than one place before reaching the final care facility (Sequences C and D), with nearly a third seeking care from two or more formal health facilities. This reflects complexity of care-seeking in urban areas, and may be influenced by the type/capacity of health facilities visited by women.23 This finding could further reflect deficiencies in health facilities’ capacity to manage obstetric complications. However, it could also be attributed to women’s preferences, perceived severity of their condition, or unacceptability and un-affordability of the facility to which they were referred.40,44

Surprisingly, we found that 60% of women referred from one facility to another returned home before going to the final care facility. This was common among women with pre-eclampsia (71%) and obstructed labor (70%). This pattern points to inadequate birth preparedness and complication readiness, as women reported that they had to first prepare themselves, get baby clothes or money during this time. Poor birth preparedness practices among urban residents have also been reported in Nigeria.45 This is likely a result of financial constraints and limited decision-making power of women regarding money for pregnancy/childbirth which would force women to delay care seeking.13 It could also be attributed to the system inadequately supporting pregnant women with childbirth essentials during their stay in health facilities, or a lack of national health insurance, as is the case in Uganda. Inevitably, women bear this burden which potentially delays their care-seeking.

We found that the majority of women seeking care for complications in Kampala reside in neighboring districts or further away. Several factors contribute to this, such as inadequate numbers or readiness of EmOC facilities in neighboring districts, or proximity to Kampala which facilitates physical access to facilities. Critically, we note the lack of a gatekeeping practice in Uganda’s health system (one can go wherever they want). While districts in Uganda are mandated to manage their population’s healthcare under the decentralization policy, there is limited guidance on collaboration across district health systems which is crucial for the metropolitan context of Kampala. Among ongoing efforts to address urbanization challenges in Uganda is the development of an Urban Health Strategy. It is critical that this strategy includes a system to coordinate service delivery including referral care in metropolitan areas.

This could leverage the networks of care approach where health facilities in the metropolitan area are purposefully linked, to optimize patient experiences and outcomes.46

Strengths and limitations

With a large sample of 433 women who had common six obstetric complications, our study provides useful insights into different features of pathways to EmOC in an urban setting, which can be used to improve the provision of EmOC services and optimize health outcomes in urban health facilities. We included eight of the high obstetric volume facilities in Kampala. However, the study also had some limitations. First, due to recruiting from health facilities, the pathways identified may not represent journeys of women who fail to reach skilled routine or emergency obstetric care. However, given the high facility delivery rate of 94% in Kampala 47, the identified pathways provide a good reflection of how women possibly reach care. Also, our recruitment sites included only two private hospitals which might not be representative of all women seeking care in the several private health facilities across the city. Second, we used self-reported data which could have been affected by response bias. Third, we did not include women who experienced an emergency due to fetal complications and other maternal complications. Similarly, we did not include pathways of women who experienced maternal deaths and some women, particularly those who experienced poor pregnancy outcomes such as stillbirths, declined to participate in the study.

Implications for future research

Our study focused on care-seeking processes of women with complications. Future studies should examine processes of care provision at the health facilities where women sought care to further understand the alignment between care-seeking and service delivery. Also, future studies could use objective measures for some pathway attributes such as time spent within and between different steps and integrate quality, experience, and cost of care assessment along women’s pathways as well as women’s preferences. We also suggest that studies including women who experienced poor perinatal outcomes be conducted to map their care-seeking processes and establish pathway attributes which are more likely to contribute to poor outcomes.

Conclusions

We found four common pathway sequences to care in Kampala, with most women going directly to the final care facility. The majority of women referred between health facilities lived in neighboring districts, and bypassing was common. Also, most referred women delayed reaching final facility as they had to first prepare themselves. Increasing availability and readiness of CEmOC facilities is crucial to reduce CEmOC-to-CEmOC referrals or bypassing of nearest CEmOC facilities by lower-level facilities. Referral care and healthcare seeking competency in Kampala and surrounding districts should be improved at all system levels to achieve effective patient care pathways. The “networks of care” approach could be adopted, which would need capacity building of peripheral facilities, streamlining and strengthening the quality of facility linkages, and increasing availability of affordable emergency transport options, especially for emergencies originating in peripheral facilities. The development of a national health insurance system that covers the urban poor could greatly improve timely access to care. Future studies should investigate technical quality and experience of care received in these pathways. Last, women’s social context should be integrated in interventions addressing delays in seeking/reaching EmOC.

Data Availability

All data produced in the present study are available upon reasonable request to the authors.

Author contributions

CB conceptualized the study, analyzed the data and produced initial drafts of the paper. ABT, JVO, LB and PW reviewed the design and provided overall study supervision. LB and AS participated in the analysis and visualization of results. All authors reviewed and edited the initial drafts and approved the final version that was submitted.

Funding statement

CB was funded by the Belgian Directorate for Development Co-operation PhD sandwich programme. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests

Authors declare no competing interests.

Acknowledgements

We thank the study participants and research assistants. We thank Dr Dinah Amongin for reviewing and providing feedback on the manuscript.

Appendix

Figure 1
Figure 1 Questionnaire excerpt used to capture care-seeking pathways.
Figure 2
Figure 2 Unique pathway sequences identified across all 433 respondents
Figure 3
Figure 3 Pathways by obstetric complication (n=433)
View this table:
Table 1 Description of the common pathways to EmOC identified among women who received treatment for obstetric complications in Kamapala (n=433, 9 heath facilities)

Footnotes

  • ↵† co-senior authors

References

  1. 1.↵
    United Nations D of E and SAPD. World Urbanization: The 2018 Revision (ST/ESA/SER.A/420).; 2019.
  2. 2.↵
    Zhang XQ. The trends, promises and challenges of urbanisation in the world. Habitat Int. 2016;54:241–252. doi:10.1016/j.habitatint.2015.11.018
    OpenUrlCrossRef
  3. 3.↵
    Förster T, Ammann C. African Cities and the Development Conundrum: Actors and Agency in the Urban Grey Zone. In: Ammann C, Förster T, eds. African Cities and the Development Conundrum. Vol 10. Brill and Nijhoff; 2018:3-25.
    OpenUrl
  4. 4.↵
    Kamana AA, Radoine H, Nyasulu C. Urban challenges and strategies in African cities – A systematic literature review. City and Environment Interactions. 2024;21:100132. doi:10.1016/j.cacint.2023.100132
    OpenUrlCrossRef
  5. 5.↵
    Elsey H, Agyepong I, Huque R, et al. Rethinking health systems in the context of urbanisation: Challenges from four rapidly urbanising low-income and middle-income countries. BMJ Glob Health. 2019;4(3). doi:10.1136/bmjgh-2019-001501
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    Vlahov D, Freudenberg N, Proietti F, et al. Urban as a determinant of health. Journal of Urban Health. 2007;84(SUPPL. 1). doi:10.1007/s11524-007-9169-3
    OpenUrlCrossRefPubMed
  7. 7.↵
    Wong KLM, Banke-Thomas A, Sholkamy H, et al. A tale of 22 cities: utilisation patterns and content of maternal care in large African cities. BMJ Glob Health. 2022;7(3). doi:10.1136/bmjgh-2021-007803
    OpenUrlAbstract/FREE Full Text
  8. 8.↵
    Birabwa C, Banke-Thomas A, Waiswa P, et al. Maternal health in cities: analysis of institutional maternal mortality and health system bottlenecks in Kampala City Uganda 2016-2021. (in press). J Glob Health Rep. Published online 2023.
  9. 9.↵
    Berman P, Pallas S, Smith AL, Curry L, Bradley EH. Improving the Delivery of Health Services: A Guide to Choosing Strategies.; 2011.
  10. 10.↵
    Banke-Thomas A, Wright K, Collins L. Assessing geographical distribution and accessibility of emergency obstetric care in sub-Saharan Africa: A systematic review. J Glob Health. 2019;9(1). doi:10.7189/jogh.09.010414
    OpenUrlCrossRef
  11. 11.↵
    Banke-Thomas A, Balogun M, Wright O, et al. Reaching health facilities in situations of emergency: qualitative study capturing experiences of pregnant women in Africa’s largest megacity. Reprod Health. 2020;17(1):145. doi:10.1186/s12978-020-00996-7
    OpenUrlCrossRefPubMed
  12. 12.↵
    Ezeh A, Oyebode O, Satterthwaite D, et al. The history, geography, and sociology of slums and the health problems of people who live in slums. The Lancet. 2017;389(10068):547–558. doi:10.1016/S0140-6736(16)31650-6
    OpenUrlCrossRefPubMed
  13. 13.↵
    Magunda A, Ononge S, Balaba D, et al. Maternal and newborn healthcare utilisation in Kampala urban slums: perspectives of women, their spouses, and healthcare providers. BMC Pregnancy Childbirth. 2023;23(1):321. doi:10.1186/s12884-023-05643-0
    OpenUrlCrossRef
  14. 14.↵
    Banke-Thomas A, Wong KLM, Ayomoh FI, Giwa-Ayedun RO, Benova L. “in cities, it’s not far, but it takes long”: Comparing estimated and replicated travel times to reach life-saving obstetric care in Lagos, Nigeria. BMJ Glob Health. 2021;6(1). doi:10.1136/bmjgh-2020-004318
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    World Health Organization. Trends in Maternal Mortality 2000 to 2020: Estimates by WHO, UNICEF, UNFPA, Work Bank Group and UNDESA/Population Division.; 2023. Accessed February 23, 2023. https://www.who.int/publications/i/item/9789240068759
  16. 16.↵
    Musarandega R, Nyakura M, Machekano R, Pattinson R, Munjanja SP. Causes of maternal mortality in Sub-Saharan Africa: A systematic review of studies published from 2015 to 2020. J Glob Health. 2021;11:04048. doi:10.7189/jogh.11.04048
    OpenUrlCrossRef
  17. 17.↵
    Paxton A, Maine D, Freedman L, Fry D, Lobis S. The evidence for emergency obstetric care. International Journal of Gynecology & Obstetrics. 2005;88(2):181–193. doi:10.1016/j.ijgo.2004.11.026
    OpenUrlCrossRefPubMed
  18. 18.↵
    World Health Organization. Monitoring Emergency Obstetric Care: A Handbook.; 2009.
  19. 19.↵
  20. 20.↵
    Clarke-Deelder E, Afriyie DO, Nseluke M, Masiye F, Fink G. Health care seeking in modern urban LMIC settings: evidence from Lusaka, Zambia. BMC Public Health. 2022;22(1):1205. doi:10.1186/s12889-022-13549-3
    OpenUrlCrossRef
  21. 21.↵
    Geleto A, Chojenta C, Musa A, Loxton D. Barriers to access and utilization of emergency obstetric care at health facilities in sub-Saharan Africa: a systematic review of literature. Syst Rev. 2018;7(1):183. doi:10.1186/s13643-018-0842-2
    OpenUrlCrossRefPubMed
  22. 22.↵
    Banke-Thomas A, Avoka C, Olaniran A, et al. Patterns, travel to care and factors influencing obstetric referral: Evidence from Nigeria’s most urbanised state. Soc Sci Med. 2021;291:114492. doi:10.1016/j.socscimed.2021.114492
    OpenUrlCrossRefPubMed
  23. 23.↵
    Elmusharaf K, Byrne E, AbuAgla A, et al. Patterns and determinants of pathways to reach comprehensive emergency obstetric and neonatal care (CEmONC) in South Sudan: qualitative diagrammatic pathway analysis. BMC Pregnancy Childbirth. 2017;17(1):278. doi:10.1186/s12884-017-1463-9
    OpenUrlCrossRefPubMed
  24. 24.↵
    Alobo G, Ochola E, Bayo P, Muhereza A, Nahurira V, Byamugisha J. Why women die after reaching the hospital: a qualitative critical incident analysis of the “third delay” in postconflict northern Uganda. BMJ Open. 2021;11(3):e042909. doi:10.1136/bmjopen-2020-042909
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    Avoka CK on, Banke-Thomas A, Beňová L, Radovich E, Campbell OMR. Use of motorised transport and pathways to childbirth care in health facilities: Evidence from the 2018 Nigeria Demographic and Health Survey. PLOS Global Public Health. 2022;2(9):e0000868. doi:10.1371/journal.pgph.0000868
    OpenUrlCrossRef
  26. 26.↵
    Radovich E, Banke-Thomas A, Campbell OMR, et al. Critical comparative analysis of data sources toward understanding referral during pregnancy and childbirth: three perspectives from Nigeria. BMC Health Serv Res. 2021;21(1):927. doi:10.1186/s12913-021-06945-9
    OpenUrlCrossRefPubMed
  27. 27.↵
    Papanicolas I, Rajan D, Karanikolos M, Soucat A, Figueras J. Health System Performance Assessment: A Framework for Policy Analysis.; 2022.
  28. 28.↵
    UBOS UB of S. The demographic and health survey 2022 key findings. Published September 2023. Accessed September 28, 2023. https://www.ubos.org/wp-content/uploads/publications/09_2023UDHS_2022_Key_Findings_Presentation_A.pdf
  29. 29.↵
    Ministry of Health. National Annual Maternal and Perinatal Death Surveillance and Response (MPDSR) Report FY 2022/2023.; 2023.
  30. 30.↵
    Kananura RM, Waiswa P, Wasswa R, Boerma T, Blumenberg C, Maiga A. Deprivation and Its Association with Child Health and Nutrition in the Greater Kampala Metropolitan Area of Uganda. Journal of Urban Health. Published online March 13, 2024. doi:10.1007/s11524-023-00804-0
    OpenUrlCrossRef
  31. 31.↵
    Ministry of Health. Health Sector Development Plan 2020/21-2024/25.; 2020.
  32. 32.↵
    Okello DA, Ssemujju R, Tumusiime A, Nakyanzi T, Jordanwood IV, Chaitkin M. Magnitude and Consequences of MNCH and FP Service Congestion in KCCA Public Health Facilities. Findings from a Rapid Assessment of Utilization and Capacity at Kawaala and Kisenyi Health Centre IVs.; 2022.
  33. 33.↵
    KCCA KCCA. Directorate of Public Health and Environment: Annual Report, Financial Year 2021/22.; 2022.
  34. 34.↵
    Kyozira C, Walugembe P, Kharono B. A Case Study of Admissions Data from Kawempe National Referral Hospital: Further Evidence to Inform the Strategy to Relieve Overcrowding in Kampala.; 2021. Accessed February 9, 2024. https://www.ugandamchn.org/resource/case-study-admissions-data-kawempe-national-referral-hospital-further-evidence-inform
  35. 35.↵
    Uganda Bureau of Statistics. Poverty Maps of Uganda. Mapping the Spatial Distribution of Poor Households and Child Poverty Based on Data from the 2016/17 Uganda National Household Survey and the 2014 National Housing and Population Census. Technical Report.; 2019. Accessed February 20, 2024. https://www.ubos.org/wp-content/uploads/publications/02_2020Poverty_Map_report_Oct_2019.pdf
  36. 36.↵
    Ministry of Health. Essential Maternal and Newborn Clinical Care Guidelines for Uganda.; 2022.
  37. 37.↵
    Ministry of Health Uganda. Comprehensive Health Service Standards Manual.; 2021.
  38. 38.↵
    Haenssgen MJ, Ariana P. Healthcare access: A sequence-sensitive approach. SSM Popul Health. 2017;3:37–47. doi:10.1016/j.ssmph.2016.11.008
    OpenUrlCrossRef
  39. 39.
    MEASURE Evaluation. Referral Systems Assessment and Monitoring Toolkit.; 2013.
  40. 40.↵
    Mubiri P, Kajjo D, Okuga M, et al. Bypassing or successful referral? A population-based study of reasons why women travel far for childbirth in Eastern Uganda. BMC Pregnancy Childbirth. 2020;20(1):497. doi:10.1186/s12884-020-03194-2
    OpenUrlCrossRefPubMed
  41. 41.↵
    Selemani S, Mwakyusa MO, Bashiri S, et al. “Are all referrals necessary?” Experiences and perceptions of maternity healthcare providers on emergency intrapartum referrals in Dar es Salaam, Tanzania. PLoS One. 2024;19(2):e0298103. doi:10.1371/journal.pone.0298103
    OpenUrlCrossRef
  42. 42.↵
    Nkamba DM, Vangu R, Elongi M, et al. Health facility readiness and provider knowledge as correlates of adequate diagnosis and management of pre-eclampsia in Kinshasa, Democratic Republic of Congo. BMC Health Serv Res. 2020;20(1):926. doi:10.1186/s12913-020-05795-1
    OpenUrlCrossRef
  43. 43.↵
    Ononge S, Magunda A, Balaba D, et al. Strengthening Kampala’s Urban Referral System for Maternal and Newborn Care Through Establishment of an Emergency Call and Dispatch Center. www.ghspjournal.org
  44. 44.↵
    Stein DT, Golub G, Rothschild CW, Nyakora G, Cohen J, McConnell M. Bypassing high-quality maternity facilities: evidence from pregnant women in peri-urban Nairobi. Health Policy Plan. 2021;36(1):84–92. doi:10.1093/heapol/czaa092
    OpenUrlCrossRefPubMed
  45. 45.↵
    Obionu IM, Ajuba M, Aguwa EN. Preparation for birth and complication readiness: rural-urban disparities among pregnant women in communities in Enugu State, Nigeria. Pan African Medical Journal. 2022;42. doi:10.11604/pamj.2022.42.310.33015
    OpenUrlCrossRef
  46. 46.↵
    Kalaris K, Radovich E, Carmone AE, et al. Networks of Care: An Approach to Improving Maternal and Newborn Health. Glob Health Sci Pract. 2022;10(6):e2200162. doi:10.9745/GHSP-D-22-00162
    OpenUrlFREE Full Text
  47. 47.↵
    UBOS UB of S, ICF. Uganda Demographic and Health Survey 2016: Key Indicators Report.; 2017.
Back to top
Posted May 16, 2024.
Download PDF
Data/Code
Email
Emergency obstetric care access dynamics in Africa’s cities: Analysis of women’s self-reported care-seeking pathways in Kampala city, Uganda
Birabwa Catherine, Beňová Lenka, van Olmen Josefien, Semaan Aline, Waiswa Peter, Banke-Thomas Aduragbemi
medRxiv 2024.05.15.24307462; doi: https://doi.org/10.1101/2024.05.15.24307462
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area