Multi Disciplinary Sickle Cell Disease Obstetrics Care Program in Ghana

Sub-Saharan Africa has a high burden of sickle cell disease (SCD). Approximately 300,000 children are born with SCD in Sub-Saharan Africa yearly (Piel et al. 2013). Ghana is one of the most SCD-burdened countries, with 2% (approximately 18,000) newborns delivered with SCD each year (Ohene-Frempong et al. 2008). Compared with the United States of America and the United Kingdom, 2,400 and 300 newborns, respectively, are born with SCD each year.

Many children with SCD living in low-resource settings, particularly urban areas, are now surviving into childbearing years and adulthood due to general improvements in healthcare. With increasing life expectancy in Africa, pregnancy has become an emerging life-threatening complication in women with sickle cell disease (SCD). In low and middle-income countries (LMICs), the odds ratio of maternal death associated with SCD is 22.81 (95% CI 14.67-35.46) (Boafor et al. 2016). The nearly 23-fold increased odds of death are primarily due to SCD-related severe complications (Boafor et al. 2016). There were no evidence-based guidelines for managing pregnant women with SCD in low-resource settings. To address this, from 2012-2014, the Obstetrics Department at Korle-Bu Teaching Hospital (the flagship academic hospital) in Accra, Ghana, established a dedicated SCD Obstetrics clinic based on guidelines adapted from the American College of Obstetrics and Gynecology. The Institutional maternal mortality for women with SCD over these three years (2012-2014) was approximately 12%.

In January 2015, a multidisciplinary SCD obstetrics team, composed of local and international experts, was established at the flagship academic hospital to provide comprehensive care for this high-risk population. The local team included two obstetricians, three hematologists, two midwives, two laboratory scientists, a pediatrician, and a dual-certified anesthesiologist/ pulmonologist. The team conducted a combined retrospective/ prospective case series of all maternal deaths in women with SCD at Korle-Bu Teaching Hospital (the flagship academic hospital in Ghana) over seven years (2010-2016) (Asare et al. 2018). This retrospective data collection highlighted challenges to delivering improved care. Acute chest syndrome, preceded by acute pain episodes, was the leading cause of death in nearly 87% of women (Asare et al. 2018). In May 2015, the team established a joint obstetrics/hematology clinic, instituted close maternal and fetal monitoring, and implemented clinical guidelines/ protocols (including protocols for antenatal care, labor and delivery, postnatal care, and the management of SCD-related complications). In addition, the multidisciplinary SCD obstetrics team established a weekly communication system to adjudicate the management of challenging cases. These interventions resulted in an 89.1% relative risk reduction in maternal mortality (from 10,791 to 1,176 deaths per 100,000 live births; p=0.007) over 13 months (May 2015 - May 2016) (Asare et al. 2017). Since initiating the multidisciplinary SCD obstetrics program, the investigators have consistently decreased maternal mortality in this cohort by approximately 90% compared to before the team was established (Oppong et al. 2019; Swarray-Deen et al. 2022).

Despite overwhelming evidence of the positive impact of the multidisciplinary SCD obstetrics team, the team is restricted to the flagship academic hospital and faces numerous competing demands. Before the creation of the multidisciplinary SCD obstetrics team at Korle-Bu Teaching Hospital, the care of pregnant women with SCD in Ghana, including the investigators' study site, was obstetrician-led. After 7 years, the intervention has increased reach (the proportion of pregnant women with SCD who have received multidisciplinary SCD obstetric care divided by the total number of eligible pregnant women with SCD receiving care in facilities in the Greater Accra Region) from 0% to 20%. While there is a reduction in the maternal mortality ratio in women with SCD at Korle-Bu Teaching Hospital (the flagship academic hospital), the investigators' current reach of approximately 20% is too small to achieve the public health impact of decreased maternal mortality in this cohort in other parts of the Greater Accra Region, where more than 80% of pregnant women with SCD are seen. To reduce maternal mortality in women with SCD living in Ghana (with 16 regions) and elsewhere, the investigators must have a better implementation and scale-up strategy to increase the reach of multidisciplinary SCD obstetrics care in the Greater Accra Region.

There is an unequal distribution of human resources for health in Ghana (Asamani et al. 2021). Academic hospitals have a more significant proportion of highly trained medical personnel than non-academic hospitals. Conducting research and implementing findings into usual care is far more feasible in academic hospitals because of the availability of trained health personnel and other resources. For equitable translation of "research to practice" to be achieved, non-academic hospitals need to participate in conducting these research studies.

To rapidly translate the findings from the multidisciplinary SCD obstetrics program, the investigators proposed transferring these evidence-based practices from the academic hospital in Accra, with a population density of approximately 1,300 people/Km2, to the non-academic hospital. The non-academic hospital, similar to a city or county hospital in the United States of America for the indigent population, is run under the Ghana Health Service (GHS), under the supervision of the Ministry of Health (MoH), Ghana, and is situated approximately 6.6km South-East and 11 minutes from the academic hospital by road, with a population density of 1,500 people/Km2.

The premise of this feasibility study is that task-shifting and improving the care of pregnant women with SCD in a non-academic hospital with a maternal mortality rate of 7,921 deaths per 100,000 live births (2018-2021) in women with SCD will dramatically reduce maternal mortality in this cohort.

Task shifting is crucial to improving the reach of the care of pregnant women with SCD in Ghana, where approximately 2% of the children with SCD are born each year. The findings of the investigators indicate that multidisciplinary care for pregnant women with SCD in academic centers alone is insufficient to increase the reach for the health care of pregnant women with SCD living in Accra or the rest of Ghana and reduce maternal mortality in this cohort by at least 50%. Training healthcare providers (obstetricians, generalists/physicians, anesthesiologists, and nurses/midwives) already working in the non-academic hospital on the protocols used in the management of pregnant women with SCD (from antenatal care through to postnatal care and during admission for acute events) will be far more efficient and sustainable than total dependence on the multidisciplinary SCD obstetrics program at the academic hospital. As of 2022, there were 15 hematologists available for a population of approximately three million in the Greater Accra Region, and approximately 20 hematologists served approximately 34 million people in Ghana. Given the small number of regional hematologists, training efforts focused on generalists/physicians caring for pregnant women with SCD as part of a multidisciplinary approach to care in a non-academic hospital will reduce the maternal mortality. In addition, several pregnant women with SCD (n=150) from the South-Eastern part of the Greater Accra Region attend the non-academic hospital because of its location and accessibility; hence, maternal mortality can be reduced.

The World Health Organization (WHO) defines task shifting as the rational redistribution of tasks among health workforce teams (World Health Organization, 2007). Specific tasks are moved, where appropriate, from highly qualified health workers with shorter training and fewer qualifications to make more efficient use of available human resources for health (World Health Organization, 2007). In Ghana, like many other resource-constrained settings, there is a shortage of well-trained health workers (Asamani et al. 2018). The WHO estimates 1.1 doctors per 10,000 population in Ghana compared to 26 doctors per 10,000 population in the United States of America (World Health Organization, 2007). Even if Ghana embarks on an emergency training program for hematologists focused on the care of individuals with SCD, it will take at least 5 years to certify a hematologist. Due to the shortage of hematologists, multidisciplinary SCD obstetrics care has been limited to academic sites. Other alternatives are needed to address this shortage of health personnel. Task shifting provides a viable option for improving healthcare by making more efficient use of already available human resources while rapidly expanding the human resource pool and building more sustainable capacity.

This project aims to fill the current research-to-practice gap by creating a multidisciplinary SCD obstetrics program to prevent maternal mortality among women with SCD in a resource-constrained country, using implementation science. As described by Powell, the proposed implementation strategies (Plan, Educate, Restructure, and Quality Management) will be employed within the Consolidated Framework for Implementation Research (Damschroder et al. 2009; King et al. 2017). The Consolidated Framework for Implementation Research (CFIR) presents several constructs that acknowledge the multiple levels of systems needed to implement evidence-based interventions. The investigators will use the CFIR to inform the domains and questions for the needs assessment (semi-structured qualitative interviews) for Aim 1. The focus of the investigators will be on the following: intervention characteristics, individuals involved, inner and outer settings, knowledge and beliefs about the intervention, patient needs and resources, self-efficacy, adaptability, and the implementation climate within the organization's structure. Another area the investigators will focus on is the implementation process to assess whether stakeholder engagement, including the leadership of the non-academic hospital in planning and executing the program, will improve support and eventual sustainability (Aim 2). The project also aims to conduct a Hybrid Type 1 Feasibility Study (Aim 3) to extend the investigators' results on decreasing maternal mortality from an academic hospital to a non-academic hospital setting in Accra, Ghana, where the maternal mortality ratio is 7,921 deaths per 100,000 live births among women with SCD (2018-2021). The investigators will test the hypothesis that a task-shifted non-academic hospital site for pregnant women with SCD will be non-inferior to a multidisciplinary SCD obstetrics program in an academic hospital in Accra, Ghana, in reducing maternal mortality in this cohort.

Approach

To test the hypothesis, the investigators will complete the following specific aims:

1. Identify contextual determinants (barriers and facilitators) that influence the adaptability of the evidence-based practice of establishing a multidisciplinary SCD obstetrics team as an intervention in the non-academic hospital, including the implementation process (Years 1-2). The goal of this aim was to better understand current knowledge gaps, the perceptions of healthcare providers and hospital administrators regarding the management of pregnant women with SCD, and capacity-strengthening needs to inform future interventions in the non-academic hospital. The investigators conducted semi-structured qualitative interviews with healthcare providers [nurses/midwives and doctors (obstetricians, generalists/physicians, and anesthesiologists)], pregnant women with SCD, and structured interviews with hospital administrators.
2. Build capacity for a multidisciplinary sickle cell disease obstetrics program in a non-academic hospital (Years 3-5). The purpose of this aim was to better describe the implementation experience (what worked/what didn't work; the barriers/facilitators) and how the intervention needs to be adapted going forward. This aim addressed the identified barriers to managing pregnant women with SCD identified in Aim 1 by filling knowledge gaps through education and training. The investigators applied previously established education programs and evidence-based protocols developed by the multidisciplinary SCD obstetrics team at Korle-Bu Teaching Hospital (the flagship academic hospital) for caring for pregnant women with SCD, including obstetricians, generalists/physicians, anesthesiologists, and nurses/midwives, in the non-academic hospital. To guide the application of these established educational programs and evidence-based protocols, the investigators used findings from Aim 1, applying the CFIR framework and focusing mainly on the characteristics of the individuals' domain to address knowledge and beliefs about the "intervention" construct.
3. Conduct a Hybrid type 1 feasibility study comparing the effectiveness of a physician-based multidisciplinary sickle cell disease obstetrics program in a non-academic hospital (Years 3-5).

The goal of this aim is to determine the feasibility of initiating a multidisciplinary SCD obstetrics program for women with SCD in a non-academic hospital through 'task-shifting'. In a before-and-after study design, the investigators will test the hypothesis that multidisciplinary care for pregnant women with SCD in a non-academic hospital will result in a 50% relative risk reduction in mortality compared to the mortality rate in pregnant women with SCD in the same hospital before the multidisciplinary care.

Method Study design: A Hybrid type 1 feasibility study Study setting: The Greater Accra Regional Hospital (a non-academic hospital and a regional referral facility) is run under the Ghana Health Service (GHS), under the supervision of the Ministry of Health (MoH), Ghana, and is situated approximately 6.6km South-East and 11 minutes from the academic hospital by road. The non-academic hospital has 420 beds with a 110-bed obstetrics unit, and conducts about 8,500 deliveries per year, of which approximately 150 involve women with SCD. At the non-academic hospital, SCD contributes 1.2% of all deliveries and yet accounts for 7.1% of maternal mortality and is a leading indirect cause of maternal death. The maternal mortality ratio in this cohort is 7,921 deaths per 100,00 live births (2018-2021). At the non-academic hospital, the obstetrics unit runs a daily obstetrics clinic with an average weekly attendance of 250 pregnant women (including approximately 10 pregnant women with SCD). Despite the well-staffed obstetrics unit, the non-academic hospital still refers some challenging and high-risk obstetrics cases (including pregnant women with SCD) to the academic hospital. In the investigators' prior studies, >95% of all the pregnant women approached enrolled in the prospective cohort study.

Sample size determination: At the flagship academic hospital, the multidisciplinary SCD obstetrics care program consistently decreased maternal mortality by approximately 90% compared to the obstetrician-led care approach. The investigators expect that the task-shifted program will attain a significant reduction in maternal mortality, but not to the extent of the multidisciplinary SCD obstetrics care program at the academic hospital. If the baseline maternal mortality is 7,921 deaths per 100,000 live births (2018-2021), the investigators expect that a reduction of approximately 50%, or approximately 3,961 deaths per 100,000 live births, will be a measure of success. The investigators plan to recruit and follow up on 198 pregnant women with SCD attending the non-academic hospital to accomplish this aim. Each year, the investigators plan to recruit and follow up 66 pregnant women with SCD from the non-academic hospital.

Data Analysis Plan:

Categorical variables will be summarized with counts and percentages, and continuous variables will be summarized with the mean and standard deviation, or median and inter-quartile range, depending on the distribution. A one-sample proportion test will be performed to compare the observed proportion of maternal mortality to the current rate. Mortality will be further investigated by examining associations with demographic and clinical covariates.