Analysis of the effects of the Health Voucher Program on maternal and neonatal health care indicators in the Adamawa Region, Cameroon

¹Higher Institute of Science, Technology, Management and Sustainable Development, Ngaoundéré, Cameroon, ²Department of Biomedical Sciences, Faculty of Science, University of Ngaoundéré, Ngaoundéré, Cameroon, ³Department of Public Health, Cosendaï Adventist University, Douala, Cameroon, ⁴Department of Biological Sciences, Faculty of Science, University of Ngaoundéré, Ngaoundéré, Cameroon

^&Corresponding author
Odile Ndeh Mafongang, Higher Institute of Science, Technology, Management and Sustainable Development, Ngaoundéré, Cameroon

Introduction    

The fight against maternal, neonatal, and infant mortality is a primary objective of health policy in Cameroon, like in other countries. It aligns with two of the three health-related Sustainable Development Goals (SDGs), specifically reducing child mortality and improving maternal health [1]. These goals have spurred numerous commitments from governments in economically disadvantaged countries and their partners to improve maternal and child health care indicators. In Cameroon, these commitments led to the implementation of several programs and projects, such as the Health Voucher program, which was implemented from 2015 to 2022 in the northern regions of the country. Before and during its implementation, a notable evolution in maternal health care indicators was recorded. From 2011 to 2018, the maternal mortality ratio decreased from 782 to 406 per 100,000 live births, the under-five mortality rate fell from 122 to 48 per 1,000 live births, and the neonatal mortality rate dropped from 31 to 28 per 1,000 live births [2,3]. Although these figures indicated an improvement, they remain well above the thresholds set by the SDGs, which aim by 2030 to reduce the global maternal mortality ratio to below 70 per 100,000 live births, neonatal mortality to 12 per 1,000 live births, and under-five mortality to 25 per 1,000 live births [1].

To enhance the health system's effectiveness in achieving these goals, the Health Voucher project was considered as one of the most suitable solutions in this context. Its strategic objectives were to contribute to the reduction of maternal and neonatal mortality in Cameroon by improving the quality of care, ensuring the availability of inputs, and reducing inequalities in access to obstetric care related to economic barriers [4]. However, its impact on maternal and neonatal health care indicators has not been clearly described across the different regions of the country, despite its integration into the Universal Health Coverage (UHC) framework in 2023 [5,6]. In a study conducted between 2021 and 2022 in two district hospitals in the North region of Cameroon, Bita et al. [7] revealed that the maternal death rate was not significantly different between users and non-users of the Health Voucher, while the neonatal death rate was higher among pregnant women who did not use the voucher compared to those who did. In another study conducted in the three northern regions between January 2013 and May 2018, Sieleunou and Bonong [8] described a significant improvement in the rate of deliveries assisted by skilled personnel in the early years of the Health Voucher project's implementation, but they did not find a similar improvement for first antenatal care visits. This information shows that mastering these indicators remains a major challenge for the Cameroonian health system. The objective of this study was therefore to analyse the evolution of maternal and neonatal health care indicators in the pilot districts of the Health Voucher program in the Adamawa region of Cameroon from 2016 to 2022.

Methods     

Study design, setting, and period: this was a retrospective quantitative study conducted from April 1^st, 2024, to September 10^th, 2024, in the health districts of Meiganga, Ngaoundéré Rural, and Ngaoundéré Urban in the Adamawa region of Cameroon.

Study population and selection of health facilities: the study population consisted of the 20 health facilities enrolled in the pilot phase of the Health Voucher project in the three specified health districts. All these health facilities were selected consecutively, and their data were analysed systematically. The health map of the study area is presented in Figure 1 [9].

Study variables: the study variables consisted of maternal and neonatal health care indicators, grouped into effect and impact indicators. They were defined as indicated by the Global Reference List of 100 Core Health Care Indicators published by the World Health Organization [10], the 2021 monitoring report of 100 key indicators in Cameroon [11], and the practices of the Health Voucher project in the country. The effect indicators included coverage rates for the first and fourth antenatal care (ANC) visits, births assisted by skilled health personnel, and the caesarean section rate. ANC coverage rates were based on the number of expected pregnant women as assigned from the national health information system data [12-18]. The first and fourth ANC coverage rates were defined as the ratio of the number of pregnant women who completed their first ANC visit and the number who completed their fourth ANC visit, respectively, over the total number of expected pregnant women. Births assisted by skilled health personnel were defined as the percentage of live births attended by skilled health personnel during a given period [10,11], and the caesarean section rate was defined as the percentage of deliveries by caesarean section [10,11]. In addition to these effect indicators, the adhesion rate was estimated by dividing the number of vouchers sold by the expected pregnant women population over a period [19]. The impact indicators included the maternal mortality ratio and the neonatal mortality rate. The maternal mortality ratio was defined as the annual number of female deaths from any cause related to or aggravated by pregnancy or its management (excluding accidental or incidental causes) during pregnancy and childbirth or within 42 days of pregnancy termination, irrespective of the duration and site of the pregnancy, expressed per 100,000 live births, for a specified time period [10,11]. The neonatal mortality rate was defined as the probability of a child born in a specific place and year or period dying during the first 28 days of life, expressed per 1,000 live births [10,11]. All twenty health facilities recruited in the pilot phase of the Health Voucher program were included in the study due to the availability of data, which were synthesised for all these facilities to determine the annual rates.

Data collection and analysis: data were collected from monthly on completed in the health facilities, then validated and compiled in CS DATA 3.0 [20] and CS DATA 4.0 [21] software over 7 years, from 2016 to 2022. The various indicators were determined according to the WHO's Global Reference List of 100 Core Health Indicators [10], the 2021 monitoring report of 100 key indicators in Cameroon [11], and the practices of the Health Voucher project. To analyse the evolution of the studied indicators during this period, their annual values were calculated using Microsoft Excel 2013, which was also used to create the corresponding graphs.

Ethical considerations: the study protocol was approved both by the ethics committee of the Ngaoundere University (Ref. M2SP006/ISSTMADD/P/D/DAARS of the 05^th August 2024) and the regional ethics committee (Ref. N° 682/L/RA/DSP/BFP/NGD of the 08^th August 2024). Before data collection, administrative consent was obtained from health facilities and health district managers. All procedures adhered to the ethical standards of the International ethical guidelines for epidemiological studies.

Results     

Effect indicators: the evolution of effect indicators is presented in Table 1. The adhesion rate, which reflects the project's acceptance by beneficiaries, was also included. During the study period, this rate increased sharply in the first three years, decreased in the following two years, and then resumed growth in the last two years. Overall, the adhesion rate showed a slight increase (β = 0.1) over the study period, remaining above 100%. The first antenatal care (ANC1) coverage rate gradually increased from 2016 to 2022, with a peak of 98% in 2018 and a regression coefficient of 1.3, indicating a positive increase over the entire period. The fourth antenatal care (ANC4) coverage rate experienced stronger growth than ANC1 (β = 4.7), but with lower absolute figures, ranging from 18.6% in 2016 to 48.1% in 2022. Unlike all other effect indicators, the rate of births assisted by skilled personnel showed a significant decline over the study period (β = -1.4). Relatively high at the beginning of the project, this rate began to fall from 2019 to 2021. The caesarean section rate was estimated only in the three health facilities that offered this service as part of the complementary package of activities covering emergency obstetric and neonatal care. During the study period, this rate showed a gradual increase across all studied health districts (β = 0.3). It rose from 2.1% in 2016 to a maximum of 4.7% in 2022.

Impact indicators: the impact indicators are expressed by the maternal mortality ratio and the neonatal mortality rate. The evolution of the maternal mortality ratio revealed an overall increasing trend during the study period (β = 0.1), marked by peaks and troughs from one year to the next, with a maximum of 2.09% in 2022, a minimum of 1.12% in 2017, and an average of 1.78% over the entire period (Figure 2). In contrast to the maternal mortality ratio, the neonatal mortality rate showed an overall decline over the study period, with a maximum of 14.1‰ in 2016, a minimum of 6.3‰ in 2020, followed by an upturn from 2021 onwards (Figure 3).

Discussion     

This study highlights the evolution of health indicators during the implementation of the Health Voucher project in the pilot districts of Meiganga, Ngaoundéré Rural, and Ngaoundéré Urban in the Adamawa region, Cameroon. The analysis provides an assessment of the project's contribution to maternal and neonatal care, particularly in relation to its key indicators.

The adhesion rate to the Health Voucher was studied as an indicator of the project's acceptance by its beneficiaries, pregnant women. The fact that this rate remained above 100% throughout the project's implementation period suggests massive adoption, exceeding the threshold of expected users according to national health information system data [12-18]. This high adhesion can be attributed to the influx of populations from neighbouring, non-enrolled health areas, reinforced by the project's social marketing approach. This approach included systematic enrollment of women, a communication system involving various resources such as community health workers, local associations, and health personnel themselves, and financial incentives for enrollment activities [22]. The significant variations observed in the enrollment rate during the study period are likely determined by the different phases of the project's implementation and their corresponding managerial approaches. This evolution has been described in three phases [23]. The first operational phase, from June 2015 to May 2018, was characterised by direct fund mobilisation that respected the prescribed 45-day deadlines, thus promoting fluidity in data processing and transmission. This phase corresponded to a significant increase in project adhesion. The second phase, from June 2018 to July 2020, was marked by complexities in fund mobilisation, leading to payment delays for various activities. This resulted in the demotivation of field actors, reflected in a progressive decline in the adhesion rate. This phase also coincided with the onset of the COVID-19 pandemic, which contributed to keeping beneficiaries away from health facilities due to the widespread fear it caused in communities. The third phase, from July 2020 to the end of the project, was marked by the payment of old invoices, which re-incentivised providers and consequently revived beneficiary adhesion.

During the study period, the evolution of effect indicators was marked by discrepancies. The increased coverage rates for the first and fourth antenatal care visits could be seen as a result of effective communication that progressively made users aware of the facilities offered by the voucher and encouraged them to use its services [22]. The observed results indicate a strengthening of antenatal care coverage, consistent with trends reported by Bita et al. [7] and in North Cameroon and by Sieleunou et al. [8]. In a time-series analysis of the Health Voucher in the northern regions of Cameroon. However, the disparity in the magnitude of these two indicators showed that users were significantly more diligent in attending the first ANC visit than the fourth. This discrepancy illustrates a low utilisation of purchased vouchers, attributable to various reasons: vouchers being purchased by individuals who did not fully commit to the project, the perception among some women whose care quality was lower for voucher holders, and the long distances for some women coming from other health areas. The overall decline in the coverage of births assisted by skilled personnel suggests a desertion of services, which shows a similarity to the trend in adherence to the Health Voucher project. It is plausible to consider the influence of COVID-19, but also a potential decline in the quality of delivery services linked to the difficulties of the second implementation phase. Despite this decline, the rate of skilled birth attendance remained significantly higher than that of fourth ANC visits throughout the study period. This could be due to two phenomena: the higher perceived risk associated with childbirth compared to pregnancy, which compels voucher holders to move to accredited health facilities for delivery, and the financing of referrals made by traditional birth attendants to encourage them to limit home births [24]. The rate of skilled birth attendance observed in this study was close to the 69.3% reported by Audibert et al. [25] among 2,406 women surveyed in the 2015 MICS survey in Mauritania, which focused on adherence to an obstetric package, a corollary to the Health Voucher in that country. However, it was lower than the 88.85% reported by Bita et al. [7] in a study on the contribution of the Health Voucher to improving maternal and neonatal health in the Guider and Mayo-Oulo districts of the North Region of Cameroon. This difference may be due to varying levels of information dissemination in the two regions, as well as adaptive practices developed by personnel during the project's second phase. Due to payment delays for invoices during this phase, beneficiaries were not always systematically registered as such, but were often required to pay directly for most services.

One of the tangible effects of the Health Voucher project on the provision of complementary obstetric and neonatal care is clearly visible in the steady increase in the caesarean section coverage rate, which rose from 2.1% in 2016 to a maximum of 4.7% in 2022. The financial mechanisms established to ensure referrals to facilities performing caesarean sections were highly incentivising for personnel, ensuring that women were referred promptly and systematically.

The fluctuating evolution of the maternal mortality ratio could be linked to the fidelity of reporting and the events observed during the different implementation phases of the project. The arrival of the Health Voucher project seems to have brought a substantial reduction in maternal deaths between 2016 and 2017. During this period, death reporting did not yet receive special attention. With the implementation of Performance-Based Financing, which included a bonus for death reporting, this activity gained momentum in 2018, explaining the increase in the maternal death rate. The drop observed in 2019 could be linked to the emergence of the COVID-19 pandemic, which led to a decrease in hospital attendance and concealed some deaths occurring in the community. Following extensive public awareness campaigns about the pandemic and the implementation of barrier measures, populations regained confidence, and health facility attendance increased, explaining the second peak in the reported maternal mortality ratio. It is possible that the drop in 2021 was linked to staff disengagement due to payment delays in the implementation of Performance-Based Financing and the Health Voucher, and that the recovery after this date corresponded to the back-payment of old invoices. The trend in the maternal death ratio indicates an overall increase in this indicator since 2016, which contrasts with the objectives of the Health Voucher program [5].

Conversely to the maternal death ratio, the downward trend in the neonatal death rate was reassuring. Indeed, with this project, special emphasis was placed on the technical capacity of health facilities caring for newborns, as well as on the scrupulous respect for hygiene measures. The positive effect of the Health Voucher project on neonatal mortality aligns with the conclusions of Noukeu et al. [26] and the UNICEF-IGME report [27], which indicated an improvement in neonatal survival linked to the increased availability of emergency obstetric and neonatal care. These results argue for strengthening geographical and financial targeting in the context of Universal Health Coverage in Cameroon [4].

Study limitations: the adhesion rate estimation was based on the number of vouchers sold and not on the effective enrollment of women who purchased these vouchers. It is possible that vouchers were sold to individuals who were not subsequently enrolled. Furthermore, the analysed figures were derived from activity reports filled out by providers. They were susceptible to underestimation, depending on the reporting fidelity of these providers. As this data did not cover all districts in the region, the focus on only the project's pilot districts reduced the representativeness of the districts. Nevertheless, the study offers an analysis of trends in maternal and neonatal health indicators over seven years, thereby highlighting variations in the indicators and raising questions about the real impact of the Health Voucher project on the health status of the concerned populations.

Conclusion     

Data analysed in this study revealed an overall increase in pregnant women's access to prenatal care, visible through massive adherence and a high coverage rate of antenatal consultations. It also highlighted a reduction in the neonatal mortality rate. However, the rate of births assisted by skilled personnel experienced a significant decline, and the maternal mortality ratio increased during the study period. These results challenge the various stakeholders involved in reducing maternal and neonatal mortality to rethink the implementation mechanisms of projects aimed at this reduction, such as Universal Health Coverage, and to analyse the causes of maternal deaths in these communities.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Study design and planning: Odile Mafongang Ndeh, Djibrilla Yaouba. Data collection: Odile Mafongang Ndeh, Raoul Adala Hayatou, Yaouba Djanabou. Data analysis and interpretation: Odile Mafongang Ndeh, Jacques Olivier Ngoufack Tsougmo, Albert Ngakou. Manuscript writing: Odile Mafongang Ndeh, Jacques Olivier Ngoufack Tsougmo. Critical revision of the manuscript: Odile Mafongang Ndeh, Jacques Olivier Ngoufack Tsougmo, Djibrilla Yaouba, Yaouba Djanabou, Raoul Adala Hayatou, Albert Ngakou. Guarantor of the study: Albert Ngakou. All authors have read and approved the final version of the manuscript.

Acknowledgments     

The authors warmly thank the staff of the health facilities accredited to the Health Voucher project and the regional health information system officials for their contribution in making the data available. They are also thankful to the regional officials of the Health Voucher project for their support and the useful information they provided. They express their gratitude to the officials of the Higher Institute of Science, Technology, Management and Sustainable Development and the ethics committee for their validation and authorisation.

Figures     

Figure 1: health map of the study area

Figure 2: evolution of the maternal mortality ratio

Figure 3: evolution of the neonatal mortality rate

References