Cost-effectiveness of conditional cash transfers to retain women in the continuum of care during pregnancy, birth and the postnatal period: Protocol for an economic evaluation of the Afya trial in Kenya

listen audio

Study Justification:
– The study aims to assess the impact, cost-effectiveness, and scalability of conditional cash transfers to promote increased utilization of antenatal care, facility delivery, and postnatal care services in rural Kenya.
– The utilization of these services by pregnant women in low/middle-income country settings is well below the recommended levels by the World Health Organization.
– The study addresses the need to reduce maternal and child mortality and morbidity in high-burden settings by evaluating the effectiveness of a specific intervention.
Study Highlights:
– The study will estimate the total and incremental costs and cost-effectiveness of the intervention, focusing on provider costs and health service costs.
– It will be conducted in Siaya County, western Kenya, which has poor indicators of maternal and child health.
– The intervention involves conditional cash transfers to pregnant women for attending ANC visits, facility-based delivery, PNC, and childhood immunization.
– The study will compare the intervention arm with a control arm that receives a nominal gratuity.
– Primary outcomes include the proportion of eligible ANC visits, facility-based deliveries, PNC appointments, and immunization appointments attended.
– Cost-effectiveness analysis and cost-consequence analysis will be conducted, presenting results in terms of incremental cost-effectiveness ratios (ICERs) and cost-consequence tables.
– Sensitivity analyses will be performed to assess uncertainty and robustness of the results.
– The equity impact of the intervention will be analyzed to determine if the benefits are shared equitably among different socioeconomic groups.
– Affordability of the intervention will be explored through fiscal space analysis and cost-effectiveness thresholds.
Recommendations for Lay Reader and Policy Maker:
– The study aims to evaluate the effectiveness and cost-effectiveness of conditional cash transfers to improve maternal and child health in rural Kenya.
– The findings will provide evidence on the impact of the intervention and its potential scalability.
– The study will help policymakers make informed decisions regarding resource allocation for maternal and child health interventions.
– The results will be disseminated in a peer-reviewed journal and presented at an international conference.
Key Role Players:
– Safe Water and AIDS Project, Kenya
– Stockholm Environment Institute, Sweden
– University College London, UK
– Ministry of Health, Kenya
Cost Items for Planning Recommendations:
– Program costs: CCT payments, gratuity payments, setting up and maintaining the payment system, community and health facility sensitization.
– Health service costs: ANC, facility-based delivery, PNC, child immunization, and related referral visits.
– Start-up costs and implementation costs will be differentiated.
– Donated goods and volunteer time will be assigned a current market value.
– Costs will be presented in Kenyan Shillings and International Dollars (INT$).
– Costs will be adjusted for inflation and converted to present values using an annual discount rate of 3% in the base-case, with sensitivity analyses using 0% and 6% discount rates.
– Affordability measures include total cost as a proportion of national GDP, proportion of public health expenditure, and proportion of public health expenditure on maternal and child health.

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong because it provides a clear description of the study objectives, methods, and expected outcomes. The abstract mentions that the study has received ethics approval and provides details about the sample size and primary outcomes. However, to improve the evidence, the abstract could include more information about the study design, data collection methods, and statistical analysis plan.

Introduction A wealth of evidence from a range of country settings indicates that antenatal care, facility delivery and postnatal care can reduce maternal and child mortality and morbidity in high-burden settings. However, the utilisation of these services by pregnant women, particularly in low/middle-income country settings, is well below that recommended by the WHO. The Afya trial aims to assess the impact, cost-effectiveness and scalability of conditional cash transfers to promote increased utilisation of these services in rural Kenya and thus retain women in the continuum of care during pregnancy, birth and the postnatal period. This protocol describes the planned economic evaluation of the Afya trial. Methods and analysis The economic evaluation will be conducted from the provider perspective as a within-trial analysis to evaluate the incremental costs and health outcomes of the cash transfer programme compared with the status quo. Incremental cost-effectiveness ratios will be presented along with a cost-consequence analysis where the incremental costs and all statistically significant outcomes will be listed separately. Sensitivity analyses will be undertaken to explore uncertainty and to ensure that results are robust. A fiscal space assessment will explore the affordability of the intervention. In addition, an analysis of equity impact of the intervention will be conducted. Ethics and dissemination The study has received ethics approval from the Maseno University Ethics Review Committee, REF MSU/DRPI/MUERC/00294/16. The results of the economic evaluation will be disseminated in a peer-reviewed journal and presented at a relevant international conference. Trial registration number NCT03021070.

A cost-effectiveness analysis of the Afya trial will estimate the total and incremental costs and cost-effectiveness of the intervention prospectively from the provider perspective, measuring provider (programme and health service) costs. Afya is being implemented in Siaya County, western Kenya. In 2016, Siaya County had an estimated population of 984 069 and a Human Development Index score of 0.46, significantly below the national average of 0.56.29 The county has very poor indicators of maternal and child health.30 Estimates from a 2011 survey show that the infant mortality rate was 111 per 1000 live births, far higher than the national rate of 49 per 1000. Similarly, the maternal mortality rate in Siaya is 695 per 100 000 live births, again much higher than the national maternity mortality rate of 488 per 100 000.30 Data from a survey conducted in 2012 indicate that only 52% of mothers completed four WHO recommended ANC visits, and 49% delivered at a health facility, of which only 40% reported receiving any PNC services 48 hours after delivery. Only 18% of women reported all services along the continuum of care (ANC attendance, health facility delivery, PNC and newborn assessment) indicating low levels of service utilisation.31 In Kenya, health services are organised in six levels of care (table 1). In 2015 in Siaya, there were 174 health facilities of which 123 were public health facilities, 7 non-governmental, 16 faith-based and 28 private.32 Among these, there is one level 5 facility and each subcounty has one level 4 facility.33 The rest are mainly level 3 and 2 health facilities staffed by nurses or clinical officers, and level 1 community facilities staffed by community health volunteers. Overall, there is low coverage of healthcare, with a doctor to population ratio of 1:62 000, and nurse to population ratio of 1:2500.32 Types of health facilities and health services offered, Kenya Source: Ministry of Health (2017). ANC, antenatal care. All women recruited into the study are given an ANC clinic book as is standard practice and are provided with an enrolment card (henceforth, Afya card) attached to the clinic book. The Afya card is linked to a card reader installed at all participating health facilities. The Afya card is the size of a credit card and stores holder data such as authentication information (study ID, study arm, clinic at which enrolled) and pregnancy-related information (pregnancy stage at enrolment, expected delivery date, parity). The Afya card also allows visits to be tracked by touching the card on the card reader, which informs the payment (or not) of the CCT. In the intervention arm, a CCT payment is made to pregnant women for each facility appointment attended for ANC, facility-based delivery, PNC and childhood immunisation; and referrals related to any of these visits. For each verified health visit made on time, a woman receives a cash transfer of KSh 450 (~US$4.5). In the control arm, women are also provided a nominal gratuity of KSh 50 (~US$0.5) in the form of mobile phone airtime to ensure that women carry the Afya card to ANC visits, health facility deliveries and PNC visits. The gratuity is transferred through the same system used to issue the incentives. The unit of randomisation for the trial is a level 2 or 3 health facility. Level 2 and 3 health facilities are comparable and only those offering the full profile of ANC services were considered for inclusion in the study. The criteria for eligibility were: (1) women attending their first ANC visit; (2) long-term resident of the catchment area served by the health facility, with long-term residence defined as living in the area for at least 6 months; and (3) women with access to a mobile phone that belongs either to themselves or to a member of their household or person whom they trust. The total sample of enrolled women is 5488 women. The primary outcomes of the trial are the proportion of eligible ANC visits made after recruitment; participants delivering at a health facility; eligible health appointments attended for PNC; expected immunisation appointments attended by children. We will test for differences between intervention and control arms in the primary outcomes using logistic regression for binary outcomes, and ordinal logistic regression for ordinal outcomes, adjusting for clustering using random effect models. Further details on the list of secondary outcomes and power calculations can be found in the trial protocol.23 Cost and cost-effectiveness analyses will be conducted from the provider perspective, which takes into account the costs incurred by the provider in the provision of the health programme or intervention.34 An overview of cost data is presented in table 2. Description of costs, data sources and sample sizes ANC, antenatal care; CCT, conditional cash transfer; PNC, postnatal care. Provider costs are incurred by the institutions implementing the Afya intervention, namely Safe Water and AIDS Project, Kenya; the Stockholm Environment Institute, Sweden; and University College London, UK. These costs (programme costs, henceforth) data will be sourced from the financial project accounts of these institutions. These programme costs include the costs associated with starting up and implementing the intervention, which include but are not limited to the cost of CCT payments in the intervention arm; gratuity payments in the control arm; setting up, implementing, and maintaining the Afya card payment system; community and health facility sensitisation. A step-down costing methodology will be used whereby costs from programme accounts are entered into a customised tool created in MS Excel.35 The programme cost data are entered annually into the tool, which is adapted each year to reflect the changing cost structure of the trial at different phases of activity. Financial costs will be converted to economic costs, that is, any donated goods or volunteer time that do not appear in the programme accounting data will be added to the cost sheets and assigned a current market value.34 36 Key informant interviews with programme leads will assist in identifying donated or subsidised items and in allocating joint costs between programme components. The allocation of joint staff costs is informed by monthly staff time sheets. Summary Excel worksheets present the costs by programme component and a single summary worksheet also summarises the total cost data, allows effect data to be entered and calculates the cost-effectiveness results. Research costs will not be included in the cost-effectiveness analyses. However, start-up costs will be reported and differentiated from implementation costs. Provider costs are also incurred by the Ministry of Health in Kenya who provide ANC, facility-based delivery, PNC, child immunisation and related referral visits (health service costs). In the intervention arm, the CCT payment is likely to increase the demand for the ANC, facility-based delivery, PNC and immunisation services. We will estimate any change in the demand for these services resulting from the intervention, and the concomitant value of any additional care provided. The routine monitoring of service statistics at the health facilities will capture the number of visits and any differences in visits between intervention and control arms will be attributed to the intervention. Primary data on the average unit cost of care will be collected from 48 facilities in the study area. A simple cost-capture form will be developed for facility data collection. The data from the facilities will be collected though interview with the facility manager, complemented with health facility records and accounts. Data from the cost-capture form will be used to complement existing data from centre reports, patients’ records and published national and state reports relating to ANC, facility-based delivery, PNC, child immunisation and related referral visits. The economic evaluation of the Afya intervention will involve both cost-effectiveness analysis and cost-consequence analysis. The cost-effectiveness analysis will be conducted as a within-trial analysis using the trial results. Results will be presented in terms of incremental cost-effectiveness ratios (ICERs), calculated as the arithmetic mean difference in cost between the intervention and control arms, divided by the arithmetic mean difference in effect. ICERs will be calculated for statistically significant primary outcomes as well as selected secondary outcomes, along with estimates of total cost at scale. Sensitivity analyses will be conducted to assess the impact on the cost-effectiveness, of changes in parameters with the greatest uncertainty, or with the greatest impact on the total costs. Cost-effectiveness acceptability curves will be generated to further describe uncertainty around the cost estimates.37 The results will also be presented as a cost-consequence analysis. All relevant costs and outcomes of the interventions will be listed in a tabular format, without aggregating into ratios. This allows policymakers to compare the incremental costs with the incremental consequences of different interventions. All statistically significant primary and secondary trial outcomes will be reported. Cost-consequence analysis is recommended for complex public health interventions, such as Afya, that have multiple health and non-health effects, which may be difficult to measure in a common unit.34 38 Costs will be presented in current prices in Kenyan Shillings and International Dollars (INT$). All costs will be adjusted for inflation using the consumer price index for Kenya and will be converted to 2020 INT$ using the 2020 purchasing power parity conversion factor for Kenya. Costs and outcomes will be converted to present values using an annual discount rate of 3% in the base-case, and annual rates of 0% and 6% in sensitivity analyses. The equity impact of the intervention will be analysed within the economic evaluation to investigate whether the gains from such an intervention are equitably shared among the target population, that is, to investigate the extent to which different socioeconomic groups benefit from the intervention. The premise that underlies this component of the economic evaluation is that the intervention should disproportionately benefit the poorest, who also tend to have the highest need for health services.39 The primary and secondary trial outcomes will be decomposed according to the socioeconomic status of participants. A multidimensional poverty index (MDPI) will be used to measure households’ socioeconomic status. The MDPI allows us a more nuanced understanding of socioeconomic status. It takes monetary and non-monetary dimensions of deprivation into account and enables differentiation between population groups who may all be relatively poor in monetary dimensions such as income or asset ownership.40 Thus, the consideration of other non-monetary attributes such as housing, literacy and so on in addition to income or asset ownership allows us to to distinguish between households that are homogenously asset or cash poor in this study setting.41 Data at the household level on indicators in three dimensions of deprivation—health, education and living standards—will be collected during the enrolment survey from the trial participants. If a household is deprived if in a third or more of indicators, they are identified as ‘MDPI poor’. The extent of household poverty is measured by the percentage of deprivations experienced, which also provides indications of relative poverty in this study setting.42 The affordability of the intervention will be explored using an analysis of fiscal space for programme delivery, a generalised fiscal space assessment method43 44 and probabilistic analyses to determine a set of cost-effectiveness thresholds.37 45 These analyses will also enable the exploration of a multicriteria decision analysis framework for resource allocation to this and other similar interventions to improve maternal and child health.46 A wide range of affordability measures have been selected such as the total cost as a proportion of national gross domestic product (GDP), proportion of public health expenditure and proportion of public health expenditure on maternal and child health, in part, due to the paucity of evidence on the cost-effectiveness of comparable interventions.

The Afya trial in rural Kenya is implementing conditional cash transfers (CCT) as a recommendation to improve access to maternal health. The trial aims to assess the impact, cost-effectiveness, and scalability of CCTs in promoting increased utilization of antenatal care, facility delivery, and postnatal care services.

Under the intervention arm, pregnant women receive a cash transfer of KSh 450 (~US$4.5) for each verified health visit made on time, including ANC, facility-based delivery, PNC, and childhood immunization visits. In the control arm, women receive a nominal gratuity of KSh 50 (~US$0.5) in the form of mobile phone airtime to encourage carrying the Afya card to health facility visits.

The trial will measure the proportion of eligible ANC visits, facility-based deliveries, PNC appointments, and immunization appointments attended by children in both the intervention and control arms. The cost-effectiveness analysis will be conducted from the provider perspective, considering program costs incurred by the implementing institutions and health service costs incurred by the Ministry of Health in Kenya.

The economic evaluation will also assess the equity impact of the intervention by analyzing whether the benefits are equitably shared among different socioeconomic groups. The results of the evaluation will be disseminated in a peer-reviewed journal and presented at an international conference.

The publication of the trial’s economic evaluation can be found in BMJ Open, Volume 9, No. 11, Year 2019.
AI Innovations Description
The recommendation to improve access to maternal health is the implementation of conditional cash transfers (CCT) to incentivize pregnant women to utilize antenatal care, facility delivery, and postnatal care services. The Afya trial in rural Kenya aims to assess the impact, cost-effectiveness, and scalability of CCTs in promoting increased utilization of these services. Under the intervention arm, pregnant women receive a cash transfer for each facility appointment attended, including ANC, facility-based delivery, PNC, and childhood immunization visits. The cash transfer amount is KSh 450 (~US$4.5) for each verified health visit made on time. In the control arm, women receive a nominal gratuity of KSh 50 (~US$0.5) in the form of mobile phone airtime to encourage carrying the Afya card to health facility visits. The trial will measure the proportion of eligible ANC visits, facility-based deliveries, PNC appointments, and immunization appointments attended by children in both the intervention and control arms. The cost-effectiveness analysis will be conducted from the provider perspective, considering program costs incurred by the implementing institutions and health service costs incurred by the Ministry of Health in Kenya. The economic evaluation will also assess the equity impact of the intervention by analyzing whether the benefits are equitably shared among different socioeconomic groups. The results of the evaluation will be disseminated in a peer-reviewed journal and presented at an international conference.
AI Innovations Methodology
The methodology described in the abstract aims to simulate the impact of conditional cash transfers (CCTs) on improving access to maternal health. Here is a summary of the methodology:

1. Study Setting: The Afya trial is being conducted in Siaya County, western Kenya, which has poor indicators of maternal and child health.

2. Study Design: The trial randomly assigns eligible women to either the intervention arm or the control arm. Women in the intervention arm receive cash transfers for attending antenatal care (ANC), facility-based delivery, postnatal care (PNC), and childhood immunization visits. Women in the control arm receive a nominal gratuity in the form of mobile phone airtime.

3. Data Collection: The trial collects data on the proportion of eligible ANC visits, facility-based deliveries, PNC appointments, and immunization appointments attended by children in both the intervention and control arms.

4. Cost-effectiveness Analysis: The cost-effectiveness analysis is conducted from the provider perspective, considering program costs incurred by the implementing institutions and health service costs incurred by the Ministry of Health in Kenya. Provider costs include the costs associated with implementing the intervention, such as CCT payments and setting up the Afya card payment system. Health service costs are estimated based on changes in demand for ANC, facility-based delivery, PNC, and immunization services resulting from the intervention.

5. Equity Analysis: The economic evaluation also includes an analysis of the equity impact of the intervention. The socioeconomic status of participants is measured using a multidimensional poverty index, and the outcomes are analyzed to determine if the intervention benefits different socioeconomic groups equitably.

6. Dissemination: The results of the evaluation will be disseminated in a peer-reviewed journal and presented at an international conference.

Overall, the methodology involves collecting data on service utilization and costs, conducting cost-effectiveness and equity analyses, and disseminating the findings to inform policy and practice in improving access to maternal health.

Share this:
Facebook
Twitter
LinkedIn
WhatsApp
Email