The Policy Process, Quality and Cost of Free Maternal Healthcare in Kenya: A Mixed Methods Analysis of Maternity Policy

Background: Reducing both global maternal and neonatal mortalities are essential parts of meeting the Sustainable Development Goals (SDGs) of ending preventable deaths caused by poor access to, and utilisation of maternal and family planning services. Several countries, particularly low/low and middle-income countries such as Kenya, are eliminating the financial barriers to access of quality maternal and infant health care and enhancing the utilisation of skilled birth attendance (SBA) through incentives such as free (non-user fee) delivery (birth) policies to achieve universal health coverage (UHC). This study sought to explore the policy process, implementation, and effects on quality and cost of maternal care of the free maternity policy as implemented in Kenya.

Methods: A convergent parallel mixed methods case study underpinned by pragmatism informed the study design and allowed for an in-depth exploration of the policy, thereby strengthening construct validity. The study was set in one county and three county hospitals (high volume, medium volume, and low volume) in Kenya. Data sources were exit interviews (n=553) with mothers who utilised delivery services under the policy in the three hospitals, focus group discussions (n=10) with mothers who sought postnatal services, in-depth interviews (n=20) with healthcare workers and the county officials, key informant interviews with the policymakers (n=15), document reviews, and demographic health survey data which were collected between November 20, 2018 to September 13, 2019. Data analysis was drawn from three approaches: thematic analysis for the qualitative data, descriptive statistical analysis for the exits interviews, and difference in difference analysis of maternal and neonatal outcomes using Kenya Demographic Health Survey (KDHS) data. Qualitative data was managed through NVivo 12, while quantitative and econometrics analysis were managed by STATA 15.Results: The introduction of the current policy was done to overcome the challenges from the previous free maternity policy and was both a political and a technical initiative. There was adequate consultation, costing, and evaluation of sustainability at the onset of the policy. Analysis of the implementation components showed that there were reimbursement delays; the claims system was fraught with challenges; there was poor communication of the policy plans; reversal of the referral system, and poorly incentivised accreditation and contracting of the providers. The County and HCWs utilised street-level bureaucrat (SLB) tacts to reshape the working practice of the policy, repurpose the policy processes, and provide services amidst challenges in order to meet the ethical and professional concerns. The majority of the mothers had a positive perception about the policy despite still bearing a mean out-of-pocket (OOP) payment of US$3.0 on the free services that were catastrophic at a threshold of 5% of the annual income. Overall, the policy had a significant positive impact on early neonatal mortality and neonatal mortality; but no significant impact on delivery through a caesarean section or other intermediate outcomes (skilled delivery, birth in a public facility (hospital), and low birth weight.Recommendation: Implementation of the current policy as it is, has only partially solved the challenges from the previous free maternity policy. There is a need to promote awareness of the policy to the poor and disadvantaged women in rural areas to help narrow the inequality gap on utilisation and reduce the impoverishment of households. The implementation processes such as claims management, accreditation and contracting, fragmentation of the benefits package should be re-evaluated through multi-stakeholder consultation if the policy is to achieve its objective fully. The policy should be sustainably funded in order to achieve the UHC agenda.