By Parfait Uwaliraye, Andrew Muhire, Emmanuel Ntawuyirusha, Valencia Lyle, Godfrey Ngoboka, James Mwanza, Carla AbouZahr
Every public health leader broods over the same fundamental question: With multiple demands, yet limited resources, where should we invest our resources to maximize our people’s health? How do we prioritize? One strategy is to identify and focus on the illnesses and injuries that cause the largest number of premature deaths. It may seem counterintuitive to focus on the dead when the primary responsibility for a policymaker is for the living, but the best way to understand the health challenges facing the living is to know why people die.
In recent years, Rwanda has invested considerable effort to understand patterns of mortality and tackle the causes. This has helped reduce maternal mortality, from 1,071 deaths per 100,000 live births in 2000 to 210 deaths per 100,000 live births in 2014, as well as improving the mortality rate for children under 5, which has fallen from 152 deaths per 1,000 live births in 2005 to 50 deaths per 1,000 live births in 2014.[i]
Progress in addressing maternal and child mortality is characteristic of a pattern in which mortality shifts towards adulthood, with an increasing share of deaths caused by non- communicable diseases such as cancers, cardiovascular diseases and injuries. New strategies will be needed to track and respond to these developments and answer the critical question about where to invest health sector resources.
In response, the Government of Rwanda has introduced two interventions to improve its understanding of why people die. First, the Government, in collaboration with Bloomberg Philanthropies Data for Health Initiative, and with support from Vital Strategies, has introduced verbal autopsy for deaths that occur outside a health facility. Verbal autopsy trains community members to interview family members and other witnesses to determine and register the underlying cause of death when a person has died without medical care. Rwanda conducted more than 1,500 verbal autopsies in the piloted area of 100 cells out of 2,148 cells under the home-based care practitioners (HBCP) program that started in July 2017.
Second, for the deaths that occur in health facilities (some 40 percent of total deaths in Rwanda), the country has implemented the internationally recognized process of Medical Certification of Cause of Death (MCCOD), an ambitious process that involved training every medical doctor in the country.
In Rwanda, prior to October 2017, hospitals were not using the internationally accepted standard cause-of-death form, medical doctors had not been trained to correctly complete the form, and the quality of cause-of-death data in the country was poor, with many deaths attributed to ill-defined causes which are of limited use for public health decision-making.
Introducing MCCOD into the health system, training physicians to use it appropriately, and building statistical coding capacities among data management officers, are major undertakings, widely assumed to take many years to implement. Undeterred, in October 2017, the Ministry of Health initiated an ambitious program to integrate MCCOD into all hospitals.
As a first step, the Government commissioned a legal review and adjusted the 2016 WHO International Death Certificate to the Rwandan context, while remaining compliant with the International Classification of Diseases system. The Ministry of Health introduced cascade training: Eight clinical directors from referral and provincial hospitals were trained as MCCOD master trainers and took on the responsibility for training 44 clinical directors from district hospitals representing the nation’s public hospitals and 27 physicians representing private health facilities at a MCCOD training of trainers. All public and private health facility trainees were required to train their colleagues upon return to their respective places of work under the support of Ministry of Health. Providing on-the-job training at the physicians’ place of employment was believed to be more effective than providing training in locations outside of their workplaces.
At regular intervals, follow-up sessions were organized to track the quality of cause-of-death certification and coding, both among trainers and trainees. To ensure sustainability and institutionalization of the standards, MCCOD training was introduced into the medical training core curriculum for all trainee doctors as well as into obligatory continuing medical education. The MCCOD e-Learning was introduced to offer flexible in-service refresher trainings and offer continuous professional development points to medical doctors through Rwanda Medical and Dental Council.
Up to the end of 2018, 5,000 deaths in public and private hospitals were medically certified according to international standards. Once the MCCOD forms have been completed by hospitals, the information is shared with National Institute of Statistics of Rwanda and the Ministry of Health to be entered into a national – and district-level health information system. These data are analyzed and will be disseminated to the public each year by way of the Annual Health Statistics Booklet.
Initially, physicians voiced concerns that the MCCOD might take time from their clinical responsibilities. To counter this sentiment, trainers emphasized the public health benefits of correctly recording causes of death. By the end of the first year of implementation, physicians reported feeling empowered as a result of their increased knowledge and skills, and they felt they had gained a deeper capacity to participate in the regular hospital reviews of adverse events, such as maternal and perinatal death audits. An online group was established by the Ministry of Health to encourage physicians to share their experiences and receive answers to any questions that they have regarding MCCOD.
The cascade training approach reached all physicians in public and private hospitals and MCCOD was integrated into the health system within six months. A formal evaluation of the quality of cause-of-death recording will take place in early 2019, but initial evidence indicates that the cascade training approach has been effective. To maintain standards, the Ministry of Health has established a process for routine auditing of death certificates to safeguard data quality and accountability. Mentoring schemes are in place in hospitals, particularly for newly trained physicians, and physicians are encouraged to complete refresher training through an MCCOD online distance learning course.
Rwanda’s experience can inspire other countries to introduce MCCOD into their health systems using a cost-effective, time-saving methodology. Strong leadership and support provided by the Ministry of Health garnered enthusiasm among physicians to immerse themselves in the MCCOD training and transition to the 2016 WHO International Death Certificate. The continuous improvement in the quality of mortality data recorded will strengthen data-driven policymaking and empower public health leaders in Rwanda to implement actions to address the causes of premature mortality.
Counting the Uncounted is a blog series that examines topics related to civil registration and vital statistics. The blog series is a part of the Bloomberg Philanthropies Data for Health Initiative. The views expressed are not necessarily those of Bloomberg Philanthropies. Partners in the initiative include: WHO, the U.S. Centers for Disease Control and Prevention (CDC) and the CDC Foundation, the University of Melbourne, and Johns Hopkins University’s Bloomberg School of Public Health. For more information visit: Bloomberg Philanthropies Data for Health Initiative.
[i] Republic of Rwanda. Rwanda Demographic and Health Survey 2014-15. 2016. The DHS Program, ICF International. Rockville, Maryland, USA.