Page 52 - Demo

Certification of Coroners’ cases by pathologists would improve the completeness of death registration in Jamaica
52
The Vital Statistics Commission (VSC), established in 2003 to improve death registration, commissioned this review to identify priority areas for improvement. Their objectives were to (1) determine the completeness of death registration; (2) identify factors associated with under-reporting and how under- reporting affected overall (crude) and age-specific mortality rates, including maternal and child indicators (MDG 4,5) and; (3) assess the quality of certification and coding of deaths.
Study design
Electronic listings from the Registrar General’s Department (RGD), the police (traffic accidents, violence, suicide) and the Ministry of Health (hospital deaths) yielded a preliminary universe of deaths, which was updated with records from hospitals, the Ministry of Justice (forensic pathologists) and the Coroners courts. A 10% sample was used to evaluate the quality of certification and coding.
What Have We Learned?
• Of 19,286 deaths identified, 76% were registered, a deterioration from the 89% noted in 1998. While under- five year-old death registration improved over the decade, age-specific mortality rates declined in the 5-44 years age groups (Figure 1), over-estimating male life expectancy by six years (one year for females).
• Most (82%) stillbirth certificates lacked cause of death information, and 76% of maternal deaths was not accounted for (Table).
• Deaths certified by attending physicians achieved complete registration (94%) while only 22% of Coroners’ cases (sudden deaths from natural causes, accidents and violence -- Figure 2) were registered.
• One in six (15%) deaths were coded to ill-defined
causes, thereby under-reporting deaths from
prematurity, maternal causes, certain injuries and violence, selected cardiovascular conditions, and HIV/AIDS.
• Few doctors recorded duration of illness (19%) or correctly sequenced the causes of death (16%).
• The leading causes of death were cerebrovascular disease, diabetes mellitus and homicide.
Benefits
In 2015, the RGD added physician’s registration numbers to the medical certificate of cause of death (MCCD) and revised the pregnancy check box to objectively identify certifiers, and improve notation of maternal deaths.
Potential impact & applications
• Data quality
• Train clinicians to correctly complete MCCDs, and
nosologists (coders) to accurately code maternal,
young child and cardiovascular deaths.
• Quality control officers should return poorly completed
MCCDs, without delaying the registration.
• Legislation
• The VSC should be established in law and empowered to enable the required changes.
• Empower pathologists to issue MCCDs on conclusion of the autopsy, but ahead of the medico-legal post mortem report, to enable timely registration of Coroners’ cases.
• Improved monitoring
• MCCDs should be reviewed by medical records officers
to correct sequencing errors and omissions.
• The JCF should notify the Registrar of all Coroners’
cases (not just accidents and violence).
• The VSC should oversee a monitoring and evaluation system with indicators of effectiveness, for example,
- - -
% Coroners’ case files transferred to court within 30 days
% Coroners’ case files transferred from the courts to the RGD within 5 days
% Coroners’ cases (accidents, violence, natural causes) registered within 90 days.
• Public education
• Improve health- seeking behaviour to reduce premature
deaths
• Training in how to complete the registration of all
sudden deaths.
Delays in registering Coroners’ cases were the main source of under-reporting, exacerbated by a 50% increase in fatal injury deaths (from 1792 to 2692) over the decade. However, with 15% of deaths certified to conditions of no public health value and 24% under-reporting, the 2008 data were of medium, not limited quality. The measure of an effective health system is improved population survival. For routine vital data to usefully monitor health trends, it must provide reliable, unbiased indicators.
Acknowledgements
The study was supported by an Inter-American Development Bank grant ATN/OC-11745-JA. Special thanks to demographer, Dr Valerie Nam, for life-expectancy estimates.
Recognising Outstanding Researchers 2016