Improving quality of cause of death reporting in New York City hospitals

Objectives: Mortality statistics are used to determine and evaluate public health efforts. However, imprecise death certificate (DC) information results in unreliable statistics. Validation studies have found heart disease (HD) over-reporting throughout the US including New York City (NYC). We describe a within-hospital intervention that provides sustained improvement in cause of death (COD) reporting.

Methods: NYC Department of Health and Mental Hygiene (DOHMH) intervened from July 2009 to January 2010 at 8 of the 12 hospitals with the highest proportion of HD deaths. These institutions were required to: 1) participate in a conference call, 2) provide a DC workflow protocol, 2) review medical records for 30 DCs with HD documented as the COD, 3) necessitate staff complete an online training, and 4) require staff attend a DOHMH in-service covering COD, legal requirements, and hospital-specific examples. Impact was assessed by measuring changes in the average proportion of HD deaths and number of conditions reported in intervention vs. non-intervention facilities from pre- (January–June 2009) to post-intervention (July–October 2010).

Results: The average proportion of HD deaths reported at intervention hospitals decreased from 0.68 to 0.35 (48.2% reduction; p<0.0001), while non-intervention hospitals were stable (0.26 to 0.24; p=0.071). The average number of conditions per DC increased from 2.3 to 3.3 (42.9%; p<0.0001) in intervention hospitals while non-intervention hospitals remained at 2.9 (p=0.583).

Conclusion: This multi-faceted, institution-level intervention provided sustained improvements in the quality of COD reporting at targeted hospitals. This intervention can help ensure accurate mortality data often used to drive policy, program, and resource allocation decisions.