Preeclampsia mortality in Texas: a capture-recapture analysis

Public health professionals and clinicians rely on hospital discharge data to explore the epidemiology of preeclampsia. Our objective was to compare the concordance in preeclampsia inpatient mortality between hospital inpatient discharge data and death certificate data in Texas during a 2-year period using the capture-recapture technique.

The statewide Texas hospital inpatient discharge dataset and a statewide death certificate dataset were searched for fatal episodes of preeclampsia or eclampsia. Women who were Texas residents and expired in a hospital in 2006 or 2007 were identified using ICD-9-CM and ICD-10 codes. Using the capture-recapture sampling technique, we estimated the number of inpatient preeclamptic deaths that were missed by both databases, the death certificate database and the hospital inpatient database.

The hospital inpatient dataset captured 17 fatal episodes of preeclampsia while the inpatient death certificate database captured 9. Six patients were found in both datasets. Application of the capture-recapture method revealed that the total number of statewide inpatient preeclamptic deaths was 26 during this period.

This pilot study found that 2 prominent sources of statewide health data underestimated the frequency of inpatient deaths in Texas due to preeclampsia.