Background: From 1987-92 in Izumo City, Japan, we diagnosed 123 patients as having subarachnoid hemorrhage (SAH) by computed tomography, autopsy, or surgery (proven SAH); the crude incidence rate was 25/100,000/year for all ages. However, to estimate the actual incidence and mortality rates, we should take into account the decedents who died without confirmation by these methods but were presumed to have died of SAH.
Methods: From 1987-92, we reviewed all of 3562 death certificates for the city of Izumo (population 82,679), and calculated the incidence and mortality rates of SAH by combining proven and possible SAH.
Results: We diagnosed 36 patients as having possible SAH on death certificates. When adding these 36 patients to the 123 with proven SAH, the crude and the age-adjusted and sex-adjusted incidence rates for all ages became 32/ 100,000/year and 29/100,000/year, respectively. Of these, 40% (64) died by day 3 (day 0 defined as the day of hemorrhage), 43% (69) within 1 week, and 53% (84) within 1-6 months, respectively.
Conclusions: When including the patients who may have died of SAH, the actual incidence rate of SAH is much higher than that which has been reported to date, and the actual mortality rate is still very high.