Analyses of stroke morbidity or mortality are usually based on official sta
tistics. A reduction in stroke mortality rates has been shown for many coun
tries. It is not clear, however, whether this is due to declining morbidity
or case fatality (or both). For this purposes population-based register da
ta are required. Using the standardized methodology of the WHO-MONICA Proje
ct, stroke cases were also registered in Germany from 1984 to 1993 (7,435 f
irst-ever and recurrent stroke cases). The data colt lection was almost res
tricted to East Germany. The age-specific stroke rates in males/females sho
wed an increase from 9/11 per 100,000 population in the youngest age group
(25-34) to 1,005/779 cases per year in the oldest group under study (65-74)
. If one tries to classify stroke types, which is not always possible in a
population-based register, the best estimate for men (women) would be: 63(6
2%) thromboembolic stroke, 25(22)% intracerebral hemorrhage,and 12(17)% sub
arachnoid hemorrhage. The 28-day case fatality of the 25-74 year old stroke
patients was found to be about 40%. Neither for stroke attacks nor for cas
e fatality was a convincing time trend over the 10-year period found. The v
ery small changes observed over 10 years time should lead to increased atte
ntion to strokes, particularly primary and secondary prevention, and this n
ot only in East Germany. This applies also for treatment in the acute phase
, because the case fatality before admission in the hospital and during the
first few days is still very high. Population-based studies of the long-te
rm prognosis of stroke patients in Germany are also missing, i.e., includin
g the effectiveness of various forms of treatment and rehabilitation. Syste
matic monitoring of the development in this field is an important part of t
he assessment of the quality and effectiveness of the health care service.