P. Primatesta et al., CHANGING PATTERNS IN THE EPIDEMIOLOGY AND HOSPITAL-CARE OF PEPTIC-ULCER, International journal of epidemiology, 23(6), 1994, pp. 1206-1217
Background. Our aim was to study trends in hospital admission rates fo
r peptic ulcer in a geographically defined population, and to distingu
ish the effects of period, age and birth cohort on the rates. Methods.
Analysis of linked, routinely collected abstracts of hospital inpatie
nt care held by the Oxford record linkage study for the period 1970-19
86. Age- and sex-specific and age-standardized hospitalization, readmi
ssion and operation rates were calculated for patients with peptic ulc
er. Age, cohort and period effects were examined using log-linear mode
ls. Results. Records for a total of 5462 people with gastric ulcer and
10 186 with duodenal ulcer were identified. Overall, the age-standard
ized admission rates for both gastric and duodenal ulcer declined over
the study period. The decrease was confined to people <65 years of ag
e. Among elderly patients admission rates for peptic ulcer increased o
ver time, more so in females than in males. Admission rates were highe
r in the elderly than in young people for both gastric and duodenal ul
cer. The apparent age effect was, in fact, mainly attributable to a bi
rth cohort effect: age-specific admission Fates were lower in people b
orn after 1925 than in people born at the beginning of the century. Th
is was more marked for males than females. There was a considerable de
cline in major operations undertaken on peptic ulcer; admission rates
for endoscopy increased; and readmission rates did not show significan
t changes. Conclusions. The overall decline found for hospital care of
peptic ulcer during the study period is consistent with that found in
England for mortality rates ascribed to peptic ulcer. The cohort effe
ct found in the data for hospitalized morbidity supports that reported
by others for mortality. The cohort effect indicates that the decline
was related more to changes in risk factors in the cohorts born in di
fferent periods than to the introduction of new pharmacological treatm
ents since the 1970s.