Skip navigation

Low dose aspirin and haemorrhage in older people


Clinical bottom line

In the 1990s prescriptions of low dose (75 mg) aspirin in England rose dramatically, along with proton pump inhibitors and some other drugs. Gastrointestinal haemorrhage in older people also increased. Is there a link with aspirin?


J Higham et al. Recent trends in admissions and mortality due to peptic ulcer in England: increasing frequency of haemorrhage among older subjects. Gut 2002 50: 460-464.


This study determined the rate of finished consultant episodes for gastric and peptic ulcer, and for haemorrhage and perforation from 1989/90 to 1998/9. The figures were standardised against the European standard population. Trends in the age groups of 25-44 years, 45-64, 65-74 and 75 years or older. The information excluded day cases.

Information on the number of items dispensed in the community between 1990 and 1999 were supplied by the statistics division of the DoH. From 1991 onwards were from prescription cost analysis.


Hospital admissions for gastric and peptic ulcer did not change over the period for men and women under 65 years, but rose in older age groups for both men and women (by about 10-30% in 65-74s and 30-40% in over 75s.

For duodenal ulcer, finished consultant episodes rose by 25% for men and women aged 65-74 years, and 30-50% for men and women over 75 years or over. The figures for duodenal ulcer haemorrhage are shown in Figure 1. Rates for duodenal ulcer perforation showed no systematic change over the period.

Figure 1: Duodenal ulcer haemorrhage in over 65s

Mortality rates fell over the period 1958 to 1998 for men and women, but were higher in over 65s. Deaths for this cause were 20-30 per 100,000 in 1998.

prescribing over the period showed major changes. For instance, there was a huge increase in prescribing of proton pump inhibitors and selective serotonin reuptake inhibitors, and also in the 75 mg dose of aspirin (Figure 2). NSAID prescribing did not change over the period.

Figure 2: Prescribing of NSAIDs, low dose aspirin and PPI over the 1990s


This ecological study cannot link cause and effect between any particular prescribing and changes in duodenal ulcer haemorrhage. The number of prescriptions, if each was for 30 tablets (a reasonably if conservative assumption) and one 75 mg tablet was taken a day, then by 1999 there would have been 900,000 more person years of exposure. Given that one or two gastrointestinal bleeds per 1000 patients on low dose aspirin, then low dose aspirin could account for 1000 to 2000 episodes of gastrointestinal bleeding, and that approximates the 1000 excess admissions actually observed.