Skip navigation

COXIB, NSAID, paracetamol and gastric protection

 

Clinical bottom line

This study is complicated, but confirms lower levels of co-prescription of GPA with COXIBs than NSAIDs, despite patients being at somewhat higher risk.


Reference

E Rahme et al. Use of NSAIDs, Cox-2 inhibitors, and acetaminophen and associated coprescriptions of gastroprotective agents in an elderly population. Arthritis & Rheumatism 2002 47: 595-602.

Study

In Quebec all people aged 65 and older by the province, and information on prescriptions is held in a database. A 25% random sample of all who had a prescription for COXIB, NSAID or paracetamol over eight months in 2000 was obtained, with medical and demographic records. The definition was according the first prescription filled in the period. Information on prescriptions and medical history in the previous year was obtained. The main outcome was prescription of gastroprotective medication.

The main analysis related to the use of gastroprotective agents in all the groups whether continuing or changing medication, or starting it. There was also a secondary analysis for patients not on continuing medication, but who first started therapy, and who were without prior gastrointestinal medical history or cancer, and with no prescription for any COXIB, NSAID or paracetamol in the previous year.

Results

There were 42,000 people prescribed a COXIB, 20,000 prescribed an NSAID and 8,000 prescribed paracetamol. About 60% were women, 80-90% had at least one GI risk factor. The groups were by no means identical, though. For instance, 40% of those prescribed paracetamol had had a hospital admission in the previous year, compared with 28% with COXIBs and 22% with NSAIDs. There were differences also in the medical conditions suffered and other drugs taken.

In all patients, prescribing of GPAs was 8.7% in those prescribed a COXIB, compared with 102% in those prescribed NSAID. For those with prior GI events the percentages were 19% and 27% respectively.

Predictors of prescription for gastroprotective medication in the COXIB cohort rather than the NSAID cohort were several. It was much more likely in patients with a prior history of gastrointestinal events (odd ratio 15). Other criteria increasing the likelihood of a GPA prescription, but slightly, were age over 85 years, more than three co-medications, ischaemic heart disease and prior use of paracetamol or COXIBs. Factors negatively associated with gastroprotective medication in the COXIB cohort rather than the NSAID cohort were more than 14 physician claims, COXIB versus NSAID prescription, and COXIB versus paracetamol prescription.

New users

Results for new users are shown in Table 1. Patients prescribed paracetamol tended to be more often older than 84 years and were more likely to have had a hospital admission in the previous year, but there was little difference between people prescribed COXIB or NSAID. Co-prescription of GPAs for those prescribed COXIBs was less than half that in those prescribed NSAID (Figure 1).

Table 1: Patients newly prescribed

 

Paracetamol

COXIB

NSAID

Number

3537

9084

1824

Age over 84 (%)

9.7

4.2

4.0

Hospital admission in previous year (%)

36

14

14

Steroids (%)

14

12

10

GPA (%)

2.4

2.1

5.2

Figure 1: GPA with new prescriptions


Comment

This study is complicated, but confirms lower levels of co-prescription of GPA with COXIBs than NSAIDs, despite patients being at somewhat higher risk.