Skip navigation

Underuse of GPAs with NSAIDs

 

Clinical bottom line

Use of gastroprotective therapy for patients using NSAIDs was grossly underused even in patients with several risk factors.

Reference

W Smalley et al. Underutilization of gastyroprotective measures in patients receiving nonsteroidal antiinflammatory drugs. Arthritis & Rheumatism 2002 46: 2195-2200.

Study

The study cohort was of all persons aged 50 years or older enrolled with Tennessee medicaid for at least one year and who filled an NSAID or COXIB prescription during 1999 and half of 2000. Information on prescriptions, including gastroprotective agents, and medical conditions was available in a database. Information was sought on prolonged use, patients had to have at least one additional prescription in the previous year, or in the 60 days after the identified prescription.

Persons at high risk of NSAID-associated gastrointestinal bleeding were identified as:

Other comorbisities were identified.

Results

Of 319,000 people, 107,000 (34%) received at least one prescription for NSAID or COXIB, and of those 77,000 (24%) were defined as being prolonged users. There were 72,000 users of NSAID and 5,000 users of COXIBs.

The mean age was 63 years, and women predominated (70%). Patients prescribed COXIBs had more non-ulcer hospital admissions in the previous year (29%, compared with 19% with NSAID), were more likely to be using antihypertensives (75%) and other cardiac drugs (50%), bronchodilators (32%), non-NSAID analgesics, and psychotropic drugs (71%). TMore of them had ulcer bleeding in the previous year (7.5% versus 4.3%) and were more likely to have had gastroprotective agents in the preceding year (65% versus 41%).

The relationship between risk factors and use of recommended antiulcer therapy with traditional NSAIDs, and use of COXIBs, is shown in Table 1. The use of recommended gastroprotective therapy, or COXIB, was the exception rather than the rule. In no group, either with a single risk factor or more than one risk factor did the sum of NSAID plus therapy and COXIB use approach 50% (Figure 1).

Table 1: Prescribing practice in cohort

 

Percent on

 

Number

NSAID + recommended antiulcer therapy

COXIB

No risk factor

76,765

10

5
Specific single risk factors

59,486

    
Age 75+

10.788

5

7
Previous ulcer

2,053

24

10
GI bleeding

329

21

6
Anticoagulants

1,322

7

29
Steroids

1,095

14

15
Any single risk factor

15,587

9

9
Two or more risk factors

1,692

11

19

Figure 1: Risk factors and good prescribing

When analysis was by higher doses of NSAIDs, or new users, no better result emerged.

Anti-ulcer drug therapy was prescribed at effective doses in 10% of those on NSAIDs and 27% of those on COXIBs. Proton pump inhibitors were most often used with people prescribed COXIBs and histamine antagonists most often used with those prescribed NSAIDs.

Comment

Confusing, isn't it. This study of course reflects one geographic location at one point in time, but several important things emerge. Firstly that in those people with risk factors for gastrointestinal bleeding with NSAIDs, use of gastroprotective therapy at recommended doses is massively underused. Curiously, better prescribing exists for COXIBs, a safer alternative to NSAID, perhaps reflecting the "sicker" status of those patients at baseline.

Things will change, and perhaps the single important lesson is that if practice isn't audited, you don't know how bad it is.