Skip to main content

Table 3 Demographic, health, and health care use characteristics of COX-2 switchers and COX-2 stoppers in the first 100 days after discrediting of rofecoxib

From: An observational study of the discrediting of COX-2 NSAIDs in Australia: Vioxx or class effect?

Characteristic COX-2 Switchersa(N = 189)
%
COX-2 Stoppersa
(N = 81)
%
χ2, df, p-value
Continuous Rofecoxib User 47 14 61.90, 1, < 0.0001
Urban Area 57 45 7.54, 1, 0.01
Able to Manage on Income 80 78 0.47, 1, 0.49
No Formal Education 72 73 0.03, 1, 0.87
Marital Married/De facto 40 36  
  Sep/Divorced 4 6 2.29, 1, 0.32
  Widowed/Single 56 58  
BMI Underweight 1 3  
  Normal 44 48 0.02b
  O/weight 55 49  
Alcohol Non-drinker 63 68  
  Rarely/Low drinker 32 31 0.05b
  Risky/High Risk 5 1  
Smoker 4 5 0.77, 1, 0.38
Excellent/Very Good/Good General Health 63 59 0.60, 1, 0.44
Consulted Family Doctor ≥5 Visits 70 79 5.15, 1, 0.02
Hospital Doctor Visit in 12 months 19 26 3.36, 1, 0.07
Specialist Doctor Visit in 12 months 55 62 2.56, 1, 0.11
Conditions:    
Ever Reported Arthritis 86 77 9.03, 1, 0.003
Hypertension 61 54 2.93, 1, 0.09
Depression 6 8 1.29, 1, 0.26
Anxiety/nervous 7 7 0.15, 1, 0.70
Asthma 9 11 0.24, 1, 0.62
Diabetes 14 11 0.93, 1, 0.33
Bronchitis/Emphysema 5 9 3.3, 1, 0.07
Heart Disease 21 27 2.16, 1, 0.14
Cancer 4 6 1.97, 1, 0.16
  1. a In the first 100 days after rofecoxib discrediting
  2. b Fisher's exact test used due to small cell sizes, p-value only reported