Table 1

Sequence symmetry results of the 1000 simulations for different scenarios of medicine utilisation trends and effect sizes, for a population size of 1 million and 20% population prevalence of use of each drug
Crude sequence ratio Adjusted sequence ratio
Medicine utilisation scenario Pairs (N) Null SR Crude SR (95% CI) Cover-age (%) Power* (%) Relative bias (%) Adjusted SR (95% CI) Cover-age (%) Power* (%) Relative bias (%)
Expected ASR = 1.0
No trend 3597 1.00 1.00 (0.94-1.07) 95.5 2.2 0.38 1.00 (0.94-1.07) 95.6 2.1 0.38
Gradual 656 0.85 0.86 (0.73-1.00) 47.6 0.0 −13.85 1.00 (0.85-1.17) 94.4 2.1 1.02
Constant 1604 0.84 0.84 (0.76-0.93) 7.3 0.0 −15.34 1.00 (0.90-1.11) 95.2 1.9 0.46
Rapid 1811 0.88 0.88 (0.80-0.97) 22.4 0.0 −11.71 1.00 (0.91-1.10) 96.1 2.2 0.46
Expected ASR = 1.2
No trend 3956 1.00 1.20 (1.13-1.28) 94.2 100.0 0.35 1.20 (1.13-1.28) 94.7 100.0 0.36
Gradual 717 0.85 1.03 (0.88-1.19) 42.8 5.1.0 −13.51 1.21 (1.03-1.40) 94.6 66.4 1.43
Constant 1754 0.84 1.01 (0.92-1.11) 4.8 2.9 −15.52 1.20 (1.09-1.32) 94.8 96.0 0.23
Rapid 1983 0.88 1.05 (0.96-1.15) 18.9 19.8 −11.85 1.20 (1.09-1.31) 94.2 96.8 0.31
Expected ASR = 1.5
No trend 4498 1.00 1.50 (1.41-1.59) 95.2 100.0 0.19 1.50 (1.41-1.59) 95.6 100.0 0.18
Gradual 806 0.85 1.29 (1.11-1.48) 41.0 92.2 −13.82 1.51 (1.30-1.73) 95.6 100.0 1.07
Constant 1973 0.84 1.27 (1.15-1.39) 3.9 100.0 −15.29 1.50 (1.37-1.65) 96.0 100.0 0.50
Rapid 2238 0.88 1.32 (1.21-1.44) 16.6 100.0 −11.67 1.50 (1.38-1.64) 95.6 100.0 0.50
Expected ASR = 2.0
No trend 5395 1.00 2.00 (1.90-2.12) 94.2 100.0 0.42 2.00 (1.90-2.12) 94.7 100.0 0.42
Gradual 957 0.85 1.72 (1.49-1.96) 36.7 100.0 −13.92 2.01 (1.74-2.29) 95.2 100.0 0.95
Constant 2338 0.84 1.69 (1.55-1.84) 2.4 100.0 −15.29 2.00 (1.83-2.19) 95.7 100.0 0.50
Rapid 2660 0.88 1.76 (1.62-1.91) 11.2 100.0 −11.69 2.00 (1.85-2.17) 96.1 100.0 0.48
Expected ASR = 3.0
No trend 7197 1.00 3.00 (2.85-3.17) 95.1 100.0 0.26 3.00 (2.85-3.17) 95.1 100.0 0.27
Gradual 1260 0.85 2.57 (2.26-2.90) 29.2 100.0 −13.92 3.01 (2.65-3.41) 95.8 100.0 0.98
Constant 3074 0.84 2.54 (2.35-2.75) 0.9 100.0 −15.00 3.01 (2.79-3.26) 95.4 100.0 0.82
Rapid 3510 0.88 2.63 (2.45-2.84) 8.6 100.0 −11.72 3.00 (2.79-3.24) 93.5 100.0 0.44
Expected ASR = 0.6
No trend 4799 1.00 0.60 (0.56-0.64) 94.0 100.0 −0.03 0.60 (0.56-0.64) 94.0 100.0 −0.03
Gradual 893 0.85 0.51 (0.44-0.58) 32.6 100.0 −14.61 0.60 (0.52-0.68) 95.7 100.0 0.13
Constant 2187 0.84 0.51 (0.46-0.55) 23 100.0 −15.54 0.60 (0.55-0.66) 96.3 100.0 0.21
Rapid 2456 0.88 0.53 (0.48-0.57) 12.1 100.0 −11.90 0.60 (0.55-0.65) 96.3 100.0 0.24
Expected ASR =0.80
No trend 4048 1.00 0.80 (0.75-0.85) 95.7 100.0 0.11 0.80 (0.75-0.85) 95.7 100.0 0.10
Gradual 745 0.85 0.68 (0.58-0.79) 39.4 100.0 −14.30 0.80 (0.68-0.92) 95.5 86.3 0.51
Constant 1823 0.84 0.68 (0.61-0.74) 4.8 100.0 −15.49 0.80 (0.73-0.88) 95.0 99.9 0.27
Rapid 2054 0.88 0.70 (0.64-0.77) 15.8 100.0 −12.17 0.80 (0.73-0.87) 95.6 99.9 −0.07

*Note: When the expected Sequence ratio =1, Power is the type I error.

Pratt et al.

Pratt et al. BMC Medical Research Methodology 2014 14:66   doi:10.1186/1471-2288-14-66

Open Data