Skip to main content

Table 3 Results of the segmented linear regression models for the volume of Cefuroxime and its alternative drugs

From: The impacts of Chinese drug volume-based procurement policy on the use of policy-related antibiotic drugs in Shenzhen, 2018–2019: an interrupted time-series analysis

  Coefficient Standard Error t p-value 95 % CI
Lower Upper
Model 1, Cefuroxime
Secular trend, β1 4.12 3.09 1.33 0.199 -2.35 10.59
Change in level, β2 161.16 48.61 3.32 0.004 59.43 262.90
Change in trend, β3 -5.41 7.52 -0.72 0.480 -21.16 10.33
Cold, β4 89.67 32.04 2.80 0.011 22.62 156.72
Constant, β0 149.22 28.04 5.32 0.000 90.54 207.89
Model 2, Alternative drugs
Secular trend, β1 20.52 5.42 3.79 0.001 9.18 31.86
Change in level, β2 273.65 87.66 3.12 0.006 90.17 457.12
Change in trend, β3 -47.57 12.91 -3.69 0.002 -74.59 -20.55
Cold, β4 313.94 65.63 4.78 0.000 176.56 451.31
Constant, β0 634.46 47.66 13.31 0.000 534.70 734.22
Model 3, Total
Secular trend, β1 24.70 7.25 3.41 0.003 9.52 39.88
Change in level, β2 436.31 117.29 3.72 0.001 190.81 681.81
Change in trend, β3 -54.09 17.30 -3.13 0.006 -90.31 -17.88
Cold, β4 385.09 87.21 4.42 0.000 202.55 567.63
Constant, β0 786.20 63.89 12.30 0.000 652.47 919.94
  1. Model 1, F = 19.63, p-value < 0.001, R2 = 0.805, Adjusted R2 = 0.764; Model 2, F = 32.63, p-value < 0.001, R2 = 0.873, Adjusted R2 = 0.846; Model 3, F = 36.12, p-value < 0.001, R2 = 0.884, Adjusted R2 = 0.859