Table 2.

Number of drugs and prevalence of potentially inappropriate prescribing by quartiles of continuity of GP care

QuartileNumber of prescriptionsPotentially inappropriate prescribing
Mean (SD)P-valueaIRR (95% CI)bP-valuebMean (SD)P value aIRR (95% CI)bP-valueb
Usual Provider of Care Index
  Lowest quartile9.67 (5.31)ref2.50 (2.28)ref
  Low intermediate quartile9.47 (5.17)0.200.99 (0.96 to 1.01)0.332.47 (2.18)1.001.00 (0.95 to 1.04)0.89
  High intermediate quartile8.98 (4.93)<0.010.97 (0.94 to 0.99)0.012.29 (2.17)<0.010.96 (0.92 to 1.01)0.09
  High quartile8.52 (4.75)<0.010.96 (0.93 to 0.98)<0.012.09 (2.06)<0.010.93 (0.88 to 0.97)<0.01
Bice–Boxerman Continuity of Care Index
  Lowest quartile9.35 (5.23)ref2.40 (2.23)ref
  Low intermediate quartile9.29 (5.08)0.560.98 (0.95 to 1.01)0.112.41 (2.22)0.710.99 (0.94 to 1.04)0.60
  High intermediate quartile9.47 (5.12)0.360.98 (0.96 to 1.01)0.132.44 (2.19)0.250.98 (0.93 to 1.03)0.41
  High quartile8.56 (4.77)<0.010.96 (0.93 to 0.99)<0.012.09 (2.06)<0.010.93 (0.88 to 0.97)<0.01
Sequential Continuity Index
  Lowest quartile9.10 (5.10)ref2.30 (2.20)ref
  Low intermediate quartile9.70 (5.15)<0.010.98 (0.95 to 1.00)0.112.56 (2.22)<0.010.99 (0.95 to 1.04)0.80
  High intermediate quartile9.36 (5.22)0.120.98 (0.96 to 1.01)0.152.41 (2.20)0.040.98 (0.93 to 1.02)0.33
  High quartile8.51 (4.71)<0.010.96 (0.93 to 0.98)<0.012.08 (2.07)<0.010.93 (0.89 to 0.98)<0.01
  • a Two-sided student t-test.

  • b IRRs estimated using negative binomial regression model, stratified by quartile of CGPC and adjusted for age, sex, 14 comorbidities, and frailty status. CGPC = continuity of GP care. CI = confidence interval. IRR = incidence rate ratio. ref = reference. SD = standard deviation.