Analysis of the impact of practice characteristics and social deprivation on prescribing quality for asthma

Abstract

Background: Several studies have shown that there is a relationship between demographic and practice characteristics with the pattern of drugs used/prescribed by/to patients. This study was designed to investigate the causal effects of the different soft-economic and practice characteristics including Jarman score (JS) reflecting the potential workload on the services of general practitioners (GPs); Townsend index (TI) as the indicator of material deprivation; fund-holding status (FHS) of the practices; percentage of south Asians (PSA) registered in the practices and the number of patients per GP (PPGP), on the respiratory drugs prescribing. Method: This study is a retrospective analysis of PACT level 3 data of April 1996 to March 1997 for BNF category 3 (chapters 3.1-3.3), prescribed by GPs in Bradford. Simple linear regression and ridge regression analysis models were applied to describe the relationships between the characteristics of practices and of the practice population with the preventive to bronchodilator ratio (P:B), based on Defined Daiy Doses, as the quality indicator of asthma prescribing. Results: PACT data of 62 (64%) consenting practices located in Bradford have been analysed. 36 (58%) practices were fund-holding and 26 (42%) were not. The summary of the distribution of the variables and regression results is shown below. Median Mean (SD) Range P Value Ratio of P:B 0.49 0.51 (0.15) 0.05, 0.85 NA JS 20.23 23.47 (21.22) -12.84, 58.92 0.049 TI 2.87 3.49 (3.71) -3.70, 9.86 0.062 PPGP 1755.9 1934.2 (635.1) 1082, 3750 0.287 FHS NA NA NA 0.329 PSA 6.18 25.11 (34.31) 0.20, 96.43 0.373 The resuts of the regression analysis showed that the JS has a significant inverse relationship with the P:B ratio (P=0.049, r= -0.25, r 2= 0.06). There was also an inverse non-significant association between TI and the P:B ratio (P=0.062, r= -0.24, r 2= 0.06). Analyses of the effects of the four other variables were not significant. Conclusion: In the more deprived areas using the P:B ratio suggests that the respiratory drugs usage pattern may be less appropriate in comparison with more affluent areas. This might be because of the higher workload of the practices in these areas (indicated by higher Jarman scores). To achieve more accurate results, the use of individual prescribing information and the socio-economic situation of asthmatics with links them to outcomes of the treatment is in progress.