Measuring clinical performance and outcomes from diabetes information systems: an observational study

Aims/hypothesis. To examine changes in diabetes care provision after the in troduction of a district diabetes information system. Methods. All patients with diabetes registered on the system between 1993 a nd 1998 (n = 6544) were included in the analysis. Drop-out cohort analysis was used to handle population changes, logistic regression models with gene ral estimating equations were used to examine changes in clinical performan ce over time. Results. After the introduction of the system, care processes improved appr eciably, in both primary and secondary care. The proportion of patients rec eiving a preventative care review within the calendar year rose from 56 % i n 1993 to 67 % in 1998. The proportion of these in whom each process was co mpleted improved in all categories from 1993 to 1998: blood pressure 96 % t o 98 %; glycaemic check 67 % to 93 %; lipid check 31 % to 68 %; renal check 46 % to 87 %; fundoscopy 79 % to 92 %; foot screen 87 % to 87 %. Similarly there was an increase in the proportion of patients achieving intermediate outcome treatment targets (HbA(1c) less than or equal to 9.0 % from 29 % t o 43 %; cholesterol less than or equal to 5.5 mmol/l 5 % to 19 %; blood pre ssure less than or equal to 160/90 37 % to 46 %). Conclusion/interpretation. Our results suggest appreciable improvements in diabetes care between 1993 and 1998. These changes apply to an entire popul ation of patients across primary and shared care. We believe that these imp rovements could, in part, be attributable to the way in which The district diabetes information system has facilitated the structured cascade of diabe tes care.