Development of a quality use of medicines coding system to rate clinical pharmacists' medication review recommendations

Objective: To develop a 'quality use of medicines' coding system for the assessment of pharmacists' medication reviews and to apply it to an appropriate cohort. Method: A 'quality use of medicines' coding system was developed based on findings in the literature. These codes were then applied to 216 (111 intervention, 105 control) veterans' medication profiles by an independent clinical pharmacist who was supported by a clinical pharmacologist with the aim to assess the appropriateness of pharmacy interventions. The profiles were provided for veterans participating in a randomised, controlled trial in private hospitals evaluating the effect of medication review and discharge counselling. The reliability of the coding was tested by two independent clinical pharmacists in a random sample of 23 veterans from the study population. Main outcome measure: Interrater reliability was assessed by applying Cohen's kappa score on aggregated codes. Results: The coding system based on the literature consisted of 19 codes. The results from the three clinical pharmacists suggested that the original coding system had two major problems: (a) a lack of discrimination for certain recommendations e. g. adverse drug reactions, toxicity and mortality may be seen as variations in degree of a single effect and (b) certain codes e. g. essential therapy were in low prevalence. The interrater reliability for an aggregation of all codes into positive, negative and clinically non-significant codes ranged from 0.49-0.58 (good to fair). The interrater reliability increased to 0.72-0.79 (excellent) when all negative codes were excluded. Analysis of the sample of 216 profiles showed that the most prevalent recommendations from the clinical pharmacists were a positive impact in reducing adverse responses (31.9%), an improvement in good clinical pharmacy practice (25.5%) and a positive impact in reducing drug toxicity (11.1%). Most medications were assigned the clinically non-significant code (96.6%). In fact, the interventions led to a statistically significant difference in pharmacist recommendations in the categories; adverse response, toxicity and good clinical pharmacy practice measured by the quality use of medicine coding system. Conclusion: It was possible to use the quality use of medicine coding system to rate the quality and potential health impact of pharmacists' medication reviews, and the system did pick up differences between intervention and control patients. The interrater reliability for the summarised coding system was fair, but a larger sample of medication regimens is needed to assess the non-summarised quality use of medicines coding system.

API documentation with executable examples

The rise of component-based software development has created an urgent need for effective application program interface (API) documentation. Experience has shown that it is hard to create precise and readable documentation. Prose documentation can provide a good overview but lacks precision. Formal methods offer precision but the resulting documentation is expensive to develop. Worse, few developers have the skill or inclination to read formal documentation. We present a pragmatic solution to the problem of API documentation. We augment the prose documentation with executable test cases, including expected outputs, and use the prose plus the test cases as the documentation. With appropriate tool support, the test cases are easy to develop and read. Such test cases constitute a completely formal, albeit partial, specification of input/output behavior. Equally important, consistency between code and documentation is demonstrated by running the test cases. This approach provides an attractive bridge between formal and informal documentation. We also present a tool that supports compact and readable test cases; and generation of test drivers and documentation, and illustrate the approach with detailed case studies. (C) 2002 Elsevier Science Inc. All rights reserved.

Effects of physical activity on emotional well-being among older Australian women - Cross-sectional and longitudinal analyses

Objective: To explore relationships between physical activity and mental health cross-sectionally and longitudinally in a large cohort of older Australian women. Method: Women in their 70s participating in the Australian Longitudinal Study on Women's Health responded in 1996 (aged 70-75) and in 1999 (aged 73-78). Cross-sectional data were analyzed for 10,063 women and longitudinal data for 6472. Self-reports were used to categorize women into four categories of physical activity at each time point as well as to define four physical activity transition categories across the 3-year period. Outcome variables for the cross-sectional analyses were the mental health component score (MCS) and mental health subscales of the Medical Outcomes Study Short Form (SF-36). The longitudinal analyses focused on changes in these variables. Confounders included the physical health component scale (PCS) of the SF-36, marital status, body mass index (BMI) and life events. Adjustment for baseline scores was included for the longitudinal analyses. Results: Cross-sectionally, higher levels of physical activity were associated with higher scores on all dependent variables, both with and without adjustment for confounders. Longitudinally, the effects were weaker, but women who had made a transition from some physical activity to none generally showed more negative changes in emotional well-being than those who had always been sedentary, while those who maintained or adopted physical activity had better outcomes. Conclusion: Physical activity is associated with emotional well-being among a population cohort of older women both cross-sectionally and longitudinally, supporting the need for the promotion of appropriate physical activity in this age group. (C) 2003 Elsevier Science Inc. All rights reserved.