Thimothy Oke

Quality of drug prescribing in elderly people in nursing homes and special care units for dementia: a cross-sectional computerized pharmacy register analysis.

AbstractBackground: Drug prescribing to the elderly is extensive and often inappropriate. Furthermore, the number of drugs used is the most important risk factor for adverse drug reactions. Despite this, drug prescribing in the elderly in Sweden is high and increasing. In 2003 the Swedish National Board of Health and Welfare launched a set of indicators to evaluate the quality of drug therapy in the elderly. Use of this tool in combination with the Swedish computerized national register covering all persons receiving multi-dose drug dispensing (drugs dispensed in one dose unit bag for each dose occasion) would enable detection of inappropriate drug prescribing and could help reduce the risk of drug-related problems among the elderly. Objectives: To assess the extent and quality of drug prescribing in younger and older elderly residents receiving multi-dose drug dispensing in ordinary nursing homes (NHs) and special care units for dementia (NHDs), and to evaluate the relationship between the quality of prescribing and the number of prescribers per resident, in a Swedish county. Methods: The computerized national pharmacy drug register provided the database and a cross-sectional design was used. Selected drug-specific quality indicators proposed by the Swedish National Board of Health and Welfare in 2003 were used to assess the quality of drug prescribing. Results: This study included 3705 residents. Their mean age was 85 years and 72% were women. The mean number of prescribed drugs was 10.3 per resident. The proportion of residents with prescriptions for psychotropic drugs was 80% in NHs and 85% in NHDs. The prevalence of each drug-specific quality indicator was as follows: long-acting benzodiazepines 16.4% (NHs) versus 11.7% (NHDs), anticholinergic drugs 20.7% versus 18.5%, drug duplication 14.6% versus 13.6%, three or more psychotropic drugs 25.6% versus 35.3%, class C interactions (drug combinations that may require dose adjustment) 41.9% versus 38.7% and class D interactions (drug combinations that should be avoided) 8.1% versus 5.6%. Younger elderly residents (age 65–79 years) had a lower quality of drug prescribing. An increasing number of prescribers per resident was associated with a lower quality of drug therapy. Conclusions: We found a lower quality of drug prescribing, e.g. anticholinergic drugs prescribed to approximately 20% of residents of NHs and NHDs, and a higher rate of psychotropic drug use (≥80%) compared with previous studies in NHs. Our results also demonstrated a negative correlation between quality of prescribing and number of prescribers per resident.

Influencing Prescribing for Urinary Tract Infection and Asthma in Primary Care in Sweden: A Randomized Controlled Trial of an Interactive Educational Intervention

Our aim was to evaluate effects on prescribing for urinary tract infection (UTI) and asthma, of an education with messages based on national guidelines, aimed at improving prescribing in primary care in Sweden. The study is part of the European Drug Education Project. A randomized controlled trial, with groups of general practitioners (GPs) allocated to education on UTI (18 groups, 104 GPs) or asthma (18 groups, 100 GPs), the two parallel intervention arms being controls for each other. Feedback was provided on the GPs judgments of simulated cases and prescribing. Prescribing indicators were developed and measured before and after the intervention. Analysis was performed by multi-level technique. Prescribing of first choice UTI drugs increased in the intervention arm from 52% to 70% and remained constant in the control arm (P < 0.001). The proportion of patients receiving an inhaled corticosteroid increased insignificantly in both study arms. The educational model can be used to improve prescribing. Further studies are needed to define when the model is effective.

An Educational Experiment in Primary Care in Sweden

Objectives: An educational experiment was conducted in Swedish primary care to evaluate the effects of noncommercial treatment information based on national guidelines on the management of hypercholesterolemia.Methods: Community health centres (n=134), with 570 doctors, were randomly allocated to an intervention or a control group. The information was conveyed by community pharmacists and discussed during four group sessions. Clinical performance was studied after the intervention through a retrospective review of 2883 medical records for patients aged 30-65 years with hypercholesterolemia (≥6.5 mmol/1) at each of 110 consenting health centres. Diagnostic investigations, and non-pharmacological and drug treatment actions were used as outcome measures.Results: For patients with more than one other risk factor for coronary heart disease (CHD) and severe hypercholesterolemia (>7.8 mmol/1), the prescription rate of a lipid-lowering drug was 17% higher in the intervention group compared with the control group ...

Small-sample properties of some tests for unit root with data-based choice of the degree of augmentation

Abstract In the augmented Dickey–Fuller (ADF) regression one usually decides on the level of the “augmentation” prior to the performing of unit root test. This is a purely data-dependent method that uses either some information criteria or some sequential test of significance on parameter estimates. Contrary to earlier beliefs, our analyses reveal that the presence and/or absence of a drift and a time trend in the data generating process has a remarkable effect on the behaviour of the subsequent tests for unit root.

Some results on specification search and pre-testing in an ARMA(1,1) process.

In this study we consider simple autoregressive moving‐average (ARMA) models of order at most 1. Pre‐testing, on the moving‐average coefficient θ, is used to choose between an ARMA(1,1) and an AR(1) in a Monte Carlo design. We find that the pre‐test estimator is not always dominated by the others, and that the bias and the mean square error of the estimate of the autoregressive coefficient φ very often depend on the sign of the autoregressive and moving‐average parameters of the ARMA(1,1) model in the data‐generating process. Further, we note that the degrees of size and power distortion of the t test on φ, after pre‐testing for θ, are generally associated with model misspecification.