1. A SIMPLE METHOD FOR ASSESSING IRRATIONAL PRESCRIBING AND PRIORITIZING PRESCRIBING PROBLEMS FOR INTERVENTION: A PILOT STUDY IN UGANDA Hansen EH, Trap B and Anfinnsen M Danish University of Pharmaceutical Sciences, Copenhagen, Denmark & Euro Health Group, Denmark

2. Abstract ID: 262 Problem Statement: Irrational prescribing is common worldwide. Besides having adverse effects on health, it results in unnecessary costs; therefore, the problem is of utmost importance in developing countries, where resources are especially limited. Methods do exist to assess clinical aspects of prescribing appropriateness, but we have identified no generic methods that include costs in the assessment. Objective: To develop and field-test a simple method for assessment of prescribing appropriateness, including both clinical and cost aspects. Design: Analytical, register-based study. Setting and Study Population: Data were collected at the Jinja Regional and Referral Hospital, Uganda. Co-trimoxazole, chloroquine, and erythromycin were selected as tracer drugs, based on prevalence of use, risk of resistance, and cost of treatment. Stratified random samples of prescriptions were drawn from outpatient department log books. A sample fraction of prescriptions was selected from each month in 2002. A panel of four physicians evaluated drug choice and dosage (dose multiplied by duration) in relation to symptoms/diagnoses. Scores were based on the degree of accordance with Standard Treatment Guidelines. Scores given by panel members were used to create a Correction Score List. A Prioritization Factor was developed to assess in which cases interventions would have the greatest impact. Prioritization Factor = Actual Costs/Rationality Factor – Actual Costs. The Rationality Factor encompasses issues of correct drug choice and dosage. Outcome Measures: Percentage of correct drug choices for each drug for each diagnostic group; percentage of correct dosages for each drug for each diagnostic group; the Rationality Factor for each drug and dosage form; the Prioritization Factor for each drug and dosage form. Results: The method was applied to differentiate between prescribing problems with co-trimoxazole, chloroquine, and erythromycin in relation to diagnostic groups. Co- trimoxazole was prescribed most inappropriately with respect to drug choice; chloroquine was prescribed irrationally mostly with respect to dosing schedule. The Prioritization Factor showed that an intervention would be most cost-effective if it were directed toward erythromycin prescriptions for sexually transmitted diseases (Prioritization Factor: 264,021); the second-most cost-effective intervention would target co-trimoxazole prescriptions for “other” symptoms/diagnoses (Prioritization Factor: 120,836). Conclusions: A simple method to assess appropriateness of prescriptions and categorize prescribing problems in order of importance has been developed and field- tested. The method can be applied to direct interventions for prescribing problems toward areas where interventions would have the greatest clinical and cost impact. Study Funding: Danish University of Pharmaceutical Sciences.

3. BACKGROUND 1  Irrational prescribing is prevailing worldwide  Consequences are e.g. adverse effects on health and unnecessary costs  In developing countries resources are especially scarce and rigorous prioritisation of how to spend resources should be made  Published studies on assessments of rationality of prescribing do not report any generic methods that include costs in the assessment

4. BACKGROUND 2  There is a need for developing a method which is cheap and simple to apply in practice settings Methods applied in assessment of prescribing rationality SampleResultStudiesComment RandomNumber/ frequency Multiple e.g. WHO core indicators Simple Dot not address specific problems Drug(s)Conditions/ diseases with inappropriate use Very fewSimple Addresses diseases that are treated incorrectly Possible to compare across drugs Disease(s)Adherence to standard list e.g. STG FewSimple Only treatments of selected disease are addressed

5. OBJECTIVES  To develop and field-test a simple method for assessment of prescribing appropriateness, including both clinical and cost aspects

6. DESIGN  Analytical, register-based study SETTING  Jinja Regional and Referral Hospital, Uganda SAMPLE 3 tracer drugs based on  prevalence of use  risk of resistance  cost of treatment  Co-trimoxazole  Chloroquine  Erythromycin  Sample fraction of prescriptions each month 2002

7. METHODS 1  Modified nominal group technique: Expert panel of 4 local physicians evaluated drug choice and dosage (dose x duration) for stated diagnoses/symptoms based on degree of accordance with Standard Treatment Guidelines (STG)  Scores for drug choice: 1)In accordance with STG 2)Acceptable, but not in accordance with STC 3)Incorrect  Dosing schedule assessment C)Curative S)Sub-curative

8. METHODS 2 Prioritisation factor: Actual costs Rationality factor Actual costs: Cost x Total consumption x Prevalence of diagnostic group Rationality factor: Correct drug choice Total nu cases – Actual costs x Curative dosage Nu cases evaluated for dosage

9. RESULTS 1 Cotrimoxazole was the drug prescribed least appropriately Prescriptions evaluated as correct choice of drug (%) Diagnostic groupErythromycin (N=214)Cotrimoxazole (N=418) Sexually transmitted diseases9974 Gastro- intestinal diseases5042 Respiratory diseases10081 Genito-urinary diseases10086 Infectious and parasitic diseases6814 Other06 Total8854 Correct drug choice of erythromycin and cotrimoxazole (in %)

10. RESULTS 2 The dosage for chloroquine was inappropriate in more than half of cases, even if the drug was correct (for malaria) The dosing of erythromycin was also poor Percent of cases evaluated as curative dose x duration

11. RESULTS 3 Overall prioritization factors x 1000 Chloroquine inj. 49 Chloroquine tabs 4 Erythromycin311 Cotrimoxazole 57 Prioritization factors (PF) for erythromycin x 1000 (FF) Prioritization factors differ between diagnoses e.g. STD264 Infections etc. 33 G-U diseases 2 G-I diseases 11 Respiratory diseases 4 Diagnostic group PF

12. CONCLUSIONS  A simple method to assess appropriateness of prescriptions was developed  The method can serve as a tool to categorise prescribing problems in order of importance  The method addresses both clinical and cost aspects and can, therefore, be applied to direct interventions towards areas where the greatest impact can be achieved  The field test applied to the prescribing of three tracer drugs showed that interventions would be most cost-effective if directed towards erythromycin prescriptions for STD

13. POLICY IMPLICATIONS AND RECOMMENDATIONS  There is a great need for cheap and simple methods to assess rationality of prescribing in ‘real life contexts’. The methods should include both clinical and cost aspects  The developed method serves all these purposes. The method should be applied in other settings with the necessary adaptation to address specific local problems  WHO, universities and organisations teaching rational prescribing of medicines should address this or similar methods as an additional step on top of the well-known indicators