1 Abstract Problem statement: An in-service training programme in Effective Prescribing, adapted from the World Health Organization’s “Guide to Good Prescribing”, was presented to Primary Health Care (PHC) clinic staff. Objectives: To assess the impact of the training programme on prescribing patterns. Design: Randomised, controlled, pre-and post-intervention study. Setting and Study Population: Twenty-four PHC clinics in Region 5 (Lowveld) of the Northern Province, now Limpopo Province, each allocated to a study group or a control group by a combination of stratified and random sampling (12 clinics in each group). Intervention: Effective prescribing training for primary health care prescribers. The study consisted of four phases: pre-intervention survey, four-day training workshop for staff from the study group clinics, first post-intervention survey one month after the workshop and second post-intervention survey three months after the work-shop. During each survey, at each clinic 30 prescriptions were analysed for each of two target conditions, namely upper respiratory tract infections (URTI) and diarrhoea and vomiting (DV). Outcome Measures: Number of items per prescription, generic prescribing, antibiotic prescribing, treatment with ORS, correct choice of treatment according to the national Standard Treatment Guidelines (STG). Other measures of prescribing and patient care were investigated, but are not presented here. Results: Improvements were found both for the target condition addressed by the workshop (URTI) and for a condition which was not covered in the workshop (diarrhoea and vomiting). Improvements were sustained to the second post- intervention survey. Conclusions: Improvements in prescribing for both an example and a non- example condition showed that the principles of effective prescribing were internalised during the training and then applied in everyday practice. In the World Health Organization’s “Teacher’s Guide to Good Prescribing” this training programme has been described in Chapter 6, “Application in Primary Care Settings”, and examples of patient cases are presented in Annex 1. Funding sources: Operating budgets of the MEDUNSA School of Pharmacy and provincial pharmaceutical services of Limpopo Province. Health Systems Trust (HST). IMPACT OF TRAINING IN EFFECTIVE PRESCRIBING IN PRIMARY CARE Summers RS 1, Meyer JC 1, Möller H 2 (1) School of Pharmacy, Medical University of Southern Africa (MEDUNSA) (2) South African Drug Action Programme (SADAP)

2 Introduction Prescribing practices impact greatly on drug use and expenditure. The situation in developing countries is often compounded by a limited health budget. Furthermore, due to role substitution in these countries, prescribers are often not formally trained in rational prescribing. As a result of a study carried out in the Northern Province of South Africa (1), the provincial Department of Health implemented a training cascade in 1997, with an initial focus on the improvement of prescribing practices at primary care level. Objectives of the study To assess the effect of a prescribing training intervention for primary health care nurses.

3 Method An Effective Prescribing training cascade was implemented in the Northern Province (now Limpopo Province) of South Africa. This study investigated the impact of the training on prescribing for two target conditions (see Study Plan). WHO- recommended indicators and supplementary “drug use” indicators for the target conditions were used. Data were collected by patient interviews and prescription reviews. Data collector interviewing patients

4 Training course Effective prescribing training was based on two major sources. 1. The WHO’s “Guide to Good Prescribing” (2), which describes six steps in rational pharmacotherapy: 2. A problem-based learning approach, adapted from the Problem-Based Pharmacotherapy Teaching course of the Department of Clinical Pharmacology of Groningen University in the Netherlands. Trainers also introduced the participants to some key aspects of the National Drug Policy of South Africa, as well as to the underlying principles of Standard Treatment Guidelines (STGs) and the Essential Drugs List (EDL). Between 1997 and 2000, 946 prescribers, mostly nurses, were trained in this way.

5 Training cascade Training of trainers course Effective prescribing training in districts Effective Prescribing course Regional and district trainers attended a one- week generic Training of Trainers’ course, and a one-week trainers’ workshop on Effective Prescribing. They subsequently conducted similar 4-day Effective Prescribing courses for approximately twenty Primary Health Care workers per workshop under MEDUNSA’s supervision. Trainers of the Lowveld region

6 4-day workshop Study plan 2 clinics excluded: prescribers transferred to study group clinics 47 clinics considered 23 clinics excluded 24 clinics selected and randomised 1 clinic to control group, prescriber could not attend training Post-intervention evaluation 2 No training Post-intervention evaluation 1 11 clinics Pre-intervention evaluation 11 clinics 12 clinics Apr-Sep clinics Control groupStudy group = 1 month

7 Indicators: A Average items per prescription B % encounters with no drugs C% items prescribed by generic name D % items from National Essential Drugs List E % prescriptions with antibiotic(s) F % prescription with injection(s) G % prescriptions with antidiarrhoeal(s) H % prescriptions with antimicrobial(s) I % prescriptions without Oral Rehydration Solution J % prescriptions with only Oral Rehydrat. Solution K % items according to Standard Treatm’t Guidelines Key to tables and figures Significant differences (p<0.05) are shown as follows: within groups (paired T-test) between groups (T-test for independent means)  between groups, changes (T-test for independent means) Results We found numerous significant differences within and between groups (see figures). The comparison between pre-test and post-test 1 results (one month after training) showed similar patterns to those between pre-test and post-test 2 (three months after training). Only results of the latter comparison are shown.

8 Results: Prescribing for Upper Resp. Tract Infection % Improvements Deteriorations A†BCDEFKA†BCDEFK † as % of baseline       Control groupStudy group PrePost2PrePost2 Scripts reviewed: A Ave items/Rx B % non drug Tx C % items generic D % items EDL E % Rx with AB F % Rx with inj K % items STG

9 Prescribing for Diarrhoea and Vomiting % Improvements Deteriorations A†BCDEFGHIJKA†BCDEFGHIJK † as % of baseline Control groupStudy group    PrePost2PrePost2 Scripts reviewed: A Ave items/Rx B % non drug Tx C % items generic D % items EDL E % Rx with AB F % Rx with inj G % Rx with AD H % Rx with AM I % Rx with no ORS J % Rx w. only ORS K % items STG

10 Discussion Prescribing practices for the two conditions examined were improved by the training. Changed behaviour was not only seen in prescribing for Upper Respiratory Tract Infections, used as an example condition, but also for Diarrhoea and/or Vomiting, which was not included in the training programme. Prescribers thus applied their new skills to other conditions. The results show that it is not necessary to use qualified teachers to do the training, as the people who successfully conducted the training workshops had never trained before. Teaching must be conducted in small groups of four to six people. This approach, in which the WHO’s “Guide to Good Prescribing” was adapted to problem- based training, has proved successful. The WHO’s “Teacher’s guide to good prescribing” (3) describes this training programme in Chapter 6, “Application in Primary Care Settings”, and presents examples of patient cases in Annex 1.

11 Limitations 1.The sample was selected from a single region and may not be fully representative of the entire Province. 2.To prove a retention effect it would have been ideal if post-test 2 had been carried out six months, instead of three months, after the intervention. However, if post-test 2 had been postponed to six months after the initial training took place, the control group would have been lost. A typical scene outside a PHC clinic

12 Contributors Cascade training concept developed by D Meyer, Pharmaceutical Services, Northern Province, and H Möller Training developed, coordinated and supervised by H Möller, RS Summers, A Joubert and P Jack from the School of Pharmacy, Medical University of Southern Africa (MEDUNSA) Study coordinator: JC Meyer Investigators: JC Meyer, H Möller, M Rasengane (team leaders); M Pholoane, I Maakana, E Makoala, F Mosetsa, SF Maphalle, P Jack, C Nobela, R Hlatswayo Acknowledgements This work was supported by Health Systems Trust (HST), Durban, South Africa, Grant Number 186/97. Prof H Schoeman provided statistical consultancy. Monika Zweygarth assisted with this presentation. References (1)Möller H, Summers RS, Hlongwane E, Nobela C, Thupana M. A Study of Pharmaceutical and Related Services in Northern Province and a Follow-up Action Plan. CHASA Journal of Comprehensive Health 1997; 8(3/4), (2)De Vries TP, Henning RH, Hogerzeil HV, Fresle DA. Guide to Good Prescribing. WHO/DAP/95.1. Geneva: World Health Organization; (3)Hogerzeil HV, Barnes KI, Henning RH et al. Teacher’s Guide to Good Prescribing. WHO/EDM/PAR/ Geneva: World Health Organization; 2001.