Drugs are a valuable resource in developing countries and in view of their scarcity, often considered an indicator of quality of care. Worldwide, the reliable supply of pharmaceuticals alone is often used as a criterion for quality of care. Obviously, the reliable supply of pharmaceuticals, implying effective management of drug supply should not be the only indicator that characterizes a high standard of care. Without an evaluation of the actual use of the pharmaceuticals the indicator is incomplete. In fact, critics assert that the failure to recognize and appreciate the appropriate use of drugs by both the patient and the provider will undermine any positive effect that a reliable supply of drugs may have (Laing 1989, Hogerzeil et al 1989, WHO 1994, Ross-Degnan et al 1992). Consequently, the Action Programme on Essential Drugs (DAP) at the WHO and the International Network for the Rational Use of Drugs (INRUD) have collaborated on methods to systematically identify inappropriate drug use and implement and evaluate interventions to promote appropriate drug use (WHO 1993).

The inappropriate use of drugs is characterized by the use of drugs when no therapy is indicated, the use of the wrong drug for a specific condition requiring drug therapy, the use of drugs with doubtful/unproved efficacy, the use of drugs of uncertain safety status, failure to provide available, safe and effective drugs, and the use of correct drugs with incorrect administration, dosages and duration (Ross-Degnan et al 1992, WHO 1994, Hogerzeil 1995). The criteria used to determine appropriate drug use include:

• Appropriate indication
• Appropriate drug
• Appropriate patient
• Appropriate information
• Appropriate monitoring

Previous studies in Malawi, Nigeria, Tanzania and Uganda provide a possible range of values for "drug indicators"; indicators determined by DAP/WHO and INRUD as reliable measurements of appropriate drug use (Walker et al 1990, Hogerzeil et al 1993). Nigeria, when compared to other African countries such as Malawi, Tanzania, and Uganda appears to prescribe an excessive number of drugs, including antibiotics, per prescription (Table 1). However, Nigerian prescribers spend considerably more time with their patients. Interestingly, dispensing times, short in Nigeria yet relatively long in Tanzania, do not appear to correlate with patient knowledge of drug regimen. This is a small sample of indicator studies, however, and serves only as an example for comparison. In addition, without knowing disease prevalence it is difficult to estimate the appropriateness of drug prescription, especially when considering antibiotic rates.

Table 1. Drug indicator studies in Africa

N/A = not available

To the authors' knowledge no studies to date have explored WHO complementary indicators such as prescriber compliance with standard treatment guidelines, average cost of prescription or average dispensary waiting times. Neither have there been studies examining prescriber utilization of diagnostic tools (i.e. laboratory referral rates) nor a modified patient knowledge of drug regimen that includes the name, purpose and side-effect of drugs dispensed.

Of the many factors that can contribute to inappropriate drug use, in both developed and developing countries, the poor prescription practices of health care providers have received considerable attention. Designing effective interventions that necessarily imply prescriber behaviour change in order to improve prescribing practices, is critical now that more countries are seeking ways in which to stop this potentially dangerous and costly phenomenon. To date, strategies and interventions designed to address prescribing practices have included: information/education (the distribution of printed materials, group education, feedback on prescribing patterns, and face-to-face outreach), management (structured supervision) and regulation (development of protocols and guidelines) (Oxman et al 1995, Ross-Degnan et al 1997).

As poor prescription practices were originally attributed to a lack of prescriber information, the majority of interventions focused on providing up-to-date information regarding appropriate prescription. In a comprehensive review by Oxman et al (1995) of 102 studies designed to promote better practice in developed countries, those interventions found to be most effective were face-to-face visits or academic detailing (McConnell et al 1982, Avorn & Soumerai 1983, Avorn et al 1992, Soumerai et al 1993), patient-mediated interventions wherein the patient, once educated, demanded a specific treatment (Vinicor et al 1987, Cohen et al 1989, Cummings et al 1989), objective needs assessments and corresponding marketing (White et al 1985, Jennett et al 1988), audit and feedback (Gehlbach et al 1984) and any combination of the above (Putnam & Curry 1989). Those interventions found to be not at all or less effective were the distribution of educational materials without discussion and follow-up (Avorn & Soumerai 1983, Evans et al 1986), conferences that were didactic and not necessarily interactive, and reminders. The effectiveness of using local opinion leaders was also evaluated but without concluding whether or not the method was effective (Stross & Bole 1983, 1985, Stross et al 1986).

A similar review by Ross-Degnan et al (1997) of studies done in developing countries reiterates Oxman's findings. The study found that educational interventions, i.e. workshops or training courses (Angunawela et al 1991, Gani, Tangkilisan & Pujilestary 1995, de Vries et al, 1995), and community case management (Delacollette et al 1996, Fauveau et al 1992) were the most frequently used methods of promoting appropriate drug use. Administration (group processing, norm setting or performance review) (Guiscafre et al 1988, Guiterrez et al 1994, Agyepong et al 1996), supervision (audit and feedback) (Chowdhury et al 1996), and regulatory measures (Essential Drugs Programme) (Christensen et al 1990, Chalker 1995, ZEDAP 1996) were used less frequently. Those interventions evaluated as most effective were repeated, focused, multiple modality (lectures, group problem-solving, role-playing) workshops and courses that were done on-site and using opinion leaders or supervisors as trainers (Bexell et al 1995, Kafuko et al 1996, Gonzalez Ochoa et al 1996, Thomas 1996). Supervision, monitoring, regular audit and feedback were considered to have a moderate effect. However, it still has not been established if these methods are necessarily effective when considering a variety of problem practices (Kafle et al 1995, Sunartono & Darminto 1995, Chowdhury et al 1996). As found in Oxman's study, the distribution of educational materials without any complementary intervention was ineffective (Ross-Degnan et al 1997).

To the authors' knowledge, there have been few or no studies to date that examine the effectiveness of varying lengths of training, non-health-related formal education, off-site versus on-site supervision, doctor's supervision of prescribers and the corresponding frequency of different types of supervision. Nor have there been any studies that examine the context in which a prescription is written or dispensed, including average outpatient visits per day and number of prescriber/dispensers.

What has been established, however, is that inappropriate drug prescription is affected by a variety of complex, underlying factors which can be categorized as deriving from patients, prescribers, facility administration, supply system, regulation, drug information and/or misinformation (Ross-Degnan et al 1992, WHO 1994, Hogerzeil 1995). Globally, regulation through the implementation of protocols (standard treatment guidelines and essential drug lists), as the CBC has done, has been one of the most popular methods of counteracting these forces. However, while this method may initially reduce inappropriate drug use, the effect has not been shown to be sustainable. Okwaare et al's (1994) study of three interventions concluded that standard treatment guidelines alone were insufficient in affecting change in areas such as generic drug use and patient treatment for malaria and diarrhoea. Hogerzeil (1994), who corroborated this finding, has stated that:

"Treatment guidelines developed without wide consultation, distributed without proper introduction and training, not accompanied with a system to make the same drugs available in the health system, and without mechanism for continuous supervision and medical audit are unlikely to have an impact on prescribing".

Other factors identified as contributing to inappropriate drug prescription include: lack of training combined with poor prescriber supervision and monitoring (Bapna, Shewade & Pradham, 1994), drug availability (Hogerzeil & Walker 1989, Chowdhury 1994, Ofori-Adeji 1994), patient expectations and beliefs, (Homedes & Ugalde, 1993, Wolff, 1993), and prescriber beliefs and attitudes (Hamm, Hicks & Bemben 1996). Given the diverse influences on inappropriate drug use, a mixture of the strategies mentioned earlier is needed to promote appropriate drug use among prescribers (Soumerai, McLaughlin & Avorn 1989, Greco & Eisenberg 1993, Guiterrez et al 1994).

Of the various strategies used to address inappropriate drug use, improved prescriber supervision and monitoring is receiving increasing attention, with particular interest in the development of tools for self-monitoring within health centres. An intervention in the Gunungkidul District in Java, Indonesia, identified self-learning and active participation by prescribers as instrumental in changing prescribing practices (Sunartono & Darminto 1995). Using baseline drug use indicators to identify problem areas, the research team worked with local staff to develop and implement appropriate tools for self-monitoring in the health centres. Subsequent district-wide implementation of self-monitoring produced considerable decreases in polypharmacy (26% reduction), antibiotic use (51% reduction) and injection use (74% reduction), and an overall 17% reduction in the number of drugs ordered.

In Cameroon, drug use and misuse has best been described by Sjaak Van der Geest, a Dutch anthropologist, who wrote extensively on pharmaceutical use in the South West Province in the 1980s. He attributed the inappropriate use of drugs in Cameroon to the poor example of formal health care in government-run health care facilities (Van der Geest 1982a, 1982b, 1987a, 1987b, 1987c). Government health care facilities at that time were considered by the Cameroonians to be under-staffed due to "frequent absences of health workers," under-supplied, "characterized by bureaucratization and poor management," and constantly short of drugs (Van der Geest 1987c). Furthermore the financial incentive to prescribe greater quantities of drugs, as well as the relative convenience of informal sources of drugs prescribed, were undermining health services delivered through government channels. By 1995, a World Bank report noted that instead of going immediately to the formal sector when ill as was the practice before 1987, Cameroonians chose to first self-medicate. Other options included visiting a "quack doctor", street vendor, traditional or faith healer and often, only when very ill would people resort to the hospital, (World Bank 1995). Each of these factors has exacerbated inappropriate drug use in Cameroon.

Fortunately, alternative sources of formal health care are still an attractive option for Cameroonians. In fact, since 1987 there has been a steady increase in the number of outpatients seen at the CBC (CBC 1996). The CBC contributes to maintaining the trust between the formal health care facility and the people. Subsequently, ensuring appropriate prescribing and dispensing practices within the CBC is a step towards ensuring quality health care for Cameroonians.