In general, prescribers' diagnoses of the four case scenarios were fairly accurate (67.9% diagnosed all four correctly). However, not a single prescriber prescribed a treatment for all four illnesses which was consistent with standard treatment guidelines. Overall only 11.3 % of the treatments prescribed were in accordance with standard treatment guidelines. Examples included: not prescribing Fansidar for chloroquine-quinine resistant malaria before resorting to quinine and not prescribing oral rehydration solution for dehydration. Dosage, especially for quinine, was calculated inconsistently or incorrectly. Few prescribers included the total number of tablets needed and often the regimen was given using 250 mg tablets, while the dispensary only stocked tablets of 200 mg or 300 mg. There was also considerable variation in the case scenarios with regards to the regimen prescribed (e.g. ½ tablet two or three times a day, etc.) further indicating that not all prescribers were prescribing in accordance with the standard treatment guidelines. A copy of the appropriate diagnosis and treatment for the case scenarios is included for further clarification (Appendices 4 and 5).

As mentioned previously, qualitative results from the prescriber interviews provided insight into the presumably inappropriate prescribing for malaria. Prescribers mentioned that they were taught to always consider the economic position of a patient before prescribing. Therefore, they would often prescribe the least expensive anti-malarial, quinine, and not chloroquine, followed by Fansidar. Prescribers stated that not only was quinine less expensive, but that it was more likely to cure the patient the first time, potentially saving more time and money in the long run.

The average amount of time a prescriber at the CBC facilities spent with a patient before either writing a prescription or referring the patient to the doctor, hospital or laboratory was 5.5 minutes (Nigeria 6.3 min; Tanzania 3.0 min). There was considerable variation among facilities (3.5-8.0 min). Doctors spent slightly less time with patients (3.4 min) than did screening nurses. An indicator not previously studied, dispensary waiting times, also varied. At the hospitals, patients waited on average 21.3 and 29.4 minutes for a drug prescription to be filled, while at equally busy outpatient departments patients waited only 10.7 and 10.3 minutes (Fig. 4). The shortest dispensary waiting time (5.4 min) was at a moderately busy facility. Dispensing times also varied between facilities. The CBC average dispensing time was 1.1 minutes, with a range of 0.6 to 1.9 (equally busy out-patient departments), (Nigeria 12.5 seconds; Tanzania 77.8 seconds, see Fig. 5).

Figure. 4 - Average dispensary waiting times

Figure 5 - Average dispensing times

Patients knew how and when to take their medication 84.9% of the time (range 48.3-100%) Nigeria 84.8%; Tanzania 75%). However, they knew the name, purpose, regimen and precautions only 38.4% of the time (range 0-88.9%).

On average, drug packages were adequately labelled with the name of the drug and regimen 92.1% of the time, with some variation between facilities (range 77.4-100%). Drug packages were considered adequately labelled if they contained the name of the drug and the regimen. These findings are summarized in Table 4 and presented in full in Appendix 10.

Table 4. Patient care indicators