The Nepalese health care delivery system operates at different levels. The sub-health post (SHP) often serves as the first level of contact of the rural population with the health care system. Three or four SHPs are under the control of a health post (HP) and in turn three or four HPs are under the administrative control of a primary health centre (PHC). Due to the shortage of doctors in the rural areas of Nepal, auxiliary health workers (AHWs) usually man the sub-health posts. AHWs undergo a one-year course after matriculation and the last three months of the course are spent attached to a health post. AHWs rarely receive advice or supervision from physicians and are the ones making diagnostic and therapeutic decisions.(2)

Shortage of medicines continues to be a major problem in the HPs and SHPs of Nepal.(3) The annual consignment of drug supplies for government health institutions usually lasts only 3-5 months and there is no provision made for reordering until the next years supply is due. To overcome the shortage, the Community Drug Programme (CDP) was formulated and implemented from the start of the fiscal year 1996/97. (4) The mechanism of the revolving drug fund (RDF) for the replenishment of drugs is in operation in most of the community health facilities of Kaski district, Western Nepal where the Hemja SHP is located. His Majesty's government (HMG) provides seed money to the SHPs to buy medicines. The SHP sells the medicines to the patients and recoups the expenses. The money obtained is used to replenish the stock of medicines in the SHP pharmacy. 

The running of the Hemja SHP has been handed over to the Hemja village development committee (VDC). A committee of 8 persons supervises the day to day running of the SHP. The committee consists of the SHP-in-charge, the pharmacist, a representative of HMG and 5 elected members of the VDC of whom at least two are women. 

The objectives of our study were to:
1) obtain information on the age and sex distribution of patients attending the Hemja SHP during the study period
2) obtain information on the morbidity pattern and the age and sex distribution of morbidity
3) obtain information on the drug prescribing patterns, rationality of prescriptions and average cost per prescription and 
4) apply the Anatomical therapeutic chemical (ATC) classification and to calculate the defined daily dose (DDD) of the ten most commonly prescribed drugs in the Hemja SHP.

Methods:

Location where the work was carried out: Hemja sub-health post, Kaski district, Western Nepal 'and the Departments of Pharmacology and Community Medicine, Manipal College of Medical Sciences, Pokhara.

All drug prescriptions over a three-month period (01/05/2000 to 31/07/2000) from the Hemja SHP were collected and entered into a patient indicator form (PIF) for further analysis. The Hemja SHP serves a population of 9622 in the Kaski district of Western Nepal. Information about patients who were not prescribed any medicines was obtained from the outpatient register of the SHP. Information about the age and sex of the patient was recorded. The morbidity pattern during the study period was recorded. Age and sex differences in the morbidity patterns were analyzed using the _X2 test. A p value of less than 0.05 was taken as statistically significant.
Information on the number of drugs prescribed, the name of the drugs and the duration of prescription was noted. The percentage of drugs prescribed by generic names and from the Essential drug list of Nepal(5) was noted. The cost of drugs per prescription and the overall cost per prescription including the non-drug costs were calculated. 
The rationality of the prescriptions was determined by two of us (PRS and PP) in consultation with the medical officers of the Community Medicine department who regularly visit the different health posts. A prescribed textbook used during the training course for AHWs was  also consulted.(2) Sex differences and age-wise differences in the prescribing of drugs were analyzed using the _X2 test (p<0.05).

Results: 
Seven hundred and twenty eight patients visited the sub health post during the study period. Three hundred and eighteen (43.7%) were males. Four hundred and thirty nine patients (60.3%) were below the age of 30 years. The detailed age and sex distribution is shown in Table 1. 

The average number of drugs per prescription is shown in Figure 1. Most of the prescriptions were for one or two drugs. The mean ± SD number of drugs per prescription was 1.53 ± 0.9. Wounds and wound infection were the most common reason for visiting the SHP [69 out of the total of 728 patients (9.5%)]. Acute respiratory infections (ARI) [67 out of 728 patients (9.2%)], worm infestation (9.1%) and diarrhea/dysentery (7.4%) were also common. The sex-wise morbidity pattern is shown in Table 2. Wounds and ARI were significantly more common in males. Visits for family planning and antenatal checkups (ANC) accounted for 11.7% and 4.9% of the total visits made by women to the SHP. Diarrhea and worm infestation were more common in men but the difference was not statistically significant. 

Figure1The average number of drugs per prescription

Under-fives accounted for 118 out of the total of 728 patients (16.2%). ARI and diarrhea/dysentery were the most common illnesses in under-fives accounting for 38 and 20 patients respectively. The incidence of ARI/1000 children (<5 years) during the study period was 322 while that of diarrhea was 169.

A total of 1117 drugs were prescribed during the study period. Antimicrobials were the most commonly prescribed class of drugs in the study. The other commonly prescribed groups of drugs in descending order of frequency were analgesics, anthelminthics, injectable contraceptives and vitamin preparations. The most commonly prescribed drug was paracetamol accounting for 21.1% of the total prescriptions. The most commonly prescribed antimicrobials were cotrimoxazole, amoxicillin, metronidazole and chloramphenicol. Topical antimicrobials accounted for 11.8% of the total antimicrobials prescribed. The other commonly prescribed drugs were antacids (5.5%) and mebendazole (5.2%). Parenteral preparations accounted for 50 out of the 1117 drugs (4.5%) while topical preparations accounted for 8.5% of the total drugs prescribed. 'Eight hundred and thirty-eight drugs (75%) were prescribed from the Essential drug list of Nepal(5).

The sex-wise prescribing patterns of individual drugs are shown in Table 3. Amoxicillin and metronidazole were more commonly prescribed in males (p<0.05). Cotrimoxazole and mebendazole were also more frequently prescribed in males but the difference was not significant. Vitamins were more frequently prescribed in females. Injection depot medroxyprogesterone acetate was used for family planning in 48 females. 

The age-wise prescribing patterns are shown in Table 4. The frequency of prescribing of paracetamol was high in the young age groups, declined in the age groups from 10 to 40 years and then rose again in the age group = 40 years. Amoxicillin was more frequently prescribed in the age groups < 10 years and the frequency declined thereafter. Prescribing frequency of antacids and vitamins, in general showed an increasing trend with age. Mebendazole was most frequently prescribed in the age groups < 20 years and showed a declining trend thereafter. 

The ATC classification and the DDD of the ten most commonly prescribed drugs in the Hemja SHP are shown in Table 5. The mean ± SD cost of drugs per prescription was 18.6 ± 15.8 Nepalese rupees (0.24 ± 0.2 US$). If the cost of the outpatient ticket and the different procedures undergone by the patient were included the cost was 20.8 ± 16.2 Nepalese rupees (0.27 ± 0.21 US$). Ten out of the 728 prescriptions (1.4%) were irrational. The common reasons for irrationality were the prescribing of antibiotics in common cold and the prescribing of antibiotics in pyrexia of unknown origin without proper investigation.

Discussion: 
Studies of prescribing patterns usually measure drug utilization by counting of prescriptions which indicates the frequency of prescribing. Estimates would be more relevant and meaningful on combining the frequency of prescribing with the total quantities of drugs prescribed per patient. The DDD concept was developed to overcome objections against traditional units of measurement of drug consumption. (6) DDD for a given drug is established on the basis of the assumed average use per day of the drug used for its main indication in adults.(6,7)

The average number of drugs per prescription is an important index of the scope for review and educational intervention in prescribing practices. In our study the average number of drugs per prescription was 1.53 which is less than 4.3 drugs reported in a Taiwanese study.(8) In an Indian study the mean ± SD number of drugs per prescription was 1.99 ± 0.71. (9) The prescription  of the lesser number of drugs in our study should be encouraged as it is associated with lesser risk of drug interactions and errors of prescribing.

Wounds and wound infections, ARI, worm infestations and diarrhea/dysentery were the common illnesses encountered in the SHP. The five common causes of morbidity in Kaski district in which the SHP is situated were skin diseases, ARI, diarrheal diseases, worm infestation and gastritis.(4) In the Western development region in which Kaski district is situated the five most common causes of morbidity were skin diseases, diarrheal diseases, ARI, worm infestation and gastritis.(4) Skin diseases, diarrheal disease, ARI, worm infestation and pyrexia of unknown origin (PUO) were the commonest diseases at the National level. (4) In the Taiwanese study (8), ARI, musculoskeletal disorders, hypertension, gastrointestinal disorders and diabetes mellitus were the common disorders. 

Under-fives constituted around 14.9% of the male population and 14.4% of the female population of Nepal in 1995. (10) Under-fives constitute a vulnerable section of the population as regards morbidity and mortality. The under-five mortality rate for Nepal is 91/1000 live births. (11) ARI and diarrheal diseases were the commonest causes for under-five morbidity in our study. These diseases are a major problem in under-fives in Nepal.(4) 

The most commonly prescribed drug categories were antimicrobials, analgesics, antacids, anthelminthics, injectable contraceptives and vitamins. In the Taiwanese study the top five most commonly prescribed drug categories were antacids, anti-cough and anti-cold preparations, vitamins, analgesics and non-steroidal anti-inflammatory drugs (NSAIDs).(8) These differences could be due to differences in the morbidity patterns in the two studies. A high percentage of drugs were prescribed from the Essential drug list. This is a welcome sign and has to be encouraged.

We have calculated the DDDs for the ten most commonly prescribed drugs in the SHP. The DDD for ORS is not defined. We have used the DDD per 1000 inhabitants per day (DID) as the indicator of drug consumption. The DIDs observed in our study were less than that observed in the literature.(12) We have studied drug utilization for only a three-month period and so seasonal influences may have a role to play. Also the entire population of Hemja VDC does not turn to the SHP for their health care needs. We have not looked at the percentage of the population visiting private and complementary medicine practitioners in our study. Also some of the more sick patients directly go to hospitals in Pokhara City without first going to the SHP. So our values of DID may be an under-estimate of the true extent of drug use in the Hemja VDC. Amoxicillin was the major amino penicillin prescribed in our study. Our DID for the aminopenicillins was less than that reported in a recent study.(13) However, antibiotic utilization varies between countries and between different time periods in the same country. In a recent study,(14) the DID for antibiotics in Ravenna, Italy was 16.5 while in Funen, Denmark the DID was 10.4. There was age-wise difference in antibiotic utilization also.

Amoxicillin and metronidazole were more commonly prescribed in males. Wounds, wound infection and ARI were significantly more common in males and this maybe, one of the factors accounting for the difference in prescribing patterns. Vitamins were more frequently prescribed in females. The increased prevalence of ARI and worm infestation in the younger age groups may account for the increased prescription of amoxicillin and mebendazole in these age groups.

 
Only 1.4% of the prescriptions were irrational. Antibiotics were the drugs accounting for the majority of the irrational prescriptions. The percentage of irrationality was lower than that reported in a Chinese study.(15) But ARI was one of the major causes of morbidity and we have not strictly applied the WHO criteria for diagnosis and treatment of ARI in developing countries.

Sub-health posts frequently serve as the first level of contact of the Nepalese population with the healthcare delivery system. Knowledge of the prescribing patterns at different SHPs can be helpful in pinpointing lacunae and improving the training of the AHWs who mainly man the SHPs. Knowledge of the commonly used drugs will be helpful to the CDP managers in formulating drug programmes relevant to the needs of their particular community. The CDP is run by the committee in charge of the SHP. They call tenders for the supply of medicines and usually accept the lowest quoted tender. The patient is assured of a supply of essential drugs at a reasonable price. A 10% quota is available for free treatment of patients who are unable to pay. More detailed studies on drug utilization at the district level in Kaski district are required and are being planned with the help of the Department of Community Medicine of our institution and the Western regional health directorate.