Menu
Menu
Menu
Menu
Background Irrational prescription of drugs can lead to high cost of treatment thus limiting access to essential medicines. We assessed the affordability and appropriateness of prescriptions written for diabetic patients in Eastern Uganda. Methods We collected secondary data from the health management information system registers of patients who attended the outpatient medical clinic at Mbale regional referral hospital from January 2019 to December 2019. The average cost of the prescriptions was calculated and adjusted odds ratios for predictors for unaffordability estimated using logistic regression. Computed scores for indicators of rational drug prescription were used to assess the extent of rational prescribing. Results The median cost per prescription was USD 11.34 (IQR 8.1, 20.2). Majority of the diabetic patients (n = 2462; 94.3%, 95% CI: 93.3–95.1%) could not afford the prescribed drugs. Predictors for unaffordability were if a prescription contained: ≥ 4 medicines (AOR = 12.45; 95% CI: 3.9–39.7); an injectable (AOR = 5.47; 95%CI: 1.47–20.32) and a diagnosis of diabetes mellitus with other comorbidities (AOR = 3.36; 95%CI: 1.95–5.78). Having no antidiabetic drug prescribed was protective for non-affordability (AOR = 0.38; 95%CI: 0.24–0.61). The average number of drugs per prescription was 2.8. The percentage prescription of drugs by generic name and from the essential medicine and health supplies list of Uganda were (6160/7461; 82.6%, 96% CI: 81.7%-83.4%) and (6092/7461; 81.7%, 95% CI: 80.8%-82.5%) respectively against WHO standard of 100%. Conclusion The majority of diabetic patients (94.3%) in Eastern Uganda cannot afford to buy prescribed medicines. The government should therefore ensure that essential medicines are readily accessible in public health facilities.
This study was conducted at the Outpatient medical clinic of Mbale Regional Referral Hospital (MRRH) located in Mbale Municipality, Eastern Uganda. MRRH, one of the fourteen (14) regional referral (tertiary) hospitals in Uganda serves sixteen (16) surrounding districts of Eastern Uganda. These are Mbale, Budaka, Pallisa, Kibuku, Butebu, Butalejja, Tororo, Manafwa, Namisindwa, Bududa, Bulambuli, Sironko, Bukedea, Kapchorwa, Bukwa and Busia. The hospital has a total bed capacity of 548 and provides specialized health services to over five million people in Elgon region and even beyond. Besides hosting medical interns from the Ministry of Health, MRRH is also the major teaching hospital for surrounding medical and nursing schools including Busitema University Faculty of Health Sciences. The diabetic clinic is a specialized clinic hosted within the general outpatient medical clinic of the MRRH. An estimated 10,000 patients are managed by this clinic annually. Uganda’s healthcare system is hierarchical in nature with chronologically increasing cadres of healthcare facilities. The lowest cadre is the Village Health Teams (VHTs), also known as a health centre HCI and predominantly offers health education, preventive and simple curative services in communities. The next level is HCII which offers out-patient services. Next in level is HCIII, which in addition to HCII services offers in-patient, simple diagnostic and maternal health services. Above HCIII is the HCIV which provides surgical services in addition to all the services provided at HCIII. Beyond HCIV we have the district hospitals. At the national level, there are national referral hospitals, regional referral hospitals and semi-autonomous institutions in respective hierarchy [11]. At all these cadres of health care, all prescribed medicines are provided to the patient at no cost regardless of being outpatients or inpatients. However, due to the inadequate supply of essential medicine to public health facilities coupled with high patient turn up, public health care facilities usually suffer from prolonged drug stock outs [4]. Since many patients do not have health insurance coverage, they usually buy these essential medicines from private pharmacies and drug shops to access primary health care [11,16]. This was a retrospective cross-sectional study that utilized secondary data from the health management information system (HMISFORM 031) registers of the outpatient medical clinic at MRRH. The study population was diabetic patients attending the outpatient medical clinic at MRRH. This population did not include paediatric patients because this nature of patients receive their care from the paediatric clinic. Neither did this study include pregnant women because these receive care from the antenatal clinic. We followed the WHO guidelines of including atleast 600 prescriptions while investigating medicine use in health facilities [12]. In this study, we used a total of 2612 prescriptions of diabetic patients that were sampled from the register of the outpatient medical clinic, MRRHfrom January 2019 to December 2019.On average, two hundred and twenty (220) observations were systematically randomly selected from each month and included in the study. Observations (entries) with complete prescription data in the registers were extracted and analysed for affordability and rational prescribing. All prescriptions having DM as one of the diseases diagnosed were considered. Observations with illegible information were excluded from the study. Research assistants with pharmacy training background were recruited to assist in data collection. The whole research team was then trained on the data collection process to minimise interpretation bias of data to be collected. A data collection tool was designed in Microsoft Excel (Microsoft Corporation, USA) to capture secondary data on different variables of each patient prescription entry as it appeared in the register. Retrospective data from January 2019 to December 2019 were then collected manually from handwritten registers between February 2020 and May 2020. Following entry, data was checked periodically for completeness by the research team. An observation was considered complete if it contained all the required variables of interest which included gender, age, location, disease diagnosis and drugs prescribed. The primary outcome of this study was affordability of the prescription. This was determined by calculating the total cost of the prescribed medicines in the prescription and computing the number of days it would take to pay off the cost based on the average income of people in Eastern Uganda. The total cost was obtained by summing up the individual costs of each drug in the prescription. The cost of each drug in the prescription was obtained by multiplying the total quantity of that drug with the average unit cost based on the average retail prices of the drugs in pharmacies and drug shops in Mbale district (S1 Table). The quantity of each drug prescribed was first calculated basing on the prescribed dose and frequency [12]. The average retail prices were calculated from a survey done regarding the unit cost of different drugs as sold from selected pharmacies and retail shops around Mbale town. The obtained price list is attached as a S1 Table. The calculated costs of prescriptions were compared with the average monthly income of lowest government paid servant as extracted from the Uganda National Household survey (UNHS) 2016/2017 [13]. From the UNHS 2016/2017 report, the average monthly income of lowest government employed person in 2019 was $44.5 (exchange rate 3704/ =). This on average translates into approximately $1.5 per day. Affordability of prescription was categorized into two levels. All prescriptions that required a maximum of three (3) days were collectively categorized as affordable and coded 0. The rest of the prescriptions that required more than three days were categorized as “unaffordable” and coded 1 [6]. Other variables were categorized as shown in Table 1 for comparison purposes. Prescribed medicines and diagnoses were classified according to the Anatomical Therapeutic Chemical (ATC) classification system and international system of classification of disease respectively. The secondary outcome variables were appropriateness of the prescriptions in reference to guidelines set by WHO in collaboration with the International Network of Rational Use of Drugs (INRUD). To assess this, indicators recommended by the WHO/INRUD were calculated. These are; (1) average number of medicines per prescription, (2) percentage encounter with antibiotics, (3) percentage of medicines prescribed by generic name, (4) percentage of injectable medicines in the medicines prescribed and (5) percentages of medicines prescribed from the Essential Medicine and Health Supplies List for Uganda (EMHSLU). These secondary outcomes were calculated using Eqs 1–5: The WHO prescription parameters were put in place to improve the appropriateness of prescriptions during patient care. An average number of two medicines per prescription are recommended to reduce polypharmacy. In an effort to curb antibiotic drug resistance, the percentage encounter with antibiotics per prescription should be less than 30%. Different brands of drugs exist on market, hence it is recommended to practice 100% generic prescribing, this ensures effective communication and information exchange amongst health care providers, additionally helps tame the cost of treatment that may be escalated by prices of the medicine brands. The use of injectable medicine is often associated with a number of challenges, some of which may include the need for trained personnel to administer the medicine, pain, nerve injury and potential exposure of a patient to infections hence these should make up less than 10% of the total prescriptions. Every country has an Essential Medicine and Health Supplies List; this entails a list of drugs that have been proved safe, efficacious and cost effective in that specific region, hence 100% of the drugs prescribed should be from that list. Data were entered into an Excel spread sheet by two independent data entrants and exported for analysis into STATA version 14.0 (StataCorp, College Station, Texas, USA). Continuous data were summarised into means and standard deviations if normally distributed. Otherwise, they were summarised into medians with interquartile ranges if not normally distributed. Categorical variables were presented as frequencies and proportions. The proportion of patients that could not afford the prescribed medicines was estimated and the confidence limits were calculated using the exact method. Multivariable logistic regression analysis was used to estimate the adjusted odds ratios of the independent variables on unaffordability of prescribed medicines while controlling for confounding. All variables with p<0.25 at the bivariate level were included in the initial model at the multivariate analysis. All variables with p<0.1 and those of biological or epidemiologic plausibility (from previous studies) were included in the second model. Ethical approval was obtained from CURE–Children’s Hospital Uganda Research and Ethics Committee (CCHU-REC/10/019), administrative clearance from Mbale regional referral Hospital and the Uganda National Council of Science and Technology (HS2686). A waiver of consent was applied for and granted by the Research and Ethics committee of MRRH to use the prescriptions records in this study. Patient confidentiality was ensured by giving a specific number code to each patient data instead of their names.
