Access to Essential Medicines for Diabetes Care: Availability, Price, and Affordability in Central Ethiopia

Background Diabetes is a major global public health burden. Effective diabetes management is highly dependent on the availability of affordable and quality-assured essential medicines (EMs) which is a challenge especially in low-and-middle-income countries such as Ethiopia. Methods A cross-sectional study was conducted in 60 selected public and private medicine outlets in central Ethiopia from January to February 2022 using the World Health Organization/Health Action International (WHO/HAI) standard tool to assess access to EMs. We included EMs that lower glucose, blood pressure, and cholesterol as these are all critical for diabetes care. Availability was determined as the percentage of surveyed outlets per sector in which the selected lowest-priced generic (LPG) and originator brand (OB) products were found. The number of days’ wages required by the lowest paid government worker (LPGW) to purchase a one month’s supply of medicines was used to measure affordability while median price was determined to assess patient price and price markup difference between public procurement and retail prices. Results Across all facilities, availability of LPG and OB medicines were 34.6% and 2.5% respectively. Only two glucose-lowering (glibenclamide 5mg, metformin 500mg) and two blood pressure-lowering medications (nifedipine 20mg and hydrochlorothiazide 25mg) surpassed the WHO’s target of 80% availability. The median price based on the least measurable unit of LPG diabetes EMs was 1.6 ETB (0.033 USD) in public and 4.65 ETB (0.095 USD) in private outlets, respectively. The cost of one month’s supply of diabetes EMs was equivalent to 0.3 to 3.1 days wages in public and 1.0 to 11.0 days wages in private outlets, respectively, for a typical LPGW. Thus, 58.8% and 84.6% of LPG diabetes EMs included in the price analysis were unaffordable in private and public outlets, respectively. Conclusion There are big gaps in availability and affordability of EMs used for diabetes in central Ethiopia. Relevant stakeholders should work to improve access to EMs.


BACKGROUND
Non-communicable diseases (NCDs) have become a major public health problem worldwide.In 2019, NCDs represented 74.5% out of total deaths that occurred globally of which 47% of NCD mortality in low-and-middle-income countries (LMICs) were premature, i.e., before the age of 70 [1].Diabetes mellitus (DM), one of the most common NCDs, leads to chronic hyperglycemia which is associated with long-term damage and dysfunction of different organs such as the heart, brain, eyes, nerves, and kidneys.Still, with early diagnosis and treatment, many of the harmful effects of the disease can be delayed or even avoided [2].Globally, it has become a health challenge due to its high prevalence and its cardiovascular (CV) complications [3].In 2021, International Diabetes Federation (IDF) estimated that there were 536.6 million adults aged 20-79 years with diabetes.For the same year, IDF estimated 6.7 million deaths due to diabetes or its complications corresponding to 12.2 % of globl mortality from all causes in this age group.The corresponding number of adults with diabetes for Ethiopia is 1.92 million, which places it among the top four countries in the African region [4].
According to the World Health Organization (WHO) package of essentials for non-communicable disease (PEN) interventions in low-resource countries, early diagnosis and the provision of affordable and effective medicines are major strategies in reducing the burden of NCDs [5].Wellfunctioning health systems are critical for preventing, controlling, and managing the steadily rising diabetes and for improving health outcomes [6].Access to affordable essential medicines (EMs) and their availability for patients with DM is very important because such patients need these medications life-long [7].But access to EMs is multifactorial and associated with rational selection and use of medicines, availability and affordability of medicines, sustainable health care nancing and reliable supply system of quality products [8].Failure in one portion of the framework results in malfunctioning of the other [9].According to the WHO, nearly 2 billion people have no access to EMs, with many people in Africa facing the problem [10,11].Taking initiative, WHO had voluntary set 80% targets for availability of EMs and other health technologies to control major NCDs in health facilities (HFs) by 2025 [12].Like many LMICs, Ethiopia is facing a devastating burden of increasing NCD morbidity and mortality, especially from DM [13].As a result, Ethiopia started putting several initiatives such as developing the rst national guideline on clinical and programmatic management of major NCDs in 2016 [14].Studies addressing access to EMs for DM care in the country are however scarce.This study aimed to assess the accessibility of EMs used for diabetes care in central Ethiopia's public and private medicine outlets with respect to availability and affordability parameters.

Study area, design, and period
The study was conducted in health facilities operating in central Ethiopia where 15% of the country's population lives [15].An institution-based cross-sectional survey was utilized to collect data regarding availability, affordability, and pricing of EMs.Estimates were prepared through collecting and analyzing data using the WHO/Health Action International (WHO/HAI) format from January 1, 2022, to February 30, 2022 [16].

Study facilities selection
The study area has seven administrative districts.Considering Addis Ababa, the capital of the country, as a center for the study, six districts that can be reached within 1 day, have public health facilities (PHFs) that have provided diabetes care services for at least one year, handle selected EMs for diabetes care, and have pharmacy professionals and physicians to manage the interest of patients with NCDs were selected.According to the standardized WHO/HAI methodology, hospitals and health centers' outpatient pharmacies from the public sector, and retail pharmacies and drug stores from the private sector (closer to the selected public health facilities) were identi ed and used as study settings [16].

Medicine outlets selection
The country's three-tier system (primary, secondary, and tertiary level categorization of HFs) of healthcare served as a baseline for selecting medicine outlets [17].Purposively taking one main hospital from the higher level of the framework for each selected study area, the remaining PHFs (2 public hospitals and 2 health centers (HCs)) within three hours of travel from it were randomly selected from the lists of PHFs obtained from the health bureaus of Addis Ababa, Oromia, and Amhara regions for the public sector [16].Five licensed and private medicine outlets (PMOs) which were proximate to selected PHFs in each study area were also chosen by simple random sampling.In total, 60 medicine outlets were included, 30 from the public and 30 from the private sector.

Study medicines selection
Thirty-ve EMs were identi ed and selected based on the (i) 2019 WHO 21st list of EMs for adults, (ii) medicines commonly used for treatment of DM and medicines for CV risk management that are listed in the current Ethiopian Essential Medicine List (EML), and (iii) those medicines speci ed in the WHO core medicine list used for comparative measurement of availability across the countries [11,18].Two forms of products were selected and surveyed for each medicine; namely, the originated brand (OB), more speci cally the brand-name proprietary product, and the lowest-priced generic (LPG) product, the cheapest generic equivalent that was present at each pharmacy during the time of the survey [16].

Data collection and analysis
Data collection was adapted from the WHO/HAI's standardized methodology on measuring medicines availability, prices, and affordability [16].Three experienced pharmacists were appointed and trained as data collectors for this study.They received one-day training on the study's purpose, the different names, strengths, and dosage forms of selected medicines, how to complete the data collection form, and how to compute unit costs.Data on the availability of EMs was determined by direct observation: a medicine was considered available if it was on the shelf and ready to be dispensed at the time of the visit.Price data (selling prices of medicines for end users) was recorded for medicines in stock.Public-sector procurement prices were gathered from Ethiopia's public procurement agency, i.e., Ethiopian Pharmaceutical Supply Services during the previous two years.
For tracking quality of data collection, processing, and statistical analysis, data were entered into a customized MS Excel from the workbook provided as part of the WHO/HAI methodology.All medicine outlets surveyed ful lled the WHO/HAI recommendation criteria to collect data on the selected 35 medicines (Table 1) [16].

De nitions of availability, price, and affordability of medicines
The availability of each studied EM was measured by its physical presence in the medicine outlets by their speci ed strength and dosage form on the survey date.It was determined as the mean percentage (%) availability of individual medicines, availability across groups of medicines, variations between product types such as (LPG vs OBs), and of individual medicines between sectors [19][20][21].The current study utilized percentage ranges: 0% -absent-not found in any retail outlet surveyed; <30% -very low-very di cult to nd; 30%-40% -low-somewhat di cult to nd; 50%-80% fairly high-available in some retail outlet; and > 80% very high-good availability to describe the extent of availability of medicine for diabetes care [19,20].
Prices for products were taken as unit prices and de ned as price per capsule or tablet or vial (least measurable unit).It was computed using the following equation.
Both price lists and prices on the pack of medicine were used to ll in the data for each surveyed medicine physically found in each sampled facility.The prices were converted to US dollars using the buying exchange rate, i.e., 1 USD = 49.1482Ethiopian Birr (ETB) which was taken from the Ethiopian National Bank website on January 1 st, 2022, the rst day of data collection [22].In the analysis of price data, both LPG and OB medicines were analyzed separately.The median value of retail price, interquartile price ranges, and minimum and maximum prices were used to describe individual medicine prices in local currency (ETB).Price data of medicines that were found in less than four medicine retail outlets were not included in the price analysis, given the small sample size and low precision of potential estimates.
Affordability was estimated by comparing the total cost required to cover the complete course of standard treatment based on the lowest-paid government worker's (LPGW) daily wage [16].Assessment of affordability for standard treatment of each medicine used the de ned daily dose (DDD) of each EM, which is the expected average maintenance dose per day for adults and serves as a standard dose unit of measurement [23].Accordingly, affordability was calculated by applying the following equation.
If the cost of a course of treatment of an anti-diabetic medicine is no more than one day's wage or income, it is considered affordable.The treatment courses that cost more than one day's wage were classi ed as unaffordable.Thus, daily wages were used to express affordability and calculated by dividing the cost of the treatment course by the LPGW's daily wage.As of January 2022, the Ethiopian Civil Service Authority paid 1409 birr per month or 28.67 USD per month to the LPGW.As a result, the daily wage was calculated by dividing the monthly salary for the previous 30 days, which was ETB 46.97 per day (USD 0.96 per day).

Price of diabetes care EMs
Seventeen LPG medicines were found in ≥ 4 medicine outlets and hence eligible for measurement of the median price (Table 2).Accordingly, the median prices (based on the medicine price for the least measurable unit) were 1.6 ETB (0.033 USD) and 4.65 ETB (0.095 USD) for PHFs and PMOs respectively.For 16 LPG medicines, patients were paying much higher in PMOs than their public counterparts.For instance, for the second lowest priced medicine among the surveyed EMs-hydrochlorothiazide 25mg tablet, patients paid more than two times as much from PMOs as they would from PHFs (Table 2).Since there were not enough retail outlets where OB-type medicines could be purchased, the median patient price for OB version was not estimated.

U nitP rice = P riceofP ackageofM edicineF ound P ackSizeofM edicineF ound T reatmentcoursecost = N umberofunitdoserequiredforDDDofEM xM edianU nitpriceofEM xDaysofatreatmentcourse
In addition, using the public procurement price and the retail price, an inter-sectoral pricing comparison was conducted for 13 LPG medicines commonly found in both sectors.As per the comparison, a markup difference between the patient price and the procurement price registered in the PHFs and PMOs, respectively, was found to be 55.74% and 145.9%.The markup difference between the two sectors was also found to be 56.9%.

Affordability of diabetes care EMs
The affordability of EMs for diabetes care varied by type of medicine and sector (Table 3).In both sectors, four medicines (glibenclamide 5mg, enalapril 5mg, hydrochlorothiazide 25mg, and amlodipine 5mg) were found to be affordable.All OB medicines observed in PMOs during the study time were found to require payment between 2.9-and 74.1-days' wages for LPGWs.Eight LPG medicines in PHFs and 18 LPG medicines in PMOs cost more than two days' wage for a monthly supply of medicines.All Insulin products surveyed across the study area required 3 or more days' wage for diabetes patients in both sectors.Insulin products were almost 2 times more unaffordable in PMOs than in PHFs.

Comprehensive analysis of diabetes care EMs availability and affordability
The comprehensive analysis of the availability and affordability of LPG medicines across all surveyed medicine outlets is presented in Fig. 4. The availability score for each medicine is depicted on the X-axis where as the Y-axis was labeled with the number of days wages required to purchase a one-month supply of medicine-affordability.The gure illustrates four quadrants based on WHO/HAI cutpoints: one-day wages for affordability and 80% for high availability.Metformin 500 mg and glibenclamide 5 mg were discovered in Quadrants III and VI, respectively, indicative of good availability although the former required patients to pay more than a day's wage for a month's supply.Unfortunately, 73.1% of the medications that were eligible for price analysis were found in Quadrant II, where products' availability fall behind WHO/HIA availability target and are unaffordable for a typical LPGW.

DISCUSSION
Access to EMs for the management of DM and CV risk management in terms of availability, patient prices, and affordability was low in all 60 surveyed facilities.The availability of LPGs and OB medicines were 34.6% and 2.5%, respectively, with most medicines well below the WHOrecommended target of 80% availability.The median patient price of LPG diabetes EMs was 0.033 USD in the PHFs and 0.095 USD in the PMOs with most LPG diabetes EMs being unaffordable in both the PHFs and PMOs based on a typical LPGWs' daily wages.
Our ndings are consistent with previous national assessments indicative of low availability of EMs for DM and other NCDs, which is one major determinant for poor glycemic control, and associated morbidity and mortality among patients with diabetes in Ethiopia.While there could be different reasons for the observed low availability, some of the reported ones include lack of attention to NCDs, limited nancial resources, and weak supply systems including staff capacity and logistics management information systems, among others [24,25].Nevertheless, the nding is a further warning for the healthcare system to address EM availability for DM and related conditions.
Insulin is required for the survival of people with type 1 diabetes and for the enhanced control of diabetes in some patients with type 2 diabetes.It is listed as an EM in the WHO EM list as well as in the Ethiopian EML, demanding that it should be available at all the times [11,18].In this study, the availability of insulin products was relatively better compared to oral DM agents and was more available than previous reports from Ethiopia suggest [26-28].The availability of insulin is comparable to those reported from other LMICs such as Uganda and Brazil [12,29,30].
The ndings also revealed that the availability of insulin was much better in hospitals (85.2%) and pharmacies (65%) compared to HCs (8.3%) and drug stores (33.3%), similar to ndings reported by M. Ewen et al [31].This low availability in or near primary healthcare facilities which serve the majority of the population may lead to unequitable access to EMs for patients with diabetes [32].This could lead to patients forgoing care at nearby primary healthcare facilities and travelling longer distances to hospitals and pharmacies due to the lack of EMs in nearby outlets.This could create additional barriers to access and in turn lead to lower adherence to follow up and medications [14].This calls for efforts to ensure distribution of EMs to locations that are accessible to patients with DM.
The median price of 26 LPG medicines was 4.65 ETB in PMOs while that of 17 LPG medicines was 1.6 ETB in PHFs which indicates products in PMOs were sold at three times of median price in PHFs.Price being a key determinant of affordability, it may have indirect role in ensuring access [33].As the cost of medicines take up the major share for DM and CV risk factors management services, developing policies that address medicine prices for patients and enhance access are critical.Some of these include introducing mechanisms to make pricing transparent, reduce medicines prices such as tax reduction and regulate prices [12].
Most EMs in both sectors were considered unaffordable, costing between 1.2-and 3.2-days' and 1.3-and 74.1-days' wage to cover a month's treatment in the PHFs and PMOs respectively.When EMs remain unaffordable, patients may forgo their treatment especially if they are paying out-of-pocket for their medicines, which increases the burden of DM and its complications [34].In order to enhance access to EMs and protect citizens from the devastating diabetes complications and nancial risk that have huge effect on public health, the country should work to make health care nancing sustainable for its e cient operation.These may include expanding the existing community-based health insurance, initiate the social health insurance and introduce other types of health insurance [12,35,36].

Study Limitations
This study has some limitations.Among these is the cross-sectional nature of the study whereby the availability of data reported in this study was based on a one-day visit to surveyed medicine outlets.Hence, it is unable to re ect the average monthly or annual, or overtime availability of medicines at the outlets.Moreover, medicines median price ratios for international price comparison are not reported in this study as the existing international price reference guideline was outdated.Lastly, the affordability of medicines reported was determined by the government's lowest salary scale for an unskilled worker.Thus, this implies that medications that appear to be relatively affordable in this study may be unaffordable when other expenditures are considered.Therefore, all these limitations need to be considered while generalizing the outcomes of this study.
This study can however provide an important and clear picture to national policymakers on access to EMs for DM and other non-communicable diseases.Different strengths and dosage forms of speci c medications were included in this study to circumvent WHO/HAI availability underestimation which occurs by the inclusion of limited strength of medicines.The clinical importance of surveyed medicines has been triangulated between the national EML, the WHO EML, and national and international standard treatment guidelines.

CONCLUSION
The mean availability of LPG EMs used for diabetes care in central Ethiopia falls far below the WHO target of 80% and the median patient prices for most of the EMs were unaffordable in both PHFs and PMOs based on the LPGWs' daily wages.
Based on the ndings, this study recommends increased government attention to availing affordable EMs for diabetes care.This may require strengthening EMs supply systems, expanding and strengthening nancing sources, and pricing mechanisms to mitigate and avoid catastrophic expenditures.Given that these ndings are limited to Central Ethiopia, it is recommended that similar studies be conducted in different parts of the country.

Table 1
Availability of diabetes care essential medicines (EMs) in the central Ethiopia EMs name, strengths and dosage Percentage of outlets where anti-diabetic and CV risk management EMs were found

Table 2 Median
Price [the 25 th -75 th Percentile] of LPG diabetes care EMs (PHFs, n = 17; PMOs, n=26) * *-Median Price Ratio (MPR) for international price comparison were not reported in this study as MSH 2015 international price reference (IPR) guideline was outdated to use.Therefore, the study took Health Action International's recommendation to present the price outcome by median price in the local currency, ETB (0.0203 USD).

Table 3
Affordability of EMs for diabetic care: money needed to cover a monthly treatment against days' wage of LPGW Total*-Total unit for 30 days treatment; DDD-de ned daily dose; MTP-Median treatment price; when data was absent, the cells of table was shaded; LPGW-lowest-paid unskilled government worker's daily salary; EMs-essential medicines.