Availability and affordability of essential medicines and diagnostic tests for diabetes mellitus in sub-Saharan Africa: A systematic review.

Background Currently, sub-Saharan Africa (SSA) is experiencing a steady increase in the prevalence of diabetes mellitus (DM) coupled with a prevailing high burden of communicable diseases. To effectively address this burgeoning burden of DM, optimal access to affordable essential medicines and diagnostic tests for DM in healthcare systems should be prioritised. We conducted a systematic review of the evidence on the availability and affordability of essential medicines and diagnostic tests for DM in SSA as recommended by the World Health Organization Package of Essential Non-communicable Disease Interventions for Primary Health Care in Low-Resource Settings. Methods PubMed, Science Direct and African Journals Online databases were searched for original research articles conducted in sub-Saharan Africa and published between 2000 and 2018 reporting availability and affordability of essential medicines and diagnostic tests for diabetes mellitus. Results Twenty one original cross-sectional studies were included in the systematic review, with the majority conducted in Eastern Africa (n=11, 58%). The availability of essential medicines and diagnostic tests was largely sub-optimal. For oral hypoglycaemic agents and insulin, angiotensin-converting-enzyme inhibitors, statins and aspirin, availability ranged from 0-100%, 0-96.5%, 0-84% and 53%-100% respectively. Considering diagnostic tests, availability of blood glucose tests, urine protein and ketone tests, serum creatinine tests, lipid profile tests and electrocardiography ranged from 6-100%, 33.3-100%, 0-86.4%, 0-65.9% and 5.7-54.6% respectively. The lowest priced generic (LPG) glibenclamide, metformin and aspirin cost <1.2 days’ wages. However, the cost of LPG insulin (any type), captopril and simvastatin ranged from 3.85-18.7 days’ wages, 1.2-6.41 days’ wages and 6.5-30 days’ wages respectively. Blood glucose tests, urine protein and ketone tests and serum creatinine tests cost <3.3 days’ wages. Structured summary Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.

Conclusions Optimal access to affordable essential medicines and diagnostic tests for DM remains a significant challenge in SSA. This represents a significant barrier towards the attainment of sustainable development goals and universal health coverage. Pragmatic region-specific solutions are urgently needed to address this challenge. Background Sub-Saharan Africa (SSA) is currently experiencing a rapidly increasing burden of noncommunicable diseases (NCD) such as diabetes mellitus (DM) coupled with a high burden of communicable diseases such as HIV and tuberculosis [1,2]. This high dual burden of NCD and communicable diseases poses a significant economic strain on the poorly structured healthcare systems and meagre health resources.
According to the 2019 International Diabetes Federation (IDF) estimates, Africa has about 19 million adults living with DM. This translates to a regional prevalence of 3.9%. The region has the highest proportion of people with undiagnosed DM (60%) and it is estimated to have the greatest future increase in the burden of diabetes (about 47 million adults with diabetes in 2045 which is a 143% increase) [3].
Despite this projected significant increase in the regional burden of DM, SSA still faces an insurmountable challenge of equitable access to affordable essential medicines and diagnostic tests for DM in healthcare systems [4]. In September 2011, the United Nations General Assembly high-level meeting recognized the magnitude of the NCD epidemic globally and its threat to national economic development. One of its commitments was to improve access to medicines to treat NCD (DM inclusive) [5].
Optimal availability of affordable essential medicines and diagnostic tests for NCD in healthcare systems is fundamental in addressing the growing burden of NCD, DM inclusive. As part of its 2013-2020 Global Action Plan (GAP) for prevention and control of NCD, the WHO included a target of ≥ 80% availability of affordable essential medicines and basic technologies required to treat major NCD [6].
There is need for contemporary evidence on the extent of availability and affordability of essential medicines and diagnostic tests integral in the management of DM in SSA. These findings will help inform appropriate intervention strategies to address the challenge of inequitable access.
To-date, no systematic review has been published on availability and affordability of essential medicines and diagnostic tests for DM in SSA. This systematic review collated data from original research studies that investigated the availability and affordability of essential medicines and diagnostic tests for DM in SSA, as recommended by the WHO Package of Essential Non-communicable Disease Interventions for Primary Health Care in Low-Resource Settings (WHO-PEN) [7].

Methods
Search strategy and study selection The references of the selected research articles and published review articles were also searched for additional original studies to include in the systematic review.
The search terms used were: "Access OR availability OR pricing OR cost OR affordability AND "essential medicines" OR medicines OR drugs OR therapies AND tests OR "laboratory tests" OR diagnostic OR "point of care" AND "diabetes mellitus" OR diabetes OR "type 2 diabetes" OR "type 1 diabetes" AND Africa OR "Sub-Saharan Africa".
Only original research articles with information about availability and affordability of essential medicines and diagnostic tests for DM, conducted in any sub-Saharan African country and published between 2000 and 2018 were included in the systematic review. We excluded published review articles, original research articles whose full texts were not accessible and those published research articles published in languages other than English.
Essential medicines and diagnostic tests of interest were those recommended in the management of DM by the WHO Package of Essential Non-communicable Disease Interventions for Primary Health Care in Low-Resource Settings (WHO-PEN) [7] (summarised in Table 1). Despite the exclusion of glycated haemoglobin (HbA1c) and serum ketone tests in the WHO-PEN, we also obtained information from the studies about their availability and affordability because of their relevance in optimal diabetes care in clinical practice. Table 1 Package of Essential Non-communicable disease (PEN) disease interventions for primary healthcare in low resource settings: Essential medicines and diagnostic tools for diabetes. We obtained the information on study country, study period, type and number of health facilities surveyed, type of essential medicine and diagnostic test studied and estimates of availability and affordability as reported in the original studies involving primary data collection. In all these studies, availability was defined as the proportion of health facilities where the essential medicine (s) and/ diagnostic test (s) of interest was found present at the time of primary data collection in the study. Affordability was defined as the estimated total number of days' wages the lowest-paid government worker would be required to pay to purchase a full monthly standard dose of the medicine or to pay for the diagnostic test as recommended by the WHO and Health Action International (HAI). We considered the availability of any essential medicine or diagnostic test of ≥ 80% as optimal, as recommended by the WHO GAP for prevention and control of NCD [6]. We did not define optimal affordability because no internationally recognised definition exists. Assessment of methodological quality, bias and data extraction process The methodological quality of the identified studies was assessed by three independent reviewers (DK, RES and DA) using the adapted Newcastle-Ottawa Scale (NOS). A maximum score of 8 was considered for the selected cross-sectional studies (Table 3) [8]. The quality assessment tool for observational cohort and cross-sectional studies published by the National Heart, Lung and Blood Institute was used to evaluate the risk of bias of cohort and cross-sectional studies (Table 4) [9]. The PRISMA guidelines for the reporting of systematic reviews and meta-analysis were followed (Table 5) [10]. The studies were rated as either having a low, moderate or high risk of bias. Rating of studies was independently performed by two reviewers (DK and RES) and inconsistencies were resolved by consulting a third reviewer (DA).

Results
The search yielded a total of 5,487 published articles. Eighty-five duplicates were identified and removed. Of the remaining 5,402 articles, a total of 12 articles that were editorials, review articles, systematic reviews, study protocols and a study published in French were excluded leaving 5,390 articles whose titles and abstracts were screened.
Of these, 5,329 articles were excluded leaving 61 articles whose full texts were extracted and assessed for eligibility. Only 21 original research articles met the inclusion criteria described above and were included in the final systematic review   (Fig. 1).
According to the quality assessment tool for observational cohort and cross sectional studies, all the original cross-sectional studies included in the systematic review were considered as having a low risk of bias (Tables 4). The majority of the studies had low methodological quality (Table 3).

Availability Of Essential Medicines For Dm
The availability of the different essential medicines for DM as recommended by the WHO PEN are shown in Table 2.

Insulin
Generally, availability of insulin as reported by the majority of the studies was sub-optimal basing on the recommended WHO GAP goal of ≥ 80%. Availability of insulin of any type ranged from 0% in Mozambique [22], Benin and Eritrea [31] to 100% in Zambia [22].

Oral hypoglycaemic agents (OHA)
Availability of oral hypoglycaemic agents (OHA) in the surveyed health facilities ranged from 0% in Eritrea (metformin and glibenclamide) [31] to 100% in Uganda (metformin and glibenclamide/glimepiride) [11] and Swaziland (metformin) [25]. The most studied OHA were glibenclamide and metformin. In addition to those two drugs, one study from Ethiopia also assessed the availability of gliclazide or glipizide which was reported to be very low (4%) [19].

Availability Of Diagnostic Tests For Dm
A total of 14 studies (66.7%) investigated the availability of at least one diagnostic test for DM. Among the recommended diagnostic tests for DM in the WHO PEN, availability of blood glucose tests was the most investigated (13 studies, 92.9%) [11, 13, 14, 17-22, 26-29, 31] while microalbuminuria tests were the least investigated (2 studies, 14.3%) [11,20].
The availability of the diagnostic tests for DM as recommended by the WHO PEN is as shown in Table 2.

Blood glucose tests
Availability of blood glucose tests ranged from 6% in a study conducted in Mozambique [22] to 100% in a study conducted in Cameroon [27]. Only two studies reported optimal levels of availability of blood glucose tests of ≥ 80% [11,27] while one study conducted in Sudan documented close to optimal levels of availability of 75% [31]. Availability of blood glucose tests in the remaining studies was < 70% [13, 17-22, 26, 28, 29, 31].

Serum creatinine tests
Availability of serum creatinine tests was also reported by five studies [11,20,27,28,31]. Optimal availability of ≥ 80% was only noted in two studies conducted in Uganda [11] and Cameroon [27].

Electrocardiography (ECG)
The documented availability of ECG as reported by four studies was 5.7% and 54.6% in Uganda [11,20], 9.1% in a study conducted in Uganda and Kenya [14] and 10% in Cameroon [27].

Microalbuminuria tests
Availability of microalbuminuria tests was reported by only two studies conducted in Uganda both documenting low levels of 6.8% and 13.2% respectively [11,20].

Serum troponin tests
Availability of serum troponin tests as reported by three studies was 0% in Ghana [28] and Eritrea [31], 8% in Benin and Sudan [31] and 43.2% in Uganda [11].

Serum ketone tests
Availability of the serum ketone tests was investigated by only one study which reported them available in only 11.4% of the surveyed health facilities, which were all private hospitals [11].

Affordability Of Essential Medicines For DM
Affordability of any of the essential medicines for DM was investigated by five studies [11,23,25,27,30] (summarised in Table 2).

Oral hypoglycaemic agents
The lowest priced generic (LPG) glibenclamide cost less than 2 days' wages in all the studies [11,23,25,27,30]. Except for the study conducted in Uganda [11], LPG metformin cost ≤ 1.2 days' wages in the rest of the studies [23,25,27]. One study assessed the cost of the newer generation sulphonylurea-glimepiride whose monthly dose cost 3.2 days' wages [11].

Insulin
Affordability of insulin was assessed in three studies [11,23,27]. Short-acting and intermediate-acting insulin cost 4.7 and 4.9 days' wages respectively in Uganda [11]. The cost of short-and intermediate-acting insulin in Cameroon was similar (3.85 days' wages) [27]. A high cost of the innovator brand of intermediate-acting insulin was reported in Malawi (19.6 days' wages) [23]. A high cost of pre-mixed insulin was noted in Cameroon (18.7 days' wages) [27] compared to Uganda (4.9 days' wages) [11].
Aspirin, ACEI (captopril) and statins (simvastatin/atorvastatin/rosuvastatin) Affordability of any of the three classes of essential medicines above was investigated by four studies [11,23,25,27]. Only two studies assessed affordability of all the three classes of essential medicines above [11,27].
Aspirin cost less than a days' wages in the two studies [11,27]. The cost of captopril and statins greatly varied among the countries. It cost < 1.3 days' wages in Malawi [23] and Swaziland [25], 2.8 days' wages in Uganda [11] and 6.41 days' wages in Cameroon [27].

Affordability Of Diagnostic Tests
Affordability of diagnostic tests for DM was assessed by only two studies conducted in Uganda [11] and Cameroon [27]. Blood glucose, urine protein and ketone, serum creatinine tests cost < 3.3 days' wages in both countries [11,27]. In comparison, the cost of lipid profile testing in Uganda [11] was twice the cost in Cameroon [27] (7.5 and 3.59 days' wages respectively).
Both ECG and HbA1c cost > 8 days' wages in both countries with a higher cost documented in Cameroon. The cost of serum ketone, microalbuminuria and serum troponin tests was investigated by only one study reporting costs of 2.1 days' wages, 9.6 days' wages and 11.3 days' wages respectively [11].

Discussion
This systematic review presents, to our knowledge, the first comprehensive assessment of availability and affordability of essential medicines and diagnostic tests for DM as recommended by the WHO-PEN in SSA.
Universally, the availability of essential medicines and diagnostic tests for DM remains sub-optimal in SSA based on the WHO GAP goal, particularly intermediate-acting insulin, statins, ECG, HbA1c, microalbuminuria and serum troponins tests. All the three types of insulin (short-acting, intermediate-acting and pre-mixed) and statins are largely costly.
With the exception of blood glucose, urine protein, urine ketone and serum creatinine tests, the cost of the remaining WHO PEN recommended diagnostic tests was high.
Similar findings of sub-optimal availability and high costs of essential medicines and diagnostic tests for DM has been widely reported in most low-and middle-income countries [32][33][34][35]. There are several plausible explanations for this. Limited funding of the health sector by the respective governments to procure the appropriate essential medicines and diagnostic tests for DM could explain the low availability observed in the public sector [36].
Substantial attention is given to communicable diseases like malaria, tuberculosis and HIV in SSA with minimal consideration to NCD. Healthcare systems are mainly structured to test, treat and cure communicable diseases, with little emphasis given to longitudinal NCD care. Few global funding initiatives or programs exist in SSA to support equitable access to affordable essential medicines and diagnostic tests for DM as seen with malaria, tuberculosis and HIV.
Glaring knowledge-practice gaps in DM care among healthcare practitioners creating less demand for essential medicines for NCD, absence of some essential medicines on the national essential medicine lists (NEML), lack of incentives to maintain optimal medicine stocks at the health facilities, forecast inaccuracy and inefficient purchasing or distribution systems are other plausible explanations for the sub-optimal availability of essential medicines and diagnostic tests for DM [36][37][38].
The high costs of all types of insulin (key medicine for all patients with type 1 DM and some patients with type 2 DM) and statins (an essential medicine for primary and secondary prophylaxis of cardiovascular diseases in patients with type 2 DM) in SSA is of great concern. One plausible explanation for the high costs of insulin in SSA is the market monopoly in insulin production and volume sales by a limited number of multinational companies with minimal generic insulin production. The current shift from the use of human insulin to the newer insulin analogues has further increased insulin costs in SSA [37][38][39].
The Lancet NCD Action Group in their seminal paper on promoting access to essential medicines for NCD, including DM, acknowledge that improving access to affordable medicines requires a comprehensive health system approach including pharmaceutical sector governance, appropriate pharmaceutical workforce training, pharmaceutical management information systems, procurement planning and sustained financing of medicines [40]. They propose several key strategies to improve access to affordable medicines like legislation to promote generic market entry and submission, appropriate pricing for generic medicines, reduced patient co-payments for generics, rational selection and use of medicines for NCD, good monitoring electronic systems to avert stock-outs and increased financing for NCD medicines from domestic and international sources [40].
To improve access to medicines for NCD, 21 global biopharmaceutical companies have established access to medicine initiatives mainly in LMIC. One systematic review identified 120 of these initiatives with 52% focused on NCD. A worthwhile example is the Novartis Access program which is currently running in 7 countries in SSA (Cameroon, Ethiopia, Kenya, Uganda, Rwanda, Malawi and Tanzania). Its objective is to offer a portfolio of medicines for NCD like metformin to the public sector at a subsided fee of one US dollar and also build health system capacity in preventing, diagnosing and treating NCD, including DM [41]. Novo Nordisk, one of the key multi-national insulin manufacturing companies also adopted an equity pricing initiative to supply insulin at a much-subsidised fee to a selected number of low-income countries in SSA. The company also supports the Changing Diabetes in Children (CDiC) program in 10 countries in SSA in partnership with Roche Pharmaceuticals, the International Society of Paediatric and Adolescent Diabetes and World Diabetes Federation by offering free insulin and glucometers to children and adolescents living with type 1 DM [42].
In addition to the above recommendations, we propose these broad practical solutions. Despite the SARA tool being nationally representative and has been frequently used globally, it doesn't include price data and hence, cannot be used to obtain information about affordability. The majority of the eligible studies had low methodological quality. All of the reviewed studies were cross-sectional studies that considered a one-time point assessment of availability and affordability. This does not put into consideration time variations in the availability and cost of essential medicines and diagnostic tests.

Conclusion
This systematic review shows that some essential medicines and diagnostic tests for DM in SSA remain unaffordable and their availability still falls substantially below the WHO defined GAP goal. This underscores the need for concerted national and international efforts to improve equitable access to affordable essential medicines and diagnostic tests for DM in SSA, as part of strategies to achieve the defined sustainable development and universal health coverage goals. Figure 1 Flow diagram for the systematic review