Background: Cardiovascular disease, including hyperten- sion and diabetes, is a leading cause of the disease burden in African countries. The World Health Organization has targeted that basic technologies and essential medicines should be available in >80% of health facilities. We exam- ined the availability of protocols, equipment and medicines for the management of CVD risk in primary care health facil- ities in the public sector in 9 countries in the African region.

      Methods: Under the coordination of the African Heart Network, an affiliate of the World Heart Federation, survey officers visited 5-10 health centers purposively selected in and around an urban center in 9 African countries (Burundi, Niger, Uganda, Benin, Kenya, Sudan, Tunisia, South Africa, Seychelles). They assessed the availability of protocols, ba- sic diagnosis equipment and medications related to CVD risk reduction.

      Results: Protocols for hypertension, diabetes, rheumatic heart diseases, CVD or WHO PEN were available consistent- ly (i.e. in >50% of health centers in a country) in 9 countries. A sphygmomanometer was found in nearly all health centers but a large cuff was consistently available in only 4 countries. A glucometer was available consistently in 4 countries. Medicines from the four main classes for hy- pertension were available consistently in only 4 countries. Availability of these items was strongly associated with a country’s Gross Domestic Product (GDP) per capita and ad- equate availability was found in the 3 countries with the highest GDP (>Int $ 10’000).

      Conclusions: Availability of protocols, basic diagnosis equipment and main medicines for CVD risk management was poor in primary health care facilities in the public sec- tor in countries with a low GDP but was adequate in coun- tries with a higher GDP. These findings underlie the need to strengthen the capacity of primary health care to address CVD reduction in the region.

      Cardio Vascular Disease (CVD) is the leading cause of the dis- ease burden in Africa [1] and the prevalence of CVD risk factors such as hypertension and diabetes is high in the region [2,3]. The World Health Organization (WHO) Global Action Plan for the Prevention and Control of Non Communicable Disease (NCD) 2013-2020 has formulated a set of cost-effective inter- ventions for reducing CVD risk in populations and among high- risk individuals. WHO has targeted that medicines to prevent major NCDs (including CVD) are available in 80% of public and private health facilities and that 50% of eligible individuals re- ceive treatment to prevent heart attacks and strokes by 2015 (including blood pressure and glycemic control).

      Suboptimal medicine use is a main barrier to adequate CVD risk reduction. A number of policy variables need to be ad- dressed to improve the availability of CVD medicines in LMICs in relation to rational selection, affordable prices, sustainable financing and reorganization of health systems toward man- agement of chronic diseases [4-6]. A particular issue it to maxi- mize CVD risk reduction in the population while minimizing the number of individuals requiring medications. The number of individuals eligible for treatment can extend, for example, to all individuals with blood pressure ≥140/90 mmHg according to various guidelines [7] or be limited to individuals with a total 10-year CVD risk ≥30% according to WHO recommendations for low-resource settings [8-10]. Yet, irrespective of how CVD risk is stratified, the numbers of persons requiring medicines for CVD risk reduction will inevitably be large because of the high prev- alence of hypertension, diabetes and other risk factors in the population. This underlies that it is crucial that management of CV risk is handled mainly at primary health care level.

      Several studies have shown a low availability of medicines for CVD reduction (e.g. medicines for hypertension, diabetes, medicines for secondary CVD prevention) in low and middle- income countries (LMICs) [11-14]. A recent study in 18 coun- tries, based on medicines found in pharmacies located in the surveyed communities, indicated that four main CVD medi- cines needed for secondary CVD prevention (aspirin, beta- blockers, angiotensin-converting enzyme inhibitors and statins) were available, in urban/rural settings, in 95%/90% of high in- come countries, 80%/73% of upper middle income countries, 62%/37% of lower middle income countries, and only 25%/3% of low income countries [14].

      Few studies have documented the availability of diagnosis equipment, protocols and medicines related to CVD in the Af- rican region. Hence, this survey examined these variables in primary health care facilities in the public sector in nine Afri- can countries. It is expected that the findings can be useful to guide policy aimed at strengthening health system capacity for addressing CVDs in the region. The findings can also help inform progress toward achieving the WHO target of 80% availability of essential medicines in health facilities in the region.

      The survey was organized through the African Heart Network (AHN) (www.ahnetwork.org), a coalition of heart foundations, cardiac societies and individuals working in the field of CVD in Africa, and an affiliate of the World Heat Federation. The sur- veys took place between 2015 and 2016 in 9 countries in which AHN was active. In each country, 6-10 primary care health facili- ties of the public sector and including at least one medical doc- tor were selected. Selection of the health facilities was based on a purposive rather than representative sampling, aiming to strike a balance between representativeness and practically, consistent with the very low budget available. Around half of the facilities were located within an urban center (Bujumbura, Burundi; Niamey, Niger; Kampala, Uganda; Nairobi, Kenya; Cotonou, Benin; Khartoum, Sudan; Monastir, Tunisia, Johannes- burg, South Africa; and Victoria, Seychelles) and the other half were located in smaller towns at up to 100 km distance. During the visits in the facilities, survey officers, working in pairs, ad- ministered a structured questionnaire to two senior officers of the health facility. Questions asked about characteristics of the health facility, the estimated numbers of patients attending the health center, and selected procedures for CVD management. Survey officers assessed the presence of selected pre-defined items, including guidelines or protocols for different CVD condi- tions (CVD risk reduction, hypertension, diabetes, dyslipidemia, and rheumatic heart disease), basic equipment to measure blood pressure and blood glucose and pre-defined classes of medicines for the treatment of CVD, hypertension, diabetes, dyslipidemia, and Rheumatic Heart Disease (RHD). Authoriza- tion to conduct the surveys was obtained from administrative authorities in each country.

      Data were tabulated according to country, sorted by their GDP per capita, expressed in international $ in 2015, as an indi- cator of overall wealth of the country. The small number of ob- servations within each country (5-10 health facilities per coun- try) and the fact that multiple variables were examined make statistical tests neither relevant nor helpful in this context.

      Table 1 Shows the characteristics of the health facilities in the 9 countries. Health facilities had an average of 7-15 doc- tors and nurses, except for lower numbers in Benin. The average number of adult patients attending health centers in one day, as estimated by the senior health personnel of each facility, ranged from 54 to 233 in average per country, except in Benin where it was lower. These data suggest that health facilities were of a significant size in all countries except in Benin and that all health facilities acknowledged examining several hypertensive or dia- betic adult patients every day.

      Table 2 shows that a list of essential medicines was available in nearly all health centers. However, the presence of protocols for the management of CVD, hypertension, diabetes, dyslipidemia or rheumatic disease, were present in a majority of health facilities only in Tunisia, South Africa and Seychelles, which cor- responds to the countries with the highest GDP per capita (e.g. exceeding Int $ 10’000). The WHO PEN package for manage- ment of NCDs (9), which includes treatment protocols for CVD, hypertension and diabetes, were found in only a few health centers in 4 countries (Uganda, Benin, Kenya, and Tunisia).

      Table 3 shows the availability of selected medicines in the health facilities in the 9 countries. Medicines in all main cat- egories of hypertensive medications (thiazide diuretics, calcium channel blockers, ACE inhibitors/angiotensin receptor blockers and beta-blockers) were present in a majority of health facili- ties only in 4 countries, which also were those with the high- est GDP per capita. Availability of oral antidiabetic medicines

      or insulin, or that of cholesterol lowering medicines, was found consistently only in around half of the countries, and particu- larly in those with a higher GDP. These findings suggest that a consistent availability of the main classes of medicines for CVD risk reduction in health facilities in the public sector strongly correlates with a country’s GDP per capita.

      Table 4 shows that a sphygmomanometer was present in virtually all health centers in all countries but a large cuff was consistently available in health facilities only in a few countries. A glucometer was found in a majority of health facilities in half of the countries. Medicines for hypertension or diabetes were prescribed to well-controlled (stable) patients at adequately large time intervals (e.g. every 2-4 months) in only 3 countries, which also were with the highest GDP per capita.

      We found that the availability in primary care facilities, in the public sector, of basic diagnosis equipment, protocols, and medicines for CVD risk management was strongly correlated with a country’s wealth (i.e. GDP per capita), consistent with other studies [15]. An adequate availability of these items was found only in the three countries which had a GDP per capita

      > Int$10’000. According to these findings, only a few countries in the region would meet the WHO target of 80% availability of essential medicines. Of note, availability of diagnostic material, protocols and medicines does not imply that eligible patients are adequately treated but this study did not examine this ques- tion.

      The availability of diagnosis equipment in health facilities is a necessary, although not sufficient, condition for management of CVD risk. Basic diagnosis material, such as blood pressure de- vices or glucometers were available in a majority of health facili- ties, which is reassuring. However, a large cuff was not available in many health centers, including in countries where all medi- cines were available, which makes diagnosis of hypertension in obese persons difficult or biased [16]. Of note, the study did not assess whether sufficient amounts of batteries were available for electronic blood pressure devices or sufficient amounts of strips for glucometers.

      Similarly, protocols for CVD risk management (including for hypertension and diabetes) were not consistently available in several countries, which generally were the same countries in which basic diagnostic devices and medications for CVD reduc- tion were not also consistently available. While some countries may face difficulty to develop their own national guidelines or protocols for CVD risk reduction, WHO has developed the PEN protocols [9], and more recently the HEART package [16], which are designed to provide guidance for diagnosis and manage- ment of CVD risk, including diabetes and other chronic condi- tions.

      Several classes of the main medicines for hypertension treat- ment were only consistently available in the 3-4 countries with the highest GDP per capita. This suggests that CVD risk man- agement has not been prioritized and integrated into primary health care in these countries, despite the fact hypertensive or diabetic patients were reported to attend health care services in all countries, which is consistent with the high prevalence of these conditions in all countries. This calls for strengthening the health care systems toward management of NCDs and CVDs, and particularly at the primary health care level. Lifelong treat- ment for hypertension or diabetes incurs substantial costs to both the health systems and the patients, but some medicines from all antihypertensive and antidiabetic main classes are now available under generic low-cost forms (e.g. < Int$ 0.1 per daily dose) and procurement can be optimized by different means, in- cluding pooled purchasing and purchasing by generic name [12].

      Inefficient management of CVD risk is also apparent from our data. For example, medicines for hypertension and diabe- tes were prescribed at short intervals (e.g. every 1-2 months) in several countries, particularly in countries where only few med- ications were available in health facilities. A high frequency of visits, when applied for patients with stable conditions, increas- es workload of health professionals and loss to follow up chron- ic patients who often cannot afford frequent visits and subse- quent loss of wages on these days, particularly if waiting time in health facilities is long. Efficiency of CVD risk management can be improved by spacing out appointments to every 3-6 months for patients with stable conditions; setting fixed time appoint- ments to minimize waiting time; relying on good information systems (e.g. forms for CVD patients or electronic medical files); and sharing tasks (nurses can attend stable patients) [15] Coun- tries that do not have yet integrated hypertension and diabetes in routine health may consider entry points through programs for other diseases. Malawi has successfully initiated hyperten- sion and diabetes control through HIV programs [17].

      Robertson et al have called for standardized monitoring of availability of medicines [18]. They found that medicines avail- ability differed between 3 facility surveys made in Tanzania, a rare country where several facility-based survey had been done (with the help of donors), but the overall pattern was fairly consistent for some medicines. Some protocols for assessment or availability and affordability of medicines exist [19,20], but they require resource intensive data collection with subsequent substantial funding which may not be easily available in low re- source settings. It was also suggested that facility-based surveys of medicines availability should be integrated in routine infor- mation systems to provide a reliable platform to best spearhead progress in improving access to medicines, which is a valid con- cern. Yet, pending optimal surveillance systems are functioning in low-resource settings, rapid, inexpensive and simple surveys (as our survey relying on good will of AHN members) may pro- vide useful information that can assist health policy at coun- try level. For example, our finding that some important hyper- tension medicines were available in nearly none of the survey health facilities in some countries and, inversely, in all health facilities of some other countries likely represents a real situa- tion and estimates might be mot markedly altered if our coun- try surveys had included larger numbers of health facilities in larger country areas.

      There are a few limitations in this study. First, the number of health centers was limited (5-10 per country) and their lo- cations limited to a main urban center and nearby areas. The finding may therefore not necessarily represent the situation in the whole country. Second, the study only involved health cen- ters in the public sector with no comparison with other types of facilities (e.g. hospital, dispensary) or private premises. Third, some results (e.g. numbers of hypertensive or diabetic patients seen every day) relied on estimates made by the referents. Fourth, the study only assessed the presence of medicines but not amounts available. Fifth, the study did not assess treatment costs incurred to the patients. In the Seychelles, medicines are supplied at no cost to all inhabitants but patients must pay part or all of their medication cost in several other countries. Finally, the study did not assess whether treatment was given adequately to patients, e.g. according to CVD risk. One may ex- pect some inadequacy, particularly in countries where proto- cols were not available. The study also has several strengths. It is based on same methods in the 9 countries, allowing direct comparison of the point-in-time findings. Importantly, several variables were assessed de visu in this facility survey.

      In conclusion, we found that availability of essential diagno- sis tools, guidelines and medications in primary health care fa- cilities in the public sector was poor in several countries in the African region but adequate in the few countries with a higher GDP per capita. Although our findings are based on small sam- ple sizes, it is unlikely that our overall findings would be mark- edly altered had our survey included more health facilities in each country. The findings may provide useful information on gaps in availability of basic equipment and essential medicines in primary health facilities in the public sector in the region. More generally, it would be important that monitoring of avail- ability and affordability, through standardized facility surveys, are regularly conducted and integrated in national health infor- mation systems to spearhead efforts to improve access to CVD and other medicines. In addition, studies should examine fac- tors associated with availability and affordability of medicines in the region.

      Survey officers benefited from small grants from the African Heart Network to cover for transportation.

      The authors thank Bola Ojo and Christelle Crickmore, past executive secretaries, and professor Pamela Naido, current president of AHN.