Fifty Years of Global Cardiovascular Research in Africa: A Scientometric Analysis, 1971 to 2021

Abstract

Background

To analyze the quantity and impact of cardiovascular research done in Africa or coauthored by researchers based in Africa, their determinants, and the patterns of research collaboration.

Methods and Results

We retrieved data from Web of Science and additional sources. We analyzed temporal trends from 1971 to 2021 and geographical distribution, research impact using country‐level h‐index, international research cooperation, and associations of research quantity and quality using linear regression. The annual volume of cardiovascular research from Africa has increased from 4 publications in 1971 to 3867 in 2020 and currently represents ~3% of the global cardiovascular research output. Authors from South Africa (28.1%) and Egypt (24.1%) accounted for more than half of all publications from African countries, and they had the highest h‐index (209 and 111, respectively). Important collaborators outside Africa included the United States, United Kingdom, France, Germany, and Australia. The country's publication count was associated with larger population size (P<0.001), whereas the country's h‐index was associated with larger population size (P=0.001) and higher human development index (P=0.023). International collaboration was dominated by the United States, South Africa, United Kingdom, Egypt, and Canada. The level of collaboration between African countries was lower than their collaboration with non‐African countries.

Conclusions

Cardiovascular research output from African authors remains low, despite marked progress over the past 5 decades. These findings highlight the urgent need to improve the quantity and quality of cardiovascular research in Africa through increased investments, training of human resources, improved infrastructures, and expansion of collaborative research networks, particularly within Africa.

Nonstandard Abbreviations and Acronyms

GDP

gross domestic product

HDI

human development index

Clinical Perspective.

What Is New?

• Cardiovascular research output from Africa has progressed over the past 50 years and currently represents ~3% of the global cardiovascular research output.

• Despite this progress, cardiovascular research output from Africa remains low and out of proportion to the increasing burden of cardiovascular disease on the continent.

• The extent of collaboration within Africa is much lower than the level of collaboration with non‐African countries.

What Are the Clinical Implications?

• Efforts are needed to improve the quantity and impact of cardiovascular research in Africa.

• Increased resources to train cardiovascular health professionals and researchers, build infrastructures, and fund research on cardiovascular disease priorities for Africans are highly needed.

• Collaborative research networks within Africa should be created and expanded.

Africa, home to >1 billion people, is experiencing a surge in the burden of cardiovascular disease (CVD). ¹ In 2019, >1 million deaths were attributable to CVD in sub‐Saharan Africa alone. ² It is estimated that CVD will overtake infectious diseases as the leading cause of death on the continent by 2030. ³ This rise in the prevalence of CVD represents a significant health and socioeconomic challenge for African countries. Health systems in Africa, especially in sub‐Saharan Africa, which are still highly burdened by communicable, maternal, neonatal, and nutritional diseases, ¹ are ill‐prepared to cope with the CVD epidemic. ⁴ High‐quality research is crucial to inform evidence‐based strategies to curb the burden of CVD in Africa.

Although Africa carries the largest proportion of the global burden of disease, ¹ it has the lowest contributions to medical progress as judged from biomedical publications. ⁵ , ⁶ There are several potential reasons for the limited health research output in African countries, including a limited number of trained researchers; scarce and poorly‐equipped research institutions; dearth of financial resources, investments, and incentives to pursue research; and a weak political willingness to build research capacities.

Over the past few decades, there has been a sharp increase in international research collaborations, because connecting with colleagues across geographic boundaries has become much easier than ever before. ⁷ These collaborations are an effective way to share and advance knowledge. In African countries, international collaborations, especially with overseas colleagues, represent an opportunity to overcome several local challenges by bringing research funding and sharing expertise. Researchers based in Africa also contribute to a global research agenda beyond their local health priorities. A previous report showed some progress in cardiovascular research productivity from African countries between 1999 and 2008, with the largest number of publications coming from South Africa, Egypt, Tunisia, and Nigeria. ⁸ However, the current contribution of Africa to the global cardiovascular research and its determinants are unknown. Therefore, this study aimed to (1) analyze the trends and geographic distribution of articles in cardiovascular research with contributions from Africa, either conducted in Africa or coauthored by researchers from Africa; (2) determine the factors associated with the quality and quantity of cardiovascular research output across countries; and (3) examine the patterns of research collaborations between researchers based in Africa and those outside Africa over the past 50 years.

METHODS

Availability of Data and Material

Most data generated or analyzed in this study are included in this article and its supplemental material. Additional information can be provided upon request to the corresponding author.

Registration and Ethical Approval

This project was registered with PROSPERO (international prospective register of systematic reviews, www.crd.york.ac.uk/prospero) (CRD42021260613). Because the project was based on published data, ethical approval was not sought nor required.

Data Search

We aimed to include all articles on CVD either from studies conducted in Africa or authored by researchers based in Africa, published between 1971 and 2021, and excluding animal studies. We searched Web of Science (Clarivate Analytics, Philadelphia, PA) to identify all relevant articles. Search strategies were developed based on terms related to the broad spectrum of CVD, the cardiovascular risk factors hypertension and dyslipidemia, and an African filter including the names of all African countries (Tables S1 through S3). Risk factors such as obesity, diabetes, or physical inactivity were not included because they are not specific to cardiovascular research and therefore could have reduced the precision of the searches. Searches were performed on September 16, 2021.

Data Acquisition and Management

Data were extracted from Web of Science and processed automatically using the web application SciPE (Science Performance Evaluation; Saarland University, Saarbrücken, Germany), which performs various scientometric analyses specified by the user, as described previously. ⁹ These data were used to set up an undirected multipartite graph with distinct partitioned sets of nodes, including countries, institutions, publications (categorized as general authorship, first authorship, and last authorship), citations, country‐ and author‐level Hirsch (h)‐indices, authors' first and last names, and authors' sex. First and last authorships are related to the first and last author on a article, whereas general authorship corresponds to any position as coauthor. Additional information for each African country was extracted manually from various sources. Data on the human development index (HDI) for the year 2020 were obtained from the United Nations Development Programme. ¹⁰ Data on adult literacy rate (percent of people aged 15 years and above) and physicians per 1000 people for the most recent available year, gross domestic product (GDP) per capita for the year 2020, health expenditure per capita for the most recent year, and total population for the year 2020 were obtained from the World Bank, ¹¹ the number of universities in the country from UniRank, ¹² and CVD mortality rate for 2019 from the Global Burden of Disease. ¹³

The h‐index, coined by the physicist Jorge E. Hirsch in 2005, ¹⁴ was used to measure the productivity and citation impact of the publications of authors and countries. It was calculated as the number of articles h published by a researcher (for author's h‐index), or by researchers in a specific country (for country's h‐index), that have at least h citations each. International collaborations were analyzed between the first author's country and the countries of all coauthors. Each country that was different from the first author's country accounted for 1 collaboration and was visualized in the chord diagram.

We examined the representation of selected cardiovascular research domains in the overall research output. These domains included specifically CVDs such as coronary artery disease, cardiomyopathies, endocarditis, pericardial disease, rheumatic heart disease, nonrheumatic valvular disease, congenital heart disease, arrhythmias, heart failure, pulmonary hypertension, peripheral artery disease, venous thromboembolism, and cerebrovascular disease. For each of these domains, we conducted a specific search on Web of Science (Table S3). Cardiovascular risk factors, such as hypertension, were not included as specific domains. The proportion of each domain was calculated as its specific search yield divided by the sum of yields of all domains and expressed in percentage.

We analyzed the characteristics of the top 200 CVD researchers from Africa. Several authors had publications with different names from various combinations of their first names, middle names, and last names. Such authors had their publications summed up and reported under a single name, which was performed manually by 1 author (J.J.N.). This could not be done for the h‐index because of the potential inaccuracy that can emerge when attempting to resolve multiple h‐indexes attributed to several names of the same author. For this reason, authors were ranked based on the total number of publications rather than the h‐index.

Statistical Analysis

Categorical variables were summarized using frequency and percentage, and quantitative variables using median and interquartile range (IQR). Linear regression analysis was used to investigate the factors associated with each African country's number of publications and country's h‐index (based on general authorship). Explanatory variables included the following country characteristics were: population size, number of universities, literacy rate, HDI, GDP per capita, health expenditure per capita, number of physicians per population, and CVD mortality rate. The strength of unadjusted and adjusted associations was measured with the β coefficient reported with a 95% CI. All variables were included in the multivariable regression model, except GDP per capita, literacy rate, and health expenditure per capita because of collinearity with HDI. These variables are represented in the HDI, which is a composite of index of life expectancy, education, and per capita income indicators. A 2‐sided P value <0.05 was considered statistically significant without correction for multiple testing. A complete case analysis was performed. Data were analyzed using IBM SPSS Statistics version 27.0 (Armonk, NY). Figures were generated using SciPE and Microsoft Excel version 2019 for Windows (Redmond, WA).

RESULTS

Overall Output From Africa

We identified 35 368 publications in cardiovascular research done in Africa or coauthored by researchers based in Africa, representing 1.7% of the global cardiovascular research output from 1971 to 2021. The annual number of publications from Africa has increased over time from 4 publications in 1971 to 3867 in 2020 (Figure 1). The proportion of global CVD publications from Africa has also risen from 0.1% in 1971 to 3.1% in 2021, with an accelerated trend in the past decade (Figure S1).

Figure 1. Temporal trends in cardiovascular research output in Africa and worldwide.

Research Domains

The most common publications were related to coronary artery disease (22.6%), heart failure (19.1%), cerebrovascular disease (11.9%), cardiomyopathies (8.3%), arrhythmias (7.7%), and venous thromboembolism (7.0%) (Figure 2).

Figure 2. Distribution of publications among selected research domains for cardiovascular diseases.

Nations' Publication Quantity

Cardiovascular research from Africa was authored by researchers both within and outside of Africa. The leading contributing countries by number of publications based on general authorship were South Africa (n=9055), Egypt (n=7777), United States (n=5559), United Kingdom (n=3900), Nigeria (n=2824), Tunisia (n=2352), France (n=2041), Germany (n=1591), Morocco (n=1510), and Australia (n=1495) (Figure 3A). The overall pattern was similar for publications based on first and last authorships (Figures S2 and S3). In regard to specifically African countries, South Africa (28.1%) and Egypt (24.1%) accounted for more than one‐half of the total number of publications based on general authorship, followed by Nigeria (8.8%), Tunisia (7.3%), and Morocco (4.7%), which collectively accounted for nearly three‐quarters of this total (Table 1 and Table S4). This distribution was similar for publications based on first authorship and last authorship (Table S5). At the regional level, most publications were from Northern Africa (42.3%) and Southern Africa (28.8%) (Table S6). When considering country's population size, Seychelles (n=731), Tunisia (n=199), South Africa (n=153), Mauritius (n=78), and Egypt (n=76) had the highest number of publications (based on general authorship) per 1 million population (Table S7).

Figure 3. Geographical distribution of cardiovascular publications involving Africa (Panel A) and related country‐level h‐index (Panel B).

Table 1.

African Countries Ranked by the Total Number of Publications

Country	Publications	Country‐level h‐index	Citations	Universities	Population	GDP per capita	HDI	Literacy rate (%)	Physicians per 1000 population	Health expenditure per capita	CVD death rate
South Africa	9055	209	94 404	25	59 308 690	5090.7	0.709	87	0.9	525.96	148.7
Egypt	7777	111	35 449	54	102 334 400	3547.9	0.707	71	0.79	125.55	266.35
Nigeria	2824	87	15 310	160	206 139 590	2097.1	0.539	62	0.4	83.75	76.11
Tunisia	2352	63	11 885	71	11 818 620	3319.8	0.740	79	1.27	251.55	302.52
Morocco	1510	50	6543	37	36 910 560	3009.2	0.686	74	0.73	174.78	325.53
Ethiopia	984	55	4419	36	114 963 580	936.3	0.485	52	0.1	24.23	71.54
Ghana	829	63	3619	67	31 072 940	2328.5	0.611	79	0.18	77.91	125.73
Kenya	738	72	3211	62	53 771 300	1838.2	0.601	82	0.2	88.39	80.85
Algeria	679	47	2384	91	43 851 040	3310.4	0.748	81	1.83	255.87	234.02
Cameroon	673	61	3612	15	26 545 860	1499.4	0.563	77	0.08	54.14	91.03
Uganda	667	59	3036	47	45 741 000	817	0.544	77	0.09	43.14	68.46
Tanzania	519	55	2302	32	59 734 210	1076.5	0.529	78	0.04	36.82	96.04
Sudan	396	32	1011	52	43 849 270	595.5	0.510	61	0.41	60.17	182.24
Mozambique	316	52	878	11	31 255 440	448.6	0.456	61	0.07	40.26	107.33
Senegal	266	26	862	28	16 743 930	1487.8	0.512	52	0.07	58.9	105.6
Malawi	229	35	1142	22	19 129 960	625.3	0.483	62	0.02	35.5	84.86
Zimbabwe	226	47	838	18	14 862 930	1128.2	0.571	89	0.08	140.32	118.65
DRC	192	27	896	55	89 561 404	556.8	0.480	77	0.074	18.51	102.55
Burkina Faso	162	20	381	17	20 903 280	830.9	0.452	41	0.06	40.25	97.62
Rwanda	158	32	708	26	12 952 210	797.9	0.543	73	0.14	58.31	93.2
Benin	151	34	254	26	12 123 200	1291	0.545	42	0.16	30.94	90.15
Zambia	138	27	323	50	18 383 960	1050.9	0.584	87	0.09	75.99	99.56
Cote D'Ivoire	125	22	311	14	2 637 828	2325.7	0.538	47	0.23	71.88	85.5
Botswana	122	23	423	15	2 351 630	6711	0.735	87	0.37	482.96	150.47
Libya	111	23	549	29	6 871 290	3699.2	0.724	86	2.16	309.88	189.26
Guinea	105	20	687	1	13 132 790	11 940	0.477	32	0.08	38.32	115.06
Mauritius	99	32	1993	5	1 265 740	8622.7	0.804	91	2.25	653.35	259.2
Congo	85	16	79	1	5 518 090	1972.5	0.574	80	0.165	47.52	134.76
Gambia	84	26	533	2	2 416 660	787	0.496	51	0.11	22.16	115.92
Angola	82	13	218	18	32 866 270	1895.8	0.581	66	0.21	87.62	85.35
Togo	73	16	109	4	8 278 740	915	0.515	64	0.05	41.84	105.55
Seychelles	72	25	128	1	98 460	11425.1	0.796	96	0.95	833.08	236.72
Mali	60	17	79	10	20 250 830	858.9	0.434	35	0.14	34.95	89.06
Namibia	55	16	56	3	2 540 920	4211.1	0.646	92	0.6	471.49	155.92
Niger	51	14	155	2	24 206 640	565.1	0.394	35	0.05	30.36	73.84
Madagascar	45	8	109	6	27 691 020	465.5	0.528	75	0.18	22.05	132.62
Gabon	33	11	37	23	2 225 730	7005.9	0.703	85	0.36	218.37	145.21
Sierra Leone	33	10	28	7	7 976 980	484.5	0.452	43	0.02	85.78	108.74
Eritrea	21	10	174	7	3 213 970	642.5	0.459	77	0.06	23.79	99.24
Burundi	18	5	6	21	11 890 780	274	0.433	68	0.05	24.03	90.52
Liberia	17	9	4	4	5 057 680	583.3	0.480	48	0.04	45.42	89.76
Mauritania	17	7	0	4	4 649 660	1672.9	0.546	53	0.2	54.49	98.44
CAR	14	5	13	1	4 829 760	476.9	0.397	37	0.06	53.66	136.03
Swaziland	13	5	0	5	1 160 160	3415.5	0.611	88	0.08	271.14	132.97
Cape Verde	12	4	2	9	555 990	3064.3	0.665	87	0.82	194.86	190.48
Somalia	11	4	29	33	15 893 220	309.4	NA	NA	0.02	NA	91.04
Chad	10	4	14	9	16 425 860	614.5	0.398	22	0.05	29.24	84.38
Lesotho	10	4	29	2	2 142 250	861	0.527	77	0.1	124.79	192.58
Comoros	7	3	0	NA	869 600	1402.6	0.554	59	0.27	65.23	159.96
Djibouti	2	1	0	1	988 000	3425.5	0.524	NA	0.22	70.86	103.11

HDI is a composite index of life expectancy, education, and per capita income indicators. CAR indicates Central African Republic; CVD, cardiovascular disease (deaths per 100 000 people); DRC, Democratic Republic of Congo; GDP, gross domestic product (per capita in US dollars); HDI, human development index (health expenditure per capita in US dollars); and NA, not available.

Nations' Publication Quality

Countries with the highest h‐index based on general authorship were the United States (h=223), South Africa (h=209), United Kingdom (h=209), Germany (h=162), Canada (h=162), Australia (h=160), Italy (h=154), France (h=144), and the Netherlands (h=142) (Figure 3B). The overall picture was similar for the h‐indices based on first and last authorship publications (Figures S4 and S5). In regard to African countries, the top 5 with the highest h‐index were South Africa (h=209), Egypt (h=111), Nigeria (h=87), Kenya (h=72), and Tunisia (h=63) (Table S4). When adjusting for population size, Seychelles (h=254), Mauritius (h=25), The Gambia (h=11), Botswana (h=10), and Ivory Coast (h=8) had the highest h‐index (based on general authorship) per million population (Table S7).

Factors Associated With Research Quantity and Quality Among African Countries

Multiple linear regression analysis revealed that every 1 million or larger population size was associated with 37.9 (95% CI, 19.3–56.4; P<0.001) more publications in the country's output. Furthermore, every 1 million or larger population size and every unit in HDI were associated with 0.6 (95% CI, 0.2–1.0; P=0.001) and 152.0 (95% CI, 22.2–281.7; P=0.023) higher country‐level h‐index, respectively. The number of universities, literacy rate, GDP per capita, health expenditure per capita, number of physicians per population, and cardiovascular disease mortality were not associated with either a country's number of publications or its h‐index (Table 2).

Table 2.

Factors Associated With the Country's Number of Publications and h‐Index Based on General Authorship Among African Countries

Factors	Country's no. of publications				Country's h‐index
	Unadjusted β coefficient (95% CI)	P value	Adjusted* β coefficient (95% CI)	P value	Unadjusted β coefficient (95% CI)	P value	Adjusted* β coefficient (95% CI)	P value
Population	22.98 (11.34 to 34.62)	<0.0001	37.85 (19.26 to 56.44)	<0.0001	0.54 (0.31 to 0.77)	<0.0001	0.63 (0.26 to 1.00)	0.001
GDP per capita	0.09 (−0.09 to 0.28)	0.319			0.002 (−0.002 to 0.005)	0.386
HDI	5396.31 (971.65 to 9820.97)	0.018	4983.19 (−1477.64 to 11444.03)	0.127	113.77 (23.16 to 20 4.37)	0.015	152.01 (22.24 to 281.78)	0.023
Literacy rate, %	15.35 (−11.43 to 42.12)	0.255			0.49 (−0.05 to 1.02)	0.074
No. of universities	20.43 (4.34 to 36.53)	0.014	−21.91 (−45.97 to 2.16)	0.73	0.57 (0.26 to 0.88)	0.001	−0.16 (−0.64 to 0.33)	0.520
No. of physicians	0.96 (−0.03 to 1.95)	0.056	−0.17 (−1.45 to 1.12)	0.792	0.02 (−0.003 to 0.038)	0.098	−0.001 (−0.027 to 0.025)	0.922
Health expenditure per capita	2.18 (−0.61 to 4.96)	0.122			0.05 (−0.01 to 0.10)	0.111
Cardiovascular disease mortality	8.45 (0.80 to 16.09)	0.031	6.48 (−3.59 to 16.55)	0.201	0.10 (−0.07 to 0.26)	0.231	−0.04 (−0.25 to 0.16)	0.671

Number of publications and h‐index are based on general authorship. Cardiovascular disease mortality represents the number of deaths per 100 000 people. Number of physicians per 100 000 people. Population in millions of people. GDP indicates gross domestic product (per capita in US dollars); and HDI, human development index (health expenditure per capita in US dollars).

^*GDP per capital, literacy rate, and health expenditure per capita were not included in the multiple linear regression because of collinearity with HDI. Only HDI was included.

International Collaborations

The chord diagrams (Figure 4A and 4B) depict the collaborations between the country of the first author and the countries of all coauthors. The width of each connecting line is proportional to the cooperation in terms of collaborative publications with the connected country. Among publications from research done in Africa or coauthored by researchers from Africa, the United States, South Africa, the United Kingdom, Egypt, and Canada had the most intensive research collaboration (Figure 4A). The level of international collaboration between African countries was lower than their collaboration with non‐African countries (Figure 4A). South Africa, Nigeria, and Cameroon had the highest level of international collaboration among African countries (Figure 4B).

Figure 4. Network of cardiovascular research collaboration among African countries and with non‐African countries (Panel A) and among only African countries (Panel B).

Characteristics of the Most Prolific Authors

The total number of publications (general authorship) of the top 200 researchers from Africa ranged from 26 to 376, with a median of 45.0 (IQR, 32.0–66.8). Men represented 72.5% (n=145) of the top researchers. They were mostly from South Africa (51.5%, n=103) and Tunisia (19.0%, n=38). The University of Cape Town (15.0%, n=30), University of the Witwatersrand (13.5%, n=27), North‐West University (7.0%, n=14) in South Africa, and the University of Monastir in Tunisia (6.5%, n=13) had the highest number of top researchers (Table S8).

DISCUSSION

This study analyzed the quantity and impact of cardiovascular research from Africa between 1971 and 2021, their determinants, and the patterns of research collaboration. We observed that cardiovascular research done in Africa or coauthored by researchers based in Africa has progressed over the past 50 years and currently represents ~3% of the global cardiovascular research output. South Africa and Egypt accounted for more than half of the total number of publications from African countries, and their research had the highest impact. The quantity and impact of research positively correlated with the country's population size and HDI. The level of international collaboration among African countries was lower than their collaboration with non‐African countries.

Our study confirms and extends the data from 2 previous studies that analyzed data from 1999 to 2008 and showed a steady progression in cardiovascular research productivity in Africa ⁸ and globally. ⁶ We observed an accelerated publication output in the past decade, with the total number of publications that has quadrupled between 2010 and 2020. Similarly, the proportion of global CVD publications from Africa has risen from 0.1% in 1971 to 3.1% in 2021, with a substantially higher increase in the past decade. However, the cardiovascular research output from Africa remains low, especially in view of the current and projected CVD burden on the continent. Furthermore, this research output is highly skewed, with 5 countries (South Africa, Egypt, Nigeria, Tunisia, and Morocco) accounting for nearly three‐quarters of the publication share in Africa. These disparities are not specific to cardiovascular research. Similar pictures were reported in an analysis of 1996 to 2005 PubMed articles from Africa ¹⁵ and in a bibliographic study of nephrology research in Africa between 1960 and 2017. ¹⁶

Differences in population size are the major reasons for the differences in cardiovascular research output in Africa. Highly populated countries like Nigeria or Egypt have high publication counts. Linear regression analysis revealed that a country's population size has a substantial positive correlation not only with research quantity, but also with impact. Countries with larger populations are more likely to have a larger research workforce. We observed that HDI was associated with higher research impact, whereas literacy rate was associated with neither research quantity nor impact. HDI is a summary measure of average achievement in key dimensions of human development, including a long and healthy life, being knowledgeable, and having a decent standard of living. ¹⁷ Our findings highlight the importance of overall human development beyond education as a determinant of scientific research output. Furthermore, the number of universities within countries was not associated with research quantity and quality. This suggests that the quality of universities or research institutions might have higher impact on countries' research productivity than their quantity. The African countries with the highest number of publications and h‐index, especially South Africa and Egypt, have the highest‐rated universities on the continent (https://www.topuniversities.com/). More than one‐quarter of the most prolific cardiovascular researchers in Africa were affiliated with the University of Cape Town and University of the Witwatersrand (South Africa), the 2 most highly rated universities in Africa. ¹⁸ South Africa has had increasing investments in biomedical research, ¹⁹ and research infrastructure in the country is highly supported by universities, the National Research Foundation of South Africa, and the Medical Research Council, as well as by private companies and foundations. ²⁰ The country has established institutes dedicated to cardiovascular research, such as the Hatter Institute for Cardiovascular Research in Africa, located at the University of Cape Town, and several programs to address the rising burden of CVD have been developed. ²⁰

GDP per capita has been shown to be a major determinant of biomedical research productivity. ²¹ However, GDP per capita and health expenditure per capita were not associated with research quantity or impact in the current study, in keeping with another bibliographic analysis focusing on nephrology research. ¹⁶ This suggests that, in Africa, greater national economic wealth does not necessarily translate into investment in health and in medical research, though it is possible that there is a threshold or lag effect. National investment in health research has historically been low in most African countries. ¹⁹ In 2005, 15 years after the call of the Commission on Health Research for Development to governments in developing countries to devote 1% of national GDP to research, only South Africa had almost met this target. ¹⁹ In 2001, African Union heads of state committed in the Abuja Declaration to devote at least 15% of their annual national budget to the health sector. ²² Although many African countries have marginally increased health spending overall, in 2018, average government health spending hovered around 7.2% of the national budget, less than half of the target set in Abuja, with only 2 countries (Madagascar and Zimbabwe) having met the target. ²³

Expenditure on health research in Africa is not only low, but it is also unequally distributed and does not match the burden of disease, ²⁴ with most of the health expenditure allocated to infectious diseases. ²⁵ Because CVD is set to overtake infectious diseases as the leading cause of death in Africa by 2030, ³ there is the need for larger resources to be allocated to cardiovascular research and subsequently health promotion and service provision activities.

This study shows extensive collaborations among African and non‐African countries, with patterns that are consistent with previous studies on international collaborations clusters in biomedical research in Africa. ²⁶ , ²⁷ The top non‐African countries involved in coauthorships in cardiovascular publications with African countries are the United States, United Kingdom, France, Germany, and Australia. These countries are leaders in research globally, especially in the cardiovascular field. ²⁸ Collaboration clustering between African and non‐African countries is highly influenced by colonial languages, as well as historical and political ties. ²⁶ , ²⁷ This was apparent in the current study; there are clusters of collaboration between France and French‐speaking African countries such as Tunisia, Morocco, Algeria, and Cameroon; between the United Kingdom and United States, and English‐speaking African countries such as South Africa, Nigeria, Egypt, Ghana, and Kenya; and between Arab countries, mainly Saudi Arabia and Egypt. Furthermore, countries like the United States and United Kingdom are important sources of public research funding for African countries through the National Institutes of Health, Wellcome Trust, and UK Medical Research Council, as well as private research funding such as the Bill and Melinda Gates Foundation, which has not invested heavily in cardiovascular research to date.

Research collaborations between African countries remains limited. However, there is an increasing number of pan‐African research initiatives. Notable examples include CVD registries launched by the Pan‐African Society, ²⁹ the VALVAFRIC study, ³⁰ the Sub‐Saharan Africa Survey on Heart Failure, ³¹ Pan African Pulmonary Hypertension Cohort study, ³² cardiovascular studies within the H3 (Africa Human Heredity and Health in Africa) initiative, ³³ and Investigation of the Management of Pericarditis and Comparison of Three Combination Therapies in Lowering Blood Pressure in Black Africans trials, ³⁴ , ³⁵ among others. Such pan‐African projects have the potential to provide more robust scientific evidence for diseases that affect the continent, while sharing expertise on how to set up cardiovascular research in low‐resource settings and how to maximize research funding for individual institutions. Recently, African cardiovascular researchers are taking the lead in multicountry global studies that have a focus on Africa but go far beyond the African continent such as the European Cardiac Society Global Peripartum Cardiomyopathy study ³⁶ and the World Heart Federation Global Study on COVID‐19 and Cardiovascular Disease. ³⁷ The May Measurement Month is another initiative in which African researchers have been significantly involved. ³⁸

Brain drain, which is the emigration of skilled nationals, is considered a contributor to low‐quantity and low‐quality research in Africa. It has been estimated that 25% of African physicians and 10% of African nurses work in a high‐income country. ³⁹ Although brain drain remains a major challenge for developing sustainable research programs in Africa, this can be turned into an opportunity. ¹⁶ Cardiovascular researchers who have left Africa can be and have been recruited and incentivized to build partnerships between their host institutions abroad and African institutions for research training, collaborative research projects, joint applications for research grants, and knowledge sharing through scientific meetings.

Our study has some limitations. First, the analysis included only publications in journals indexed on Web of Science; therefore, publications in local African journals may be underrepresented. Second, the web application SciPE used to extract and process data from Web of Science could not appropriately differentiate some countries with similar names such as Sudan and South Sudan, or Guinea, Guinea‐Bissau, and Equatorial Guinea. However, the nonrepresentation of some of these countries likely had a limited impact on the overall estimates. South Sudan is a young country (founded in 2011), and Guinea‐Bissau and Equatorial Guinea are small countries with population sizes <4 million combined (https://data.worldbank.org/) and with low research output as previously reported. ⁸ Despite these limitations, the current study provides the most extensive and up‐to‐date report on cardiovascular research in Africa over the past 5 decades. Most importantly, this study is unique in that it is not focused on Africa as an isolated entity but presents the place of Africa in the global cardiovascular research community and the extent of contribution from non‐African countries.

CONCLUSIONS

Cardiovascular research output from Africa is rising but remains low and out of proportion to the rising burden of CVD on the continent, despite some progress over the past 50 years. The extent of collaboration within Africa is much lower than the level of collaboration with non‐African countries. These findings advocate for strategies to improve the quantity and impact of cardiovascular research, including increased resources to train cardiovascular health professionals and researchers, build infrastructures, and fund research on CVD priorities for Africans. Creating and expanding collaborative research networks within Africa will be pivotal to improve global cardiovascular health.

Sources of Funding

We acknowledge support by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) and Saarland University within the Open Access Publication Funding programme.

Disclosures

Dr Noubiap is supported by a postgraduate scholarship from the University of Adelaide. In the past 3 years, Dr Huffman has received support from the American Heart Association, Verily, and AstraZeneca for work unrelated to this research. Dr Huffman has planned patents for combination therapy for the treatment of heart failure. The George Institute for Global Health has a patent, license, and has received investment funding with the intent to commercialize fixed‐dose combination therapy through its social enterprise business, George Medicines. The other authors have no conflicts of interest. Mahfoud is supported by Deutsche Gesellschaft für Kardiologie (DGK), Deutsche Forschungsgemeinschaft (SFB TRR219), and Deutsche Herzstiftung. He has received scientific support from Medronic and ReCor Medical and speaker honoraria from Astra‐Zeneca, Bayer, Boehringer Ingelheim, Inari, Medtronic, Merck, and ReCor Medical.

Supporting information

Tables S1–S8