Current status on health sciences research productivity pertaining to Angola up to 2014

Searching the BVS on June 8, 2014, with the keyword “Angola”, showed 1,029 results,
of which 74.6 % were also in MEDLINE/PubMed database. Consequently, it was considered
that further search in MEDLINE/PubMed would produce representative research publication
data on health in Angola. Therefore, we undertook a further search for relevant Angolan
publications in MEDLINE/PubMed, keeping the keyword “Angola”. To describe the number
of publications across time we included all publications retrieved from BVS without
the abstract filter. All other variables were taken into account using the availability
of an abstract as the main inclusion criteria, and this yielded 658 abstracts. Applying
the exclusion criteria we selected 301 abstracts for further description.

Publication rate

As shown in Fig. 1, the publications in MEDLINE/PubMed addressing health issues related to Angola date
back to 1909. MEDLINE/PubMed, the first bibliographic database in the life sciences,
with a focus on biomedicine, mainly covers the literature published from 1946 onwards,
but it also includes more dated publications 22]. There has been a remarkable, albeit irregular, increase in the number of publications
during the 1990s, with a more sustained increase from 2004 onwards, and a peak in
2013. From 1979 to 2003 (34 years) there were 62 publications, but in the last 10 years
(2004–2013) the number increased four-fold (n?=?232), suggesting a rampant increment
in publication rate (Fig. 1). Additionally, it is noteworthy that the number of publications by the beginning
of June 2014 already reached half the figure recorded in 2013, supporting the notion
that Angola is motivating a steady increase in health-related scientific publications.
This exponential trend was also observed in Palestine, a war zone, where a total of
770 publications were retrieved in the medical and biomedical field across a 10-year
period (01 January 2002 to 31 December 2011), averaging approximately 80 articles
per year. Interestingly, the number of publications has also increased four-fold during
the period 2002–2011, with a stabilization in the last 3 years of the study period
23].

Fig. 1. Between 1909 and 1973, none of the publications deposited in MEDLINE/PubMed up to
June 8, 2014, complied with the inclusion criteria for abstracts about health research
in Angola

A study published in 2007, which analyzed the geography of the PubMed biomedical publications
in Africa between 1996 and 2005, concluded that the contribution of Africa, in particular,
was considerably less than that of other continents. Indeed, it was evident that there
was a continuous increase in publication output during this period in all African
sub-regions, although Angola was located in the lowest quintile with less than two
publications per year 17]. In the present study, we found, in the same period, five publications that included
researchers that were affiliated with Angolan institutions. However, this did not
change the lower position of Angola. Another publication about the scientific production
in Public Health and Epidemiology in the WHO African region, in the period 1991–2010,
obtained very similar results for Angola, placing it in the lowest quintile, with
11–50 publications 18]. This increase in health research in Angola matches with a recent study about the
WHO African Region between 2000 and 2014 24]. In this study, Angola was located in quintile three, with 100 to 499 articles.

Health research topics

Malaria was the disease that stood out, with 45 publications, followed by HIV infection
(n?=?26), trypanosomiasis (n?=?24), and themes of epidemiology/public health (n?=?24). Tuberculosis occupied the tenth position with nine papers (Fig. 2). As a whole, infectious diseases was a topic of 59 % of the publications under analysis
(n?=?178).

Fig. 2. Publications are deposited in MEDLINE/PubMed up to June 8, 2014

The most investigated topics are aligned with the epidemiological profile of Angola,
regarding the main causes of morbidity and mortality (communicable diseases), with
a large emphasis on malaria 25]. Interestingly, there was a disproportionate volume of publications on HIV/AIDS relative
to tuberculosis, which is unusual given the frequent association of both morbidities
26]. In an assessment of scientific output in public health and epidemiology in Africa,
in the period 1991–2010, HIV/AIDS infection was the predominant topic (11.3 %) while
malaria accounted for 8.6 % and tuberculosis for 7.1 % 16]. This analysis also considers the issue of the importance of research funding concerning
infectious diseases 27]. The trend towards epidemiological transition in sub-Saharan Africa, owing to the
increase of chronic non-communicable diseases, is not yet reflected at the scientific
publication level, neither in the present study nor in the aforementioned African
publications 16], 28]–30].

Authorship and affiliation

The primary affiliation of the first author was most frequently in an Angolan institution
(45 publications) (Fig. 3). A total of 150 institutions were listed as the first author’s primary affiliation,
revealing a high institutional dispersion; 14.7 % were Angolan institutions. Furthermore,
65 % of publications represented research conducted in universities and institutes
or research centres (Fig. 4). The majority of publications by first author affiliation country and the number
of academic research institutions were located in Portugal, United States of America,
or Brazil (Fig. 5). The high number of publications with first author affiliation in Angola denotes
the concern of researchers in linking their work to institutions from the country
of study. Nevertheless, this involvement is dispersed across many different institutions,
many of which have relatively low track records in publications, namely hospitals
and non-governmental organizations. Furthermore, academic research in universities
and institutes (or research centres) related to institutional affiliations abroad
indicate the need for reinforcing the academic research capacity in Angola.

Fig. 3. Country denotes the location in which the primary institution of the first author
affiliation was based in publications up to June 8, 2014

Fig. 4. Publications deposited in MEDLINE/PubMed up to June 8, 2014

Fig. 5. Country denotes the location in which the primary institution of the first author
affiliation was based concerning publications on health research in Angola deposited
in MEDLINE/PubMed up to June 8, 2014

Publications by Angolan researchers

An Angolan researcher was the first author in 58 (19 %) abstracts. The frequency of
Angolan first authors or co-authors increased when the first author was affiliated
to an Angolan institution (Fig. 6). In the present study, it is noteworthy that one-fifth of all publications had an
Angolan as first author. The number of African first authors on publications relating
to Africa between 1991 and 2010 has been increasing 16]. This finding was reinforced in the period between 2000 and 2014 in the WHO African
Region, and it was found that the contributions of first authors from Africa doubled
in this period, although it is recognized that it is still minimal (0.7 % in 2000
and 1.3 % in 2014) 24].

Fig. 6. Publications on health research in Angola deposited in MEDLINE/PubMed up to June 8,
2014, according to country in which the primary institution of the first author affiliation
was based

Scientific fields

Epidemiological research was by far the most common research area (n?=?165) (Fig. 7). Although a few population genetics studies (data not shown) are included in this
topic, this finding is in accordance to the burden of infectious diseases in the country.
However, the increasing frequency of non-communicable diseases in sub-Saharan Africa
is likely to be a matter of concern for research in Africa in coming years.

Fig. 7. Publications deposited in MEDLINE/PubMed up to June 8, 2014

On the other hand, a small number of publications (n?=?32; 10.6 %) on socioeconomic and professional research, including governance, health
policies, health systems management, and human resources is noted. Considering that
Angola was a war zone for more than 50 % of its existence as an independent state,
we might have expected a larger number of research publications on relevant topics
such as the socioeconomic determinants of health (e.g., unemployment, poverty, education,
family dissolution, and lifestyles) and the performance of health unities and health
systems. However, we recognize the possibility that studies concerning some of those
socioeconomic determinants of health may be part of research in social and economic
sciences and, therefore, more accessible in other bibliographic databases. Further,
the lack of studies on human resources in the various health professions, including
studies in medical education, is also noticeable.

We expect that the increasing number of health professional schools in Angola, including
six new public medical schools, particularly since 2009, will raise interest in these
topics, especially if educational curricula at the undergraduate and postgraduate
levels in Angola incorporate more programs focused on improving research skills and
providing on-the-job training in epidemiology 31].

Journals choice

We also found that, despite publications being widely spread in a variety of journals,
11.4 % concentrated in PLoS ONE and Malaria Journal, both of which are free access journals indexed by common databases (Fig. 8). This finding is not unusual given that malaria was the most common research topic
among Angolan publications observed in the present study. Notwithstanding the debate
around open access publishing, our findings reinforce the importance of providing
readers in financially disadvantaged countries with a wide range of relevant research
and information 32], 33].

Fig. 8. Publications on health research in Angola deposited in MEDLINE/PubMed up to June 8,
2014

Limitations of the study

The fact that Angola is a Portuguese speaking country and the majority of the journals
indexed by MEDLINE/PubMed are in English may have introduced a selection bias due
to idiom barriers. Though MEDLINE/PubMed represented the vast majority of the publications
of the study and the search began at BVS, which included papers published in journals
indexed by Scielo, it does not embody all the scientific and biomedical journals.
Furthermore, we did not search “grey literature”. Thus, the number of Angolan authors
and coauthors might be underestimated. Nevertheless, Angola still does not have an
exhaustive electronic database of researchers containing individual publication records.
In addition, the incorrect reporting of authors’ nationality and affiliation, observed
in a few papers in the present study, may further contribute to underestimation of
Angolan-authored publications. Finally, the study did not include the analysis of
journal impact factors and the authors did not have access to any citation index database,
which would allow the performance of bibliometric citation analyses.