Systematic review of prevalence and risk factors of transfusion transmissible infections among blood donors, and blood safety improvements in Southern Africa

Abstract

Blood and blood products are listed as one of the essential medicines by the World Health Organization (WHO). In addition to inadequate supply, most sub-Saharan Africa (SSA) nations fail to meet their blood needs because many donated blood units are discarded because they are contaminated with transfusion-transmitted infections (TTIs). We sought to estimate the prevalence of TTIs, identify the risk factors for TTIs among blood donors, and identify the efforts and interventions that have been made to improve blood safety in Southern African nations, particularly the nations of the South African Development Community (SADC). We investigated the prevalence and risk factors for TTIs, blood safety interventions, and blood quality improvement in the SADC region from major PubMed/MEDLINE, Cochrane Library, and HINARI databases from 1 January 2011 to 31 April 2021. All investigations followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). In meta-analysis, we estimated the pooled TTIs prevalence and summarised the same using forest plots. A total of 180 articles published from the SSA region were identified covering our three targeted themes: TTI prevalence, risk factors for TTIs, and blood safety improvements. Of these 180 articles, only 27 (15%) focused on the SADC region. The overall pooled TTI prevalence estimate was 2.0% (95% CI: 1.0–3.0) and hepatitis B was the most prevalent TTI in the region (prevalence = 3.0; 95% CI: 2.0–5.0). The prevalence of HIV, HCV, and syphilis was 2.0% (95% CI: 1.0–4.0), 1.0% (95% CI: 0.0–2.0), and 2.0% (95% CI: 0.0–8.0), respectively. In general, replacement donors and first-time donors were more likely to be infected with TTIs than repeat donors. Twelve articles explored blood safety research in the region; however, they vary greatly highlighting the need for consistent and more comprehensive research. Few publications were identified that were from the SADC region, indicating lack of research or resources towards improving both quantity and quality of blood donation. TTI prevalence remains one of the highest in the world and blood safety recommendations vary across the region. More effort should be directed towards developing a cohesive regional blood transfusion policy and effective blood monitoring and evaluation strategies.

1 |. INTRODUCTION

Sufficient supply of high-quality blood and blood products is a crucial aspect of a successful health system and is listed as one of the World Health Organization (WHO)’s recommended essential medicines. A safe and adequate blood supply can alleviate the burden on health systems and economic development caused by critical blood shortages in developing nations.¹ The 16 nations that comprise the Southern African Development Community (SADC) face unique challenges, such as high burden of infectious disease, maternal mortality, and injury and traffic accidents that create a substantial demand for blood. However, lacking infrastructure, inadequate management, and poor quality contribute to chronic blood shortages. This higher demand for blood coupled with lacking supply from ineffective systems affects vulnerable populations the most, such as women with post-partum haemorrhage and children with malaria-induced anaemia.¹ These challenges have led to significant need for blood safety regulation.

Blood services should ideally provide universal access to safe blood through recruitment and retention of blood donors and proper TTI screening of donated blood products. WHO recommends a centralised blood transfusion service instead of fragmented decentralised systems to address blood shortages and increase blood safety. Through establishment of a national blood transfusion service, replacing an in-hospital model, countries can better ensure consistent blood supply and donor pool. For example, in 2003 Malawi established the Malawi Blood Transfusion Service (MBTS) in accordance with WHO recommendations. Recently, MBTS increased annual blood collections by 22% (56 802 in 2015 to 69 148 in 2019). The rate then decreased by 15%, due to the Covid-19 pandemic (58 836 whole blood units in 2020). However, despite being successful in increasing blood collection in the past years, the country remains with a 27% deficit of national demand for blood products.² Even with a centralised WHO-approved transfusion system, a country can still suffer from significant shortages of safe blood highlighting the need to modify blood systems to function effectively in a low-income setting.³

Poor quality and maintenance of donated blood further limits the available usable blood for transfusions. Transfusion-transmitted infections (TTIs) are a major threat to supply since they are unusable and discarded, wasting time and resources. The WHO recommends mandatory screening of all donated blood units for 4 TTIs: human immunodeficiency virus (HIV), hepatitis C virus (HCV), hepatitis B virus (HBV), and syphilis. These represent the most common TTIs and are prioritised for screening, although countries are encouraged to adopt mandatory screening of additional TTIs according to national burden, such as malaria.⁴ Since only four countries have an accredited National Blood Transfusion Service in the SADC region, countries looking to develop blood hemovigilance guidelines can begin with these four TTIs that are most common in the sub-Saharan region and one of the main reasons for discarding blood donations.⁵

One approach to alleviate shortages is to bolster national surveillance of TTI distribution. Frequent TTI monitoring in combination with considering risk factors for TTIs can inform strategic blood collection.^6,7 National blood services can target regions with low TTI prevalence and initiate appropriate interventions in regions with high TTI prevalence. Such hemovigilance will allow for faster and more effective responses to emerging blood safety threats.⁸ In addition, identification of interventions aimed at improving blood quality through TTI prevention will also aid in strengthening the blood supply across Southern Africa. Focusing research on specific economic development blocks such as the SADC region allows for targeted and regionally coordinated policy implementations and goals. Economic regional blocks or zones emerged to progress towards economic integration and regional development and tend to exhibit distinct patterns and culture.

Though these key strategies will aid in future blood safety, the current published literature and knowledge of TTIs and blood quality interventions in the SADC region are lacking. As such, we conducted a systematic review of the literature to assess the prevalence, risk factors of TTIs, and the efforts that have been made to identify interventions for improving blood safety in SADC nations.

2 |. METHODS

2.1 |. Search strategy

This systematic review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was registered in PROSPERO (registration #: CRD42021252371, https:/www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021252371).

We retrieved articles that addressed one or more of our three thematic areas: TTI prevalence (HIV, HBV, HCV, or syphilis), TTI risk factors, TTI blood safety interventions, and blood quality management in sub-Saharan Africa (SSA) from three major databases: PubMed/MEDLINE, Cochrane Library, and HINARI.

We used the following search terms: ‘blood donor’, ‘blood donation’, ‘blood transfusions,’ ‘transfusion transmitted infection’, ‘prevalence’, ‘epidemiology’, ‘data quality improvement’, ‘blood safety’, ‘incidence’, ‘residual risk’, ‘risk factors’, and ‘sub-Saharan Africa’ and the names of all 46 countries of SSA alone or in combination using ‘OR’ and ‘AND’. These search terms were created following the PICOS framework because we wanted to include different study designs. This framework generates a greater number of search hits with greater sensitivity.⁹ The search strategy is available in Appendix A.

The search study was limited from 2011 to 2021 due to recent blood safety improvements in SSA after 2010 which is relevant to our study context.

2.2 |. Study selection

The study selection was conducted in two phases. In the first phase, we screened using a thorough review of the title and abstract, and subsequently reviewed the full-text manuscripts of eligible articles using this criteria.

Studies were eligible if they met the following inclusion criteria: (1) published from 1 January 2011 to 31 April 2021 (start of the review); (2) took place in sub-Saharan Africa; (3) included a blood donor cohort aged 16–65, and (4) covered at least 1 of our 3 topics of interest. There were no language restrictions and we included studies of all study designs to gain an overview of all literature related to our defined search terms.

Articles were excluded if they: (1) were published before 1 January 2011, (2) took place outside of SSA, (3) did not include a blood donor cohort and included ages outside of 16–65, and (4) focused on another blood component or TTI other than HIV, HBV, HCV, and syphilis.

The second phase narrowed down articles to only include those from the 16 SADC member nations (Angola, Namibia, The Democratic Republic of the Congo [DRC], Malawi, Mozambique, Tanzania, Zambia, Botswana, Zimbabwe, Eswatini, South Africa, Lesotho, Madagascar, Comoros, Mauritius, and Seychelles). Only the articles from the SADC region were included in the final analysis. We included a preliminary search for all articles in SSA to allow for both a current assessment of related articles in SSA as a whole and to compare results from the SADC search. If an article focused on SSA as a whole and included a SADC member country, it would be included in the final analysis only if data specific to that SADC country was provided. Regional categorisation alone was not enough.

2.3 |. Data extraction

From each study, the following information was collected: first authors last name, topic studied, study setting, design, year, TTI (HIV, HBV, HBC and syphilis) prevalence, median age, male/female %, blood safety recommendations and risk factors of TTIs.

The studies were categorised into three distinct themes: TTI prevalence, risk factors for TTIs, and blood safety recommendations. An article was included if it covered at least one theme. If an article covered one or more of the three themes, they were included in each relevant category. If there was a paper reporting the same donors in different publications, we did not count the same donor sample size twice in the meta-analysis. We would only include the listed data one time.

2.4 |. Quality assessment

The selection process was performed by two authors (SPM and ES) who screened and obtained articles according to our defined eligibility criteria. Disagreements or concerns were resolved through discussion and if no consensus between the two was found, the issue was presented to other experienced investigators (MC and BM) for input.

The two reviewers independently assessed the quality of the final article selection in accordance with the following checklists: PRISMA,¹⁰ STROBE,¹¹ and CONSORT¹² and ROB 2¹³ checklist. These tools contained areas such as clear definition of the primary outcome, analysis methods, indication of study period and limitation of the study. Appendix B summarises the quality assessment tools used for each study design included in the final analysis.

2.5 |. Meta-analysis

We conducted a meta-analysis of prevalence of overall TTIs from the 27 studies included in the final analysis. We also performed metaanalysis of reported prevalence estimates for each individual TTI: HIV, HBV, HCV, and syphilis. We used R package meta to summarise the pooled estimates using a forest plot. To describe the percentage of variation across studies that is due to heterogeneity we used the I² statistic.¹⁴

3 |. RESULTS

3.1 |. Eligible studies

During our search, we initially identified a total of 2390 articles and after removing duplicates we remained with 2000 articles. A total of 225 full articles were reviewed based on the predefined inclusion and exclusion criteria after removing abstracts, reports, and any non-conventional publications. One hundred and eighty articles were found in SSA. Finally, we identified 27 articles to be included in the final analysis after further screening as indicated in flowchart in Figure 1.

FIGURE 1.

PRISMA flowchart of an expanded selection process. SADC, South African Development Community; TTI, transfusion-transmitted infection.

We found 180 articles published from SSA region as a whole. There was a total of 66 prevalence articles, 42 risk factor articles, and 72 blood safety articles (Appendix C). The top three producing countries from SSA were Nigeria (15.5%; 27/180), Ethiopia (13.85%; 25/180), and Ghana (9.44%; 17/180). A total of 27 (15.0%) articles contained data from multiple countries from SSA. However, in our final analysis, we only included articles that focus solely on one SADC member country.

A total of 27 articles were found for the SADC region: 15 on prevalence, 8 on risk factors, and 12 on blood safety recommendations (Appendix C). These 27 articles were included in the final analysis.

Of the 27 articles, the Democratic Republic of the Congo had the most articles (33.3%; 9/27), while South Africa and Tanzania contributed 22.2% (6/27) and 18.5% (5/27) of articles respectively. From the remaining articles, 7.4% (2/27) were from Madagascar while Botswana, Mozambique, Namibia, and Zimbabwe and Malawi contributed 3.7% of articles or 1/27 each (Appendix C).

3.2 |. Prevalence of TTIS in the SADC region

3.2.1 |. Overall TTI prevalence

A total of 15 eligible articles studies assessed TTI prevalence in the SADC region (Tables 2A–2D). From these articles, the pooled estimate Of prevalence of all four TTIs (HIV, HBV, HCV, syphilis) was found to be 2.0% (95% CI: 1.0–3.0) (Figure 2).

TABLE 2A.

Final articles selected for assessing risk factors for HIV.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	Risk factors for HIV
Kabinda, J et al.	2014	Bukavu, DRC	Cross-sectional study	595	23	70.3/29.7	• Low level of education (OR = 5.5, 95% CI = 1.1–26.3; p = 0.04) • Replacement blood donors (OR = 6.8, 95% CI = 1.9–24.2; p = 0.001)
M’baya, B et al.	2019	Malawi	Retrospective	125 893	19	82/18	• Age greater than 25 (AOR = 1.46, 95% CI = 1.17–1.83; p = 0.001) • Being out of school:  ∘ Employed (AOR = 1.85, 95% CI = 1.42–2.40; p < 0.001)  ∘ Self-employed (AOR = 1.35, 95% CI = 1.04–1.73; p = 0.022)  ∘ Other (AOR = 1.35, 95% CI = 1.04–1.76; p = 0.026) • Repeat donors (AOR = 0.48, 95% CI = 0.42–0.55; p < 0.001)
Mremi, A et al.	2021	Tanzania	Cross-sectional study	101 616	18–65a	83.7/16.3	• Replacement donor (OR = 1.22, 95% CI = 1.10–1.35)
Namululi, B et al.	2012	Bukavu, DRC	Retrospective cohort study	3292	27.5	70.2/29.8	• New women family donors (OR = 3.75, 95% CI = 1.65–8.55; p = 0.001)
Namululi, B et al.	2013	Bukavu, DRC	Retrospective cohort study	3292	27.5	70/30	• Family donors (ARR = 7.09, 95% CI = 3.75–13.39; p < 0.001) • Female donors (ARR = 1.77, 95% CI = 1.07–2.93; p = 0.025) • Age 21–30 years (ARR = 2.32, 95% CI = 1.03–5.18; p = 0.041)
Sumbu B.M.M. et al.	2018	Kinshasa, DRC	Retrospective analysis	26 341	33.5 ± 11	80.5/19.5	• Age less than 25 (OR = 1.7, 95% CI = 1.4–2; p < 0.0001) • HCV infection (OR = 3.0; 95% CI = 1.8–4.9; p < 0.001)

Abbreviations: CI, confidence interval; DRC, Democratic Republic of the Congo; HCV, hepatitis C virus; HIV, human immunodeficiency virus; OR, odds ratio.

^aAge range.

TABLE 2D.

Final articles selected for assessing risk factors for syphilis.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	Risk factors for syphilis
M’baya, B et al	2019	Malawi	Retrospective	125 893	19	82/18	• Female (AOR = 0.61, 95% CI = 0.50–0.74; p < 0.001) • Repeat donors (AOR = 0.63, 95% CI = 0.55–0.72; p < 0.001) • Self-employed (AOR = 1.44, 95% CI = 1.09–1.90; p = 0.011) • Married (AOR = 1.54, 95% CI = 1.21–1.96; p = 0.001) • Age greater than 25 (OR = 1.37, 95% CI = 1.18–1.60; p < 0.001)
Mremi, A et al.	2021	Tanzania	Cross-sectional study	101 616	18–65a	83.7/16.3	• Replacement donor (OR = 1.33, 95% CI = 1.20–1.48)

Abbreviations: CI, confidence interval; OR, odds ratio.

^aAge range.

FIGURE 2.

Total transfusion-transmitted infection pooled prevalence in the South African Development Community region articles.

3.2.2 |. HIV prevalence

Of the 15 eligible prevalence articles in the SADC region, 10 (66.7%) included the prevalence estimates of HIV and were included in the meta-analysis (Table 1; Figure 3A). The range of prevalence across articles was between 1.0% and 4.0%^15–23 except for Stokx et al.²⁴ that found prevalence to be 9.0%. The pooled prevalence estimate of HIV in blood donors was 2.0% (95% CI: 1.0–4.0). The I² statistics (99%, p < 0.01) showed heterogeneity across the studies.

TABLE 1.

Final selected articles on TTI prevalence in the SADC region.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	TTI prevalence (%)
							HIV	HBV	HCV	Syphilis
Arivelo, R et al.	2011	Antananarivo, Madagascar	Retrospective	47 597	33.3 ± 10.4	80.5/19.5	-	3.84	-	-
Choga, W et al.	2019	Botswana	Cross-sectional	12 575	29 ± 1.7	67.7/33.3	-	1.02	-	-
Kabinda, J et al.	2014	Bukavu, DRC	Cross-sectional	1079	26	72.4/27.6	-	4.2	3.8	-
Kabinda, J et al.	2014	South Kivu, DRC	Cross-sectional	595	23	70.3/29.7	1.6	4.8	3.9	-
Lidenge, S et al.	2020	Dar es Salaam, Tanzania	Cross-sectional	504	31	85.5/14.5	4.2	7.3	3.2	14.9
M’baya, B et al.	2019	Malawi	Retrospective	125 893	19	82.0/18.0	1.9	3.6	1	2.6
Mavenyengwa, R et al.	2014	Namibia	Prospective	24 761	16–65a	52.7/47.3	0.3	0.6	0.1	0.3
Mohamed Z et al.	2019	Dar es Salaam, Tanzania	Retrospective	6402	34.7 ± 9.8	83.8/16.2	1.7	4.1	1	2.2
Namululi, B et al.	2013	Bukavu, DRC	Retrospective	3292	27.5	70/30	1.0	3.7	-	-
Nzaji, M et al.	2013	DRC	Retrospective	1015	28 ± 6	95.0/5.0	2.9	1.6	0.3	0.2
Randriamanantany, Z et al.	2012	Antananarivo, Madagascar	Retrospective	47 636	33.3	80.0/20.0	-	-	0.65	-
Shindano T, et al.	2018	DRC	Systematic Review/Meta-Analysis	142 864			-	5	-	-
Stokx, J et al.	2011	Tete, Mozambique	Prospective	750	27 ± 9	88.9/11.1	8.5	10.6	0	1.2
Sumbu B.M.M. et al.	2018	Kinshasa, DRC	Retrospective	26 341	33.5 ± 11	80.5/19.6	2.2	-	-	-
Vermeulen, M et al.	2017	South Africa	Prospective	297 640		44.7/55.3	1.13	0.66	0.03

Abbreviations: DRC, Democratic Republic of the Congo; HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus; SADC, South African Development Community; TTI, transfusion-transmitted infection.

^aAge range.

FIGURE 3.

Pooled prevalence of transfusion-transmitted infections.

3.2.3 |. Hepatitis B prevalence

A total of 13 of 15 (86.7%) eligible prevalence articles included the prevalence estimates of HBV and were included in the meta-analysis (Table 1; Figure 3B). The range of prevalence across articles was between 1.0% and 7.0%^{15–18,20–23,25–28} except for Stokx et al.²⁴ that reported a higher prevalence of 11.0%. The pooled prevalence estimate of HBV was found to be 3.0% (95% CI: 2.0–5.0). The I² statistics (100%, p < 0.001) showed heterogeneity across the studies.

3.2.4 |. Hepatitis C prevalence

Ten articles (66.7%) also included the prevalence of HCV and were included in the meta-analysis (Table 1; Figure 3C). The range of prevalence across articles was between 1.0% and 4.0%. The majority (7/10) of the articles reported HCV prevalence between 0.0% and 1.0%,^{17,18,20,22–24,29} but three articles reported prevalence between 3.2% and 3.9%.^15,16,26 Of the articles that reported higher HCV prevalence, two were from DRC and one was from Tanzania. The pooled prevalence estimate of HCV in blood donors was 1.0% (95% CI: 0.0–2.0). The I² statistics (99%, p < 0.001) showed heterogeneity across the study.

3.2.5 |. Syphilis prevalence

Prevalence estimates of syphilis were reported in 6 of 15 (40.0%) of eligible prevalence articles (Table 1). Across the articles, the prevalence ranged from 0% to 3%^{17,18,22–24} except for the Lidenge et al.’s¹⁶ article which was an outlier with a prevalence of 15%. The pooled prevalence estimate of syphilis in blood donors was found to be 2.0% (95% CI: 0.0–8.0) (Figure 3D). In addition, the I² statistics (99%, p < 0.01) showed evidence of heterogeneity across study settings.

3.3 |. Risk factors for TTIS in the SADC region

3.3.1 |. Overall risk factors

A total of 8 articles reported risk factors for general TTI transmission, with the majority of the articles done in the DRC (5/8; 71.4%) (Tables 2A–2D). The other three articles were from Madagascar, Tanzania, and Malawi. Six of the eight articles did not report risk factors for all four TTIs.^{15,19,21,26,27,30} Only two articles reported risk factors across all TTIs.^23,31 Mremi et al.³¹ found that being a replacement donor was associated with increased risk of all TTIs compared to VNRBDs. M’baya et al.²³ found that repeat blood donors were less likely to be infected with all TTIs than first time donors.

3.3.2 |. HIV risk factors

HIV risk factors were assessed in six of eight studies (Table 2A). Sumbu et al.¹⁹ report that blood donors with HCV infection are three times more likely to be HIV positive (odds ratio [OR] = 3.0; 95% CI = 1.8–4.9; p < 0.001). Two studies, Mremi et al.³¹ and Kabinda et al.¹⁵ reported higher risk for HIV among family replacement donors (OR = 1.22, 95% CI = 1.10–1.35; OR = 6.8, 95% CI = 1.9–24.2; p = 0.001 respectively). Namululi et al.³⁰ similarly found that replacement donors were more at risk. Their study found that new female women family replacement donors were 3.75 times more likely to be HIV positive than new volunteer women donors (OR = 3.75, 95% CI = 1.65–8.55; p = 0.001). M’baya et al.²³ found repeat blood donors to be at lower risk of HIV than new donors when looking at donation status (adjusted odds ratio [AOR] = 0.48, 95% CI = 0.42–0.55; p < 0.001).

When considering age there were contrasting results between articles. M’baya et al.²³ report that donors above 25 years of age are more at risk for HIV (AOR = 1.46, 95% CI = 1.17–1.83; p = 0.001) but Sumbu et al.¹⁹ report higher risk for blood donors below age 25 (OR = 1.7, 95% CI = 1.4–2; p < 0.0001).

Regarding education, Kabinda et al.¹⁵ report that a low level of education, defined as primary school or less, increases risk for HIV positivity (OR = 5.5, 95% CI = 1.1–26.3; p = 0.04). Similarly, M’baya et al.²³ report that being out of a school setting increases risk for HIV positivity.

Namululi et al.²¹ found that overall, the residual risk (RR) for HIV was higher among family donors (adjusted residual risk [ARR] = 7.09, 95% CI = 3.75–13.39; p < 0.001), female donors (ARR = 1.77, 95% CI = 1.07–2.93; p = 0.025) and those aged 21–30 years (ARR = 2.32, 95% CI = 1.03–5.18; p = 0.041).

3.3.3 |. Hepatitis B risk factors

Table 2B shows seven out of the eight studies that examined HBV risk factors in blood donors. Age of donors was reported as a risk factor by Arivelo et al.²⁷ (age > 35; p < 0.05), Kabinda et al.²⁶ (age 31–40 years, AOR = 0.2, 95% CI = 0.06–0.8; p = 0.021) and age over 40 (AOR = 0.2, 95% CI = 0.04–0.6; p = 0.007), and Kabinda et al.¹⁵ (below 30 are OR = 0.3, 95% CI = 0.1–0.6; p < 0.001).

TABLE 2B.

Final articles selected for assessing risk factors for HBV.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	Risk factors for HBV
Arivelo, R et al.	2011	Antananarivo, Madagascar	Retrospective	47 597	33.3 ± 10.4	80.5/19.5	• Male (p < 0.05) • Below age 35 (p < 0.05)
Kabinda, J, Miyanga, S. et al.	2014	Bukavu, DRC	Cross-sectional study	1079	26	72.4/27.6	• Age 31–40 years (AOR = 0.2, 95% CI = 0.06–0.8; p = 0.021) • Age over 40 (AOR = 0.2, 95% CI = 0.04–0.6; p = 0.007) • Married (AOR = 3.5, 95% CI = 1.2–10.2; p = 0.022) • Female sex group (AOR = 0.3, 95% CI = 0.1–0.8; p = 0.018) • Rural (AOR = 4.0, 95% CI = 1.8–9.0; p = 0.001)
Kabinda, J et al.	2014	Bukavu, DRC	Cross-sectional study	595	23	70.3/29.7	• Age under 30 (OR = 0.3, 95% CI = 0.1–0.6; p < 0.001) • Single (OR = 0.4, 95% CI = 0.19–0.87; p = 0.02) • Replacement blood donors (OR = 3.7, 95% CI = 1.6–8.8; p = 0.001)
M’baya, B et al.	2019	Malawi	Retrospective	125 893	19	82/18	• Female (AOR = 0.51, 95% CI = 0.43–0.60; p < 0.001) • Repeat Donors (AOR = 0.43, 95% CI = 0.38–0.48; p < 0.001)
Mremi, A et al.	2021	Tanzania	Cross-sectional study	101 616		83.7/16.3	• Replacement donor (OR = 1.35, 95% CI = 1.27–1.44)
Namululi, B et al.	2012	Bukavu, DRC	Retrospective cohort study	3292	27.5	70.2/29.8	• New women family donors (OR = 0.71, 95% CI = 0.51–0.98)
Namululi, B et al.	2013	Bukavu, DRC	Retrospective cohort study	3292	27.5	70/30	• Family donors (ARR = 4.03, 95% CI = 2.63–6.20; p < 0.001) • Rural areas (ARR = 1.65, 95% CI = 1.12–2.43; p = 0.012)

Abbreviations: CI, confidence interval; DRC, Democratic Republic of the Congo; HBV, hepatitis B virus; OR, odds ratio.

Being male was also reported as a risk for HBV by Arivelo et al.²⁷ (p < 0.05), Kabinda et al.²⁶ (AOR = 0.3, 95% CI = 0.1–0.8; p = 0.018), and M’baya et al.²³ (AOR = 0.51, 95% CI = 0.43–0.60; p < 0.001). Similarly, Namululi et al.³⁰ found that when comparing HBV prevalence in new family donors by sex that women were 0.71 less likely to be HBV positive than their male counterparts (OR = 0.71, 95% CI = 0.51–0.98).

Donor status especially family replacement donors were at higher risk for HBV positivity, with Mremi et al.³¹ reporting OR = 1.35 (95% CI = 1.27–1.44) and Kabinda et al.¹⁵ reporting a four-fold increase in odds (OR = 3.7, 95% CI = 1.6–8.8; p = 0.001).

Other risk factors reported for increased risk of HBV positivity included being first-time donor, repeat donor, living in rural environment^21,23,26,30 (Table 2B).

3.3.4 |. Hepatitis C risk factors

A total of four articles reported risk factors for HCV in blood donors (Table 2C), with rural blood donors, being a replacement donor, and age above 25 years having an increased risk of HCV. Kabinda et al.²⁶ found an increased risk for HCV for rural blood donors when compared to urban donors. M’baya et al.²³ reported that blood donors above 25 years of age had higher risk of HCV. Mremi et al.³¹ only found that being a replacement donor was associated with higher HCV risk. However, among the risk factors for HCV that Kabinda et al.¹⁵ investigated, none were statistically significant.

TABLE 2C.

Final articles selected for assessing risk factors for HCV.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	Risk factors for HCV
Kabinda, J, Miyanga, S. et al.	2014	Bukavu, DRC	Cross-sectional study	1079	26	72.4/27.6	• Rural (AOR = 2.9, 95% CI = 1.3–6.6; p = 0.009)
M’baya, B et al.	2019	Malawi	Retrospective	125 893	19	82/18	• Female (AOR = 0.47, 95% CI = 0.300.74; p = 0.001) • Repeat donors (AOR = 0.75, 95% CI = 0.60–0.92; p = 0.007)
Mremi, A et al.	2021	Tanzania	Cross-sectional study	101 616	18–65a	83.7/16.3	• Replacement donor (OR = 1.28, 95% CI = 1.12–1.46)

Abbreviations: CI, confidence interval; DRC, Democratic Republic of the Congo; HCV, hepatitis C virus; OR, odds ratio.

^aAge range.

3.3.5 |. Syphilis risk factors

Only two articles investigated risk factors for syphilis in blood donors^23,31 (Table 2D). Mremi et al.³¹ found that being a replacement blood donor was associated with higher risk for syphilis infection compared to a voluntary donor. M’baya et al.²³ found new blood donors and males were associated with higher syphilis infection.

3.4 |. Blood safety improvements in the SADC region

3.4.1 |. Overall

Twelve articles investigated blood safety in the SADC region (Table 3).^{24,28,32–41} Of the 12 studies, 41% (5/12) were done in South Africa, while Tanzania and DRC had 2 articles each. The other three articles were from Botswana, Mozambique, and Zimbabwe. The recommendations range from improving blood screening, establishment of blood transfusion committees, establishing robust donor assessment and deferral system to using infectious modelling to improve hemovigilance.

TABLE 3.

Final selected articles on blood safety improvements in the SADC region.

Author	Year	Study setting	Study design	Sample size	Median/mean age	Male/female (%)	Blood safety improvement findings
Cable et al.	2013	South Africa	Retrospective analysis	649 745	-	-	• ID-NAT has contributed to enhancing blood safety. • Anti-HBc testing of NAT nonrepeat reactive donations seems useful in identifying a subgroup of donors with OBI who may be at risk of transmitting HBV.
Choga et al.	2017	Botswana	Cross-sectional	12 575	29 ± 1.7	67.7/33.3	• There is need to establish mutations that correlate with HBV/HIV co-infection as opposed to HBV monoinfection as there is evolutionary diversity between different HBV cohorts. • Use of NATs in addition to serological screening should be introduced at national blood transfusion services (NBTS) and utilised as the standard screening technique to prevent TTIs transmission through blood transfusion services.
Kabinda et al.	2015	DRC	Review	-	-	-	• Draft all the necessary documents and standards. • Integrate blood safety system into the country’s health system. • Adopt a transversal blood safety approach. • Organise blood safety monitoring. • Establish a safe supply of reagents and supplies. • Sustain dynamics of voluntary associations of blood donors. • Provide stable funding for these blood safety activities.
Makubi et al.	2012	Dar es Salaam, Tanzania	Audit	-	-	-	• Referral hospitals should have a transfusion committee to provide an active oversight, facilitate communication between those involved with transfusion, recommend, or perform practice audits and monitor transfusion practice.
Mapako et al.	2013	Zimbabwe	Retrospective analysis	-	-	-	• Reevaluation of cost-effectiveness and utility of screening and discarding first-time donation is needed. There are comparable residual transmissions between first-time and repeat donors.
Mitchel et al.	2019	South Africa	Impact evaluation	3 169 656a	-	-	• Implementation of a scripted interview is associated with increased HIV risk deferral and decreased recent HIV infection.
Ngunza et al.	2020	DRC	Descriptive cross-sectional study	387	Not indicated	82.2/17.8	• The use of a composite classification to assess the regularity of voluntary blood donors provides more accurate information that will enable the improvement of donors’ awareness and retention as well as the possible reinstatement of former donors.
Stokx et al.	2011	Tete, Mozambique	Prospective	750	27 ± 9	88.9/11.1	• Seroprevalence of TTIs after questionnaire screening is high. • Current screening does not effectively exclude HIV- infected donor candidates. • Locally used assays led to unnecessary rejection of safe blood donations. • A contextualised questionnaire and consistent use of quality-assured assays would considerably improve the current screening procedure for blood donation.
Valerian et al.	2018	Moshi, Tanzania	Retrospective cross-sectional	14 377	-	79.1/19.9	• Appropriate preventive interventions in the community are needed to address prevalent causes of deferrals caused by TTIs.
Vermeulen et al.	2021	South Africa	Model analysis	1 238 670	-	-	• Testing HIV-positive donations using the LAg assay provides a reliable method to estimate incidence in firsttime donors for countries that collect the majority of blood from first-time donors and do not screen with NAT.
Vermeulen et al.	2012	South Africa	Prospective	149	-	-	• ID-NAT screening does not eliminate transmission risk. • HBV blood transmission can still occur even with ID- NAT testing.
Weusten et al.	2017	South Africa	Model analysis	4 019 985a	-	-	• A Gumbel distribution can describe a log viral load distribution or the OBI donations in a setting where ID- NAT is performed but anti-hepatitis B core antibody (anti-HBc) is not.

Abbreviations: DRC, Democratic Republic of the Congo; HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus; ID-NAT, individual donor-nucleic acid testing; LAg, limiting-antigen assay; OBI, occult hepatitis B infection; SADC, South African Development Community; TTIs, transfusion-transmitted infections.

^aDonation.

3.4.2 |. Improving lab-based blood screening

To improve blood screening, three articles recommended the use of individual donor-nucleic acid testing (ID-NAT) in blood screening.^28,32,33 Two of these articles were from South Africa and one from Botswana. Cable et al.³² found that ID-NAT in the Western Cape Province of South Africa has significantly enhanced blood safety for both HIV and HBV. In addition, ID-NAT was found helpful for identification of donors with occult blood infections creating risk for HBV transmission. Choga et al.²⁸ also recommended nucleic acid tests (NAT) utilisation in the National Botswana Transfusion Service to prevent TTI transmission when their study found evolutionary differences between HBV cohorts. Although ID-NAT is helpful, Ver- meulen et al.³³ highlighted the limitation of this method as they reported the first known case of HBV transmission that was screened by ID-NAT, revealing that the method does not completely eliminate risk.

3.4.3 |. Blood transfusion systems and preventive measures

A total of three (3) articles from Tanzania and the Democratic Republic of the Congo discuss blood safety recommendations for TTI preventive measures and health system organisation to reduce donor rejections and improve blood safety.^35,39,40 The two articles from Tanzania recommend preliminary measures for the first stages of blood safety efforts.^35,39 Makubi et al.³⁵ recommend the implementation of a blood transfusion monitoring systems and committee in hospitals and to conduct regular audits to monitor transfusion practice. Valerian et al.³⁹ call for appropriate preventive measures to address the main causes of deferrals focusing on HIV and HBV infections. Kabinda et al.⁴⁰ reflected on the DRC’s transition from vertical blood donor safety approach to a transversal integration of blood services in their healthcare system. They found this change led to more efficient mobilisation of blood donors and clinician training in blood donation ultimately reducing TTI prevalence over 11 years.

3.4.4 |. Donor assessments and deferral systems

The efficacy of screening questionnaires and risk assessment systems was reported by four articles.^24,36,37,41 Stokx et al.²⁴ evaluated the questionnaire used at the Tete Provincial Hospital in Mozambique after it was found that seroprevalence remained high even with screening. The study found that the questionnaire did not effectively exclude HIV-infected donors and the locally used lab assay led to unnecessary rejection of donations that were later found to be safe. They recommend a contextualised questionnaires and consistent use of quality-assured assays to ensure blood quality. Mitchel et al.³⁷ found that a scripted interview was associated with increased HIV risk deferral and decreased recent HIV infection in South Africa highlighting the potential for scripted interview as a means for TTI reduction. Beyond screening, Ngunza et al.⁴¹ suggested a composite classification system to assess the regularity of voluntary blood donors which would allow for more informed efforts towards donor retention and reinstatement.

Mapako et al.³⁶ assessed Zimbabwe’s current HIV risk management strategy that directs all blood donations from first time donors to be collected in dry packs and discarded. Facing intermittent blood shortages in the country, the article sought to compare HIV risk between first time and repeat blood donations. They found that there was comparable residual risk transmission between both donor groups indicating waste of viable blood donations.

3.4.5 |. Infectious disease modelling

Infectious disease modelling was also recommended in two articles from South Africa.^34,38 Weusten et al.³⁴ conducted a model analysis aimed at estimating TTI incidence and risk in blood donations. They adjusted an original model to estimate the residual risk of TTI transmission caused by occult hepatitis B infection (OBI) donations in a setting where ID-NAT is performed but anti-hepatitis B core antibody (anti-HBc) is not. The authors found that for the OBI donations, a log viral load distribution can be described by a Gum- bel distribution. The authors assert the model can be used to study the impact of not performing nucleic amplification testing screening as well as to compare the efficacy of different blood safety testing scenarios.

Vermeulen et al.³⁸ offer the limiting-antigen assay (LAg) as an alternative to NAT testing to measure HIV incidence among first time donors. The authors determined incidence using three models: a classic extrapolation from first time donors, a NAT yield window period (WP) model and LAg avidity testing. They found that LAg avidity testing produced similar estimates to the NAT yield WP model and lower estimates from the classic extrapolation method. The authors conclude that LAg assays can be a reliable choice for HIV incidence estimates in countries that collect mainly from first time donors and cannot use the NAT system.

4 |. DISCUSSION

Our systematic review searched for articles in SSA that investigated risk factors and prevalence of TTIs in blood donors and blood safety recommendations. Our search yielded 180 articles across the region in the past 10 years that have addressed these topics. Only 27 of these articles are from the SADC region. The Southern African region is disproportionately lacking in research in the three theme areas of blood safety that this systematic review explores and much work remains to be done. Blood safety measures and TTI surveillance are a foundational necessity for a well-functioning healthcare system.

Overall pooled prevalence for all four TTIs in the SADC region was 2% (95% CI: 1%−3%). At the time of our study, there are no systematic reviews examining national prevalence for all four WHO recommended TTIs in blood donors in the SSA region or any regional blocks. The systematic reviews on TTIs yielded by our study were either focused in one country, were not inclusive of all four TTIs, or focused on general population but including blood donor cohorts. When compared to the WHO’s African Report on Blood Safety, our result was significantly less than the 12.3% prevalence of all four TTI prevalence reported in the SADC region.⁵ In addition, the WHO report only includes 13 out of the 16 SADC members in its report. This discrepancy highlights both the lack of current research in this area and the difficulty using limited research for a representative understanding of bloody safety regionally.

Our study found a HIV prevalence in blood donors of 2% in the SADC region which is higher compared to 1.0% reported by the WHO.⁵ The SADC region has the highest HIV morbidity/mortality in the world and 34% of the total people living with HIV/AIDs world-wide reside in 10 SADC countries.⁴² Despite this statistic, few studies from this area comprehensively investigate HIV prevalence in blood donors and assess risk of transmission. Across SSA, the rate of HIV in blood donors was lower than HBV and HCV positivity overall which could be due to the global efforts to end the AIDs epidemic, relatively less focus on hepatitis, and test performance of assays.

Of the TTIs, HBV has the highest pooled prevalence at 3% in the SADC region. This was lower than the 5% reported by the only TTI systematic review from the SADC region yielded in our search. The yielded study was conducted in the DRC, which has probable HBV endemicity and cannot be extrapolated to the entire region. Our result is higher than the WHO reported 1.5% in the SADC region.⁵ Southern Africa overall has lower HBV prevalence in the general population compared to other regions in Africa and our SSA search showed higher HBV prevalence in blood donors from Western African countries and Ethiopia.⁴³

HCV has a pooled prevalence of 1% in blood donors which is similar but lower than the 1.78% reported by the systematic review of HCV in SSA by Mora et al.⁴⁴ However, our finding is higher than the 0.6% prevalence in the SADC region reported by the WHO.⁵ In general populations, HCV prevalence is lower in Southern Africa and highest in Western Africa.⁴⁴ Similarly, our results from the general SSA search also showed higher HCV prevalence in blood donors in Western Africa and lower prevalence in Southern Africa. There were reports as high as 14.6% 11.7%, 5.8%, 5.6% in Ghana, 8.6%, 6.9%, 6.3% in Burkina Faso and 12.5% in Cameroon.^45–52 In Southern Africa, the highest HCV prevalence was 3.8% in DRC.²⁶ This aligns with findings from the WHO; they report HCV prevalence to be 1.5% in the Economic Community of Central African States (ECCAS), 1.8% in the Eastern African Community (EAC), and 1.5% in the Economic Community of Western African States (ECOWAS).⁵

Syphilis in blood donors has a pooled prevalence of 2% in the SADC region which is higher than the 1.2% prevalence of syphilis in the SADC region reported by the WHO.⁵ SSA syphilis prevalence could increase up to 25% in blood donors, but little research was found focused on this TTI in this group.⁵³ The highest prevalence found in our search was 21.5% in Equatorial Guinea, 14.9% in Tanzania, and 10.42% in Ghana.^16,54,55 No geospatial pattern emerged for syphilis in blood donors; however, the higher prevalence seemed to be from Western Africa. This contrasts with the WHO reported prevalence of 0.8% in ECOWAS, which is the lowest of all economic communities.⁵ This could be due to the uneven distribution of articles across regions in Africa in our results. Attention to this TTI is important since there is increased risk of HIV infection with syphilis having an implication for the HIV/AIDs epidemic.

Of the 15 articles that examined risk factors, only 2 included all four TTIs. The other 13 articles presented individual risk factors for one to three TTIs. TTI coinfection and occult HBV infections emerged as a common topic focus among the articles in SSA, however, is outside the scope of this study. This review found that risk factors for TTIs in blood donors vary across settings and for each TTI; however, there are a few common risk factors that emerge in the SADC region. Consistent with WHO findings, first time blood donors were at greater risk for TTIs and repeat donors are a safer donor population. In addition, family blood replacement donors were found to be more at risk for TTIs than voluntary non-renumerated blood donors (VNRBDs). This is consistent with findings outside of the SADC region and is in line with the WHO recommendation for VNRBDs.⁵⁶ There was little consistency of specific risk factors such as age, gender, or sex across the studies even within the same setting. The lack of overlapping commonalities underlines the contextualised nature of risk factors. These findings emphasise the importance of directed research to determine TTI risk factors at a national level as well as regionally to reflect a country’s unique context.

Four main themes emerged from the selected SADC region studies: improving lab-based blood screening; blood transfusion systems and preventive measures; donor assessments and deferral systems; and TTI risk modelling. The 12 articles that discussed TTI blood safety interventions and recommendations were not uniformly distributed across the region with a little less than half from South Africa (5/12). The topics varied based on the level of development of the countries’ blood transfusion systems. For high-quality blood and safety, countries need to ultimately build and maintain consistent blood donors along with effective screening methods to decrease the risk of TTI and transfusion reactions. As is evident from the paucity of blood donor TTI safety research produced by the member nations in the SADC region, higher income nations can afford advanced research into modelling and assessment of advanced blood testing method as seen with the model analyses from SA and ID-NAT assessments from SA and Botswana. Only upper middle to high-income countries are investigating use of ID-NAT testing. As of 2017, the only three nations in the WHO Africa region supporting such testing are South Africa, Mauritius, and Namibia.⁵⁷

In addition, each article recommends further research in preventative and safety measures; however, very few countries have produced such articles in the past 10 years. Little literature is available on the effectiveness of national blood transfusion systems policy and research. Countries need to continually assess their blood policies for efficiency and optimisation as well as create legislature for sustainable function. Perhaps there are internal evaluations that have not been published; however, there is still cause for worry. Even though 38 out of the 46 countries in the WHO Africa region report establishing a national blood transfusion service, only 11 provide structural support with experts or advisory boards to achieve such progress.⁵⁸ Furthermore, in the SADC region, only four countries have an accredited National Blood Transfusion Service and only six have national guidelines for hemovigilance system.⁵

In this review, we focused on SADC members as it is an economic community that ‘aims to create appropriate institutions and mechanisms for the mobilisation of requisite resources for the implementation of programmes and operations of SADC and its institutions’.⁵⁹ Targeting regional blocks could serve as a solution for unified effort towards blood safety. Through the development of protocols that are intended to be adopted by each member state, economic communities guide regional integration and form a common agenda. The ninth policy of SADC’s common agenda is to ‘Combat HIV and AIDS and other deadly or communicable diseases’; however, since 1999 their Protocol on Health has not been amended and has no mention on blood safety or TTI surveillence.⁵⁹ The nations that comprise SADC have similar disease burden, TTI prevalence, and structural barriers, in addition to economic position. Utilising the power of SADC can take advantage of existing infrastructure to implement cohesive blood safety policies and mutual encouragement of commitment to blood safety priorities. The SADC body can potentially lobby larger organisations and funders to support this neglected issue and help these countries meet their national need for blood.

Comprehensive TTI prevalence studies are limited both in SSA and especially in the SADC region. Performing surveillance for different TTIs over multiple years leaves a fragmented picture of burden limiting the effectiveness of future interventions. In addition, pulling from general prevalence does not generate an accurate depiction for blood donors as they have characteristically lower risk as a cohort further highlighting need for blood donor-specific studies.⁶⁰ To bolster blood safety countries should invest in consistent and continual TTI mapping for generating an accurate epidemiological description of TTI burden specifically in blood donors to direct informed interventions.

Our study is subject to limitations. First, a main limitation of this review stems from the minimal uniformity in the screening methods used across the studies. The accuracy of detection depends on the screening method used for detection. Secondly, we could not conduct a meta-analysis for risk factors since there also was varying statistical analysis performed across the studies that did not allow for analysis, forcing a qualitative assessment of risk factors. Thirdly, a general lack of literature on the topic could leave a skewed description of the current TTI situation in the region. There were a few nations within SSA and the SADC region that produced a significant amount of the articles creating an incomplete comparison. Fourthly, articles in other languages could have been excluded if they did not include an English abstract. Despite these limitations, our study highlights the gaps in TTI research and blood safety measures in both SSA and the SADC region. Even though there may be geographically disproportionate articles being published, these countries share enough characteristics where recommendations and evidence can be applied. Our study shows the importance of TTI blood safety research and where focus needs to be shifted.

5 |. CONCLUSION

Much work remains to achieve complete blood safety in the SADC region. Overall, the prevalence of TTIs in blood donors in this region is one of the highest in the world. There is limited information on key indicators such as blood requirements, safety, and distribution of TTI prevalence that are necessary for an effective and robust national blood safety initiative. Longitudinal monitoring of TTIs, focused donor retention, and fortified blood transfusion systems, both centralised and in-hospital, are vital. TTI research across the continent is not well distributed with a high number of articles being produced from Western Africa, and relatively low publications from Southern Africa. Blood research should become a priority in the SADC region. Furthermore, there needs to be increased effort to discern contextualised risk factors within each country to properly assess TTI prevalence and adjust blood collection practices. Blood safety recommendations are varied depending on advancement of blood transfusion policy and infrastructure. Across the SADC region, more effort should be directed into creating a cohesive blood transfusion policy and bolstering monitoring and evaluation plans. Through such coordination, the region can attain sustainable and effective blood safety.

What is known about the topic?

• The SADC region and SSA face suboptimal blood donations, high demand for blood, and insufficient quality controls that lead to chronic shortages.

• Contaminated blood donations with TTIs are one of the main reasons for discarded blood in this region.

• Efforts for blood safety improvement are varied across the region, and efficacy of interventions is seldom investigated.

What is new?

• The literature shows high TTI contamination in the SADC region above the other economic communities in Africa.

• There are varying levels of research efforts across the region, with some countries producing many articles and some not at all.

• Published blood safety improvements research is lacking in Southern Africa compared to other regions of the continent.

What are the key questions for future work on the topic?

• What is the best mechanism to implement consistent TTIs screening?

• How can we implement routine monitoring and quality improvement into SADC blood transfusion system?

• How can the integrative power of regional blocks such as SADC be harnessed to implement cohesive and target blood safety surveillance systems across the economic community?

APPENDIX A: SEARCH STRATEGY

For each database, the following search strategy was used:

(((blood safety) OR (blood donor) OR (blood donation) OR (blood transfusions) OR (transfusion transmitted infection)) AND ((prevalence) OR (epidemiology) OR (data quality improvement) OR (incidence) OR (residual risk) OR (risk factors)) AND ((Sub-Saharan Africa) OR (Cameroon) OR (Central African Republic) OR (Chad) OR (Congo) OR (Democratic Republic of the Congo) OR (Equatorial Guinea) OR (Gabon) OR (Burundi) OR (Djibouti) OR (Eritrea) OR (Ethiopia) OR (Kenya) OR (Rwanda) OR (Somalia) OR (Sudan) OR (South Sudan) OR (Tanzania) OR (Uganda) OR (Angola) OR (Botswana) OR (Lesotho) OR (Mozambique) OR (Ivory Coast) OR (Namibia) OR (South Africa) OR (Swaziland) OR (Zambia) OR (Zimbabwe) OR (Benin) OR (Burkina Faso) OR (Cote D’ivoire) OR (Gambia) OR (Ghana) OR (Guinea) OR (Guinea-Bissau) OR (Liberia) OR (Mali) OR (Malawi) OR (Mauritania) OR (Mauritius) OR (Burundi) OR (Eswatini) OR (Madagascar) OR (Niger) OR (Nigeria) OR (Senegal) OR (Sierra Leone) OR (Togo)))

APPENDIX B: QUALITY ASSESSMENT TOOL DURING THE SYSTEMATIC REVIEW

Characteristics	Categorisation
Year of study	2011–2021
Region	Sub-Saharan
Study topic	HIV, syphilis, Hep B, Hep C testing in the context of blood donation or transfusion
Study design and reporting	RCT—CEBM-RCT checklist Cohort—STROBE Checklist Cross-section—STROBE Checklist Case-control—STROBE Checklist
Conflict of interest	Authors and funders
Relevance	- Is the research method/study design appropriate for answering the research question? - Are specific inclusion/exclusion criteria used?
Reliability	- Is the effect size practically relevant? How precise is the estimate of the effect? Were confidence intervals given?
Validity	From Cochrane Risk of Bias tool (Cochrane Risk of Bias [ROB] 2.0 Tool) - Were there enough subjects in the study to establish that the findings did not occur by chance? - Were subjects randomly allocated? Were the groups comparable? If not, could this have introduced bias? - Are the measurements/tools validated by other studies? - Could there be confounding factors?
Applicability	- Can the results be applied to my organisation and my patient?

APPENDIX C: COUNTRIES IN SADC REGION IN WHICH BLOOD TRANSFUSION RESEARCH WAS CONDUCTED

Region/country	No. of articles (%)	Prevalence	Risk factors	Blood safety
Total	180 (100)
SSAa	180	86	54	72
SADCa	27 (100)	15	8	12
DRC	9 (33.3)	6	5	2
South Africa	6 (22.2)	1	0	5
Tanzania	5 (18.5)	2	1	2
Madagascar	2 (7.4)	2	1	0
Botswana	1 (3.7)	1	0	1
Malawi	1 (3.7)	1	1	0
Mozambique	1 (3.7)	1	0	1
Namibia	1 (3.7)	1	0	0
Zimbabwe	1 (3.7)	0	0	1

Abbreviations: DRC, Democratic Republic of the Congo; SADC, South African Development Community; SSA, sub-Saharan Africa.

^aNumbers in article category do not add up to total article per region due to articles covering multiple topics.