The prevalence of anemia among human immunodeficiency virus-infected individuals in East Africa: A systematic review and meta-analysis

Abstract

Background: 

Human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) patients face several hematological abnormalities. Of these abnormalities, anemia is the most common one. Africa has a high prevalence of HIV/AIDS, especially in the East and South African region, which is heavily affected by the virus. Therefore, this systematic review and meta-analysis aimed to determine the pooled prevalence of anemia among patients with HIV/AIDS in East Africa.

Methods: 

This systematic review and meta-analysis was conducted based on the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis. PubMed, Google Scholar, Science Direct, Dove Press, Cochrane Online, and African journals online were searched systematically. The quality of the included studies was assessed by 2 independent reviewers using the Joanna Briggs Institute critical appraisal tools. Data were extracted into an Excel sheet and then exported to STATA version 11 for analysis. A random-effect model was fitted to estimate the pooled prevalence and Higgins I² test statistics were done to test the heterogeneity of studies. Funnel plots analysis and Egger-weighted regression tests were done to detect publication bias.

Results: 

The pooled prevalence of anemia among HIV/AIDS patients in East Africa was 25.35% (95% CI: 20.69–30.03%). A subgroup analysis by highly active antiretroviral therapy (HAART) status showed that the prevalence of anemia among HAART naive HIV/AIDS patients was 39.11% (95% CI: 29.28–48.93%) whereas the prevalence among HAART experienced was 36.72% (95% CI: 31.22–42.22%). A subgroup analysis by the study population showed that the prevalence of anemia among adult HIV/AIDS patients was 34.48% (95% CI: 29.52–39.44%) whereas the pooled prevalence among children was 36.17% (95% CI: 26.68–45.65%).

Conclusion: 

This systematic review and meta-analysis revealed that anemia is among the most common hematological abnormalities in HIV/AIDS patients in East Africa. It also underscored the importance of taking diagnostic, preventive, and therapeutic measures for the management of this abnormality.

1. Introduction

The human immunodeficiency virus (HIV) is a virus that is categorized under the genus of lentivirus, a family of Retrovirus, and a subfamily of Orthoretrovirinae.^[1] It is a causative agent of acquired immunodeficiency syndrome (AIDS).^[2] HIV is capable of being transmitted through sexual intercourse, contaminated blood products, and mother to child during birth. Postnatal transmission can also arise due to the ingestion of the virus through breast milk.^[3] After 2 to 4 weeks of entry into the body, the patient will start to develop the symptom of primary infection. Gradually, a long chronic infection develops that could last for decades.^[4]

HIV causes a gradual decline of the immune system because of a reduction in the number of cluster of differentiation 4 positive T-helper cells.^[5] As the disease becomes more advanced, it develops into AIDS where the cluster of differentiation 4 positive cell count drops below 200/mm³.^[6] A person with HIV infection usually develops complication that is directly associated with the infection or with an adverse effect of the treatment.^[7] Old age, the existence of comorbidities, and lifestyles escalate the risk of developing chronic conditions like diabetes mellitus and renal diseases.^[8]

Hematological abnormalities are common findings among HIV/AIDS patients.^[9] These abnormalities are severe in the late stage of AIDS with a high viral load.^[10] It includes defective hematopoiesis, cytopenias affecting various cell lineages, and abnormalities of coagulation. Hematological abnormalities in HIV/AIDS patients arise from immune-mediated cell destruction, direct cytopathic effects of the virus, secondary infections, and medication toxicity.^[11] Some of the hematological abnormalities that are common in HIV/AIDS patients include anemia, thrombocytopenia, leucopenia, and bone marrow dysplasia.^[12]

Anemia is the most common hematological abnormality in HIV/AIDS patients. There are several factors thought to contribute to the pathophysiology of anemia in HIV-positive patients.^[13] First, many opportunistic infections or malignancies to which the patients are vulnerable can lead to anemia.^[14] The direct effect of HIV infection also contributes to the development of anemia.^[15] In addition to these, many of the most common drugs included in standard antiretroviral therapy (ART) also harm hematopoiesis in patients taking a treatment.^[16] Generally, alterations in cytokine expression, HIV-related metabolic problems, and nutritional deficiencies result in the development of anemia.^[17]

HIV is a global health problem that affects every corner of the world. According to the Joint United Nations Program on HIV and AIDS report, by the end of 2019, 38 million adults and children were living with HIV/AIDS. The virus also caused 690,000 deaths in HIV/AIDS-infected individuals by the end of 2019. In the same year, 20.7 million Africans were infected with HIV and this is the highest number compared with the rest of the world. In the East and southern part of Africa, the number of people that live with HIV/AIDS by the end of 2019 was 20.7 million. In the same year, 300,000 people died due to HIV/AIDS.^[18] There were 1.74 billion anemia cases recorded globally in 2019, with a prevalence rate of 22.8%.^[19] In Africa, the prevalence of anemia among HIV-positive people ranges from 16.2 to 69%.^[20]

Many studies revealed that HIV/AIDS patients are at a high risk of developing anemia. Though studies have been conducted on the prevalence of anemia among HIV/AIDS patients, the findings have been inconsistent and inconclusive. Moreover, there is no previously done systematic review and meta-analysis that estimate the pooled prevalence of anemia among HIV/AIDS patients in East Africa. Therefore, this systematic review and meta-analysis was designed to estimate the pooled prevalence of anemia among HIV/AIDS patients in East Africa using the available evidence.

2. Methods

2.1. Context and protocol development of the review

This systematic review and meta-analysis was carried out following the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis.^[21] To determine the pooled prevalence of anemia among HIV/AIDS patients in East Africa, findings from published articles have been used. This systematic review and meta-analysis was registered on the international prospective register of systematic reviews (CRD42021252236).

2.2. Eligibility criteria

1.2.2. Inclusion criteria.

Studies that are accepted and published in peer-reviewed journals were included. Respondents from all ethnicities, socioeconomic backgrounds, educational statuses, and countries belonging to East Africa were included. Cross-sectional and cohort studies documenting the outcome of interest were included. Gray literature, papers archived in University repositories, and related research papers were also included. Articles published before February 28, 2021, were included. Published articles in the English language were eligible for inclusion.

2.2.2. Exclusion criteria.

Studies conducted among HIV/AIDS patients but with any other comorbidities or co-infections like tuberculosis and malaria were excluded from this systematic review and meta-analysis because these comorbidities can cause anemia by themselves. Comparative studies that reported these abnormalities in mean value but not with percent prevalence were also excluded. Studies that do not report the overall prevalence (e.g., those that report the prevalence at a different time interval [initial, 6 months…]) were excluded.

3.2.2. Study outcome.

The outcome variable in this systematic review and meta-analysis is the prevalence of anemia among HIV/AIDS patients. Anemia was defined as a hemoglobin level of <12g/dL for females and <13g/dL for males.^[22]

2.3. Information sources

Data were gathered from databases such as PubMed, Google Scholar, Science Direct, Dove Press, Cochrane Online, and African journals online by searching for previously published literature.

2.4. Searching strategy

We have conducted a wide-ranging search of eligible studies in PubMed, Google Scholar, Science Direct, Dove Press, Cochrane Online, and African journals online. The reference lists of the identified studies were manually searched to identify additional relevant studies. The search strategy was based on the combinations of keywords and medical subject heading terms. The search terms used in PubMed were ((((((((((((((((((((((((((((((Anemia) OR (anemia))) OR (cytopenia)) OR (hematological profile)) OR (hematological parameters)) OR (hematological abnormality)) AND (Human Immunodeficiency Virus (HIV))) OR (Acquired Immunodeficiency Syndrome (AIDS))) OR (Antiretroviral Therapy (ART))) OR (Highly Active Antiretroviral Therapy (HAART))) AND (Ethiopia)) OR (Somalia)) OR (Kenya)) OR (Uganda)) OR (Sudan)) OR (South Sudan)) OR (Tanzania)) OR (Djibouti))) OR (Eritrea)) OR (Burundi)) OR (Comoros)) OR (Rwanda)) OR (Seychelles). Additional filters such as language (English) and study population (Human) were used.

2.5. Study selection and quality assessment

In this systematic review and meta-analysis, retrieved articles were imported to EndNote X9 to collect and organize search outcomes and for the removal of duplicate articles. Then, the articles were screened by their titles and abstracts by 2 reviewers independently (F.G.A. and M.W.). There were active discussions and mutual agreement between the 2 reviewers. In case of disagreements, a third reviewer (M.A.) was involved. The reviewers appraised the methodological quality of included studies using Joanna Briggs Institute (JBI) critical appraisal tools.^[23] The tools consist of items to assess internal and external validity. Research that has used different study designs was appraised using different JBI critical appraisal tools. Each item was carefully used to assess the methodological quality of included studies. A value of 1 and 0 were given for each research according to JBI critical appraisal tools. A value of 1 was given for items that are stated clearly in the method whereas a value of zero (0) was given for items that are not been stated clearly in the method part of the research. Finally, the overall methodological quality of the included studies was calculated in percent. Articles with methodical quality of <50%, 50 to 75%, and >75% were considered poor, good, and high quality, respectively.^[24]

2.6. Data extraction

After assessing the methodological quality, studies that fulfilled the eligibility criteria were subjected to data extraction by all 4 reviewers through a prepared data extraction sheet. The following items were extracted for analysis: title, name of the first author, study area, study design, year of publication, study population, HAART status, sample size, and prevalence of anemia (Table 1).

Table 1 - Summary of included studies in this systematic review and meta-analysis.

Authors	Study area	Study design	Year	Study population	HAART status	Sample size	Prev. of anemia
1.Addis Z. et al, 2014 (34)	Ethiopia	Cross-sectional	2014	Adult	Naive	189	42.3
2.Fekene T. et al, 2012 (35)	Ethiopia	Cross-sectional	2012	Adult	Experienced and naive	361	51.5
3.Fenta D. et al, 2019 (36)	Ethiopia	Cross-sectional	2019	Children	Experienced	273	11.4
4.Ferede G. et al, 2012 (37)	Ethiopia	Cross-sectional	2012	Adult	Naïve	420	35
5.Gebregziabher Y. et al, 2013 (38)	Ethiopia	Cross-sectional	2013	Children	Experienced and naive	224	29.5
6.Geletaw T. et al, 2015 (39)	Ethiopia	Cross-sectional	2015	Children	Experienced and naive	240	42.8
7.Gunda, D. W. et al, 2016 (40)	Tanzania	Cross-sectional	2016	Adult	Naive	1205	58.4
8.Kamau J. et al, 2004 (41)	Kenya	Cohort study	2004	Children	Experienced	170	38.4
9.Kibaru E. et al, 2008 (42)	Kenya	Cohort study	2008	Children	Naive	337	35.9
10.Kyeyune R. et al, 2012 (43)	Uganda	Cross-sectional	2012	Adult	Experienced and naïve	400	47.8
11.Mengistu A. et al, 2019 (44)	Ethiopia	Cross-sectional	2019	Adult	Experienced	422	26.2
12.Mugisha J. O. et al, 1993 (45)	Uganda	Cohort study	1993	Adult	Naive	500	18.9
13.Ndeezi G. et al, 2007 (46)	Uganda	Cross-sectional	2007	Children	Experienced and naïve	225	85
14.Tamir Z. et al, 2016 (47)	Ethiopia	Cross-sectional	2016	Adults	Naïve	402	43.5
15.Zerihun K. et al, 2017 (48)	Ethiopia	Cross-sectional	2017	Adult	Experienced	365	34
16.Ageru T. et al, 2016 (49)	Ethiopia	Cross-sectional	2016	Adults	Experienced and naive	411	36.5
17.Enawgaw B. et al, 2012 (17)	Ethiopia	Cross-sectional	2012	Adults	Experienced and naive	290	11.7
18.Katemba C. et al, 2016 (50)	Uganda	Cross-sectional	2016	Adult	Naïve	141	67.38
19.Gebreweld A. et al, 2018 (51)	Ethiopia	Cross-sectional	2018	Adult	Experienced	499	23.2
20.Deressa T. et al, 2016 (53)	Ethiopia	Cross-sectional	2016	Adults	Experienced and naive	316	25
21.Yemane B., 2019 (54)	Ethiopia	Cross-sectional	2019	Adult	Experienced	212	33.5
22.Rahel A. et al, 2015 (55)	Ethiopia	Cross-sectional	2015	Adult	Experienced and naive	340	15.88
23.Gedefaw L. et al, 2012 (56)	Ethiopia	Cross-sectional	2012	Adult	Experienced and naive	234	23.1
24.Melese H. et al, 2015 (57)	Ethiopia	Cross-sectional	2015	Adult	Experienced and naive	385	23
25.Enawgaw B. et al, 2013 (58)	Ethiopia	Cross-sectional	2013	Children	Experienced and naive	265	16.2
26.Mihiretie H. et al, 2013 (59)	Ethiopia	Cross-sectional	2013	Children	Experienced and naive	180	22.2
27.Gebremedhin K. et al, 2018 (60)	Ethiopia	Cross-sectional	2018	Adult	Experienced	301	34.6
28.Aynalem Y. et al, 2018 (61)	Ethiopia	Cross-sectional	2018	Adult	Experienced	263	26.2
29.Munyazesa E. et al, 2005 (62)	Rwanda	Cross-sectional	2005	Adult	Naive	936	20.5
30.Zenebe W. et al, 2018 (63)	Ethiopia	Cross-sectional	2018	Adult	Experienced	422	34.8
31.Alamdo A. et al, 2013 (64)	Ethiopia	Cross-sectional	2013	All age group	Experienced	422	52.3
32.Egata G. et al, 2014 (65)	Ethiopia	Cross-sectional	2014	Adult	Experienced	425	41.2
33.Gunda D. et al, 2012 (66)	Tanzania	Cross-sectional	2012	Adult	Experienced	346	70.71
34. Teklemariam Z. et al, 2010 (67)	Ethiopia	Cross-sectional	2010	Children	Experienced	108	54.4
35. Mkumbaye S. et al, 2014 (68)	Tanzania	Cross-sectional	2014	All age group	Experienced and naïve	869	59.5
36.Ferede G. et al, 2013 (70)	Ethiopia	Cross-sectional	2013	Adult	Naïve	420	35
37.Mulaw GF. et al, 2020 (71)	Ethiopia	Cross-sectional	2020	Children	Experienced	102	53.9
38.Bayleyegn B. et al, 2020 (72)	Ethiopia	Cross-sectional	2020	Children	Experienced and naïve	255	21.2
39.Dessu S. et al, 2020 (73)	Ethiopia	Cross-sectional	2020	Adult	Experienced	422	33.5

HAART = highly active antiretroviral therapy.

2.7. Statistical analysis

The statistical software STATA version 11 was used for data analysis. Extracted data were entered into Excel and then exported to STATA for further analysis. Random-effect model meta-analysis was used to estimate the pooled effect size and effect of each study with their confidence interval. The degree of heterogeneity between the included studies within the meta-analysis was quantified using Higgin I² statistics.^[25] If the I² value was 25%, 50%, and 75% they were assumed to show low, medium, and high heterogeneity, respectively. Subgroup analysis and sensitivity analyses were employed to resolve the occurrence of high heterogeneity in the included studies. Funnel plots analysis and Egger-weighted regression test was done to detect publication bias. A P value of < 0.05 in Egger test was considered evidence of statistically significant publication bias.^[26]

3. Ethics approval

Ethical approval was not sought for this study because the study does not contain any animal or human participants.

4. Results

4.1. Literature search and identified results

A total of 5632 studies were identified through a database literature search including a manual search. After removing duplicates and irrelevant studies, there was a total of 264 studies. Then 264 articles were screened. Out of them, 194 studies were removed by reading their titles. From the remaining 70 studies, 28 studies were removed because of dissimilarity of the study area, and 3 studies were not included due to paper quality issues (^[27–29]). Finally, after excluding irrelevant articles, 39 full-text studies were identified and used for the final qualitative and quantitative analysis (Fig. 1).

Figure 1.: Flowchart of the selection of studies for the systematic review and meta-analysis on the prevalence of anemia among HIV-infected patients. HIV = human immunodeficiency virus.

4.2. Description of included studies

In this systematic review and meta-analysis, 39 articles were included. Out of them, 29 articles were included from Ethiopia, 3 studies from Tanzania, 4 from Uganda, 2 from Kenya, and 1 from Rwanda. The sample size of included study ranged from 102 to 1205 HIV-infected individuals. Of the total included studies, 36 of them were cross-sectional studies whereas, 3 of them were cohort studies. The included studies comprised a total of 14,297 HIV-infected individuals.

4.3. The prevalence of anemia among HIV/AIDS-infected patients in East Africa

The pooled prevalence of anemia using the fixed-effect model was 29.51% (95% CI: 28.62–30.41) with an I² value of 95.3% (P = .001). Since a significant heterogeneity was observed in the fixed-effect model, the Der Simonian–Laird random-effects model was used. The pooled prevalence of anemia among HIV-infected individuals in East Africa was 35.78% (95% CI: 31.54–40.03%).

4.4. Subgroup analysis of anemia among HIV/AIDS patients based on the study population

A subgroup analysis by the study population showed that the prevalence of anemia among adult HIV/AIDS patients was 34.48% (95% CI: 29.52–39.44%) whereas the prevalence among children was 36.17% (95% CI: 26.68–45.65%) (Fig. 2).

Figure 2.: The pooled estimates of the prevalence of anemia among HIV-infected patients based on the study population. HIV = human immunodeficiency virus.

Subgroup analysis of anemia among HIV/AIDS patients based on the HAART status`

A subgroup analysis by HAART status showed that the prevalence of anemia among HAART naive HIV/AIDS patients was 39.11% (95% CI: 29.28–48.93%) whereas the prevalence among HAART experienced was 36.72% (95% CI: 31.22–42.22%) (Fig. 3).

Figure 3.: The pooled estimates of the prevalence of anemia among HIV-infected patients based on their HAART status. HAART = highly active antiretroviral therapy, HIV = human immunodeficiency virus.

1.4.4. Heterogeneity and publication bias.

The heterogeneity of the included study was high according to Higgin I² statistics (95%; P < .001). The included studies were assessed for potential publication bias visually by funnel plot and Egger statistics. In this review the funnel plot of the included studies is asymmetric (Fig. 4). In addition, the Egger-weighted regression statistics showed that (P < .05) (in this case P = .001), indicating that there is publication bias (Table 2).

Table 2 - Egger test of the included studies for the determination of pooled anemia among HIV-infected patients.

Std_Eff	Coefficient	Standard error.	t	P > t	[95% confidence interval]
Slope	14.08946	4.387806	3.21	.003	5.198922	22.98
Bias	5.939884	1.542584	3.85	.001	2.814313	9.065455

HIV = human immunodeficiency virus.

Figure 4.: Funnel plot of the included studies to determine the pooled prevalence of anemia among HIV-infected patients. HIV = human immunodeficiency virus.

2.4.4. Trim and fill analysis.

As the P value is < 0.05 (0.001), there is publication bias (small study effect). Therefore, trim and fill analysis was implemented. After trim and fill analysis, the pooled estimated prevalence of anemia was 25.35% (95% CI: 20.69–30.03%).

Number of studies trimmed to fit the model = 0.

Number of studies filled to fit the model = 14.

4.5. Sensitivity analysis

A sensitivity analysis was carried out on the prevalence of anemia among HIV patients by applying random-effect models. The analysis was done to evaluate the influence of each study on the pooled estimated prevalence of anemia. The result showed that omitted studies did not show a significant effect on the pooled prevalence of anemia among HIV/AIDS patients (Table 3).

Table 3 - Sensitivity analysis of the included studies to estimate the pooled prevalence of anemia among HIV-infected patients.

Study omitted	Estimate	[95% confidence interval]
Addis Z (2014)	35.61	31.30–39.91
Fekene T (2013)	35.29	31.11–39.47
Ferede G (2012)	35.83	31.45–40.21
Gebregziabher Y (2013)	35.98	31.62–40.33
Geletaw T (2015)	35.59	31.28–39.89
Gunda D (2016)	35.29	31.02–39.55
Kamau J (2004)	35.72	31.39–40.04
Kyeyune R (2012)	35.42	31.18–39.64
Mengistu A (2019)	36.09	31.69–40.49
Mugisha, J. O. (1993)	36.28	31.93–40.62
Tamir Z (2016)	35.56	31.26–39.85
Zerihun K. W. (2017)	35.86	31.48–40.23
Ageru T. A. (2016)	35.79	31.42–40.15
Enawgaw B (2012)	36.42	32.24–40.59
Katemba C (2016)	34.85	30.76–38.93
Yemane B (2019)	35.86	31.52–40.20
Rahel A (2015)	36.34	32.05–40.63
Gedefaw L (2012)	36.16	31.80–40.50
Melese H (2015)	36.18	31.79–40.55
Enawgaw B (2013)	36.34	32.04–40.63
Mihiretie H (2013)	36.17	31.83–40.51
Gebremedhin K. B. (2018)	35.84	31.48–40.19
Aynalem, Y. A. (2018)	36.03	31.71–40.35
Munyazesa E. (2005)	36.26	31.86–40.67
Zenebe, W. A. (2018)	35.84	31.46–40.21
Egata G (2014)	35.64	31.32–39.95
Gunda D. W. (2012)	35.06	30.82–39.29
Teklemariam Z. (2010)	35.29	31.04–39.54
Fenta D. (2019)	36.43	32.17–40.70
Kibaru E. (2008)	35.79	31.48–40.09
Ndeezi G. (2007)	34.82	30.62–39.01
Gebreweld A. (2018)	36.11	31.79–40.43
Deressa T. (2016)	36.06	31.75–40.38
Alamdo A. (2013)	35.42	31.14–39.69
Mkumbaye S. (2014)	35.26	30.50–39.51
Ferede G. (2013)	35.81	31.50–40.12
Mulaw GF (2020)	35.31	31.05–39.57
Bayleyegn B (2020)	36.17	31.85–40.48
Dessu S. (2020)	35.86	31.53–40.16
Combined	35.78	31.54–40.03

HIV = human immunodeficiency virus.

5. Discussion

Anemia is the most prevalent hematological abnormality in people with HIV/AIDS. Opportunistic infections, malignancies, and drug side effects can lead to anemia in HIV/AIDS patients.^[14–16] Several investigations that had been carried out in Africa also provided proof of the presence of this abnormality in HIV-positive individuals. This systematic review and meta-analysis was conducted to determine the pooled prevalence of anemia among HIV/AIDS patients in East Africa using previously conducted research. The prevalence varies from study to study in different East African countries. This variation may arise from the difference in sample size, study design, and socioeconomic factors of the participants in the included studies.

The finding of this systematic review and meta-analysis showed that the pooled prevalence of anemia among HIV-infected individuals in East Africa was 25.35% (95% CI: 20.69–30.03%). Anemia is a major complication of HIV/AIDS disease, as evidenced by the higher prevalence of the condition among HIV/AIDS patients. This will put the affected patients at higher risk of experiencing many problems. The likely reason for this higher prevalence can be caused by opportunistic infections, the direct viral action of HIV, unfavorable effects of antiviral medicines provided to HIV patients, and nutrition deficiencies.^[17] The pooled prevalence of anemia in this systematic review and meta-analysis is in agreement with Pamela S’s literature review^[30] which reported the global prevalence of anemia range from 1.3 to 95%.

The subgroup analysis by age group showed that the pooled prevalence of anemia among adult HIV/AIDS patients was 34.48% (95% CI: 29.52–39.44%) whereas the prevalence in children was 36.17% (95% CI: 26.68–45.65%). There is no significant difference in the prevalence of anemia between children and adults. Furthermore, the result of the subgroup analysis for adult HIV/AIDS patients is in line with a systematic review and meta-analysis by Negesse A et al^[31] (31.00% [95% CI: 23.94–38.02]) in Ethiopia. Nevertheless, the result of subgroup analysis for children was higher compared to a systematic review and meta-analysis by Wagnew F. et al^[32] (22.3% [95% CI: 18.5–26.0%]) in Ethiopia. The possible reason for the increment of anemia in this review may be due to the inclusion of studies that are conducted in different East African countries. This may lead to a difference in sample size and nutritional status of the study participants. As it is known the prevalence is calculated based on the sample size and this variation in sample size greatly affects the prevalence in the 2 reviews.^[33] Moreover, in this systematic review and meta-analysis, different studies have been included from different East African countries. East Africa is a region with the highest prevalence of stunting compared to the global average both in children and adults.^[34]

Moreover, subgroup analysis by study area was also conducted. Based on this subgroup analysis, the highest prevalence of anemia among HIV/Positive patients was observed in Tanzania (62.34% [95% CI: 53.52–71.16%]) whereas the lowest prevalence was observed in Rwanda (20.50% [95% CI: 17.91–23.09%]). This wide variation in prevalence may have resulted from the difference in the sample size of the studies. The variation in the nutritional status of the population that lives in the study area also contributes to this variation in prevalence. Additionally, a subgroup analysis by HAART status showed that the prevalence of anemia among HAART naive HIV/AIDS patients was 39.11% (95% CI: 29.28–48.93%) whereas the prevalence among HAART experienced was 36.72% (95% CI: 31.22–42.22%). There is no statistically significant difference in the prevalence of anemia among HAART naive and HAART experienced patients.

6. Strengths and limitations

The strengths of this systematic review and meta-analysis comprise an inclusive search using various databases, employment of different searching strategies, critical appraisal of the methodological quality of the included studies using JBI critical appraisal tools, and application of the Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines. The limitations of this systematic review and meta-analysis were that most of the included studies are from Ethiopia and only a few studies were available from other countries. This may influence the representativeness of the pooled estimate for other East African countries. The inclusion of studies that are published only in English may also influence the representativeness of the findings. The other major limitation of this systematic review and meta-analysis was even after we performed subgroup analysis high heterogeneity was observed in all analyses.

7. Conclusion and recommendation

This systematic review and meta-analysis revealed that anemia is the most common hematological abnormality among HIV/AIDS patients in East Africa. It also underscores the importance of taking diagnostic, preventive, and therapeutic measures for the management of anemia among HIV/AIDS-infected patients. These will help to reduce morbidities and mortalities due to anemia.

Acknowledgments

First of all, we would like to acknowledge the Department of Hematology and Immunohematology, School of Biomedical and Laboratory Sciences, College of Medicine and Health Sciences, the University of Gondar for allowing us to do this Meta-analysis.

Author contributions

Conceptualization: Fasil Getu, Melak Aynalem, Muluken Walle, Bamlaku Enawgaw.

Formal analysis: Fasil Getu, Melak Aynalem, Muluken Walle.

Investigation: Bamlaku Enawgaw.

Methodology: Fasil Getu, Melak Aynalem, Muluken Walle, Bamlaku Enawgaw.

Project administration: Bamlaku Enawgaw.

Resources: Fasil Getu, Muluken Walle.

Software: Fasil Getu, Melak Aynalem, Muluken Walle, Bamlaku Enawgaw.

Supervision: Bamlaku Enawgaw.

Validation: Fasil Getu.