Search findings and study characteristics
The preliminary search retrieved 1,772 studies; from PubMed (n=1,634), Scopus (n=15), African Online journals (n=34), Google Scholar (n=62), Addis Ababa University online research repository (n=20), and by reviewing the reference lists of articles (n=7). Finally, we found 18 studies that fulfilled the eligibility criteria for further quality assessment (Figure 1).
All the included studies were cross-sectional studies. Three studies conducted by the year 2012, the other two in 2016, and four studies in 2018. For the other years, only one study in each was done. The maximum (91%) HIV status disclosure to sexual partner observed by the year 2017 and the minimum (57.4%) during 2015. While majority of studies (n=12) included both men and women study participants (28-39), the rest of the studies (n=6) six studies (40-45) included only women (Table 1).
Table 1: The characteristics of studies
First Author /Year
|
Study area
|
Study participant
|
Sample size
|
Response Rate (%)
|
Source of fund
|
Alemayehu M et al/2014 (40)
|
Mekelle, Northern Ethiopia
|
Women
|
315
|
100
|
Sheba University College, Ethiopia
|
Sendo EG et al/2013 (41)
|
Addis Ababa, Central Ethiopia
|
Women
|
107
|
95.5
|
Alkan University college and NUFU/ GEMESO Research Project on HIV/AIDS
|
Erku TA et al/2012 (28)
|
Woldia, Northern Ethiopia
|
Both sex
|
334
|
100
|
University of Gondar, Ethiopia
|
Deribe K et al/2008 (29)
|
Jimma, Southwest Ethiopia
|
Both sex
|
705
|
99.9
|
Netherlands Government Multi-Country Support Program on Social Science Research in the field of HIV/AIDS
|
Deribe B et al/2018 (42)
|
Hawassa, Southern Ethiopia
|
Women
|
207
|
100
|
Not mentioned
|
Dessalegn NG et al/2019 (30)
|
Addis Ababa, Central Ethiopia
|
Both sex
|
676
|
100
|
Australian Department of Foreign Affairs and Trade and Western Sydney Sexual Health Clinic
|
Gari T et al/2010 (43)
|
Hawassa, Southern Ethiopia
|
Women
|
384
|
100
|
EPHA-CDC project
|
Geremew TD et al/2018 (31)
|
Bale, Southern Ethiopia
|
Both sex
|
411
|
100
|
Not mentioned
|
Genet M et al/2015 (32)
|
Mekelle, Northen Ethiopia
|
Both sex
|
324
|
100
|
Not mentioned
|
Seid M et al/2012 (33)
|
Kemissie, Northen Ethiopia
|
Both sex
|
360
|
100
|
Not mentioned
|
Tesfaye T et al/2018 (34)
|
Jimma, Southwest Ethiopia
|
Both sex
|
351
|
98.1
|
Jimma University, Ethiopia
|
Gadisa T et al/2017 (35)
|
Six HIV clinic in Central Ethiopia
|
Both sex
|
1,180
|
100
|
National Institutes of Health and President’s Emergency Plan for AIDS Relief
|
Kassaye KD/2005 (45)
|
Gore and Mettu, Southern Ethiopia
|
Women
|
42
|
100
|
Menschen fur Menschen foundation IIRDP
|
Alema HB et al (36)
|
Axum, Northen Ethiopia
|
Both Sex
|
361
|
99.7
|
Bahir Dar University
|
Kassahun G et al(44)
|
Jimma, Southwest Ethiopia
|
Women
|
337
|
99.7
|
Jimma University
|
Natae et al (37)
|
West shewa, Central Ethiopia
|
Both sex
|
420
|
99.5
|
Not mentioned
|
Reda AA et al (38)
|
Eastern Ethiopia
|
Both sex
|
1537
|
98.4
|
Not mentioned
|
Koyira A (39)
|
Addis Ababa, Central Ethiopia
|
Both sex
|
341
|
100
|
Ethiopia Public Health Association-Disease Control and Prevention
|
Publication bias and quality status
In Egger’s regression test, p-value was 0.870. The distribution of original studies on the funnel plot is shown below (Figure 2). We excluded one study (45) due to poor quality status, significantly very low sample size for cross-sectional study design. The quality appraisal result is shown in the supplementary file (Supplementary file).
HIV positive disclosure status
The smallest and largest sample size considered in the analysis was 107 (41) and 1,537 (38), respectively. Seventeen studies with 8,009 individuals were included in the meta-analysis. The proportion of HIV infected people that disclose their HIV positive status to their sexual partner was 76.03 % (95% CI: 68.78, 83.27) (Figure 3).
Subgroup analysis
Based on the subgroup analysis, 75.70% of women and 76.16% of men who were infected with HIV disclosed their HIV positive status to their sexual partners (Figure 4).
Sensitivity analysis
The sensitivity analysis showed that no study leaves have a significant impact on the overall estimation (Table 2).
Table 2: The proportion (P) with 95% CI of HIV sero-disclosure practice to sexual partner when one study omitted from the analysis a step at a time
Trend analysis
Considering the year of publication, the trend graph was generated. The trend line shows HIV status disclosure through time (Figure 5).
Associated factors
Socio-demographic characteristics
According to a single study report (34), those study participants below the age of 39 years were less likely (AOR=0.014; 95% CI= 0.005, 0.037) to disclose their HIV status to sexual partner than above 39 years old. Another study (35) revealed that being 40-44 years old (AOR=0.52; 95%CI:0.44, 0.61), 45 and above age group (AOR=0.38; 95% CI: 0.22, 0.65) was less likely to disclose the HIV status to their sexual partner compared to the age of 25-29.
Based on one study report, males (AOR=3.039: 95% CI=1.164, 7.935) (34) were more likely to disclose HIV status to their sexual partners. In support of this study, female was less likely (AOR=0.25; 95% CI: 0.14, 0.45) (35) to disclose their HIV status to their sexual partner.
As one study has shown, live in urban (AOR = 1.62; 95% CI=1.0, 2.60) (31) was in support of HIV disclosure practice.
One study (36) revealed positive association between unmarried (AOR=3.71; 95% CI=1.21, 11.39) and disclosure status while another study (37) found negative association (AOR = 0.12; 95% CI= 0.036-0.39). Those who had children (AOR=9.89; 95% CI= 2.68, 36.36 ) (42) were more likely to disclose their HIV positive status to sexual partners as reported from a single study.
Well-educated study participants (AOR=0.4; 95% CI=0.17-0.92) (28) and secondary education (AOR=0.6: 95% CI= 0.39, 0.92) (35) were less likely to disclose their HIV positive status than non-educated people as evidenced each from one study. Those study participants who took the position of control-over household assets were less likely (AOR=0.21; 95% CI= 0.12, 0.36) to disclose their HIV status compared to counterparts (35).
Medical-related factors
Presence of comorbid medical illness (AOR= 2.5; 95%, CI= 1.5, 4.2) (34), and having any clinical symptoms for HIV (AOR= 2.98; 95% CI=1.72, 5.15) (34) were more likely to disclose once’s HIV positive status to sexual partner. Advanced stage HIV disease at enrollment time to care was associated (AOR=3.26; 95 % CI= 1.76–6.04) with non-disclosure status (35). The other study (29) showed that those who were on the WHO clinical stage I and II found less likely (AOR=0.22; 95% CI =0.10-0.55) to disclose their HIV status. However, another single study found being on the WHO stage I and II were more likely (AOR=2.77; 95% CI= 1.32-5.79) to disclose their HIV status (34).
The pooled effect from two studies (28, 44) showed that being on ART was positively associated with HIV status disclosure (Table 3).
Psychosocial-related factors
Having open discussion on safer sex with partner (44), using condom always (AOR= 6.20; 95% CI= 2.52–15.25) (30), having greater social support (AOR=2.98; 95%CI=1.09, 8.14) (30), being the members of close-knit social groups (AOR=2.78; 95% CI=1.1, 6.7) (35), being peer counselor (44), and low physical domain-related quality of life (AOR=3.83; 95% CI=2.01, 7.32) (34) were more likely to disclose HIV status to their partner. Those who had high social domain-related quality of life (AOR=0.053, 95% CI= 0.022, 0.125) (34) and low negative self-image (AOR=0.03; 95% CI= 0.04, 0.70) (29) were affect negatively their HIV disclosure status as showed each with single study.
Those who had a chance of seeing people with HIV positive who disclose their HIV status to the community (AOR= 2.1, 95% CI: 1.08, 4.01) (40) and knew other people living with HIV (AOR=4.76; 955CI= 2.63, 9.09) (35) were supportive factors to disclose HIV status to sexual partners.
The pooled effects of cohabiting with partner, smooth relationship with the partner, getting counseling, had a prior discussion with a partner about HIV/AIDS and HIV test, knowing partner’s HIV status, and being a member of anti-HIV association are illustrated on Table 3.
Table 3: List of variables with their pooled AOR (95% CI) and I-square percentage with its p-value
Variables
|
AOR (95% CI)
|
I-squared with p-value
|
Being on ART (28, 44)
|
6.19 (2.92, 9.49)
|
84.5%, 0.011
|
Cohabiting with partner (29, 30)
|
4.48 (1.24, 7.72)
|
0.0%, 0.454
|
Getting counseling (28, 31, 40, 42)
|
3.94 (2.08, 5.80)
|
23.9%, 0.268
|
Had discussion prior to HIV testing (29, 32, 37, 40, 41)
|
4.40 (2.11, 6.69)
|
0.0%, 0.972
|
Knowing partner’s HIV status (28, 30-32, 36, 37, 40)
|
6.08 (3.05, 9.10)
|
27.3% 0.220
|
Smooth relationship with partner (30, 34)
|
4.44 (1.28, 7.61)
|
0.0%, 0.616
|
Being member of anti-HIV association (36, 44)
|
3.70 (2.20, 5.20)
|
28.9%, 0.236
|