Abstract:
Background: HIV infection is one of the most pandemic infectious diseases in the world.
Antiretroviral therapy is a lifelong HIV/AIDS treatment for people living with human immune
deficiency virus. In Ethiopia ART service has been scaled up and increased in cohort of users but
still losses to follow-up and early mortality of patients on ART have been challenges for the
success of the program.
Objective: To assess the gender difference in survival status among ART users in Yirgalem
General Hospital, Sidama zone, SNNPR, Ethiopia.
Method: retrospective cohort study was conducted from September1
st 2010 to August 30th 2015,
in Yirgalem general hospital. A sample of 687 ART users was included in the study using simple
random sampling technique. Data were collected using structured checklist and entered to EPI
data version 3.1 then exported and analyzed by STATA version 13. Descriptive statistics:
Proportion, mean, median and standard deviation were used. The Kaplan Meier curves were
used to estimate survival function of male and female ART users and biomarkers of ART. Log
rank test was employed for the assurance of statistical difference in the Kaplan Meier curves.
The Cox-proportional hazards regression model was employed to identify predictors of survival
status. Assumption of proportional hazard model was checked using goodness of fit test and
Schoenfeld residual plot and also interaction among predictor variable were checked.
Result: The numbers of records reviewed were 412 females and 273 males with retrieval rate of
99.7%. The overall survival probability was 84.5%. Female and male had survival probability of
80.1% and 91.18% respectively. Factors: females (AHR=1.79(1.04, 3.06)), divorce
(AHR=2.09(1.10, 3.97)), no education (AHR=2.54(1.29, 4.98)) and primary education
(AHR=2.07(1.18, 3.65)), never disclosing of HIV status (AHR=3.62(1.25, 10.46)), bedridden
functional status (AHR=2.71(1.24, 5.89)) and TB-co infection (AHR=2.60(1.48, 4.45)) were
found to be independent predictors of survival status.
Conclusion and recommendation: overall survival probability was low. Predictors of
survival were: sex, educational level, disclosure of HIV status, marital status, functional status
and TB-co infection. Intervention to further reduce mortality, should focus on high risk group
and expansion of HIV counseling and testing service in facilities and community level.
