Received Date: January 30, 2014; Accepted Date: March 03, 2014; Published Date: March 15, 2014
Citation: Kedir AA, Desta A, Fesseha G (2014) Factors Affecting Survival of HIV Positive Children Taking Antiretroviral Therapy at Adama Referral Hospital and Medical College, Ethiopia. J AIDS Clin Res 5:289. doi:10.4172/2155-6113.1000289
Copyright: © 2014 Kedir AA, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of AIDS & Clinical Research
Background: The aim of this study is to explore factors affecting survival of children living with HIV/AIDS after initiation of ART. In which it highlights the need for local evidence to promote interventions that optimize survival among HIV-infected children on ART in Ethiopia.
Methods: Institution based retrospective cohort study was employed on 560 children enrolled on ART from January, 2006-December, 2010. Information on relevant variables was collected from patients` medical cards and registries. Univariate analysis was used to describe the baseline characteristics of the patients`. Life table was used to estimate survival after initiation of ART. Log rank test was used to compare survival between different categories of independent variables. Multivariable Cox proportional model was fitted to identify factors affecting survival after initiation of ART.
Results: Children on ART were followed for a median follow up period of 47 months (IQR=29, 62). At the end of follow up, 364 (65%) were alive and 43 (7.6%) were reported dead. More than three fourth of the deaths occurred within the first sixth months of starting ART. The estimated cumulative survival probabilities were 0.939, 0.928, 0.926, 0.923, 0.920, and 0.916 at 6, 12, 18, 36, 48, and 60 months, respectively. Anemia (hemoglobin level<10 gm/dl) (AHR=2.60, 95% CI=1.41, 4.84), absolute CD4 cell count below the threshold for severe immunodeficiency (AHR=3.55, 95% CI 1.48, 8.46), advanced WHO staging (stage IV) (AHR=3.08, 95% CI=1.27, 7.47), and underweight (AHR=2.49, 95% CI 1.27, 4.88) have found to be predictors of mortality after ART initiation.
Conclusions: Mortality was high especially during the first sixth months following ART initiation. Therefore, close follow up of HIV exposed children to make early diagnosis and treatment initiation before the development of severe immune deficiency and advanced clinical stage is important.
Survival status; HIV/AIDS; ART; Children; Adama; Oromia; Ethiopia
For three decades, HIV/AIDS has been overwhelming individuals and families with the disaster of untimely death and medical, financial and social burdens. In 2012, globally 3.4 million people living with HIV were children under age of 15 years; and an estimated 250,000 children under age of 15 years were died due to AIDS related cases globally in 2011 . Sub-Saharan Africa, a region with only 12% of the global population; remains the most heavily affected region by HIV. In 2011, about 69% of all people living with HIV and more than 50% of deaths from AIDS-related illnesses in adults and children`s occurred in this area .
The development of HIV into AIDS is very rapid in a proportion of infants and children; in the absence of appropriate treatment; a third of children infected with HIV will die before their first birthday . In 2009, about 260,000 deaths due to HIV were reported in under-15 years children; of which a significant number of deaths could have been averted by early diagnosis and treatment . Although the risk of early mortality is high in HIV infected children, the average age of ART initiation in low income countries stayed high . Studies conducted following the recommendation of WHO for ART in infants and children supported that early initiation of ART prevent early mortality .
Though survival of HIV positive children in Ethiopia and similar settings has improved as a result of increased access to ART, it is still low in the first six months after initiation of ART . However, studies conducted in Kenya, Zambia, and Malawi; the death of HIV positive children following ART initiation remains high, ranging from 7.5% to 15% [8-10]. Nevertheless, HIV positive children in developed countries have a substantially higher survival probability after initiation of ART compared to children in developing countries .
The effect of socio-demographic, clinical and immunological factors on survival of HIV positive children after initiation of ART were also studied in low and middle income countries. However, the findings of the studies were not consistent. For instance, in studies conducted on cohorts of HIV positive children in Cote d`Ivoire, Malawi and Zambia, only low weight-for-height and low CD4% were found to have significant negative effect on survival of children after initiation of ART [7,9,10]. However, in another large cohort of HIV positive children of Zambian, besides low CD4% and low WHZ score, younger age and low hemoglobin level were also found to have significant negative effect on survival after initiation of ART .
In addition, the scarcity of data on survival of children receiving ART beyond 2 years of follow up was another reason why this study conducted. However, in this study the children were followed for longer period to estimate their survival status. Though ART service is being provided since 2005, there are no studies or documented report on survival status and factors affecting survival of children on ART in the study area. This highlights the need for local evidence to estimate the survival status and identify factors affecting survival of children on ART at Adama referral hospital and Medical College. Therefore, the purpose of this study is to determine survival status and identify factors affecting survival of HIV positive children after initiation of ART.
Study setting and period
The setting for this study was Adama Referral Hospital and Medical College located at the center of Adama Town. Adama town, located at 99 km southeast of Addis Ababa, is an administrative center for East- Shoa Zone in Oromia Regional State. Adama Referral Hospital and medical college was one of the first three hospitals in Oromia that started providing ART services since 2005. HIV infection is confirmed by serological testing or DNA PCR used in infants under 18 months of age. At ART clinic, HIV positive children are screened for opportunistic infections, evaluated for clinical staging and eligibility for ART by ART physician or nurse. Patients were managed in accordance with World Health Organization (WHO) recommendations  and national guidelines . The ART case team in the hospital comprised of ART trained physicians; ART trained nurses, pharmacist, laboratory technicians, data clerks, and drug adherence counselors. A total of 1860 children have been enrolled since March 2005, 1350 children ever started on ART and 980 children currently on ART . The study subjects were followed from January 1st 2006 to march 6th 2013.
Study design and sampling
Institution based retrospective cohort study design was conducted. All Children living with HIV/AIDS, age ≤ 14 years and started ART at Adama referral hospital and Medical College were source population for the study. The study population was randomly selected medical records of HIV positive children that were enrolled on ART between January 1st 2006 and 30th of December 2010 at the hospital. The sample size was calculated by applying two sample proportion formula using Epi-Info version 7. The most significant predictors of survival (absolute CD4 count) were used . The calculated sample size was 560. Sampling frame containing 1030 medical card numbers of children living with HIV/AIDS that started ART between January 1st 2006 and 30th of December 2010 was prepared. Then from the sampling frame, the medical cards randomly selected using select cases randomly option in SPSS v.20 statistical software.
The data source for this study was medical records of HIV positive children enrolled on ART at Adama Referral Hospital. The time to death was the outcome variable. The time to death was assessed for subjects from the date of ART initiation to the date of death occurring before March 6th, 2013 (end of study period). Children were followed until they died; loss to follow up, transferred to other health facility, or administratively censored on March 6th, 2013. Data on deaths of the HIV positive children on ART were obtained from physician report on the medical cards. Besides, for those children who died at home, the drug adherence counselor communicated using the contact address and confirmed whether the children were alive or dead. Checklist for data collection was adopted from intake and follow up form used in the children ART clinics. Measurements were developed from WHO standards and by reviewing relevant literatures [17-19]. Pre-testing was under taken in Adama health center before data collection and some minor modification were made accordingly. The data was collected by reviewing the patient’s medical cards (follow up and ART intake form) from March 1st to 15th, 2013. Three clinical nurses trained on ART were employed for data collection after receiving one day training on the objectives and process of data collection. The most recent laboratory results before starting ART was used as a baseline value. Kaplan-Meier was used to estimate survival after the advent of ART, and log rank test was used to compare the observed survival curves. Those variables with P-value less than 20 percent in log rank test were selected as potential candidate for the multivariable cox proportional hazard model . Purposeful selection of covariates method was used to select variables for the multivariable model. No interaction was found among covariates in the final model. Only independent variables that will have statistically significant association with the outcome variable on multivariable Cox-proportional hazard will be retained in the final model. The Proportional hazard assumption was checked for variables included in the final model using re-estimation based method and none of the variables in the final model violated the proportionality assumption. The overall goodness of model fitness was checked using Cox-snail method. Alpha 5% was used as level of significance. Weightfor- age was calculated using anthropometric calculator of WHO anthro plus software.
The quality of the data was assured by caring out careful design of the check list, pre-test, cleaning of data and appropriate recruitment and training was given for data collectors. Five percent of the completed data collection checklist was selected randomly and cross checked with medical cards of the patients to check for consistency at the end of each day. The collected data was coded, entered, cleaned and analyzed using STATA version 11 statistical package.
Anemia: was defined as having hemoglobin level below 10 gm/dl.
CD4 count below threshold: was classified according to the age of a child. For infants CD4 count less than 1500/mm3, For age 12-35 months CD4 count less than 750/mm3, For age 36-59 months CD4 count less than 350/mm3, For age ≥ 60 months CD4 count less than 200/mm3 (38).
Fair adherence: if the percentage of missed dose is between 85-94%
Good adherence: if the percentage of missed dose is above >95%
Lost to follow up: if a patient discontinued ART for at least one to three month.
Regressed developmental milestone: if a child loses what has been attained for age.
Underweight: was defined as weight for age Z-score<-2 SD for under-five children and BMI-for-age score<-2 SD for children ≥5 years old.
Ethical clearance was obtained from the Mekelle University College of Health Sciences, Institutional Research Ethics Review Committee. Following the approval, Official letter of co-operation was written to concerned bodies by the School of Public Health of Mekelle University. Permission was granted from Oromia Regional Health Bureau and Adama referral hospital and Medical College. As the study was conducted through review of medical cards, the individual patients was not exposed to any harm as far as the confidentiality maintained. Moreover, no personal identifier was used on data collection form. The recorded data was not accessed by a third person except the principal investigator, and kept confidential.
Baseline socio-demographic profile of children
Five hundred sixty (560) medical cards of children living with HIV/AIDS and who started ART were reviewed for this study. Three hundred thirty three (61.5%) were above age 5 years at initiation of ART and those age less than 1 year accounts for 51 (9.1%) of the children. Two hundred seventy seven (49.4%) of the children were males. Four hundred forty one (78.7%) were from urban. Of 499 children whose orphan hood status were recorded, 161(32.2%) of the children have lost either of their parents, (Table 1).
|Care giver (n=551)|
|Orphan hood (n=499)|
|Both parents are alive||244||48.9|
Table 1: Baseline socio-demographic characteristics of children started ART at Adama referral hospital and medical college, January 1st, 2006 to December 30th, 2010.
Baseline clinical and immunological profile of children
Immunization status recorded were recorded for 347 HIV positive children; of which 294(84.7%) were vaccinated at appropriate age. Regarding feeding practice, 83(33.4%) HIV positive children were on exclusive breast feeding; whereas 154(62.1%) on mixed feeding. Four hundred twenty (75%) were on cotrimoxazol prophylaxis therapy at ART initiation. Four hundred ninety four (96.5%) of the children found to have good adherence for ART, whereas 10 (2.0%) and 8(1.5%) had fair and poor adherence level, respectively. Three hundred forty nine (62.7%) of the children had absolute CD4 count below threshold for immune deficiency. Of 522 children whose hemoglobin value at initiation were recorded, 109 (20.9%) had anemia (Hgb<10 gm/ dl). Two hundred eighty nine (51.6%) of them were malnourished (underweight). Sixty three (11.7%) and 16 (3%) of the children had a delayed and regressed developmental milestone, respectively (Table 2).
|Appropriate for age||294||84.7|
|Not appropriate for age||41||11.8|
|Nutritional history (n=248)|
|Exclusive Breastfeeding only||83||33.4|
|Infant prophylaxis (n=189)|
|sdNVP + AZT||7||3.7|
|ART Adherence on follow up (n=512)|
|Absolute CD4 count|
|CD4 count above threshold||207||37.2|
|CD4 count below threshold||349||62.7|
|≥ 10 gm/dl||413||79.1|
|< 10 gm/dl||109||20.9|
|WHO clinical staging|
|Developmental milestone (n=537)|
|Opportunistic Infections on follow up|
|ART drug side effect during follow up|
Table 2: Baseline clinical and immunological status of children started ART at Adama referral hospital and medical college, January 1st, 2006 to December 30th, 2010.
A total of 560 children on ART were followed for a median follow up time of 47 (IQR, 29-62) months. At the end of follow up, 364 (65%) were alive, 46 (8.2%) were loss to follow up, 107 (19.1%) were transferred out and 43 (7.6%) were reported dead. Thirty three (76.7%) of the total deaths have occurred within the first sixth months of starting ART. The overall mortality rate was 2.06 deaths per 100 child-years (43 deaths over a 2078 child-years). The estimated cumulative survival probability was 0.958 after 3 months, 0.939 after 6 months, and 0.920 after 12 months of ART initiation. The cumulative survival after 54 months was 0.916 (Table 3 and Figure 1).
|Time in month||No of children at start||Number of deaths||Survival function|
Table 3: Estimates of the cumulative progression to death for the cohort of children (N=560) starting ART b/n January 1, 2006 up to December 30, 2010.
Kaplan-Meier survival curves among categories of the variables
Children on ART who had normal weight for age have significantly higher survival probability compared to those under-weight at ART initiation (log rank, P<0.05). The survival probability for children with CD4 count above threshold for immunodeficiency were significantly higher than those with CD4 count below threshold at initiation of ART (log rank, P<0.05). Children on ART that had anemia at initiation of ART had lesser survival probability compared to those with no anemia (log rank, P<0.05). Children that initiated ART at advanced stage of the disease (WHO stage IV) progression had significantly lower survival probability compared to those who imitated early in the disease progression (log rank, P<0.05) (Figure 2).
Factor associated with survival rate of children living with HIV/AIDS and who started ART
In multivariable Cox regression analysis, children with Low hemoglobin level (<10 gm/dl) at ART initiation were 2.49 times (AHR=2.60, 95% CI=1.348 to 4.610, P=0.002) more likely to die as compared to those with hemoglobin level 10 gm/dl and above after adjusting for the effect of nutrition status, CD4 count, and WHO staging. Children that were presented for treatment with advanced WHO staging (stage IV) had almost three times (AHR=2.99, 95% CI=1.237 to 7.245, P=0.015) the risk of death of those presented at early staging (stage I/II). The risk of death for children with CD4 count below threshold for immunodeficiency was more than three times (AHR=3.33, 95% CI 1.401 to 7.923, P=0.004) higher than those children with CD4 count above threshold. Children with low weight for age (underweight) at ART initiation were nearly two and half times (AHR=2.42, 95% CI 1.239 to 4.744, P=0.008) more likely to die as compared to those with normal weight (Table 4).
|Variable||AHR (95% CI)||P values|
|Absolute CD4 count|
|Below threshold||3.33 (1.401-7.923)||0.006*|
|≥ 10 gm/dl||1||1|
|< 10 gm/dl||2.49 (1.348-4.610)||0.004*|
|WHO clinical staging|
|Stage IV||2.99 (1.237-7.245)||0.015*|
|Under weight||2.42 (1.239-4.744)||0.010*|
Table 4: Multivariable Cox regression analysis of socio-demographic characteristics, clinical and immunological status among children started ART at Adama referral hospital and medical college, between January 1st, 2006 and December 30th, 2010.
In this study high early mortality was observed. In fact nearly three fourth of the total deaths occurred during the first six months following ART initiation. And children who survived to this period were less likely to die in the succeeding follow up time. On multivariable Cox proportional hazard regression, factors affecting survival of children following ART initiation were CD4 count below threshold for immunodeficiency, underweight for age, WHO stage IV, and low hemoglobin level at ART initiation.
The results from this study showed a significant decrease in the estimated cumulative survival; especially during the first six months following ART initiation. This finding is comparable with reports from studies conducted in Sub-Saharan African countries [21-23] and in Ethiopia [17,24,25]. However, survival estimate in this study was higher when compared to studies from Malawi . This discrepancy could be explained in three ways. Firstly, the difference in study period as there are changes in treatment and care of children on ART through time given that the Malawi study was conducted 5 years ago. Secondly, the Malawi study was conducted in small sample size (n=258) as compared to this study; and this may affect the estimation. Thirdly, it may be due to the health care system changes in Ethiopia such as task shifting and delegation of HIV/AIDS services to low and middle-level health care providers. On the other hand, survival estimate in our study was lower when compared to children on ART in developed nation. Children in developed countries had higher CD4 level and lower baseline viral load at ART initiation. This generally indicates that children in resource rich countries initiate ART earlier in their illnesses which resulted in higher survival probability following ART initiation . Three fourth of the total deaths in our study occurred in the first sixth months following ART initiation. This finding was similar to other studies conducted in sub-Saharan Africa [16,22,23,28] which showed 69%-89% of deaths in the first sixth months of ART initiation. This could be due to delayed presentation for treatment as it is evidenced by this study that more than two third of the children had WHO stage III/IV, more than half had CD4 count below threshold for immunodeficiency, and more than half had growth failure (underweight for age) at initiation of ART.
In our study, children with absolute CD4 count below the threshold level for immunodeficiency at initiation of ART have higher risk of death than those children with CD4 count above threshold. This finding is consistent with reports from studies conducted on cohorts of children in Africa [16,21,25,29-31]. The results from this study showed that large proportion of the children had risk factor for immune reconstitution inflammatory syndrome (IRIS) such as low weight for age and severe immune suppression . Consequently, it is also possible that IRIS has a role in some of the deaths. Our result indicates that anemia as a predictor of mortality for children on ART is consistent with reports from several pediatric studies [16,23,25,32-35]. The levels of anemia in question (Hgb<10 g/dl) are modest and would not normally lead to rapid clinical deterioration . However, in children living with HIV/ AIDS it is a surrogate marker for advanced stage of HIV/AIDS . As it is evidenced in this study, more than 70% of the children started ART at advanced WHO clinical stage.
Children with growth failure (measured as low weight-for-age) were another predictor of mortality for children on ART. This finding is consistent with reports from different pediatric studies [25,28-30,34-36]. In the context of this study, in which more than two third of children classified in immune suppressed category, low weight for age would contribute additional risk of death due to nutritional deficiency. It is plausible that children with conditions, advanced HIV/AIDS and severe malnutrition will have limited capacity for immune recovery and are susceptible to life threatening opportunistic infections. Consequently, this will increase the risk of death of children on ART. We also found advanced WHO clinical staging as another predictor of mortality for children after initiation of ART. Our finding of advanced clinical stage as predictor of mortality for children on ART was also supported by reports from other studies in Africa [21,24,25,34,35,37]. It is expected that the capacity of immune recovery will be limited for children with advanced WHO clinical stage (stage IV) at initiation of ART. And this will lead to the occurrence of life threatening opportunistic infections and increased risk of early death.
As limitation of this study Mortality might be underestimated as the considerable number of children lost to follow up may include children who died. In addition, the use of secondary data source as incompleteness was inevitable. Lack of information on the possible cause of death; over representation of older children (age above 5 years) and lack of viral load information might affect survival of children under ART treatment. Therefore any interpretation of this finding within these variables shall take into account the degree of precision. The strength of this study is the use of standard measurements which is enabled to make the comparison of findings with other national and international literatures to be valid. In addition, considering long duration of follow up period of children on ART; use of analytical study (retrospective cohort); the availability of data on important predictors of mortality (CD4 count, hemoglobin, and nutritional status); adequate sample size (strong power) were the strongest side of this study. It helps increase the quality of care given for children in ART clinic/ It gives a clue how effective care and treatment of children on ART in resource limited settings like Ethiopia. It can serve as baseline information for further study, especially on the nutritional status of children on ART.
The findings of this study indicate that the estimated cumulative survival was 95.8% after 3 months, 93.9% after 6 months, and 92% after 12 months of ART initiation. The cumulative survival after 54 months was 91.6% and stayed stable. Most of the children initiated ART at older age. The overall mortality rate was 2.06 deaths per 100 child-years (43 deaths over a 2078 child-years). Three fourth of the total deaths in our study occurred in the first sixth months following ART initiation. CD4 count below threshold for immune deficiency, anemia (Hgb<10 gm/ dl), advanced WHO staging, and underweight were the most factors affecting survival of children following ART initiation. Based on the findings the ART services providing should engage close follow up of HIV exposed children to make early diagnosis and treatment initiation before the development of severe immune deficiency and advanced clinical stage. Targeted interventions should be prepared to intensify support and care for children especially during the first sixth months of ART initiation. Particular attention should be given to those presenting with anemia. Malnourished children should get proper care such as nutritional intervention to reduce the risk of death. Developing a mechanism to control the completeness of baseline and follow up data being collected. More over the Adama town health office and Oromia Regional Health Bureau should develop multi sectoral approach to improve the survival of children on ART and improving the capacity of health care providers through in service training.
We acknowledge the Oromia Regional Health Bureau and Mekelle University, College of Health sciences and Tulane University, USA for financial assistance. We extend our deepest gratitude to staffs of Adama referral hospital for facilitating good conditions to carry out the study. Lastly we thank the data collectors for their full commitment and technical support.