Skip to main content
PMC home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2020 Aug 17.
Published in final edited form as: Pediatr Infect Dis J. 2015 Aug;34(8):e191–e199. doi: 10.1097/INF.0000000000000741

Temporal Trends in Patient Characteristics and Outcomes Among Children Enrolled in Mozambique’s National Antiretroviral Therapy Program

Andrew F Auld 1, Charity Alfredo 2, Eugenia Macassa 3, Kebba Jobarteh 2, Ray W Shiraishi 1, Emilia D Rivadeneira 1, James Houston 1, Thomas J Spira 1, Tedd V Ellerbrock 1, Paula Vaz 3,4
PMCID: PMC7430037  NIHMSID: NIHMS838955  PMID: 25955836

Abstract

Background:

During 2004–2009, >12,000 children (<15 years old) initiated antiretroviral therapy (ART) in Mozambique. Nationally representative outcomes and temporal trends in outcomes were investigated.

Methods:

Rates of death, loss to follow-up (LTFU), and attrition (death or LTFU) were evaluated in a nationally representative sample of1,054 children, who initiated ART during 2004–2009 at 25 facilities randomly selected using probability-proportional-to-size sampling.

Results:

At ART initiation during 2004–2009, 50% were male, median age was 3.3 years, median CD4% was 13%, median CD4 count was 375 cells/μL, and median weight-for-age z-score was −2.1. During 2004–2009, median time from HIV diagnosis to care initiation declined from 33 to 0 days (p=0.001), median time from care to ART declined from 93 to 62 days (p=0.004), the percentage aged <2 at ART initiation increased from 16% to 48% (p=0.021), the percentage of patients with prior tuberculosis declined from 50% to 10% (p=0.009), and the percentage with prior lymphocytic interstitial pneumonia declined from 16% to 1% (p<0.001). Over 2,652 person-years of ART, 183 children became LTFU and 26died. Twelve-month attrition was 11% overall, but increased from 3% to 22% during 2004–2009, due mainly to increases in 12-month LTFU (from 3% to 18%).

Conclusion:

Declines in the prevalence of markers of advanced HIV diseaseat ART initiation probably reflect increasing ART access. However, 12-month LTFU increased during program expansion, and this negated any program improvements in outcomes that might have resulted from earlier ART initiation.

Keywords: Pediatric Antiretroviral Therapy, Treatment Outcomes, Mozambique

In Mozambique about 85 children acquireHIV daily through mother-to-child transmission.1 Without treatment, available evidence from African clinical trials suggests that about half would die before their second birthday.2 Despite global urgency to expand antiretroviral therapy (ART) access for children, progress has been sub-optimal in low- and middle-income countries.3 In Mozambique, only about 27% of theestimated 100,000 ART-eligible children were receiving ART by 2013, compared with 48% of 590,000 ART-eligible adults.4

Expanding ART access for HIV-infected children is a priority of the Mozambican Ministry of Health (MOH) and partners, including the United States Government through its President’s Emergency Plan for AIDS Relief. However, rapid expansion of a national ART program can result in decreasing care quality; for example in adult ART programs in South Africa5 and Mozambique6 rates of loss to follow-up (LTFU) increased during program expansion. In addition, a recent multi-country study of pediatric treatment outcomes reported increasing rates of LTFU among pediatric ART enrollees in certain African and South East Asian countries.7 However, nationally representative trends in pediatric ART outcomeshave not yet been evaluated in Mozambique. As Mozambique considers adoption of new 2013 World Health Organization (WHO) guidelines, with expanded ART eligibility criteria, evaluating trends in mortality and LTFU is needed to inform future scale-up.

METHODS

ARTEligibility

During 2004–2009, HIV-infected children, aged 0–14 years, with World Health Organization (WHO) clinical stage III/IV were ART-eligible, regardless of CD4+T-cell (CD4) count or percentage (%). In addition, during 2004–2008, children <18 months with CD4% <20%, and children ≥18 months with CD4%<15% were ART-eligible, while during 2009 children <36 months with CD4% <20%, and children ≥36 months with CD4%<15% were ART-eligible. Prescription of co-trimoxazole (CTX), was indicated for all pediatric ART patients until immune restoration.

Recommended first-line ART regimens includedeither stavudine (D4T) or zidovudine (AZT) with lamivudine (3TC) and either nevirapine (NVP) or efavirenz (EFV) depending on the child’s age or weight. A regimen containing ritonavir-boosted lopinavir (LPV/r) was recommended for children<2 with prior exposure to prevention of mother-to-child transmission (PMTCT) antiretrovirals.

Patient Monitoring

At baseline and every 6 months, weight measurements, staging, and CD4 counts and percentageswere recommended to monitor disease progression or improvement.8 Alanine aminotransferase (ALT) measurements were recommended at baseline and every six monthsor for suspected liver disease. Patients or their caregivers collected medications monthly from the pharmacy according to MOH guidelines, or, in rare cases, every 2–3 months, if there was a valid reason for lengthier time periods between pharmacy appointments.

Study Design

This was a retrospective cohort study using routinely collected data abstracted from paper ART medical records of children initiating ART during 2004–2009.

Sample Size

To estimate 6-month attrition with a 95% confidence interval (CI) of±3%, conservatively expecting 6-month attrition of 25%,9 and using a design effect of 1.5, a sample size of 1,130 charts was needed. We aimed to sample 1,500charts.

Sampling

During 2004–2009, 12,674 children (<15 years old) started ART at one of 210 facilities. To keep the study feasible, very small facilities (i.e., those with<20 pediatric ART enrollees) were excluded from the sample frame. Only 714 (6%) of all enrollees had enrolled at 93 very small facilities during 2004–2009. The remaining 117 study-eligible facilities were stratified by number of enrollees into three strata (1) ≥200, (2) 50–199, and (3) 20–49. Probability-proportional-to-size sampling was used to select five, 23, and seven facilities from strata one, two, and three, respectively. Simple random sampling was used to select 1,500 medical records from the 35 selected facilities, with 150 records selected from each facility in stratum one and 25 records from each facility in strata two and three.

Due to higher-than-expected costs of transport and data collection, only 25 of 35 randomly selected facilitiescould be included in the study, reducing the sample size to 1,250 medical records. In addition, 196of 1,250 charts could not be located at the 25 selected facilities. Therefore, only 1,054 medical records were abstracted between June 2010 and June 2011.

Treatment Outcomes

The primary outcomes of interest weremortality and LTFU.A childwas considered LTFU if he/she was absent from the facility in the 90days preceding data abstraction.6 In rare cases, if it was clear a child’s pharmacy pick-up appointments were scheduled at intervals of >90 days, and the child had not missed a future drug pick-up appointment, the child was considered alive on ART.

The combined outcome of attrition (death or LTFU) was a secondary outcome of interest. For all time-to-event analyses, transfers were censored at the date of transfer.

Exposure Variables

Most routinely collected variables on MOH-recommended ART records were abstracted (Table 1). Weight was recoded as weight-for-age z-score (WAZ), using Centers for Disease Control and Prevention (CDC) growth curves for children aged 5–14 years and WHO curves for children aged 0–<5 years. Severe immunodeficiency was defined using WHO-recommended age-specific thresholds.10

Table 1:

Demographic Characteristics among Children Starting Antiretroviral Therapy in Mozambique during 2004–2009

Original (N=1,054)* Imputed
n Median or %** (IQR or 95% CI) Median or %** (IQR or 95% CI) 2004 (n=18, 2%) 2005 (n=62, 5%) 2006 (n=162, 15%) 2007 (n=326, 29%) 2008 (n=289, 27%) 2009 (n=197, 19%) P-value
Age (Median)
Both Sexes 1,054 3.3 (1.7–6.5) 3.3 (1.7–6.5) 4.7 4.5 3.5 3.1 3.4 2.1 0.065
Age Categories
0-<2 years 327 33% (28–37%) 33% (28–37%) 16% 12% 22% 31% 33% 48% 0.021
2-<5 years 330 30% (27–34%) 31% (27–34%) 49% 43% 38% 33% 27% 21%
5 to <10 years 301 28% (24–32%) 28% (24–32%) 23% 28% 35% 28% 30% 20%
10 to <15 years 96 9% (7–11%) 9% (7–11%) 12% 16% 6% 9% 9% 11%
Sex
Male 523 50% (46–53%) 50% (46–53%) 46% 46% 51% 53% 44% 54% 0.597
Female 531 50% (47–54%) 50% (47–54%) 54% 54% 49% 47% 56% 46%
Maternal Vital Status
Mother deceased at ART start 173 17% (13–21%) 17% (13–21 %) 10% 29% 25% 18% 18% 9% 0.001
Observations missing data 62 6%
Paternal Vital Status
Father deceased at ART start 173 19% (15–22%) 19% (16–23 %) 10% 20% 19% 22% 18% 18% 0.848
Observations missing data 129 12%
Dual Orphan Status
Both parents deceased 51 6% (4–8%) 6% (4–8%) 0% 8% 8% 7% 6% 2% 0.167
Observations missing data 218 21%
Referral Source (recoded)
General Health Facility (outpatient) 332 34% (23–46%) 34% (22–46%) 46% 31% 34% 28% 32% 45%
TB clinic 9 1% (0–2%) 1% (0–3%) 0% 1% 0% 1% 2% 1% 0.641
PMTCT clinic 28 4% (1–6%) 4% (1–6%) 1% 0% 1% 3% 6% 6% 0.071
VCT 334 36% (20–53%) 36% (20–53%) 15% 35% 38% 40% 40% 29% 0.460
Inpatient referral 224 18% (11–25%) 18% (11–25%) 33% 28% 25% 21% 16% 10% 0.001
Other 62 6% (3–9%) 6% (3–9%) 5% 5% 3% 8% 5% 9% 0.583
Observations missing data 65 6%
Time from Diagnosis to Entry into HIV Care (days)§
Both Sexes 785 1 (0–21) 1 (0–21) 33 18 2 2 1 - 0.001*
Observations Missing Data 269 26%
Time from HIV Care Start to ART Start (days)§
Both Sexes 948 63 (28–183) 63 (29–184) 93 163 83 60 55 62 0.004*
Observations Missing Data 106 10%
Site type
Large (>100 enrollees) 662 49% (47–51%) 49% (47–51%) 39% 70% 56% 54% 49% 33% 0.202
Small (≤100) 392 51% (49–53%) 51% (49–53%) 61% 30% 44% 46% 51% 67%
Open in a new tab

Abbreviations: IQR, interquartile range; CI, confidence interval; TB, tuberculosis; PMTCT, prevention of mother-to-child transmission; VCT, voluntary counseling and testing center.

*

The denominator used to estimate percentages presented in the original data columns was 1,054 minus the number of observations missing data.

**

All percentages and medians presented in this table are weighted to account for survey design.

§

The days-to-care and days-to-ART variables were log-transformed to approximate a normal distribution, and quadratic terms for year of ART initiation were included in the linear regression models since p-values for the quadratic and unmodified ART year coefficients were <0.05 and their signs opposite.

Analytic Methods

Data were analyzed using STATA 11 (StataCorp, 2009, Stata Statistical Software, Release 11, College Station, TX). Data were weighted and survey procedures used to account for study design.

Missing data are reported for each covariate of interest. The missing at random (MAR) assumption was considered plausible based on examination of patterns of missingness. If <30% of observations were missing data for a baseline demographic or clinical covariate of interest, multiple imputation with chained equations was used to impute the missing data.11 The ice procedure in STATA was used to create 20 imputed datasets for each of three outcomes: LTFU, death, and attrition.11 The imputation model included the event indicator, all study variables, and the Nelson-Aalen estimate of cumulative hazard.12

To assess the association between baseline characteristics and year of ART initiation, linear, logistic, ordered, or multinomial logistic regression models, accounting for study design, were used for continuous, binary, ordinal, and nominal categorical variables, respectively.

A competing risk model was used toanalyze the independent risk of the two failures types: death and LTFU. LTFU is acompeting cause of death, potentially increasing the risk of death because of ARTinterruption. Thus, assumptions about the independence of these two outcomes are notrealistic.13 For this reason, we used a cumulative incidence function (CIF) to estimate thecumulative probability of each outcome during follow-up.

To estimate adjusted hazard ratios (AHRs) and 95% confidence intervals (CIs) for baseline covariates of interest, we usedCox proportional hazards regression models for each of the two competing outcomes (LTFU and mortality) and the combined outcome (attrition). Certainpatient-level covariates at ART initiation were considered apriori variables for inclusion in the multivariable models based on prior publications.9 Prior reports from Mozambique suggest that quality of WHO staging was sub-optimal and so WHO stage was not an apriori risk factor.14

The proportional hazards assumption was assessed using visual methods and the Grambsch and Therneau test.15 Estimates were combined across the imputed datasets according to Rubin’s rules using the mim procedure in STATA.16

Ethics Approval

This study was approved by the MozambicanEthics Review Committee andthe Institutional Review Board (IRB) of the U.S. Centers for Disease Control and Prevention (CDC).

RESULTS

Trends in Demographic Characteristics

Among 1,054 childrenthe median age was 3.3 years, with33% <2 years old, 31% aged 2–<5, 28% 5–<10, and 9% aged 10–<15 (Table 1). During 2004–2009, the proportion of children aged <2 at ART initiation increased from 16% to 48% (p=0.021).

During 2004–2009, about 50% of children were maleand this percentage did not change significantly over time (Table 1). At ART initiation, 17% of children were maternal orphans and 19% paternal orphans. The proportion of children who were maternal orphans changed significantly over time from 10% during 2004to 29% in 2005, 25% in 2006, 18% during 2007–2008, and 9% in 2009 (p=0.001).

Referral source was a PMTCT clinic for 4% of children, but this proportion increased marginally from 1% to 6% during 2004–2009 (p=0.071) (Table 1). Referral source was an in-patient setting in 18% of children, but this decreased from 33% to 10% during 2004–2009 (p=0.001).

Median time from HIV diagnosis to entry into care was one day overall, but declined from 33 days in 2004 to 0 days in 2009 (p=0.001) (Table 1). Median time from HIV care initiation to ART initiation was 63 days but declined from 93 to 62 days during 2004–2009 (p=0.004). About half the children were enrolled at large clinics with >100 ART enrollees at the time of sampling (49%) and half at smaller clinics with 20–100 pediatric ART enrollees (51%) and this did not change significantly over time.

Trends in Clinical Characteristics

Overall, 15% of children had active TB at ART initiation, but there was a borderline statistically significant decline from 31% to 10% during 2004–2009 (p=0.093) (Table 2). Also, 14% of children had been diagnosed with and completed treatment for TB, before ART initiation, but this declined from 50% to 10% during 2004–2009 (p=0.009). Documentation of a diagnosis of pneumonia of unspecified cause before ART was found in 19% of all records, but declined from 65% to 14% during 2004–2009 (p=0.015). Documentation of a diagnosis oflymphocytic interstitial pneumonia (LIP) before ART was found in 4% of charts, but this declined from 16% to 1% during 2004–2009 (p<0.001). Documentation of a diagnosis of Pneumocystisjirovecii pneumonia (PCP) before ART was found in 2% of charts and this did not change over time.

Table 2:

Clinical Characteristics of Children Starting Antiretroviral Therapy in Mozambique during 2004–2009

Original* Imputed
n Median/%** (IQR/95% CI) Median or %** (IQR/95% CI) 2004 (n=18, 2%) 2005 (n=62, 5%) 2006 (n=162, 15%) 2007 (n=326, 29%) 2008 (n=289, 27%) 2009* (n=197, 19%) P-value
TB treatment at ART start
Yes 163 15% (7–22%) 15% (7–22%) 31% 25% 15% 16% 13% 10% 0.093
No 887 85% (78–93%) 85% (78–93%) 69% 75% 85% 84% 87% 90%
Observations Missing Data 4 0%
Prior TB
Yes 156 14% (9–20%) 14% (9–20%) 50% 22% 14% 16% 12% 10% 0.009
No 894 86% (80–91%) 86% (80–91%) 50% 78% 86% 84% 88% 90%
Observations missing data 4 0%
Prior Pneumonia
Yes 221 19% (12–26%) 19% (12–26%) 65% 24% 19% 22% 15% 14% 0.015
No 829 81% (74–88%) 81% (74–88%) 35% 76% 81% 78% 85% 86%
Observations missing data 4 0%
Prior PCP
Yes 16 1% (1–2%) 2% (1–2%) 5% 3% 2% 2% 2% 1% 0.252
No 1034 99% (98–99%) 98% (98–99%) 95% 97% 98% 98% 98% 99%
Observations missing data 4 0%
Prior Lymphocytic Interstitial Pneumonia
Yes 48 4% (0–8%) 4% (0–8%) 16% 13% 9% 3% 2% 1% <0.001
No 1002 96% (92–100%) 96% (92–100%) 84% 87% 91% 97% 98% 99%
Observations missing data 4 0%
Maternal ARVs Received During Pregnancy/Delivery
Yes 83 13% (10–16%) 17% (12–22%) 11% 6% 9% 13% 14% 36% 0.009
No 613 87% (84–90%) 83% (78–88%) 89% 94% 91% 87% 86% 64%
Observations missing data 358 34%
Infant exposure to PMTCT ARVs after birth
Yes 35 5% (3–8%) 9% (5–14%) 4% 2% 4% 7% 9% 17% 0.028
No 696 95% (92–98%) 91% (86–95%) 96% 98% 96% 93% 91% 83%
Missing 323 31%
Weight-for-Age Z-score (Median)
Both Sexes 888 −2.1 (−3.3– −1.1) −2.1 (−3.3– −1.1) −1.8 −1.8 −2.4 −2.3 −2.1 −1.8 0.594
Observations missing data 166 16%
Weight-for-Height Z-score (Median)
Both Sexes 635 −0.7 (−2– 0.4) −0.7 (−2.1– 0.5) 0.4 0.04 −0.5 −0.9 −0.7 −0.9 0.347
Observations missing data 419 40%
CD4 cell % (Median)
All ages 766 12.6 (7.9–17.9) 13 (8–18) 10 13 12 13 13 13 0.935
Observations missing data 288 27%
CD4 cell count (Median)
All ages 791 382 (111–652) 375 (97–652) 204 406 380 400 396 307 0.871
Observations missing data 263 25%
Hemoglobin
All ages 798 9.5 (8.5–10.4) 9.5 (8.4–10.5) 9.9 9.9 9.6 9.6 9.4 9.5 0.248
Observations missing data 256 24.3
Alanine Aminotransferase Categories (ALT)
Normal (≤56/uL) 618 88% (84–91%) 83% (78–88%) 93% 91% 88% 87% 77% 79% 0.098
Moderately abnormal (57–100/uL) 32 5% (3–8%) 9% (6–13%) 5% 7% 7% 7% 12% 11%
Abnormal (>100/uL) 43 7% (4–10%) 8% (5–11%) 2% 2% 6% 6% 12% 9%
Observations missing data 361 34%
CTX prescription
Yes 305 48% (28–68%) 42% (29–56%) 54% 42% 41% 46% 38% 42% 0.684
No 345 52% (32–72%) 58% (44–71%) 46% 58% 59% 54% 62% 58%
observations missing data 404 38%
Open in a new tab

Abbreviations: IQR, inter-quartile range; CI, confidence interval; TB, tuberculosis; PMTCT, prevention of mother-to-child transmission; ARVs, antiretrovirals; CTX, co-trimoxazole; PCP, Pneumocystis jirovecii pneumonia; ART, antiretroviral therapy

*

The denominator used to estimate percentages presented in the original data columns was 1,054 minus the number of observations missing data.

**

All proportions and medians presented in this table are weighted to account for survey design.

The proportion of records documenting maternal antiretroviral prophylaxis or treatment was 17%, but this increased from 11% to 36% during 2004–2009 (p=0.009) (Table 2). The proportion of records documenting infant antiretroviral prophylaxis was 9%, but this increased from 4% to 17% during 2004–2009 (p=0.028).

Median WAZ was −2.1 and median weight-for-height z-score was −0.7 and these medians did not change significantly over time (Table 2). Median CD4% was 13% and median CD4 count was 375 cells/μL. At ART initiation during 2004–2009, median CD4% ranged from 10–13% and median CD4 count ranged from 204–307 cells/μL, but no statistically significant changes were noted. Median hemoglobin was 9.5 g/dL and this did not change over time. Severe immunodeficiency prevalence at ART initiation was 75% (95% CI, 70–81%), and this did not change significantly over time.

The proportion of children with ALT measurements within normal limits (ALT<56/μL) was 83% overall, but there was a marginally significant decline from 93% to 79% during 2004–2009 (p=0.098). Overall 42% of children were prescribed CTX at ART initiation and this did not change over time.

First-line Regimens

The most common ART regimens prescribed to children ≤3 years old were AZT, 3TC, and NVP (55%), and D4T, 3TC, and NVP (38%)(Table 3). D4T, 3TC, and NVP accounted for the majority of first-line regimens for children >3 years old (83%). Overall, 61% of children received D4T, 3TC, and NVP while 30% received AZT, 3TC, and NVP.Only 3 (<1%) of all children were prescribed LPV/r-containing regimens. Regimen distributions did not change significantly during 2004–2009 (p=0.478).

Table 3:

First-Line Antiretroviral Therapy Regimens Prescribed to HIV-infected Children in Mozambique during 2004–2009

≤ 3 years old > 3 years old Overall
n N % 95% CI n N % 95% CI n N % 95% CI
D4T-3TC-NVP 148 450 38% (27–49%) 423 514 83% (76–89%) 571 964 61% (55–68%)
D4T-3TC-EFV 6 450 1% (0–2%) 35 514 7% (3–11%) 41 964 4% (2–6%)
D4T-3TC-LPV/r 3 450 0% (0–1%) 0 514 0% (0–0%) 3 964 0% (0–1%)
AZT-3TC-ABC 20 450 4% (2–7%) 2 514 0% (0–1%) 22 964 2% (1–3%)
AZT-3TC-NVP 268 450 55% (44–66%) 40 514 8% (4–11%) 308 964 30% (24–37%)
AZT-3TC-EFV 1 450 0% (0–1%) 10 514 2% (0–4%) 11 964 1% (0–2%)
D4T-3TC-ABC 2 450 0% (0–1%) 4 514 1% (0–1%) 6 964 1% (0–1%)
D4T-3TC-other 1 450 0% (0–1%) 0 514 0% (0–0%) 1 964 0% (0–0%)
D4T-AZT-NVP 1 450 0% (0–1%) 0 514 0% (0–0%) 1 964 0% (0–0%)
missing 45 495 9% 45 559 8% 90 1054 9%
Open in a new tab

Abbreviations: CI, confidence interval; D4T, stavudine; AZT, zidovudine; 3TC, lamivudine; NVP, nevirapine; EFV, efavirenz; ABC, abacavir; LPV/r, lopinavir-ritonavir.

Mortality and LTFU

Over 2,652 person-years of follow-up, 209 children were lost through attrition;26were documented to have died, and 183became LTFU.

LTFU rates were 6.9/100 patient-years (PY) overall, but higher in days 0–90 (20.2/100 PY) compared with the time period following the first 90 days of ART (5.5/100 PY). Documented mortality rates were 0.98/100 PY overall, and were higher in days 0–90 (1.6/100 PY) compared with the time period following the first 90 days of ART (0.9/100 PY). Overall attrition rates were 7.9/100 PY but higher in days 0–90 (21.7/100 PY) compared with the time period following the first 90 days of ART (6.4/100 PY).

For all enrollees during 2004–2009, attritionproportions at 6, 12, 24, 36, 48, and 60 months were 7.4%, 11.0%, 16.6%, 20.2%, 24.2%, and 28.6%. However, attritionrates were higher for children enrolled in later calendar years compared with the 2004 and 2005 cohorts, increasing from 1.7/100 PY for 2004 and 2005 enrollees to 19.9/100 PY for 2009 ART enrollees. Similarly, 12-month attritionincreased from 3.0% for 2004 and 2005 ART enrollees to9.9%, 10.3%, 14.6%, and 21.6% for 2006, 2007, 2008, and 2009 ART enrollees, respectively (Table 4). Increases in 12-month attrition proportions during this time period were due to increases in documented 12-month mortality (from 0% to 4.0%) and LTFU (from 3.0% to 17.6%) (Table 4).

Table 4:

Cumulative Incidence of Death andLTFUamong Children Initiating Antiretroviral Therapy in Mozambiqueduring 2004–2009

Years on ART 2004–5 2006 2007 2008 2009
Death 0.5 0.0% 1.0% 0.8% 0.9% 1.9%
1 0.0% 1.4% 1.3% 1.4% 4.0%
2 0.0% 3.3% 3.5% 1.7% 4.0%
3 0.0% 4.4% 3.8% 1.7% 4.0%
4 3.6% 4.4% 3.8% 1.7% 4.0%
LTFU 0.5 3.0% 7.0% 5.9% 9.1% 12.4%
1 3.0% 8.5% 9.0% 13.2% 17.6%
2 3.0% 10.3% 13.5% 19.4% 32.1%
3 3.0% 11.7% 18.4% 25.1% 32.1%
4 4.8% 14.6% 26.2% 25.1% 32.1%
Attrition* 0.5 3.0% 8.0% 6.8% 10.0% 14.4%
1 3.0% 9.9% 10.3% 14.6% 21.6%
2 3.0% 13.6% 17.0% 21.1% 36.1%
3 3.0% 16.1% 22.2% 26.7% 36.1%
4 8.4% 19.0% 30.0% 26.7% 36.1%
Retention* 0.5 97.0% 92.0% 93.2% 90.0% 85.6%
1 97.0% 90.1% 89.7% 85.4% 78.4%
2 97.0% 86.4% 83.0% 78.9% 63.9%
3 97.0% 83.9% 77.8% 73.3% 63.9%
4 91.6% 81.0% 70.0% 73.3% 63.9%
Open in a new tab

Abbreviations: LTFU, lost to follow-up.

*

Attrition is the cumulative incidence of death and LTFU, with transfer outs censored at the time of transfer. Retention is the proportion of children alive and on ART at the relevant time point (1-attrition).

Predictors of Death and LTFU

Being one year older at ART initiation was marginally associated with lower risk of documented death (AHR 0.68; 95% CI, 0.46–1.00, p=0.051), but no reduction in LTFU rates (Table 5). Initiating ART one calendar year later during program expansion was associated with nearly two-fold higher rates of LTFU (AHR 1.90; 95% CI, 1.31–2.77), but was not associated with mortality in either crude or multivariable analysis.

Table 5:

Risk Factors for Death and LTFU among Children Starting Antiretroviral Therapy in Mozambique during 2004–2009

Original LTFU Death
N Rate/100 HR Crude (95% CI) p AHR Adjusted (95% CI) p Rate/100 Crude HR (95% CI) p AHR (95% CI) p
Age
Per year increase 1,054 — 0.93 (0.87–1.00) 0.047 0.95 (0.85–1.06) 0.308 — 0.87 (0.64–1.18) 0.360 0.68 (0.46–1.00) 0.051
Sex
Male 523 8.77 1.00 — — 1.00 — — 1.53 1.00 — — 1.00 — —
Female 531 8.18 0.94 (0.70–1.27) 0.692 1.50 (0.96–2.32) 0.070 1.13 0.74 (0.23–2.35) 0.595 2.87 (0.47–17.48) 0.232
Maternal Vital Status
Mother alive 819 8.63 1.00 — — 1.00 — — 1.46 1.00 — — 1.00 — —
Mother dead 173 7.85 0.96 (0.65–1.42) 0.826 1.64 (0.89–3.01) 0.105 0.80 0.58 (0.22–1.54) 0.255 1.33 (0.13–13.35) 0.791
ART Initiation year
Per year increase 1,054 — 1.57 (1.19–2.07) 0.003 1.90 (1.31–2.77) 0.002 — 1.18 (0.75–1.87) 0.462 1.17 (0.70–1.95) 0.523
TB treatment
No 887 9.03 1.00 — — 1.00 — — 1.32 1.00 — — 1.00 — —
Yes 163 5.76 0.67 (0.34–1.31) 0.224 0.90 (0.27–2.97) 0.849 1.33 1.04 (0.31–3.50) 0.948 0.41 (0.05–3.12) 0.360
Prior TB treatment
No 894 9.12 1.00 — — 1.00 — — 1.26 1.00 — — 1.00 — —
Yes 156 5.23 0.61 (0.38–0.96) 0.035 1.03 (0.45–2.40) 0.933 1.62 1.35 (0.39–4.71) 0.620 3.49 (0.40–30.69) 0.237
Weight-for-Age Z-score
Per unit increase 888 — 0.82 (0.75–0.91) 0.001 0.75 (0.64–0.89) 0.004 — 0.81 (0.71–0.92) 0.004 0.81 (0.52–1.25) 0.287
CD4 cell count %
>20% 125 10.20 1.00 — — 1.00 — — 2.17 1.00 — — 1.00 — —
10–20% 389 8.96 0.89 (0.52–1.51) 0.643 1.31 (0.60–2.87) 0.435 1.00 0.45 (0.10–1.99) 0.269 0.47 (0.04–5.79) 0.524
<10% 252 6.93 0.70 (0.38–1.30) 0.238 1.07 (0.39–2.88) 0.884 1.36 0.64 (0.16–2.55) 0.500 2.01 (0.12–32.71) 0.595
Hemoglobin
>=8g/dL 798 7.71 1.00 — — 1.00 — — 1.09 1.00 — — 1.00 — —
<8g/dL 256 13.03 1.61 (0.93–2.79) 0.082 1.39 (0.72–2.67) 0.293 2.74 2.41 (0.76–7.65) 0.126 1.66 (0.10–27.29) 0.699
ALT Categories
Normal (≤56/uL) 618 6.19 1.00 — — 1.00 — — 0.84 1.00 — — 1.00 — —
Abnormal (>56/uL) 32 13.32 2.04 (1.23–3.37) 0.009 1.71 (1.02–2.89) 0.044 0.87 0.91 (0.11–7.90) 0.930 0.85 (0.12–6.04) 0.859
CTX prescription
Yes 305 7.93 1.00 — — 1.00 — — 1.44 1.00 — — 1.00 — —
No 345 8.91 1.08 (0.56–2.09) 0.810 1.56 (0.63–3.90) 0.301 1.23 0.82 (0.18–3.62) 0.774 0.77 (0.14–4.21) 0.728
Site type
Large Central (>100) 662 5.74 1.00 — — 1.00 — — 0.44 1.00 — — 1.00 — —
Small-Medium (<=100) 392 12.33 1.95 (0.84–4.56) 0.116 1.85 (0.78–4.41) 0.150 2.57 5.32 (0.53–53.19) 0.146 6.46 (0.62–67.68) 0.111
Open in a new tab

Abbreviations: HR, hazard ratio; AHR, adjusted hazard ratio; ART, antiretroviral therapy; TB, tuberculosis; ALT, alanine aminotransferase; CTX, co-trimoxazole

A one-unit higher WAZ at ART initiationwas associated with a 25% reduction in LTFU rates (AHR 0.75; 95% CI, 0.64–0.89). In unadjusted analysis, a one-unit increase in WAZ was associated with a 19% reduction in documented mortality, but in adjusted analysis this association was not significant.

A raised ALT level at ART initiation was associated with increased rates of LTFU (AHR 1.71; 95% CI, 1.02–2.89) but not documented mortality.

In multivariable analysis, later year of ART initiationwas predictive of attrition (AHR 1.80; 95% CI, 1.29–2.51), whereas a one-unit higher WAZat ART initiation was protective against attrition (AHR 0.77; 95% CI, 0.66–0.89)..

DISCUSSION

This is the first nationally representative evaluation of trends in pediatric ART outcomesfrom Mozambique. The key finding is that despite declines in the prevalence of markers of advanced HIV disease at ART initiation, increasing rates of LTFU have negated any potential improvements in program outcomes that might have resulted from earlier ART initiation.

Trends in ART Enrollment Characteristics

During 2004–2009, the proportion of children starting ART aged <2 years old increased from 16% to 48%. In addition, during 2004–2009, prevalence of various markers of advanced HIV disease at ART enrollment declined.

This trend of initiating ART at younger agesand earlier disease stages is probably due to increasing access to HIV testing services, earlier enrollment in HIV care, and the change in national guidelines in January 2009 that recommended ART initiation at earlier disease stages. The trend of earlier ART initiation is encouraging for program managerssince earlierART for peri-natallyHIV-infected childrenis associated with improved survival.17 In accordance with WHO recommendations,18 in December 2009, ART was recommended for all infants, and in December 2010, all children aged <2 years old became ART-eligible in Mozambique. These changes could further lower both the median age of ART initiation and the prevalence of markers of advanced HIV disease.

ART prescribing practices did not change over time. About91% of children >3 years old were prescribed NVP-containing regimens, while the other 9% were prescribed EFV-containing regimens. Clinician preference for NVP over EFV may be related to cost differences and possibly clinician concerns about teratogenicity among the relatively few female adolescents who were sexually active.19 However, recent data suggests EFV-containing regimens may be superior to NVP-containing regimens in achieving virologic suppression, which might influence guidelines and prescribing practices.19

During 2006–2009, LPV/r-containing regimens were recommended for children <2 years old at ART initiation, who were PMTCT antiretroviral exposed, to reduce risk of virologic failure.20 Limited LPV/r use may have been due to cost or lack of cold-chains for LPV/r formulations. As more children start ART at age <2 years, using currently available LPV/r formulations, wherever the cold chain allows, will be important. In addition, development of heat-stable LPV/r-containing regimens suitable for young children is necessary21 and may soon be commercially available.22

Trends in Outcomes

Our overall reported 12-month attrition (11.0%) is similar to other reports [0–20% according to a recent meta-analysis].9 However, the increase in 12-month attrition from 3.0% to 21.6% during 2004–2009, due mainly to increases in observed LTFU (from 3.0% to 17.6%) is concerning. Increasing rates of LTFU have been observed in adult African ART programs5,6 and in a recent multi-country analysis among pediatric ART enrollees.7 However, most previous reports of trends in pediatric ART retention23,24 showed no significant changes in LTFU over time. Possible explanations for increasing LTFU in Mozambiqueinclude: (1) increasingrates of undocumented transfer of patients between health facilities, (2) declines in record keeping quality, and (3) overcrowded facilities.

In Mozambique, two initiatives combined to increase the rate of transferal of HIV-infected ART enrollees between health facilities: (1) as large central facilities became overloaded, down-referral of ART enrollees to primary health care clinics (PHCs) was encouraged during 2004–2009,25 and (2) in March 2008, the MOH, in an effort to reduce HIV stigma, decided to close 23 day hospitals that provided HIV care and treatment services exclusively to HIV-infected patients, and relocated the patients to general health services.26 During these initiatives, common problems included poor transfer documentation, which could have resulted in “silent transfers” being observed as LTFU, and lack of tracing for defaulting patients.25

Secondly, with increasing patient load, attention to timely, accurate maintenance of medical records may have been compromised, resulting in the possibility of missing entries for the most recent clinic visit.5 Implementing effective electronic monitoring systems with dedicated personnel to manage these systems could improve data quality.27

Thirdly, Mozambique has a severe healthcare worker shortage, with only 0.04 doctors and 0.21 nurses per 1,000 inhabitants,28 and with increasing patient-to-provider ratios, patient waiting times have increased, and waiting rooms have become more crowded.25 This may be associated with patient and clinician dissatisfaction with clinic conditions, which might have caused increasing LTFU.29 In addition, Mozambique is ranked 185th out of 187 countries on the United Nations Human Development Index, and for many patients in more rural areas enrolled in later program years, transport to and from ART clinic visits, may have become increasingly economically infeasible.30 Possible strategies to reduce burdens on health facilities and patients include formation of community ART groups31 or distribution of ART at community locations.32

Other Predictors of Outcomes

Similar to other reports,33 ART enrollment at younger ages was borderline predictive of increased mortality risk. This probably reflects the difficulties treating younger children with ART, including more rapid disease progression, limited repertoire of antiretroviral formulations, unpalatable liquid formulations, and adherence challenges.3 In Mozambique, earlier ART enrollment through expansion of early infant diagnosis and treatment services, and provision of optimal ART regimens (e.g. LPV/r for PMTCT-exposed infants) could improve young child outcomes.

As in other studies,9 moderate or severe under-nutrition (WAZ≤−2) was common at ART initiation, and having a lower WAZ score at ART initiation was predictive of poor outcomes. HIV-associated under-nutrition has multiple causes, including poor intake, poor absorption, increased basal metabolic requirements, and co-existing opportunistic infections.34 Food insecuritymay be an important factor in Mozambique given the high prevalence of poverty. Undiagnosed tuberculosis may also be a common cause of under-nutrition in Mozambique.35 Scale-up of the WHO-recommended HIV nutrition program,36 and regular TB screening in conjunction with use of the newXpert MTB/RIF assay for TB diagnosis,37 could help to improve ART outcomes for undernourished children.

In this study, prevalence of raised ALT (>56 U/L)increased marginally among ART enrollees over time and a raised ALT was independently associated with LTFU risk. Other studies have reported that up to 20% of pediatric ART enrollees have some evidence of liver dysfunction.38 There are many possible causes for raised ALT in HIV-infected children,39 including viral co-infections (e.g. Hepatitis B)and drug-related liver toxicity, [e.g. secondary to sulphonamides, anti-TB treatment,38 or PMTCT antiretrovirals].40 Chronic Hepatitis B infection, thought to affect about 5–10% of HIV-infected children in Africa,41 is a possible explanation for increased ALT levels. Regardless of the cause of increased ALT levels, this was significantly associated with LTFU, increasing the importance of the observed trend for program managers and further research.

Limitations

This report is subject to at least sixlimitations. Firstly, this was a retrospective cohort study utilizing routinely collected data. Secondly, missing data on patient characteristics at ART start likely introduced non-differential measurement error. Thirdly, ten of 35 ART delivery sites could not be evaluated due to financial and logistical constraints, and this may have introduced selection bias, reducing ability to generalize findings to all facilities in the country. Fourthly, mortality may be underestimated and LTFU overestimated, due to lack of active tracing.42 Fifthly, these data represent trends among children starting ART during 2004–2009 only, and do not represent more recent trends. Finally, viral load was not routinely monitored and so virologic outcomes were not captured.

CONCLUSIONS

Declines in median age and prevalence of markers of advanced HIV disease at ART initiationare encouraging for program managers as they likely indicate increasing access to HIV care and treatment services; however, despite gains in ART access, ART program outcomes have worsened over time due to increasing rates of LTFU. Causes of increasing rates of LTFU are unknown but could relate to poor documentation of increasing inter-facility transfers, sub-optimal record-keeping, or patient/caregiver frustration with overcrowded facilities.

Sources Of Support:

This Research Has Been Supported By The President’s Emergency Plan For AIDS Relief (PEPFAR) Through The U.S. Centers For Disease Control And Prevention.

Footnotes

Meetings: Presented In Part At The 5th Pediatric HIV Conference, Kuala Lumpur, Malaysia, June28–29, 2013.

Publisher's Disclaimer: Disclaimer:

Use Of Trade Names Is For Identification Purposes Only And Does Not Imply Endorsement By The U.S. Centers For Disease Control And Prevention Or The U.S. Department Of Health And Human Services. The Findings And Conclusions In This Manuscript Are Those Of The Authors And Do Not Necessarily Represent The Views Of The U.S. Centers For Disease Control And Prevention.

Disclosures: The Authors Have No Conflicts Of Interest Or Funding To Disclose.

REFERENCES

  • 1.United Nations Childrens Fund. HIV/AIDS and Children. Available at: http://www.unicef.org/mozambique/hiv_aids_2967.html. Accessed January 1, 2014.
  • 2.Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, Dabis F. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet. 2004;364:1236–1243. [DOI] [PubMed] [Google Scholar]
  • 3.Prendergast AJ, Penazzato M, Cotton M, et al. Treatment of Young Children with HIV Infection: Using Evidence to Inform Policymakers. PLoS Med. 2012;9:e1001273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Joint United Nations Programme on HIV/AIDS. Global report: UNAIDS report on the global AIDS epidemic 2013. Available at: http://www.unaids.org/en/media/unaids/contentassets/documents/epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en.pdf. Accessed February 17, 2014.
  • 5.Cornell M, Grimsrud A, Fairall L, et al. Temporal changes in programme outcomes among adult patients initiating antiretroviral therapy across South Africa, 2002–2007. AIDS. 2010;24:2263–2270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Auld AF, Mbofana F, Shiraishi RW, et al. Four-Year Treatment Outcomes of Adult Patients Enrolled in Mozambique’s Rapidly Expanding Antiretroviral Therapy Program. PLoS One. 2011;6:e18453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Leroy V, Malateste K, Rabie H, et al. Outcomes of antiretroviral therapy in children in Asia and Africa: a comparative analysis of the IeDEA pediatric multiregional collaboration. J Acquir Immune Defic Syndr. 2013;62:208–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.World Health Organization. Antiretroviral drugs for treating pregnant women and preventing HIV infections in infants: Recommendations for a public health approach, 2010 revision. Available at: http://whqlibdoc.who.int/publications/2010/9789241599818_eng.pdf. Accessed February 17, 2014. [PubMed]
  • 9.Ciaranello AL, Chang Y, Margulis AV, et al. Effectiveness of pediatric antiretroviral therapy in resource-limited settings: a systematic review and meta-analysis. Clin Infect Dis. 2009;49:1915–1927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.World Health Organization. Antiretroviral Drugs for Treating Pregnant Women and Preventing HIV Infection in Infants in Resource-Limited Settings - Towards Universal Access: Recommendations for a Public Health Approach, 2006 Version.; Geneva, Switzerland: WHO; 2006. Accessed 9 August 2010: http://www.who.int/hiv/pub/guidelines/WHOPMTCT.pdf. [Google Scholar]
  • 11.Royston P Multiple imputation of missing values: update of ice. The Stata Journal. 2005;5(4):527–536. [Google Scholar]
  • 12.White IR, Royston P. Imputing missing covariate values for the Cox model. Stat Med. 2009;28:1982–1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Schoni-Affolter F, Keiser O, Mwango A, et al. Estimating loss to follow-up in HIV-infected patients on antiretroviral therapy: the effect of the competing risk of death in Zambia and Switzerland. PLoS One. 2011;6:e27919. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Brentlinger PE, Torres JV, Martinez PM, et al. Clinical staging of HIV-related illness in Mozambique: performance of nonphysician clinicians based on direct observation of clinical care and implications for health worker training. J Acquir Immune Defic Syndr. 2010;55:351–355. [DOI] [PubMed] [Google Scholar]
  • 15.Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81:515–526. [Google Scholar]
  • 16.Royston P, Carlin JB, White IR. Multiple imputation of missing values: new features for mim. The Stata Journal. 2009;9(2):252–264. [Google Scholar]
  • 17.Violari A, Cotton MF, Gibb DM, et al. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med. 2008;359:2233–2244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.World Health Organization. Antiretroviral therapy for HIV infection in infants and children: towards universal access - Recommendations for a public health approach - 2010 revision. Available at: http://whqlibdoc.who.int/publications/2010/9789241599801_eng.pdf. Accessed April 30, 2014. [PubMed]
  • 19.Lowenthal ED, Ellenberg JH, Machine E, et al. Association between efavirenz-based compared with nevirapine-based antiretroviral regimens and virological failure in HIV-infected children. JAMA. 2013;309:1803–1809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Palumbo P, Lindsey JC, Hughes MD, et al. Antiretroviral treatment for children with peripartum nevirapine exposure. N Engl J Med. 2010;363:1510–1520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lallemant M, Chang S, Cohen R, Pecoul B. Pediatric HIV--a neglected disease? N Engl J Med. 2011;365:581–583. [DOI] [PubMed] [Google Scholar]
  • 22.Bakeera-Kitaka S, Fillekes Q, Keishanyu R, et al. Pharmacokinetics and Acceptability of a New Generic Lopinavir/Ritonavir Sprinkle Formulation in African, HIV+ Children 1–4 Years: CHAPAS-2 (Poster # 975b). Conference on Retroviruses and Opportunistic Infections (CROI), 20th Conference, March 3–6, 2013, Atlanta, USA Available at: http://www.dndina.org/component/content/article/2-events/104. Accessed April 30, 2014. [Google Scholar]
  • 23.Fatti G, Bock P, Eley B, Mothibi E, Grimwood A. Temporal trends in baseline characteristics and treatment outcomes of children starting antiretroviral treatment: an analysis in four provinces in South Africa, 2004–2009. J Acquir Immune Defic Syndr. 2011;58:e60–67. [DOI] [PubMed] [Google Scholar]
  • 24.McNairy ML, Lamb MR, Carter RJ, et al. Retention of HIV-infected children on antiretroviral treatment in HIV care and treatment programs in Kenya, Mozambique, Rwanda, and Tanzania. J Acquir Immune Defic Syndr. 2013;62:e70–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Decroo T, Panunzi I, das Dores C, et al. Lessons learned during down referral of antiretroviral treatment in Tete, Mozambique. J Int AIDS Soc. 2009;12:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.United Nations Office for the Coordination of Humanitarian Affairs. Mozambique: closing HIV day care centres brings protesters out. Available at: http://www.irinnews.org/report/85587/mozambique-closing-hiv-day-care-centres-brings-protesters-out. Accessed December 31, 2013.
  • 27.Forster M, Bailey C, Brinkhof MW, et al. Electronic medical record systems, data quality and loss to follow-up: survey of antiretroviral therapy programmes in resource-limited settings. Bull World Health Organ. 2008;86:939–947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.World Health Organization. WHO country cooperation strategy: 2009–2013. Available at: http://www.who.int/countryfocus/cooperation_strategy/ccs_moz_en.pdf. Accessed December 30, 2013.
  • 29.Musheke M, Bond V, Merten S. Individual and contextual factors influencing patient attrition from antiretroviral therapy care in an urban community of Lusaka, Zambia. J Int AIDS Soc. 2012;15 Suppl 1:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.UNDP. Human Development Index. Available: https://data.undp.org/dataset/Table-1-Human-Development-Index-and-its-components/wxub-qc5k. Accessed december 30, 2013 2012.
  • 31.Decroo T, Telfer B, Biot M, et al. Distribution of antiretroviral treatment through self-forming groups of patients in Tete Province, Mozambique. J Acquir Immune Defic Syndr. 2011;56:e39–44. [DOI] [PubMed] [Google Scholar]
  • 32.Koole O, Tsui S, Wabwire-Mangen F, et al. Retention and risk factors for attrition among adults in antiretroviral treatment programmes in Tanzania, Uganda and Zambia. Trop Med Int Health. 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Bolton-Moore C, Mubiana-Mbewe M, Cantrell RA, et al. Clinical outcomes and CD4 cell response in children receiving antiretroviral therapy at primary health care facilities in Zambia. JAMA. 2007;298:1888–1899. [DOI] [PubMed] [Google Scholar]
  • 34.Miller TL. Nutritional aspects of HIV-infected children receiving highly active antiretroviral therapy. AIDS. 2003;17 Suppl 1:S130–140. [DOI] [PubMed] [Google Scholar]
  • 35.Ansari NA, Kombe AH, Kenyon TA, et al. Pathology and causes of death in a series of human immunodeficiency virus-positive and -negative pediatric referral hospital admissions in Botswana. Pediatr Infect Dis J. 2003;22:43–47. [DOI] [PubMed] [Google Scholar]
  • 36.World Health Organization. Guidelines for an Integrated Approach to the Nutritional care of HIV-infected children (6 months-14 years). Available at: http://whqlibdoc.who.int/publications/2009/9789241597524_eng_Handbook.pdf. Accessed January 7, 2014. [PubMed]
  • 37.Nicol MP, Workman L, Isaacs W, et al. Accuracy of the Xpert MTB/RIF test for the diagnosis of pulmonary tuberculosis in children admitted to hospital in Cape Town, South Africa: a descriptive study. Lancet Infect Dis. 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Gil AC, Lorenzetti R, Mendes GB, et al. Hepatotoxicity in HIV-infected children and adolescents on antiretroviral therapy. Sao Paulo medical journal = Revista paulista de medicina. 2007;125:205–209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pol S, Lebray P, Vallet-Pichard A. HIV infection and hepatic enzyme abnormalities: intricacies of the pathogenic mechanisms. Clin Infect Dis. 2004;38 Suppl 2:S65–72. [DOI] [PubMed] [Google Scholar]
  • 40.Taha TE, Kumwenda N, Gibbons A, et al. Effect of HIV-1 antiretroviral prophylaxis on hepatic and hematological parameters of African infants. AIDS. 2002;16:851–858. [DOI] [PubMed] [Google Scholar]
  • 41.Anigilaje EA, Olutola A. Prevalence and Clinical and Immunoviralogical Profile of Human Immunodeficiency Virus-Hepatitis B Coinfection among Children in an Antiretroviral Therapy Programme in Benue State, Nigeria. ISRN pediatrics. 2013;2013:932697. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Geng EH, Emenyonu N, Bwana MB, Glidden DV, Martin JN. Sampling-based approach to determining outcomes of patients lost to follow-up in antiretroviral therapy scale-up programs in Africa. JAMA. 2008;300:506–507. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES