Abstract
Despite effective antiretroviral therapy (ART), children born to women living with HIV but who remain uninfected (CHEU) have poorer growth compared to children born to women living without HIV (CHU). Few studies have compared growth beyond early childhood. In a cross-sectional study, we enrolled CHEU and CHU aged 3–10 years. Growth Z-scores were calculated using WHO growth standards (weight-for-age [WAZ], height-for-age [HAZ], body mass index [BMIZ] [5–10 years], and head circumference-for-age [HCZ] [3–5 years]). Growth faltering was defined as Z-scores of <−2 SDs. We compared CHEU/CHU growth Z-scores and prevalence of growth faltering using generalized linear models and binomial regression, adjusting for sociodemographic characteristics. We enrolled 2001 CHEU and 200 CHU. The median age was similar in both categories. CHEU were more likely to be orphaned and vulnerable children (22%) and to have shorter breastfeeding period (12 vs. 24 months). CHEU mothers were older, had fewer years of education, and were more likely to be single parents. Most CHEU mothers were on ART during pregnancy with the enrolled child. CHEU had significantly lower HAZ (adjusted mean difference [aMD] −0.41, 95% CI −0.66, −0.17, P = .001) and WAZ (aMD −0.25, 95% CI −0.43, −0.08, P = .005). CHEU had 1.7 times higher prevalence of stunting (HAZ <−2) in unadjusted models, which was attenuated after adjustment. There were no differences in BMIZ or HCZ. Older CHEU aged 3–10 years had lower WAZ and HAZ scores than CHU. Poorer growth outcomes persisted after adjustment for socio-economic factors, suggesting a potential role of biological factors.
Keywords: HIV exposure, stunting, underweight, children, Kenya
Introduction
Globally, the number of children born to women living with HIV but who remain uninfected (CHEU) continues to rise, with an estimated 16 million CHEU in 2023 [1]. While prevention of vertical transmission programs has been tremendously successful, poorer growth and development outcomes among CHEU compared to children born to women living without HIV (CHU) persist [2–6]. Poor growth, particularly stunting, impacts cognition, educational attainment, and economic potential for individuals and societies [7, 8]. The populations of CHEU in some countries approach 20% of all children; therefore, poor growth could have a large population impact [9]. With the growing population of CHEU in low- and middle-income countries (LMICs), there is a need to integrate growth monitoring for CHEU in existing child programs and potentially reevaluate the frequency of monitoring. Early identification and referral for growth evaluation and interventions could have benefit for all children in LMICs.
Previous studies have focused on growth outcomes among CHEU in the first 2 years of life. Evidence overwhelmingly points to lower length for age (LAZ), weight for age (WAZ), and head circumference-for-age (HCZ) Z-scores and higher prevalence of stunting among CHEU compared to CHU even after adjusting for sociodemographic factors [3, 4, 10–14]. However, some studies have found no difference in growth outcomes [15]. A majority of studies were conducted before antiretroviral treatment (ART) was available for all mothers living with HIV. Among predominantly breastfed children exposed to more recent regimens and maternal antiretroviral therapy (ART) for life, there continues to be evidence of lower growth Z-scores and growth faltering among CHEU [3, 4, 10–12].
Few studies have examined growth outcomes among older CHEU (beyond age 2 years). Among children aged 6–18 years, largely exposed to maternal single-dose ART, some cross-sectional surveys found no growth differences comparing CHEU to CHU [2, 16]. In a longitudinal study assessing CHEU from birth to age 7 years, CHEU had lower WAZ, LAZ, and BMI for age (BMIZ) Z-scores in early life, and these differences persisted at school age [15]. Data from older children largely comes from cohorts of children whose parents were exposed to single-drug or two-drug combination or selected ART for prevention of vertical transmission [15, 16]. There is a dearth of evidence for growth outcomes for older children born in the era of triple ART treatment for all mothers.
Risk factors for poor growth among CHEU are multifactorial, involving an interplay of universal risk factors that affect all children, including poverty, family factors, maternal health, environmental factors, and HIV-specific factors such as maternal and fetal immune activation, exposure to HIV virions, and ART [9, 17]. Evidence points to best outcomes for CHEU survival when mothers are alive, on ART, and breastfeeding [18]. However, it is unclear if infants with early growth deficits catch up later and if other factors play a role in later growth. Understanding the differences in growth outcomes beyond early childhood may help build support for interventions throughout childhood and improve long-term outcomes for CHEU [3].
We conducted a large population-based cross-sectional survey of 2201 CHEU and CHU aged 3–10 years in Kenya to determine growth differences, prevalence of growth faltering, and correlates of poor growth among CHEU.
Materials and Methods
This was a cross-sectional study conducted in three diverse counties in Kenya (Homabay [rural], Nairobi [urban], and Kisumu [urban/periurban]) between September 2022 and July 2024. This work is part of a larger primary study that aims to understand differences in growth, neurodevelopment, and hearing among CHEU children/youth compared to age-matched CHU. The parent study is ongoing and targets to enroll 4400 CHEU and CHU children aged 3–18 years, with oversampling for CHEU (ratio 1:10). Details of the primary study are available in the published protocol manuscript [19]. The study received ethical approval from the Kenyatta National Hospital ethics and research committee (P400/07/2020) and the University of Washington institutional review board (STUDY00010495).
Eligibility criteria for this analysis were: children aged 3–10 years with growth data. Eligibility for the primary study included: HIV status of mother was known at the time of pregnancy or breastfeeding with the enrolled child, and child was known to be HIV negative through a negative HIV antibody test at least 6 weeks after breastfeeding cessation. Caregivers gave informed consent for child participation in the study, and children aged 7–10 provided assent to participate. Caregivers were recruited using a standardized script from maternal and child health (MCH) clinics for CHU and HIV clinics for CHEU as they attended routine clinic visits. Data on sociodemographic factors, ART history, and HIV testing history were collected at enrollment. Child HIV status was determined through review of MCH records where available and HIV testing was offered for the child if not tested after cessation of breastfeeding.
Growth was assessed using standard anthropometric measures. Weight was measured using calibrated Seca 803 digital flat scales with a limit of 150 kg across the sites to the nearest 0.1 kg; standing height was measured to the nearest 0.1 cm using a ShorrBoard or Seca 206 roll-up measuring tape with wall attachment and without shoes; mid-upper arm was measured using United Nations Children's Fund (UNICEF) measuring tapes; and head circumference was measured using a tailor’s tape measure. Two consecutive measures were collected for all children. If the second measurement differed from the first measurement by more than 0.5 cm for height, 0.10 cm for weight, and 0.2 cm for head circumference and mid-upper arm circumference (MUAC), a third measurement was obtained. Final growth measures were defined as the average of the first two measures or the average of the three measures if the first two measures were discrepant. Scales were standardized and frequently calibrated, and refresher training and assessments were conducted regularly. Growth Z-scores were calculated using the WHO growth standards [20] (weight-for-age [WAZ], height-for-age [HAZ], body mass index [BMIZ] [age 5–10 years], and head circumference-for-age [HCZ] [age 3–5 years]). Z-scores lower than −6 SDs or higher than +5 SDs were excluded from the analysis as previously described for HAZ and WAZ [21]. Growth faltering was defined as Z-scores of <−2 SDs and referred to as stunting (for HAZ), underweight (for WAZ), and microcephaly (for HCZ).
Sociodemographic factors collected included: age, sex (defined as male or female), and orphaned or vulnerable child (OVC) (defined as either mother or father died or not supporting the child physically, emotionally, or financially). Early life history factors included birth weight and breastfeeding. Caregiver characteristics included relationship with the child, age, sex, and years of education. Household factors included marital status and household food security assessed using the Household Food Insecurity Access Scale (HFIAS) for measurement of food access v3.0 questionnaire, which captures information on anxiety/uncertainty about food supply, insufficient quality of food, and insufficient food intake and categorizes food insecurity into four groups (food secure, mildly, moderately, and severely food insecure) [22]. Caregivers who were biological mothers were assessed for weight and height using similar procedures as child assessments, and BMI was estimated using standard BMI calculators. For CHEU mothers, data on ART timing (preconception, during pregnancy, and during breastfeeding) and regimens were collected through self-report.
Data were summarized using counts (%) for categorical data and medians (interquartile range [IQR]) for continuous data. We used generalized linear models accounting for clustering by site and families (siblings were enrolled) to estimate crude and adjusted mean differences (aMDs) and 95% confidence intervals (CIs) in growth outcomes comparing CHEU and CHU and to determine correlates of poor growth among CHEU. We used binomial regression to compare prevalence ratios and 95% CIs for growth faltering. Models were adjusted for sociodemographic factors that differed between CHEU and CHU. All analyses were conducted using Stata version 17 (StataCorp, College Station, Texas, United States).
Results
A total of 2201 children were enrolled, 2001 CHEU and 200 CHU. Compared to CHU, CHEU were significantly more likely to be OVC (22% vs. 10%). While almost all CHEU and all CHU were breastfed, CHEU breastfed for a significantly shorter duration (12 vs. 24 months). There were no differences in birth weight, early milestone attainment (walking and talking), school enrollment, or major medical diagnoses when comparing CHEU and CHU. Almost all primary caregivers were biological mothers (>99% for CHEU and 100% for CHU). CHEU caregivers were significantly older (35 vs. 31 years), had lower number of years of education (8 vs. 12 years), were more likely to be single parents (74% vs. 88%), and were more likely to have severe food insecurity (66% vs. 56%). History of anemia was more commonly reported among CHEU mothers (7% vs. 3%), and more CHU mothers had BMI scores of >30 (23% vs. 14%) (Table 1).
Table 1.
Characteristics of children and their caregivers
| Characteristic | CHEU, n = 2001 | CHU, n = 200 |
|---|---|---|
| n (%) or median (IQR) | n (%) or median (IQR) | |
| Child characteristics | ||
| Child age (years) | 6.0 (4.4, 8.3) | 6.0 (4.3, 8.3) |
| Female | 994 (50%) | 100 (50%) |
| Orphaned and vulnerable child (OVC) | 445 (22%) | 19 (10%) |
| Full-term infant (>37 weeks’ gestation) | 1924 (96%) | 193 (97%) |
| Birth weight in grams | 3300 (2900, 3600) | 3200 (2950, 3550) |
| Birth weight <2500 g | 115 (6%) | 9 (5%) |
| Ever breastfed | 1976 (99%) | 200 (100%) |
| Duration of breastfeeding in months | 12 (7, 17) | 24 (12, 24) |
| Breastfed for ≥12 months | 635 (32%) | 149 (75%) |
| Duration of breastfeeding in months | 12 (7, 15) | 24 (12, 24) |
| Walked by 18 months | 1920 (96%) | 191 (96%) |
| Said 2 or more words by 2 years | 1929 (96%) | 190 (95%) |
| Currently enrolled in school | 1701 (99%) | 180 (99%) |
| Child medical history | ||
| History of tuberculosis | 6 (<1%) | 0 (0%) |
| History of recurrent pneumonia | 68 (3%) | 11 (6%) |
| History of malnutrition or poor growth | 268 (13%) | 20 (10%) |
| Caregiver characteristics | ||
| Biological mother | 1993 (>99%) | 200 (100%) |
| Age (years) | 35 (31, 39) | 31 (29, 34) |
| Female | 1993 (>99%) | 199 (>99%) |
| Years of education | 8 (8, 12) | 12 (8, 14) |
| Household characteristics | ||
| Married or steady relationship | 1481 (74%) | 176 (88%) |
| Household food security (HFIAS) | ||
| Food secure | 106 (5%) | 19 (10%) |
| Mildly food insecure | 40 (2%) | 6 (3%) |
| Moderately food insecure | 527 (26%) | 63 (32%) |
| Severely food insecure | 1325 (66%)* | 110 (56%)* |
| Maternal health | ||
| History of tuberculosis | 18 (1%) | 0 (0%) |
| History of malnutrition | 58 (3%) | 2 (1%) |
| History of pneumonia | 126 (6%) | 14 (7%) |
| History of anemia | 132 (7%) | 6 (3%) |
| Maternal BMI | ||
| Underweight (<18.5) | 113 (6%) | 11 (6%) |
| Normal weight (18.5–24.9) | 1080 (55%) | 86 (43%) |
| Overweight (25–30) | 503 (26%) | 55 (28%) |
| Obese (>30) | 273 (14%)** | 46 (23%)** |
| Disclosure of HIV status to father of child | 1782 (90%) | 154 (77%) |
| CHEU mothers | ||
| Pre-pregnancy HIV diagnosis | 1539 (79%) | |
| Pre-pregnancy ART initiation | 1444 (74%) | |
| Child received ART prophylaxis | 1974 (99%) | |
| ART regimen during pregnancy or breastfeeding | ||
| EFV- or NVP-based | 1026 (77%) | |
| DTG-based | 130 (10%) | |
| PI-based | 41 (3%) | |
| Single-drug ART | 139 (10%) |
Bold values indicate P-value <.05.
P-value comparing food secure group to severely food insecure group.
P-value comparing normal weight group to obese group.
Among CHEU mothers, the majority had a pre-pregnancy HIV diagnosis (79%), and 74% initiated ART prior to pregnancy with the enrolled child. A majority of CHEU mothers (90%) had been on triple ART in pregnancy and breastfeeding, with regimens that were efavirenz (EVP)- or nevirapine (NVP)-based (77%), dolutegravir-based (10%), or protease inhibitor-based (3%). Overall, 10% of CHEU mothers reported being on a single-drug antiretroviral regimen at the time of pregnancy with the enrolled child (Table 1).
In unadjusted and adjusted analysis, HAZ scores were significantly lower among CHEU compared to CHU (aMD −0.41, 95% CI −0.66, −0.17, P = .001). CHEU had a significantly higher prevalence of stunting in unadjusted analysis, but the effect was attenuated in the adjusted analysis (adjusted prevalence ratio 1.52, 95% CI 0.80, 2.61, P = .122). Similarly, we observed significantly lower WAZ scores in adjusted and unadjusted analysis among CHEU compared to CHU (aMD −0.25, 95% CI −0.43, −0.08, P = .005), but no difference in the prevalence of underweight. There was no difference in BMIZ or HCAZ or the prevalence of Z-scores <2 SDs for BMI or head circumference (Table 2).
Table 2.
Unadjusted and adjusted analysis comparing growth in CHEU and CHU
| Growth measure | CHEU Mean (SD)/n (%) | CHU Mean (SD)/n (%) | Unadjusted mean difference or prevalence ratio (PR) (95% CI) |
P-value | Adjusted mean difference or prevalence ratio (PR) (95% CI) |
P-value |
|---|---|---|---|---|---|---|
| Height for age (n = 2151) | ||||||
| Z-score | −0.58 (1.44) | −0.15 (1.39) | −0.41 (−0.64, −0.19) | <.001 | −0.41 (−0.66, −0.17) | .001 |
| Stunting (HAZ ≤−2 SDs) | 274 (14%) | 16 (8%) | 1.74 (1.06, 2.87) | .030 | 1.52 (0.89, 2.61) | .122 |
| Weight for age (n = 2010) | ||||||
| Z-score | −0.48 (1.01) | −0.18 (1.11) | −0.30 (−0.46, −0.14) | <.001 | −0.25 (−0.43, −0.08) | .005 |
| Underweight (WAZ ≤−2 SDs) | 102 (6%) | 9 (5%) | 1.17 (0.60, 2.27) | .641 | 0.84 (0.42, 1.68) | .615 |
| BMI for age (n = 2145) | ||||||
| Z-score | −0.21 (1.27) | −0.21 (1.24) | −0.01 (−0.21, 0.18) | .900 | 0.03 (−0.19, 0.25) | .772 |
| BMIZ (≤−2 SDs) | 113 (6%) | 15 (8%) | 0.78 (0.43, 1.40) | .401 | 0.70 (0.34, 1.44) | .330 |
| Head circ. for age (n = 787) | ||||||
| Z-score | 0.24 (1.00) | 0.51 (0.90) | −0.28 (−0.51, 0.05) | .170 | * | * |
| Z-score ≤−2 SDs | 9 (1.3%) | 0 (0%) | .610a | |||
| Z-score ≤−1 SD | 79 (11%) | 3 (4%) | 2.98 (0.96, 9.20) | .058 | 2.96 (0.92, 9.52) | .068 |
Unadjusted mean models allow for clustering in families and county; prevalence ratio models allow for clustering in families and adjust for county.
Adjusted for maternal age, education, household food security, breastfeeding duration, and OVC status. HAZ measures were also adjusted for maternal height.
Chi-square test.
indicates model did not converge.
Bolded values are statistically significant
Among CHEU, correlates of lower HAZ scores in adjusted analysis included being from a single-parent family (aMD −0.42, 95% CI −0.63, −0.21, P < .001) and severe food insecurity (aMD −0.45, 95% CI −0.56, −0.21, P = .003), while correlates for lower WAZ scores included birth weight of <2500 g (aMD −0.46, 95% CI −0.66, −0.25, P < .001) and moderate (aMD −0.20, 95% CI −0.561, −0.08, P = .008) or severe food insecurity (aMD −0.32, 95% CI −0.56, −0.14, P = .002). Higher number of years of maternal education was associated with small but significantly higher WAZ and HCZ scores (aMD 0.02, 95% CI −0.00, −0.04, P = .031 and aMD −0.03, 95% CI −0.00, −0.05, P = .040, respectively). BMIZ scores were lower among females, OVC, and low birth weight CHEU (aMD −0.12, 95% CI −0.23, −0.01, P = .029, aMD −0.21, 95% CI −0.36, −0.06, P = .006, and aMD −0.29, 95% CI −0.54, −0.04, P = .024), respectively. Compared to maternal post-pregnancy ART, pre-pregnancy ART was associated with higher BMIZ scores (aMD 0.15, 95% CI 0.03, 0.39, P = .024). For ART regimens, compared to EFV/NVP-based regimens used in pregnancy, dolutegravir regimens and single-drug ART regimens were associated with higher BMIZ scores (aMD 0.27, 95% CI 0.03, 0.51, P = .027 and aMD 0.28, 95% CI 0.04, 0.52, P = .022, respectively) (Table 3).
Table 3.
Correlates of growth Z-scores among CHEU
| HAZ |
WAZ |
BMIZ |
HCZ |
|||||
|---|---|---|---|---|---|---|---|---|
| Unadjusted mean diff (95% CI) | Adjusted mean diff (95% CI) | Unadjusted mean diff (95% CI) | Adjusted mean diff (95% CI) | Unadjusted mean diff (95% CI) | Adjusted mean diff (95% CI) | Unadjusted mean diff (95% CI) | Adjusted mean diff (95% CI) | |
| Child characteristics | ||||||||
| Female | 0.03 (−0.09, 0.15) | 0.04 (−0.08, 0.16) | −0.04 (−0.13, 0.05) | −0.03 (−0.12, 0.06) | −0.13 (−0.24, −0.17) | −0.12 (−0.23, −0.01) | — | — |
| OVC status | 0.10 (−0.07, 0.26) | 0.11 (−0.06, 0.27) | −0.10 (−0.22, 0.02) | −0.11 (−0.22, 0.01) | −0.20 (−0.35, −0.06) | −0.21 (−0.36, −0.06) | 0.11 (−0.08, 0.30) | 0.13 (−0.06, 0.33) |
| Birthweight <2500 g | −0.34 (−0.61, −0.08) | −0.26 (−0.53, 0.01) | −0.51 (−0.71, −0.31) | −0.46 (−0.66, −0.25) | −0.36 (−0.60, −0.11) | −0.29 (−0.54, −0.04) | −0.11 (−0.42, 0.20) | −0.14 (−0.45, 0.18) |
| Ever breastfed | 0.14 (−0.42, 0.71) | −0.02 (−0.58, 0.55) | 0.43 (−0.03, 0.89) | −0.41 (−0.05, 0.87) | −0.36 (−0.15, 0.87) | 0.35 (−0.16, 0.86) | −0.70 (−1.68, 0.29) | −0.69 (−1.68, 0.30) |
| Breastfed >12 months | −0.08 (−0.22, 0.06) | −0.06 (−0.20, 0.08) | 0.00 (−0.10, 0.10) | 0.10 (−0.09, 0.11) | 0.02 (−0.10, 0.15) | 0.03 (−0.10, 0.15) | −0.05 (−0.20, 0.10) | −0.07 (−0.22, 0.09) |
| Caregiver characteristics | ||||||||
| Age (years) | 0.00 (−0.01, 0.01) | 0.00 (−0.01, 0.01) | 0.00 (−0.01, 0.01) | 0.00 (0.00, 0.01) | 0.00 (−0.01, 0.01) | 0.00 (−0.01, 0.01) | 0.00 (−0.02, 0.01) | 0.00 (−0.01, 0.01) |
| Years of education | 0.02 (−0.01, 0.04) | 0.01 (−0.01, 0.03) | 0.02 (0.00, 0.04) | 0.02 (0.00, 0.04) | 0.02 (0.00, 0.04) | 0.02 (0.00, 0.04) | 0.03 (0.00, 0.05) | 0.03 (0.00, 0.05) |
| Household characteristics | ||||||||
| No current partnera | −0.20 (−0.35, −0.04) | −0.42 (−0.63, −0.21) | −0.15 (−0.26, −0.04) | −0.11 (−0.27, 0.05) | −0.08 (−0.22, 0.05) | −0.15 (−0.05, 0.34) | 0.06 (−0.11, 0.24) | −0.01 (−0.34, 0.22) |
| Food secure | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Mildly food insecure | −0.11 (−0.68, 0.45) | −0.11 (−0.67, 0.46) | −0.11 (−0.51, 0.30) | −0.15 (−0.55, 0.25) | −0.10 (−0.60, 0.39) | −0.13 (−0.63, 0.38) | −0.11 (−0.49, 0.71) | 0.01 (−0.62, 0.63) |
| Moderately food insecure | −0.23 (−0.55, 0.08) | −0.21 (−0.52, 0.10) | −0.31 (−0.53, −0.09) | −0.30 (−0.51, −0.08) | −0.26 (−0.54, 0.02) | −0.25 (−0.53, 0.04) | −0.07 (−0.39, 0.25) | −0.06 (−0.39, 0.26) |
| Severely food insecure | −0.42 (−0.72, −0.12) | −0.45 (−0.75, −0.16) | −0.35 (−0.56, −0.14) | −0.32 (−0.53, −0.11) | −0.17 (−0.43, 0.10) | −0.11 (−0.38, 0.16) | −0.06 (−0.36, 0.24) | −0.05 (−0.35, 0.25) |
| ART characteristics | ||||||||
| Post-pregnancy ART | −0.04 (−0.18, 0.10) | −0.02 (−0.16, 0.13) | 0.09 (−0.02, 0.19) | 0.08 (−0.03, 0.18) | 0.17 (0.04, 0.29) | 0.15 (0.02, 0.28) | 0.21 (0.03, 0.39) | — |
| EFV/NVP regimen | Ref | Ref | Ref | Ref | Ref | |||
| DTG regimen | — | −0.03 (−0.30, 0.23) | 0.11 (−0.07, 0.29) | 0.10 (−0.09, 0.28) | 0.29 (0.05, 0.52) | 0.27 (0.03, 0.51) | 0.16 (−0.00, 0.39) | 0.14 (−0.10, 0.38) |
| PI regimen | — | 0.32 (−0.13, 0.77) | 0.21 (−0.12, 0.53) | 0.31 (−0.02, 0.64) | 0.10 (−0.30, 0.49) | 0.20 (−0.20, 0.61) | 0.46 (−0.04, 0.97) | 0.49 (−0.02, 0.99) |
| Single drug ART | — | −0.50 (−0.76, −0.23) | 0.00 (−0.19, 0.18) | −0.03 (−0.22, 0.15) | 0.30 (0.06, 0.53) | 0.28 (0.04, 0.52) | 0.17 (−0.16, 0.49) | 0.17 (−0.16, 0.50) |
Single, divorced, or separated.
Bolded values are statistically significant.
Among CHU, in unadjusted analysis, we found that food insecurity was associated with lower HAZ, WAZ, and BMIZ scores.
Discussion
In this large cross-sectional study, we found that CHEU children had significantly lower HAZ and WAZ scores compared to CHU peers, differences that persisted even after adjusting for sociodemographic factors. These findings are relevant to current CHEU cohorts, as a vast majority of mothers of CHEU were on optimized ART regimens, often initiated pre-pregnancy, and almost all children were breastfed. CHEU children had an almost 50% higher prevalence of stunting, which was attenuated in models adjusted for differences in sociodemographic factors. Correlates of poorer growth among CHEU included having a single parent, lower food security, OVC status, and low birth weight (<2500 g).
Few studies have evaluated growth among school-age CHEU. Among Zambian children whose median age was 7.5 years, born between 2005 and 2007—when the standard of care was perinatal nevirapine for mothers and infants and maternal ART when CD4 count was less than 200—CHEU had poor early growth (lower WAZ and borderline lower HAZ scores) that persisted to school age, though the differences were smaller at school age [16]. Among Ugandan CHEU children (mean age 11 years) born between 2000 and 2011 (prior to universal ART), Awadu et al. found poorer HAZ scores among CHEU not exposed to ART but unexpectedly, higher HAZ scores among CHEU exposed to maternal ART compared to CHU [2]. Conversely, Rosala-Hallas et al. reported no differences in growth Z-scores after adjusting for sociodemographic factors comparing CHEU and CHU ages 6–12 years born between 2001 and 2009 [15]. All these studies analyzed data from children enrolled in the pre-universal ART era and could be biased, as they potentially selected for mothers and children who were healthier, who survived, and who were enrolled in ongoing clinical trials.
Among younger children (under 2 years), most studies report lower LAZ, lower WAZ scores, and higher prevalence of stunting among CHEU compared to CHU [3, 4, 10–13, 23]. However, some studies show no growth differences or attenuation of differences after adjusting for sociodemographic factors [3, 4, 14]. Most recent studies have been conducted among breastfed CHEU exposed to optimized maternal ART regimens, suggesting that other factors may be contributing to poorer growth outcomes [3, 4, 10–12]. We observed significantly higher prevalence of stunting among CHEU compared to CHU in unadjusted models, which was attenuated in models adjusted for sociodemographic factors, underscoring the prominent role these factors play in growth of CHEU. These factors are important for all children, especially those in LMICs who face multiple vulnerabilities.
As in other cohorts of CHEU [3, 24], our study identified that CHEU were more likely to be OVC, to breastfeed for a shorter time, to live in single-parent homes, and to have a higher prevalence of food insecurity. These factors are associated with poorer growth and development independent of HIV and are exacerbated in HIV-affected families and profoundly impact health outcomes. Interventions that focus on nurturing care may provide the most benefit not only for CHEU but also for all children in LMICs by addressing multiple intersecting risk factors [25]. We found that CHEU with low birth weight were more likely to have poor growth, a finding observed in other studies [26] among both CHEU and CHU that highlights the need for close monitoring of low birth weight (LBW). Interventions that support families affected by HIV and with other vulnerabilities to develop economic skills and build more social support are important to improve child outcomes.
We did not find growth differences by ART regimen or the timing of maternal ART initiation (pre- or post-pregnancy ART). The majority of CHEU mothers (77%) were on EFV/NVP-based ART regimens, which may have limited our ability to see differences. We observed small but significant BMIZ differences by ART timing and among mothers on dolutegravir/single ART regimens; these findings could be explained by better maternal health associated with dolutegravir regimens or selection of single regimens in the pre-universal ART era.
We did not observe differences in HCAZ among CHEU or CHU aged 3–5 years. Previous studies in younger children (age 9 months) have identified lower mean HCZ and twice the prevalence of microcephaly among CHEU [10]. It is possible that these differences reduce with time as CHEU catch up in growth, as we did observe an almost three-fold higher prevalence of HCZ of <−1 SD among CHEU. Future studies on brain imaging may support a better understanding of brain growth and early versus late differences in CHEU/CHU populations. At the population level, assessment of head circumference in LMICs is often not emphasized, but it is easy to conduct and may support early identification of children who need referral.
Strengths and limitations
Our study strengths include a large sample of CHEU populations drawn from diverse settings (both rural and urban). This was a cross-sectional study, which limits our ability to infer causality. Recall bias for childhood events may result in misclassification of exposures, as we relied on caregiver self-reports on exposures. It is possible that enrolled CHEU were a select population whose caregivers were interested in and enrolled in the study, and this may not represent the most vulnerable CHEU. Our CHU population was also small, limiting CHU subgroup analysis. Unmeasured confounders, such as environmental or genetic factors, may also have influenced the results.
Conclusions
In this large cross-sectional study, we found that children aged 3–10 years born to women living with HIV but who remain uninfected had lower HAZ and WAZ Z-scores compared to CHU, even after adjusting for sociodemographic factors. This may suggest biological differences that affect growth beyond early childhood in children who remain uninfected, though they are born to women living with HIV. Stunting was more prevalent in CHEU compared to children born to women living without HIV but was attenuated after adjustment for sociodemographic differences, suggesting that addressing sociodemographic factors could have important growth benefits.
Acknowledgements
We express deep appreciation to various collaborators, the study staff, and study participants for their dedicated time and efforts to the study.
Contributor Information
Marion Muranda, Research and Programs Department, Kenyatta National Hospital, P.O Box 20723-00202, Nairobi, Kenya.
Maureen King’e, Department of Paediatrics and Child Health, University of Nairobi, P.O. Box 19676-00202, Nairobi, Kenya.
Hellen Moraa, Research and Programs Department, Kenyatta National Hospital, P.O Box 20723-00202, Nairobi, Kenya; Department of Global Health, University of Washington, Seattle, WA 98104, United States.
Alvin Onyango, Research and Programs Department, Kenyatta National Hospital, P.O Box 20723-00202, Nairobi, Kenya.
Rhoda Oricho, Research and Programs Department, Kenyatta National Hospital, P.O Box 20723-00202, Nairobi, Kenya.
Jillian Neary, Department of Global Health, University of Washington, Seattle, WA 98104, United States.
Grace John-Stewart, Department of Global Health, University of Washington, Seattle, WA 98104, United States; Department of Epidemiology, University of Washington, Seattle, WA 98104, United States; Department of Medicine, University of Washington, Seattle, WA 98104, United States; Department of Pediatrics, University of Washington, Seattle, WA 98104, United States.
Dalton Wamalwa, Department of Paediatrics and Child Health, University of Nairobi, P.O. Box 19676-00202, Nairobi, Kenya.
Christine J McGrath, Department of Global Health, University of Washington, Seattle, WA 98104, United States.
Irene Njuguna, Research and Programs Department, Kenyatta National Hospital, P.O Box 20723-00202, Nairobi, Kenya; Department of Global Health, University of Washington, Seattle, WA 98104, United States.
Author contributions
Marion Muranda (Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Software [equal], Visualization [equal], Writing—original draft [equal]), Maureen King’e (Project administration [equal], Resources [equal], Validation [equal], Writing—review & editing [equal]), Hellen Moraa (Project administration [equal], Resources [equal], Supervision [equal], Writing—review & editing [equal]), Alvin Onyango (Data curation [equal], Formal analysis [equal], Software [equal], Validation [equal], Writing—review & editing [equal]), Rhoda Oricho (Methodology [equal], Resources [equal], Validation [equal], Writing—review & editing [equal]), Jillian Neary (Data curation [equal], Formal analysis [equal], Software [equal], Validation [equal], Writing—review & editing [equal]), Grace John-Stewart (Conceptualization [equal], Funding acquisition [equal], Resources [equal], Validation [equal], Writing—review & editing [equal]), Dalton Wamalwa (Conceptualization [equal], Project administration [equal], Validation [equal], Writing—review & editing [equal]), Christine J. McGrath (Data curation [equal], Formal analysis [equal], Validation [equal], Writing—review & editing [equal]), and Irene Njuguna (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Methodology [equal], Software [equal], Validation [equal], Writing—original draft [equal])
Conflict of interest: The authors declare that they have no competing interests.
Funding
This work was supported by Eunice Kennedy Shriver National Institute of Child Health and Human Development [ R33HD10301079], National Institutes of Health (NIH), under award [1R61HD103079].
References
- 1. AIDSinfo | UNAIDS. https://www.unaids.org/sites/default/files/mediaasset/data-book-2023en.pdf (9 May 2025, date last accessed).
- 2. Awadu JE, Zalwango SK, Sikorskii A et al. Growth differences by school-age and adolescence according to in utero and peripartum antiretroviral therapy exposure among Ugandan children. Medicine (Baltimore) 2023;102:e32677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Jumare J, Datong P, Osawe S et al. ; INFANT Study Team. Compromised growth among HIV-exposed uninfected compared with unexposed children in Nigeria. Pediatr Infect Dis J 2019;38:280–6. [DOI] [PubMed] [Google Scholar]
- 4. Mabaya L, Matarira HT, Tanyanyiwa DM et al. Growth trajectories of HIV exposed and HIV unexposed infants. A prospective study in Gweru, Zimbabwe. Glob Pediatr Health 2021;8:2333794X21990338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Nyemba DC, Kalk E, Vinikoor MJ et al. Growth patterns of infants with in-utero HIV and ARV exposure in Cape Town, South Africa and Lusaka, Zambia. BMC Public Health 2022;22:55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Slogrove AL, Becquet R, Chadwick EG et al. Surviving and thriving—shifting the public health response to HIV-exposed uninfected children: report of the 3rd HIV-exposed uninfected child workshop. Front Pediatr 2018;6:157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Alam MA, Richard SA, Fahim SM et al. Impact of early-onset persistent stunting on cognitive development at 5 years of age: results from a multi-country cohort study. PLoS One 2020;15:e0227839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Akseer N, Tasic H, Nnachebe Onah M et al. Economic costs of childhood stunting to the private sector in low- and middle-income countries. EClinicalMedicine 2022;45:101320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bulterys MA, Njuguna I, Mahy M et al. Neurodevelopment among children exposed to HIV and uninfected in sub‐Saharan Africa. J Int AIDS Soc 2023;26:e26159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Neary J, Langat A, Singa B et al. Higher prevalence of stunting and poor growth outcomes in HIV-exposed uninfected than HIV-unexposed infants in Kenya. AIDS 2022;36:605–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Aizire J, Sikorskii A, Ogwang LW et al. ; PROMISE-NEURODEV study team. Decreased growth among antiretroviral drug and HIV exposed uninfected versus unexposed children in Malawi and Uganda. AIDS 2020;34:215–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Le Roux SM, Abrams EJ, Donald KA et al. Growth trajectories of breastfed HIV-exposed uninfected and HIV-unexposed children under conditions of universal maternal antiretroviral therapy: a prospective study. Lancet Child Adolesc Health 2019;3:234–44. [DOI] [PubMed] [Google Scholar]
- 13. Chen J-C, Zhang Y, Rongkavilit C et al. Growth of HIV-exposed infants in southwest China: a comparative study. Glob Pediatr Health 2019;6:2333794X19854964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Chilyabanyama ON, Chilengi R, Laban NM et al. Comparing growth velocity of HIV exposed and non-exposed infants: an observational study of infants enrolled in a randomized control trial in Zambia. PLoS One 2021;16:e0256443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Rosala-Hallas A, Bartlett JW, Filteau S. Growth of HIV-exposed uninfected, compared with HIV-unexposed, Zambian children: a longitudinal analysis from infancy to school age. BMC Pediatr 2017;17:80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Nicholson L, Chisenga M, Siame J et al. Growth and health outcomes at school age in HIV-exposed, uninfected Zambian children: follow-up of two cohorts studied in infancy. BMC Pediatr 2015;15:66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wedderburn CJ, Evans C, Yeung S et al. Growth and neurodevelopment of HIV-exposed uninfected children: a conceptual framework. Curr HIV/AIDS Rep 2019;16:501–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Arikawa S, Rollins N, Jourdain G et al. Contribution of maternal antiretroviral therapy and breastfeeding to 24-month survival in human immunodeficiency virus-exposed uninfected children: an individual pooled analysis of African and Asian studies. Clin Infect Dis 2018;66:1668–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Njuguna IN, King’e M, Moraa H et al. Cohort profile: longitudinal and population comparison of children who are HIV-exposed uninfected and children who are HIV unexposed in Kenya (HOPE study). BMJ Open 2024;14:e081975. 10.1136/bmjopen-2023-081975 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. World Health Organization. Nutrition for Health. WHO child growth standards: growth velocity based on weight, length and head circumference: methods and development. World Health Organization; 2009.
- 21. Benedict RK, Namaste S, Croft T. 2022. Evaluation of Implausible Anthropometric Values by Data Collection Team in Demographic and Health Surveys 2010-2020. DHS Methodological reports No.33. Rockville, Maryland, USA: ICF.
- 22. Coates j, Swindale A, Bilinsky P. 2007. Household Food Insecurity Access Scale (HFIAS) for Measurement of Food Access: IndicatorGuide: (v. 3). Washington, D.C.: FHI 360/FANTA.
- 23. Nyemba DC, Kalk E, Madlala HP et al. Lower birth weight-for-age and length-for-age z-scores in infants with in-utero HIV and ART exposure: a prospective study in Cape Town, South Africa. BMC Pregnancy Childbirth 2021;21:354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Filteau S, Kasonka L, Wells JCK et al. Anthropometry, body composition, early growth and chronic disease risk factors among Zambian adolescents exposed or not to perinatal maternal HIV. Br J Nutr 2022;129:1–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Britto PR, Lye SJ, Proulx K et al. ; Early Childhood Development Interventions Review Group, for the Lancet Early Childhood Development Series Steering Committee. Nurturing care: promoting early childhood development. Lancet 2017;389:91–102. [DOI] [PubMed] [Google Scholar]
- 26. Sudfeld CR, Lei Q, Chinyanga Y et al. Linear growth faltering among HIV-exposed uninfected children. J Acquir Immune Defic Syndr 2016;73:182–9. [DOI] [PMC free article] [PubMed] [Google Scholar]