The African Health Initiative includes highly diverse partnerships in five countries (Ghana, Mozambique, Rwanda, Tanzania, and Zambia), each of which is working to improve population health by strengthening health systems and to evaluate the results. One aim of the Initiative is to generate cross-site learning that can inform implementation in the five partnerships during the project period and identify lessons that may be generalizable to other countries in the region. Collaborators in the Initiative developed a common evaluation framework as a basis for this cross-site learning.
This paper describes the components of the framework; this includes the conceptual model, core metrics to be measured in all sites, and standard guidelines for reporting on the implementation of partnership activities and contextual factors that may affect implementation, or the results it produces. We also describe the systems that have been put in place for data management, data quality assessments, and cross-site analysis of results.
Results and conclusions
The conceptual model for the Initiative highlights points in the causal chain between health system strengthening activities and health impact where evidence produced by the partnerships can contribute to learning. This model represents an important advance over its predecessors by including contextual factors and implementation strength as potential determinants, and explicitly including equity as a component of both outcomes and impact. Specific measurement challenges include the prospective documentation of program implementation and contextual factors. Methodological issues addressed in the development of the framework include the aggregation of data collected using different methods and the challenge of evaluating a complex set of interventions being improved over time based on continuous monitoring and intermediate results.
Tuberculosis (TB) is a leading cause of morbidity and mortality and is frequently complicated by emergence of drug-resistant strains. Diagnosis of TB in developing countries is often based on the relatively insensitive acid-fast staining that does not enable susceptibility profiling. Microscopic observation drug susceptibility assay (MODS) is an inexpensive, simple method that enables rapid TB culture coupled with susceptibility testing. A 3-week MODS training of three Ethiopian laboratory technicians was conducted at Hadassah-Hebrew University Medical Center, Israel. Results of the trainee readings were blindly assessed by an experienced instructor. Two hundred fifty-five (255) trainee culture readings were evaluated throughout the course. The sensitivity and specificity were 75–100% and 31.5–100%, respectively. Multivariate analysis revealed that sensitivity and duration of incubation were positively correlated, although specificity was positively correlated with the length of training. MODS can be reliably performed by laboratory technicians inexperienced in culture techniques in developing countries, with high sensitivity and specificity reached after a brief learning period.
Background. We assessed the impact of 12 years of pneumococcal conjugate vaccine (PCV7) use on pneumococcal nasopharyngeal carriage and serotype-specific invasive disease potential among Native Americans.
Methods. Families were enrolled in a carriage study from 2006 to 2008; nasopharyngeal specimens and risk factor information were collected monthly for 7 visits. Pneumococcal carriage prevalence was compared with that before (1998–2000) and during (2001–2002) PCV7 introduction. We compared invasive disease incidence and carriage prevalence before and after PCV7 introduction to estimate changes in serotype-specific invasive potential.
Results. We enrolled 1077 subjects from 302 households. There was an absolute reduction in carriage prevalence of 8.0% (95% confidence interval [CI], 4.5%–11.4%) in children aged <5 years and 3.1% (95% CI, 1.1%–5.1%) in adults. In children aged <5 years, vaccine-serotype carriage prevalence decreased by 22.8% (95% CI, 20.1%–25.3%), and nonvaccine serotype (NVT) increased by 15.9% (95% CI, 12.4%–19.3%). No significant change was detected in serotype-specific invasive potential after PCV7 introduction.
Conclusions. Pneumococcal carriage prevalence decreased in all ages since PCV7 introduction; vaccine-serotype carriage has been nearly eliminated, whereas the prevalence of NVT carriage has increased. The increase in the NVT invasive disease rate seems to be proportional to the increase in colonization prevalence.
This phase III randomized clinical trial compared single dose nevirapine (sdNVP) plus HIV immunoglobulin (HIVIGLOB) to sdNVP alone for preventing maternal-to-child transmission (PMTCT) of HIV.
Primary objectives were to determine rates of HIV infection among infants, and to assess the safety of HIVIGLOB in combination with sdNVP in HIV-infected Ugandan pregnant women and their infants.
Mother-infant pairs were randomized to receive 200mg of NVP to women in labor and 2mg/kg NVP to newborns within 72 hours after birth (sdNVP arm) or to receive sdNVP plus a single intravenous 240ml dose of HIVIGLOB given to women at 36-38 weeks gestation and a single intravenous 24ml dose to newborns within 18 hours of birth (HIVIGLOB/sdNVP arm). Risk of HIV infection was determined using Kaplan-Meier and risk ratio estimates at birth, 2, 6, 14 weeks, 6 and 12 months of age.
Intent-to-treat analysis included 198 HIVIGLOB/sdNVP and 294 sdNVP mother-infant pairs. At 6 months of age, the primary endpoint, there was no statistically significant difference in HIV transmission in the HIVIGLOB/sdNVP arm versus the sdNVP arm (18.7% vs.15.0%; RR =1.240 [95% CI: 0.833-1.846]; p= 0.290). Similarly, the proportion of serious adverse events in the HIVIGLOB/sdNVP and sdNVP arms, respectively for mothers (18.9% vs. 19.3%; p= 0.91) and infants (62.6% vs. 59.5%; p=0.51), were not significantly different.
Giving mother-infant pairs an infusion of peripartum HIV hyperimmunoglobulin in addition to sdNVP for PMTCT was as safe as sdNVP alone, but was no more effective than sdNVP alone in preventing HIV transmission.
HIV; HIVIGLOB; sdNVP; breastfeeding; PMTCT; Uganda
HIV-positive lactating women may be at high risk of weight loss due to increased caloric requirements and postpartum physiological weight loss. Ten percent weight loss is associated with a higher risk of mortality in HIV-positive patients and this alone is a criterion for highly active antiretroviral therapy (HAART) initiation where CD4 counts are not available. However, no study has investigated this association in lactating postpartum women. We investigated whether 10% weight loss predicts death in postpartum HIV-positive women. A total of 9207 HIV-negative and 4495 HIV-positive mothers were recruited at delivery. Women were weighed at 6 weeks, 3 months, and every 3 months thereafter for up to 24 months postpartum and data on mortality up to 2 years were collected. The median duration of breastfeeding was longer than 18 months. Among HIV-positive women, the independent predictors of ≥10% weight loss were CD4 cell count, body mass index, and household income. Mortality was up to 7.12 (95% CI 3.47–14.61) times higher in HIV-positive women with ≥10% weight loss than those without weight loss. Ten percent weight loss in postpartum lactating HIV-positive women was significantly predictive of death. Our findings suggest that 10% weight loss is an appropriate criterion for HAART initiation among postpartum breastfeeding women.
In recent years, use of the Internet to obtain vaccine information has increased. Historical data are necessary to evaluate current vaccine information seeking trends in context. Between 2002 and 2003, surveys were mailed to 1,630 parents of fully vaccinated children and 815 parents of children with at least one vaccine exemption; 56.1% responded. Respondents were asked about their vaccine information sources, perceptions of these sources accuracy, and their beliefs about vaccination. Parents who did not view their child's healthcare provider as a reliable vaccine information source were more likely to obtain vaccine information using the Internet. Parents who were younger, more highly educated, and opposed to school immunization requirements were more likely than their counterparts to use the Internet for vaccine information. Compared to parents who did not use the Internet for vaccine information, those who sought vaccine information on the Internet were more likely to have lower perceptions of vaccine safety (adjusted odds ratio (aOR), 1.66; 95% CI, 1.18–2.35), vaccine effectiveness (aOR, 1.83; 95% CI, 1.32–2.53), and disease susceptibility (aOR, 2.08; 95% CI, 1.49–2.90) and were more likely to have a child with a nonmedical exemption (aOR 3.53, 95% CI, 2.61–4.76). These findings provide context to interpret recent vaccine information seeking research.
Objective The BED assay was developed to estimate the proportion of recent HIV infections in a population. We used the BED assay as a proxy for acute infection to quantify the associated risk of mother-to-child-transmission (MTCT) during pregnancy and delivery.
Design A total of 3773 HIV-1 sero-positive women were tested within 96 h of delivery using the BED assay, and CD4 cell count measurements were taken. Mothers were classified according to their likelihood of having recently seroconverted.
Methods The risk of MTCT in utero and intra-partum was assessed comparing different groups defined by BED and CD4 cell count, adjusting for background factors using multinomial logistic models.
Results Compared with women with BED ≥ 0.8/CD4 ≥ 350 (typical of HIV-1 chronic patients) there was insufficient evidence to conclude that women presenting with BED < 0.8/CD4 ≥ 350 (typical of recent infections) were more likely to transmit in utero [adjusted odds ratio (aOR) = 1.37, 96% confidence interval (CI) 0.90–2.08, P = 0.14], whereas women with BED < 0.8/CD4 200–349 (possibly recently infected patients) had a 2.57 (95% CI 1.39–4.77, P-value < 0.01) odds of transmitting in utero. Women who had BED < 0.8/CD4 < 200 were most likely to transmit in utero (aOR 3.73, 95% CI 1.27–10.96, P = 0.02). BED and CD4 cell count were not predictive of intra-partum infections.
Conclusions These data provide evidence that in utero transmission of HIV might be higher among women who seroconvert during pregnancy.
BED; CD4; in utero; intra-partum; seroconversion; HIV
Treatment of latent tuberculosis in patients infected with the human immunodeficiency virus (HIV) is efficacious, but few patients around the world receive such treatment. We evaluated three new regimens for latent tuberculosis that may be more potent and durable than standard isoniazid treatment.
We randomly assigned South African adults with HIV infection and a positive tuberculin skin test who were not taking antiretroviral therapy to receive rifapentine (900 mg) plus isoniazid (900 mg) weekly for 12 weeks, rifampin (600 mg) plus isoniazid (900 mg) twice weekly for 12 weeks, isoniazid (300 mg) daily for up to 6 years (continuous isoniazid), or isoniazid (300 mg) daily for 6 months (control group). The primary end point was tuberculosis-free survival.
The 1148 patients had a median age of 30 years and a median CD4 cell count of 484 per cubic millimeter. Incidence rates of active tuberculosis or death were 3.1 per 100 person-years in the rifapentine–isoniazid group, 2.9 per 100 person-years in the rifampin–isoniazid group, and 2.7 per 100 person-years in the continuous-isoniazid group, as compared with 3.6 per 100 person-years in the control group (P>0.05 for all comparisons). Serious adverse reactions were more common in the continuous-isoniazid group (18.4 per 100 person-years) than in the other treatment groups (8.7 to 15.4 per 100 person-years). Two of 58 isolates of Mycobacterium tuberculosis (3.4%) were found to have multidrug resistance.
On the basis of the expected rates of tuberculosis in this population of HIV-infected adults, all secondary prophylactic regimens were effective. Neither a 3-month course of intermittent rifapentine or rifampin with isoniazid nor continuous isoniazid was superior to 6 months of isoniazid.
Background Previously, HIV epidemic models have used a double Weibull curve to represent high initial and late mortality of HIV-infected children, without distinguishing timing of infection (peri- or post-natally). With more data on timing of infection, which may be associated with disease progression, a separate representation of children infected early and late was proposed.
Methods Paediatric survival post-HIV infection without anti-retroviral treatment was calculated using pooled data from 12 studies with known timing of HIV infection. Children were grouped into perinatally or post-natally infected. Net mortality was calculated using cause-deleted life tables to give survival as if HIV was the only competing cause of death. To extend the curve beyond the available data, children surviving beyond 2.5 years post infection were assumed to have the same survival as young adults. Double Weibull curves were fitted to both extended survival curves to represent survival of children infected perinatally or through breastfeeding.
Results Those children infected perinatally had a much higher risk of dying than those infected through breastfeeding, even allowing for background mortality. The final-fitted double Weibull curves gave 75% survival at 5 months after infection for perinatally infected, and 1.1 years for post-natally infected children. An estimated 25% of the early infected children would still be alive at 10.6 years compared with 16.9 years for those infected through breastfeeding.
Conclusions The increase in available data has enabled separation of child mortality patterns by timing of infection allowing improvement and more flexibility in modelling of paediatric HIV infection and survival.
HIV; survival; paediatric
Assumptions about survival of HIV-infected children in Africa without antiretroviral therapy need to be updated to inform ongoing UNAIDS modelling of paediatric HIV epidemics among children. Improved estimates of infant survival by timing of HIV-infection (perinatally or postnatally) are thus needed.
A pooled analysis was conducted of individual data of all available intervention cohorts and randomized trials on prevention of HIV mother-to-child transmission in Africa. Studies were right-censored at the time of infant antiretroviral initiation. Overall mortality rate per 1000 child-years of follow-up was calculated by selected maternal and infant characteristics. The Kaplan-Meier method was used to estimate survival curves by child's HIV infection status and timing of HIV infection. Individual data from 12 studies were pooled, with 12,112 children of HIV-infected women. Mortality rates per 1,000 child-years follow-up were 39.3 and 381.6 for HIV-uninfected and infected children respectively. One year after acquisition of HIV infection, an estimated 26% postnatally and 52% perinatally infected children would have died; and 4% uninfected children by age 1 year. Mortality was independently associated with maternal death (adjusted hazard ratio 2.2, 95%CI 1.6–3.0), maternal CD4<350 cells/ml (1.4, 1.1–1.7), postnatal (3.1, 2.1–4.1) or peri-partum HIV-infection (12.4, 10.1–15.3).
These results update previous work and inform future UNAIDS modelling by providing survival estimates for HIV-infected untreated African children by timing of infection. We highlight the urgent need for the prevention of peri-partum and postnatal transmission and timely assessment of HIV infection in infants to initiate antiretroviral care and support for HIV-infected children.
Non-HIV-related causes of death have been increasing after the introduction of highly active antiretroviral therapy. Underlying and contributing causes of death were assessed in respect to the presence/absence of HIV/AIDS among HIV-infected/AIDS patients in Rio de Janeiro, Brazil. Demographic variables (age, gender, ethnicity, and schooling) and CD4 cell counts closest to death were assessed through logistic regression models comparing those who did not have with those who had HIV/AIDS mentioned on the death certificate. The linkage with the two cohorts identified 1249 records, of which 370 (29.6%) did not have HIV/AIDS listed on any field of the death certificate [77 (20.8%) attributed to undefined and 72 (19.5%) to external causes]. After excluding external causes, 25.3% still did not have HIV/AIDS listed on the death certificate. Multiple logistic regression analysis showed that age >40 years (OR = 2.09; 95%CI = 1.49–2.93; p < 0.001) and CD4 cell count closest to the date of death (OR = 1.15; 95% CI = 1.07–1.23; p < 0.001 for 100 cell increase) were associated with an increased probability of not having HIV/AIDS mentioned on the death certificate, when external causes were excluded. Mortality among HIV-infected individuals is underreported in the Rio de Janeiro Mortality Registry, particularly among older individuals and those with higher CD4 counts. Physicians should be aware of the changing patterns of mortality among HIV individuals, and public health officials should regularly perform linkages between all-cause mortality and available HIV-infected patients databases, such as AIDS registries and large cohort studies.
Human papillomavirus (HPV) infection is a common sexually transmitted disease of growing public health importance, and over 40 genotypes have been identified in genital infections. Current HPV cohort studies often follow participants at pre-determined visits, such as every 6-months, and data generated from such epidemiology studies can be described as clustered longitudinal binary data where correlation arises in two ways: the directionless clustering due to the multiple genotypes tested within an individual, and the temporal correlation among the repeated measurements on the same genotype along time. Current analyses for identification of risk factors associated with HPV incidence and persistence often either do not fully utilize information in the dataset or ignore the correlation between the multiple genotypes. Given the scientific definition of incidence and persistence, conditional probability modeling provides us a natural mathematical tool. We thus present a semi-parametric regression model for such data where full specification of the joint multivariate binary distribution is avoided by using conditioning argument to handle the temporal correlation and GEE to account for the correlation between the multiple genotypes. The model is applied to the HPV data from the Rakai male circumcision (MC) trial to evaluate the as-treated efficacy of MC and also identify modifiable risk factors for incidence and persistence of oncogenic HPV types in men. A simulation study is performed to provide empirical information on the number of individuals that is needed for satisfactory power and estimation accuracy of the association parameter estimates in future studies.
clustered longitudinal binary data; conditional probability; semi-parametric model; HPV incidence; HPV persistence
In Rakai, Uganda, HIV+ men were randomized to immediate (intervention) or delayed circumcision (controls). Penile swabs were assayed for high risk human papillomavirus (HR-HPV) by Roche HPV Linear Array at enrollment and 24 months (intervention n=103, control n=107). Rate ratios (RR) of HR-HPV were estimated by Poisson regression. At 24 months, HR-HPV prevalence was intervention 55.3% and control 71.7% (RR=0.77, 95%CI 0.62–0.97). Multiple HR-HPV infections were intervention 22.4% and controls 42.5% (RR=0.53, 95%CI 0.33–0.83). New HR-HPV genotypes were acquired by 42.0% of intervention and 57.0% of control arm men (RR=0.74, 95%CI 0.54–1.01, p=0.06). Multiple new HR-HPV genotypes were acquired by 9.9% intervention and 24.7% control arm men (RR = 0.40, 95%CI 0.19–0.84, p = 0.01). Circumcision did not affect the acquisition of single HR-HPV infections (RR=1.00, 95%CI 0.65–1.53) or clearance of HR-HPV (RR=1.09, 95%CI 0.94–1.27). Circumcision of HIV+ men reduced the prevalence and incidence of multiple HR-HPV infections.
Uncircumcised HIV-negative men aged 15-49 years were randomized to immediate circumcision (n=441) or delayed circumcision (n=399). HPV was detected by Roche HPV Linear Array at enrollment, 6, 12 and 24 months. Incident HR-HPV was estimated in men who acquired a new HR-HPV genotype. HR-HPV clearance was determined in men with prior genotype-specific HR-HPV infections. Rate ratios (RR) and 95% confidence intervals (95%CI) of HR-HPV acquisition were estimated by Poisson multiple regression
Enrollment characteristics were comparable between groups. HR-HPV incidence was 19.7/100 py in the intervention (70/355.8 py) and 29.4/100 py (125/424.8 py) in the control arm (RR=0.67, 95%CI 0.51-0.89, p = 0.006.) The incidence of multiple HR-HPV infections was 6.7/100 py in the intervention and 14.8/100 py in control arm (RR = 0.45, 95%CI 0.28-0.73), but there was no significant effect on single infections (RR=0.89, 95%CI 0.60-1.30). HR-HPV incidence was lower in the intervention arm for all genotypes and demographic/behavioral subgroups. The clearance of pre-existing HR-HPV infections was 215.8/100py (205/95 py) in intervention and 159.1/100py (255/160.25 py) in control arm men (adjRR=1.39, 95%CI 1.17-1.64).
Male circumcision reduces the incidence of multiple HR-HPV infections and increases clearance of HR-HPV infections in HIV-uninfected men.
The trial was registered with ClinicalTrials.gov numbers NCT00425984
Few studies have examined the influence of individual-, household-, and community-scale risk factors on carriage of resistant commensal bacteria. We determined children's medical, agricultural, and environmental exposures by household, pharmacy, and health facility surveys and Escherichia coli cultures of children, mothers' hands, household animals, and market chickens in Peru. Among 522 children with a positive stool culture, by log-binomial regression, using “any antibiotic” and 1–14 (versus 0) sulfa doses in the past 3 months increased children's risk, respectively, for ampicillin- and sulfamethoxazole-resistant E. coli carriage (P = 0.01–0.02). Each household member taking “any antibiotic” increased children's risk for sulfamethoxazole- and multidrug-resistant E. coli carriage (P < 0.0001). Residence in a zone where a larger proportion of households served home-raised chicken (as contrasted with intensively antibiotic-raised market chicken) protected against carrying E. coli resistant to all drugs (P = 0.0004–0.04). Environmental contamination with drug-resistant bacteria appeared to significantly contribute to children's carriage of antibiotic-resistant E. coli.
CD4 counts increase during the postpartum period and may not correctly identify HAART-eligible HIV-positive women. HAART eligibility when defined by two CD4 cutoffs (<200 and <350 cells/μl) measured at two time points (within 96 h of delivery and 6 weeks) in postpartum HIV-positive women was compared. Among HIV-positive women who had CD4 at delivery and 6 weeks (n = 423), time to Stage 3 or 4 opportunistic infection or death was compared using Cox regression between three groups of women: (1) CD4 <200 cells/μl at delivery and 6 weeks, (2) CD4 <200 cells/μl at delivery but ≥200 cells/μl at 6 weeks, and (3) CD4 ≥200 cells/μl at delivery and at 6 weeks. The analysis was repeated using the CD4 <350 cells/μl cut-off. CD4 counts increased by a median (IQR) of 70 (1–178) cells/μl between delivery and 6 weeks and decreased thereafter to approximately delivery levels at 12 months. Only 60% and 61% who had CD4 <200 cells/μl and CD4 <350 cells/μl, respectively, at delivery also had those levels at 6 weeks. Among those with CD4 <350 cells/μl at both delivery and 6 weeks, the risk of death or Stage 3 or 4 disease was 5.27 (95% CI 1.85–14.96) times higher than those with CD4 <350 at delivery but ≥350 cells/μl at 6 weeks. The use of CD4 counts immediately postpartum to define HAART eligibility may lead to substantial misclassification.
The TB/HIV in Rio (THRio) study was launched in September 2005 to assess the impact of integrated tuberculosis (TB) and HIV treatment strategies in 29 HIV clinics in Rio de Janeiro, Brazil.
THRio is a cluster-randomized trial (CRT) to determine whether routine screening for and treatment of latent TB in HIV clinic patients with access to antiretroviral therapy will reduce TB incidence at the clinic level. THRio is part of the Consortium to Respond Effectively to AIDS/TB Epidemic that is implementing research studies to assess the impact of bold, new public health paradigms for controlling the AIDS/TB epidemic.
Twenty-nine public primary HIV clinics were randomly assigned a date to begin implementing TB screening procedures and provision of isoniazid preventive therapy (IPT) for TB/HIV coinfected patients. Final analysis of the CRT is expected in 2011.
Starting at date of tuberculin skin test (TST)/IPT implementation at each clinic through August 2010, 1670 HIV-infected patients initiated IPT, of which 215 are still receiving treatment. Of the remaining 1455 patients, 1230 (85%) completed therapy and only 20 (1.2%) patients initiating IPT reported adverse reactions leading to discontinuation of therapy. IPT completion was higher among HIV-infected patients receiving HAART (87%) than those not yet receiving HAART (79%, P < 0.01). Times to TST and IPT have markedly decreased postintervention, but remain considerably long. The richness of the THRio database has resulted in several analyses of this expansive cohort of HIV-infected patients that are reviewed here.
The national implementation of TST and IPT for HIV-positive patients in Brazil has been invigorated partly due to THRio’s baseline results. Expanded use of IPT in HIV patients in Rio de Janeiro is achievable with high adherence and low adverse events, although this effort requires a package of activities including training, advocacy and reorganization of services.
adherence; HIV; implementation; isoniazid preventive therapy; tuberculosis
Tuberculosis is a common complication and leading cause of death in HIV infection. Antiretroviral therapy (ART) lowers the risk of tuberculosis, but may not be sufficient to control HIV-related tuberculosis. Isoniazid preventive therapy (IPT) reduces tuberculosis incidence significantly, but is not widely used.
We analysed tuberculosis incidence in 11 026 HIV-infected patients receiving medical care at 29 public clinics in Rio de Janeiro, Brazil, between 1 September 2003 and 1 September 2005. Data were collected through a retrospective medical record review. We determined rates of tuberculosis in patients who received neither ART nor IPT, only ART, only IPT, or both ART and IPT.
The overall tuberculosis incidence was 2.28 cases/100 person-years (PY) [95% confidence interval (CI) 2.06–2.52]. Among patients who received neither ART nor IPT, incidence was 4.01/100 PY. Patients who received ART had an incidence of 1.90/100 PY (95% CI 1.66–2.17) and those treated with IPT had a rate of 1.27/100 PY (95% CI 0.41–2.95). The incidence among patients who received ART and IPT was 0.80/100 PY (95% CI 0.38–1.47). Multivariate Cox proportional hazards modeling revealed a 76% reduction in tuberculosis risk among patients receiving both ART and IPT (adjusted relative hazard 0.24; P < 0.001) after adjusting for age, previous tuberculosis diagnosis, and CD4 cell counts at baseline.
The use of both IPT and ART in HIV-infected patients is associated with significantly reduced tuberculosis incidence. In conjunction with expanded access to ART, the wider use of IPT in patients with HIV will improve tuberculosis control in high burden areas.
antiretroviral therapy; Brazil; HIV; isoniazid; tuberculosis
Objectives To estimate the rates and timing of mother to infant transmission of HIV associated with breast feeding in mothers who seroconvert postnatally, and their breast milk and plasma HIV loads during and following seroconversion, compared with women who tested HIV positive at delivery.
Design Prospective cohort study.
Setting Urban Zimbabwe.
Participants 14 110 women and infants enrolled in the Zimbabwe Vitamin A for Mothers and Babies (ZVITAMBO) trial (1997-2001).
Main outcome measures Mother to child transmission of HIV, and breast milk and maternal plasma HIV load during the postpartum period.
Results Among mothers who tested HIV positive at baseline and whose infant tested HIV negative with polymerase chain reaction (PCR) at six weeks (n=2870), breastfeeding associated transmission was responsible for an average of 8.96 infant infections per 100 child years of breast feeding (95% CI 7.92 to 10.14) and varied little over the breastfeeding period. Breastfeeding associated transmission for mothers who seroconverted postnatally (n=334) averaged 34.56 infant infections per 100 child years (95% CI 26.60 to 44.91) during the first nine months after maternal infection, declined to 9.50 (95% CI 3.07 to 29.47) during the next three months, and was zero thereafter. Among women who seroconverted postnatally and in whom the precise timing of infection was known (≤90 days between last negative and first positive test; n=51), 62% (8/13) of transmissions occurred in the first three months after maternal infection and breastfeeding associated transmission was 4.6 times higher than in mothers who tested HIV positive at baseline and whose infant tested HIV negative with PCR at six weeks. Median plasma HIV concentration in all mothers who seroconverted postnatally declined from 5.0 log10 copies/mL at the last negative enzyme linked immunosorbent assay (ELISA) to 4.1 log10 copies/mL at 9-12 months after infection. Breast milk HIV load in this group was 4.3 log10 copies/mL 0-30 days after infection, but rapidly declined to 2.0 log10 copies/mL and <1.5 log10 copies/mL by 31-90 days and more than 90 days, respectively. Among women whose plasma sample collected soon after delivery tested negative for HIV with ELISA but positive with PCR (n=17), 75% of their infants were infected or had died by 12 months. An estimated 18.6% to 20.4% of all breastfeeding associated transmission observed in the ZVITAMBO trial occurred among mothers who seroconverted postnatally.
Conclusions Breastfeeding associated transmission is high during primary maternal HIV infection and is mirrored by a high but transient peak in breast milk HIV load. Around two thirds of breastfeeding associated transmission by women who seroconvert postnatally may occur while the mother is still in the “window period” of an antibody based test, when she would test HIV negative using one of these tests.
Trial registration Clinical trials.gov NCT00198718.
A randomized trial of male circumcision (MC) was conducted among HIV-infected males to test the hypothesis that MC would reduce HIV transmission to female sexual partners.
This randomized, unblinded trial, conducted in Rakai District, Uganda, enrolled 922 uncircumcised, HIV-infected asymptomatic men aged 15–49 with CD4 counts ≥350. Men were randomly assigned to immediate circumcision (intervention) or circumcision delayed for 24 months (control). Concurrently enrolled HIV-negative female partners were followed up at 6, 12 and 24 months, to assess HIV acquisition by male MC assignment (primary outcome). An intention-to-treat analysis assessed women’s HIV acquisition using survival analysis and Cox proportional hazards modeling. The trial was registered in the Clinical Trials.gov Protocol Registration System (NCT00124878).
The trial was terminated for futility. Ninety three concurrently enrolled female partners of intervention arm men and 70 partners of control arm men provided follow up data. Cumulative probabilities of female HIV infection at 24 months were 21.7% (95% CI 12.7–33.4) in the intervention arm and 13.4% (95% CI 6.7–25.8) in the control arm (adjusted hazard ratio= 1.49, 95% CI 0.62–3.57, p = 0.368). At 6 months, intervention arm male-to-female transmission in couples who resumed intercourse ≥5 days prior to certified surgical wound healing was 27.8% (5/18), compared to 9.5% in couples who abstained longer post-surgically (6/63, p = 0.06) and 7.9% in control arm couples (5/63, p = 0.04)
Circumcision of HIV-infected men did not reduce HIV transmission to female partners over 24 months, and transmission risk may be increased with early post-surgical resumption of intercourse. Longer-term effects could not be assessed. Post surgical sexual abstinence and subsequent consistent condom are essential for HIV prevention.
Male circumcision; randomized trial; HIV-infected men; female HIV acquisition; Uganda
In Ethiopia, evidence is lacking about maternal care-taking and environmental risk factors that contribute to acute diarrhoea and the case management of diarrhoea. The aim of this study was to identify the risk factors and to understand the management of acute diarrhoea. A pretested structured questionnaire was used for interviewing mothers of 440 children in a prospective, matched, case-control study at the University of Gondar Referral and Teaching Hospital in Gondar, Ethiopia. Results of multivariate analysis demonstrated that children who were breastfed and not completely weaned and mothers who were farmers were protective factors; risk factors for diarrhoea included sharing drinking-water and introducing supplemental foods. Children presented with acute diarrhoea for 3.9 days with 4.3 stools per day. Mothers usually did not increase breastmilk and other fluids during diarrhoea episodes and generally did not take children with diarrhoea to traditional healers. Incorporating messages about the prevention and treatment of acute diarrhoea into child-health interventions will help reduce morbidity and mortality associated with this disease.
Case-control studies; Case management; Diarrhoea, Acute; Oral rehydration solution; Prospective studies; Risk factors; Ethiopia
In epidemiological studies, subjects are often followed for a period during which study outcomes are measured at selected time points, such as by diagnostic testing performed on biological samples collected at each visit. Although test results may indicate the presence or absence of a disease or condition, they cannot provide information on when exactly it occurred. Such study designs generate arbitrarily censored time-to-event data, which can include left, interval and right censoring. Adding to this complexity, the data may be clustered such that observations within the same cluster are not independent, such as time to recovery of an infectious disease of family or community members. This data structure is observed when evaluating circumcision's effect on clearance of penile high risk human papillomavirus (HR-HPV) infections using data collected from the male circumcision(MC) trial conducted in Rakai, Uganda, where the multiple infections within individual and HPV testings performed at trial follow-up visits gave rise to the clustered data with arbitrary censoring.
We describe the use of parametric proportional hazards frailty models and accelerated failure time frailty models to examine the relationship between explanatory variables and the survival outcomes that are subject to arbitrary censoring, while accounting for the correlation within clusters. Standard software such as SAS can be used for parameter estimation.
Circumcision's effect on HPV infection was a secondary end point in the Rakai MC trial, and HPV genotyping was conducted for penile samples of a subset of trial participants collected at enrollment, 6, 12 and 24-month follow up visits. At enrollment, 36.7% intervention arm men (immediate circumcision) and 36.6% control arm men (delayed circumcision at 2 years) were infected with HR-HPV, with the number of infections per man being 1-5. The proposed models were used to examine whether MC facilitated clearance of the prevalent infections. Results show that clearance of multiple infections within each man is highly correlated, and clearance was 60% faster if a man was circumcised.
Parametric frailty models provide viable ways to study the relationship between exposure variables and clustered survival outcome that is subject to arbitrary censoring, as is often observed in HPV epidemiology studies.
The risk of recurrent tuberculosis may increase in HIV-infected patients due to exogenous re-infection. We measured the frequency of and determined risk factors for recurrent tuberculosis in a cohort of HIV-infected patients in Rio de Janeiro, Brazil.
Data were abstracted from medical records of HIV-infected patients attending 29 HIV clinics between 1998 and 2007. Patients analyzed were those who had no tuberculosis history prior to their first HIV clinic visit and who had at least one episode of tuberculosis after entry. Incidence rate ratios compared incidence rates between risk groups and Cox proportional hazards regression models evaluated unadjusted and adjusted associations.
Among 1,080 HIV-infected patients with tuberculosis, 96(8.9%) developed a recurrent diagnosis. The median time between diagnoses was 2.4 years. Fewer patients with recurrent tuberculosis had completed their initial 6-month course of tuberculosis treatment compared to patients without recurrence (78% versus 86%;p=0.02). For patients who completed therapy, the incidence rate of recurrence was 2.5/100 person years (PYs) versus 9.0/100PYs for non-completers (Incidence Rate Ratio (IRR)=3.60(95%CI:1.92-6.32). In multivariate modeling, initial tuberculosis treatment completion, receipt of antiretroviral therapy (ART) and CD4 cell count >200/mm3 any time after the initial diagnosis were associated with a significantly decreased hazard of recurrence.
Tuberculosis recurrence rates were high in this HIV-infected population. Completion of initial tuberculosis therapy, use of ART and increases in CD4 cell counts were associated with lower recurrence rates. Use of secondary preventive therapy might be warranted to reduce the burden of tuberculosis in patients with HIV infection.
HIV; tuberculosis; recurrence; antiretroviral therapy; Brazil
Mortality in hospitalized, febrile patients in Sub-Saharan Africa is high due to HIV-infected, severely immunosuppressed patients with opportunistic co-infection, particularly disseminated tuberculosis (TB) and cryptococcal disease. We sought to determine if a positive lateral flow assay (LFA) result for urine lipoarabinomannan (LAM) and cryptococcal antigenuria was associated with mortality.
351 hospitalized, HIV-positive adults with symptoms consistent with TB and who were able to provide both urine and sputum specimens were prospectively enrolled at Mulago National Referral Hospital in Uganda as part of a prospective accuracy evaluation of the lateral flow Determine TB LAM test. Stored frozen urine was retrospectively tested for cryptococcal antigen (CRAG) using the LFA. We fitted a multinomial logistic regression model to analyze factors associated with death within 2 months after initial presentation.
The median CD4 of the participants was 57 (IQR: 14–179) cells/µl and 41% (145) were microbiologically confirmed TB cases. LAM LFA was positive in 38% (134), 7% (25) were CRAG positive, and 43% (151) were positive for either test in urine. Overall, 21% (75) died within the first 2 months, and a total of 32% (114) were confirmed dead by 6 months. At 2 months, 30% of LAM or CRAG positive patients were confirmed dead compared to 15.0% of those who were negative. In an adjusted model, LAM or CRAG positive results were associated with an increased risk of death (RRR 2.29, 95% CI: 1.29, 4.05; P = 0.005).
In hospitalized HIV-infected patients, LAM or CRAG LFA positivity was associated with subsequent death within 2 months. Further studies are warranted to examine the impact of POC diagnostic ‘test and treat’ approach on patient-centered outcomes.
Loss to follow-up is a major source of bias in cohorts of patients with human immunodeficiency virus (HIV) and could lead to underestimation of mortality. The authors developed a hierarchical deterministic linkage algorithm to be used primarily with cohorts of HIV-infected persons to recover vital status information for patients lost to follow-up. Data from patients known to be deceased in 2 cohorts in Rio de Janeiro, Brazil, and data from the Rio de Janeiro State mortality database for 1999–2006 were used to validate the algorithm. A fully automated procedure yielded a sensitivity of 92.9% and specificity of 100% when no information was missing. When the automated procedure was combined with clerical review, in a scenario of 5% death prevalence and 20% missing mothers’ names, sensitivity reached 96.5% and specificity 100%. In a practical application, the algorithm significantly increased death rates and decreased the rate of loss to follow-up in the cohorts. The finding that 23.9% of matched records did not give HIV or acquired immunodeficiency syndrome as the cause of death reinforces the need to search all-cause mortality databases and alerts for possible underestimation of death rates. These results indicate that the algorithm is accurate enough to recover vital status information on patients lost to follow-up in cohort studies.
cohort studies; data collection; HIV; medical record linkage; mortality; software validation