Publications

BACKGROUND

Achieving and maintaining high levels of medication adherence are required to achieve the full benefits of antiretroviral therapy (ART), yet suboptimal adherence among children is common in both developed and developing countries.

OBJECTIVES

To conduct a systematic review of the literature of evaluations of interventions for improving paediatric ART adherence.

SEARCH METHODS

We created a comprehensive search strategy in order to identify all studies relevant to this topic. In July 2010, we searched the following electronic databases: EMBASE, MEDLINE, PsycINFO, Cochrane Central Register of Controlled Trials (CENTRAL), CINAHL, LILACS, Web of Science, Web of Social Science, NLM Gateway (supplemented by a manual search of the most recent abstracts not included in the Gateway database). We searched abstracts from the International AIDS Conference from 2002 to 2010, the International AIDS Society Conference on Pathogenesis, Treatment and Prevention from 2003 to 2009, and from the Conference on Retroviruses and Opportunistic Infections from 1997 to 2010. We used search strategies determined by the Cochrane Review Group on HIV/AIDS. We also contacted researchers who work in this field and checked reference lists of related systematic reviews and of all included studies.

SELECTION CRITERIA

Randomised and non-randomised controlled trials of interventions to improve adherence to ART among children and adolescents (age ≤18 years) were included. Studies had to report adherence to ART as an outcome.

DATA COLLECTION AND ANALYSIS

After one author performed an initial screening to exclude citations that did not meet the inclusion criteria, two authors did a second screening of those citations that likely met the criteria. For all articles that passed the second screening, full articles were pulled in order to make a final determination. Two authors then extracted data and graded methodological quality independently. Differences were resolved through discussion.

MAIN RESULTS

Four studies met the inclusion criteria. No single intervention was evaluated by more than one trial. Two studies were conducted in low-income countries. Two studies were randomised controlled trials (RCT), and two were non-randomised trials. An RCT of a home-based nursing programme showed a positive effect of the intervention on knowledge and medication refills (p=.002), but no effect on CD4 count and viral load. A second RCT of caregiver medication diaries showed that the intervention group had fewer participants reporting no missed doses compared to the control group (85% vs. 92%, respectively), although this difference was not statistically significant (p=.08). The intervention had no effect on CD4 percentage or viral load. A non-randomised trial of peer support group therapy for adolescents demonstrated no change in self-reported adherence, yet the percentage of participants with suppressed viral load increased from 30% to 80% (p=.06). The second non-randomised trial found that the percentage of children achieving >80% adherence was no different between children on a lopinavir-ritonavir (LPV/r) regimen compared to children on a non-nucleoside reverse transcriptase regimen (p=.781). However, the proportion of children achieving virological suppression was significantly greater for children on the LPV/r regimen than for children on the NNRTI-containing regimen (p=.002).

AUTHORS' CONCLUSIONS

A home-based nursing intervention has the potential to improve ART adherence, but more evidence is needed. Medication diaries do not appear to have an effect on adherence or disease outcomes. Two interventions, an LPV/r-containing regimen and peer support therapy for adolescents, did not demonstrate improvements in adherence, yet demonstrated greater viral load suppression compared to control groups, suggesting a different mechanism for improved health outcomes. Well-designed evaluations of interventions to improve paediatric adherence to ART are needed.

Given the increased prevalence of non-daily smoking and changes in smoking patterns, particularly among young adults, we examined correlates of smoking level, specifically motives for smoking, and readiness to quit smoking among 2682 college undergraduates who completed an online survey. Overall, 64.7% (n = 1736) were non-smokers, 11.6% (n = 312) smoked 1-5 days, 10.5% (n = 281) smoked 6-29 days and 13.2% (n = 353) were daily smokers. Ordinal regression analyses modeling smoking level indicated that correlates of higher smoking level included having more friends who smoke (β = 0.63, 95% CI 0.57-0.69) and more frequent other tobacco use (β = 0.04, 95% CI 0.02-0.05), drinking (β = 0.04, 95% CI 0.02-0.07) and binge drinking (β = 0.09, 95% CI 0.06-0.13). Bivariate analyses indicated that daily smokers (versus the subgroups of non-daily smokers) were less likely to smoke for social reasons but more likely to smoke for self-confidence, boredom, and affect regulation. Controlling for sociodemographics, correlates of readiness to quit among current smokers included fewer friends who smoke (P = 0.002), less frequent binge drinking (P = 0.03), being a social smoker (P < 0.001), smoking less for self-confidence (P = 0.04), smoking more for boredom (P = 0.03) and less frequent smoking (P = 0.001). Specific motives for smoking and potential barriers to cessation particularly may be relevant to different groups of college student smokers.

Food insecurity is a considerable challenge in sub-Saharan Africa, disproportionately affecting persons living with HIV/AIDS. This study investigates the lived experience, determinants, and consequences of food insecurity and hunger among individuals living with HIV/AIDS on the shore of Lake Victoria in Suba District, Kenya. Parallel mixed methods included semi-structured interviews and administration of the Household Food Insecurity Access Scale among a systematic sample of 67 persons living with HIV/AIDS (49 of whom were receiving antiretroviral therapy [ART]). All respondents were either severely (79.1%) or moderately (20.9%) food insecure; no respondents were mildly food insecure or food secure. Qualitative data and simple and multiple linear regression models indicated that significant determinants of food insecurity include increased age, a greater number of children, and not being married. A number of themes related to food insecurity and ART emerged, including: (1) an increase in hunger or appetite since initiating ART; (2) exacerbation of ART-related side effects; and (3) non-adherence to ART due to hunger, food insecurity, or agricultural work responsibilities. HIV interventions should address food insecurity and hunger, particularly among at-risk populations, to promote ART adherence and better health outcomes.

BACKGROUND

Reporting of quality indicators (QIs) in Veterans Health Administration Medical Centers is complicated by estimation error caused by small numbers of eligible patients per facility. We applied multilevel modeling and empirical Bayes (EB) estimation in addressing this issue in performance reporting of stroke care quality in the Medical Centers.

METHODS AND RESULTS

We studied a retrospective cohort of 3812 veterans admitted to 106 Medical Centers with ischemic stroke during fiscal year 2007. The median number of study patients per facility was 34 (range, 12-105). Inpatient stroke care quality was measured with 13 evidence-based QIs. Eligible patients could either pass or fail each indicator. Multilevel modeling of a patient's pass/fail on individual QIs was used to produce facility-level EB-estimated QI pass rates and confidence intervals. The EB estimation reduced interfacility variation in QI rates. Small facilities and those with exceptionally high or low rates were most affected. We recommended 8 of the 13 QIs for performance reporting: dysphagia screening, National Institutes of Health Stroke Scale documentation, early ambulation, fall risk assessment, pressure ulcer risk assessment, Functional Independence Measure documentation, lipid management, and deep vein thrombosis prophylaxis. These QIs displayed sufficient variation across facilities, had room for improvement, and identified sites with performance that was significantly above or below the population average. The remaining 5 QIs were not recommended because of too few eligible patients or high pass rates with little variation.

CONCLUSIONS

Considerations of statistical uncertainty should inform the choice of QIs and their application to performance reporting.