Modeling health impact of global health programs implemented by Population Services International.

BACKGROUND: Global health implementing organizations benefit most from health impact estimation models that isolate the individual effects of distributed products and services - a feature not typically found in intervention impact models, but which allow comparisons across interventions and intervention settings. Population Services International (PSI), a social marketing organization, has developed a set of impact models covering seven health program areas, which translate product/service distribution data into impact estimates. Each model's primary output is the number of disability-adjusted life-years (DALYs) averted by an intervention within a specific country and population context. This paper aims to describe the structure and inputs for two types of DALYs averted models, considering the benefits and limitations of this methodology. METHODS: PSI employs two modeling approaches for estimating health impact: a macro approach for most interventions and a micro approach for HIV, tuberculosis (TB), and behavior change communication (BCC) interventions. Within each intervention country context, the macro approach determines the coverage that one product/service unit provides a population in person-years, whereas the micro approach estimates an individual's risk of infection with and without the product/service unit. The models use these estimations to generate per unit DALYs averted coefficients for each intervention. When multiplied by program output data, these coefficients predict the total number of DALYs averted by an intervention in a country. RESULTS: Model outputs are presented by country for two examples: Water Chlorination DALYs Averted Model, a macro model, and the HIV Condom DALYs Averted Model for heterosexual transmission, a micro model. Health impact estimates measured in DALYs averted for PSI interventions on a global level are also presented. CONCLUSIONS: The DALYs averted models offer implementing organizations practical measurement solutions for understanding an intervention's contribution to improving health. These models calculate health impact estimates that reflect the scale and diversity of program operations and intervention settings, and that enable comparisons across health areas and countries. Challenges remain in accounting for intervention synergies, attributing impact to a single organization, and sourcing and updating model inputs. Nevertheless, these models demonstrate how DALYs averted can be viably used by the global health community as a metric for predicting intervention impact using standard program output data.

Applying the disability-adjusted life year to track health impact of social franchise programs in low- and middle-income countries.

BACKGROUND: Developing effective methods for measuring the health impact of social franchising programs is vital for demonstrating the value of this innovative service delivery model, particularly given its rapid expansion worldwide. Currently, these programs define success through patient volume and number of outlets, widely acknowledged as poor reflections of true program impact. An existing metric, the disability-adjusted life years averted (DALYs averted), offers promise as a measure of projected impact. Country-specific and service-specific, DALYs averted enables impact comparisons between programs operating in different contexts. This study explores the use of DALYs averted as a social franchise performance metric. METHODS: Using data collected by the Social Franchising Compendia in 2010 and 2011, we compared franchise performance, analyzing by region and program area. Coefficients produced by Population Services International converted each franchise's service delivery data into DALYs averted. For the 32 networks with two years of data corresponding to these metrics, a paired t-test compared all metrics. Finally, to test data reporting quality, we compared services provided to patient volume. RESULTS: Social franchising programs grew considerably from 2010 to 2011, measured by services provided (215%), patient volume (31%), and impact (couple-years of protection (CYPs): 86% and DALYs averted: 519%), but not by the total number of outlets. Non-family planning services increased by 857%, with diversification centered in Asia and Africa. However, paired t-test comparisons showed no significant increase within the networks, whether categorized as family planning or non-family planning. The ratio of services provided to patient visits yielded considerable range, with one network reporting a ratio of 16,000:1. CONCLUSION: In theory, the DALYs averted metric is a more robust and comprehensive metric for social franchising than current program measures. As social franchising spreads beyond family planning, having a metric that captures the impact of a range of diverse services and allows comparisons will be increasingly important. However, standardizing reporting will be essential to make such comparisons useful. While not widespread, errors in self-reported data appear to have included social marketing distribution data in social franchising reporting, requiring clearer data collection and reporting guidelines. Differences noted above must be interpreted cautiously as a result.

Validity of verbal autopsy procedures for determining malaria deaths in different epidemiological settings in Uganda.

BACKGROUND: Verbal autopsy (VA) procedures can be used to estimate cause of death in settings with inadequate vital registries. However, the sensitivity of VA for determining malaria-specific mortality may be low, and may vary with transmission intensity. We assessed the diagnostic accuracy of VA procedures as compared to hospital medical records for determining cause of death in children under five in three different malaria transmission settings in Uganda, including Tororo (high), Kampala (medium), and Kisoro (low). METHODS AND FINDINGS: Caretakers of children who died in participating hospitals were interviewed using a standardized World Health Organization questionnaire. Medical records from the child's hospitalization were also reviewed. Causes of death based on the VA questionnaires and the medical records were assigned independently by physician reviewers and then compared. A total of 719 cases were included in the final analysis, 67 in Tororo, 600 in Kampala, and 52 in Kisoro. Malaria was classified as the underlying or contributory cause of death by review of medical records in 33 deaths in Tororo, 60 in Kampala, and 0 in Kisoro. The sensitivity of VA procedures for determining malaria deaths in Tororo was 61% (95% CI 44-78%) and 50% in Kampala (95% CI 37-63%). Specificity for determining malaria deaths in Tororo and Kampala was high (>88%), but positive predictive value varied widely, from 83% in Tororo to 34% in Kampala (difference 49%, 95% CI 31-67, p<0.001). The difference between the cause-specific mortality fraction for malaria as determined by VA procedures and medical records was -11% in Tororo, +5% in Kampala, and +14% in Kisoro. CONCLUSIONS: Our results suggest that these VA methods have an acceptable level of diagnostic accuracy for determining malaria deaths at the population level in high and medium transmission areas, but not in low transmission areas.

An assessment of Lot Quality Assurance Sampling to evaluate malaria outcome indicators: extending malaria indicator surveys.

BACKGROUND: Large investments and increased global prioritization of malaria prevention and treatment have resulted in greater emphasis on programme monitoring and evaluation (M&E) in many countries. Many countries currently use large multistage cluster sample surveys to monitor malaria outcome indicators on a regional and national level. However, these surveys often mask local-level variability important to programme management. Lot Quality Assurance Sampling (LQAS) has played a valuable role for local-level programme M&E. If incorporated into these larger surveys, it would provide a comprehensive M&E plan at little, if any, extra cost. METHODS: The Mozambique Ministry of Health conducted a Malaria Indicator Survey (MIS) in June and July 2007. We applied LQAS classification rules to the 345 sampled enumeration areas to demonstrate identifying high- and low-performing areas with respect to two malaria program indicators-'household possession of any bednet' and 'household possession of any insecticide-treated bednet (ITN)'. RESULTS: As shown by the MIS, no province in Mozambique achieved the 70% coverage target for household possession of bednets or ITNs. By applying LQAS classification rules to the data, we identify 266 of the 345 enumeration areas as having bednet coverage severely below the 70% target. An additional 73 were identified with low ITN coverage. CONCLUSIONS: This article demonstrates the feasibility of integrating LQAS into multistage cluster sampling surveys and using these results to support a comprehensive national, regional and local programme M&E system. Furthermore, in the recommendations we outlined how to integrate the Large Country-LQAS design into macro-surveys while still obtaining results available through current sampling practices.

The association between refractive cutoffs for spectacle provision and visual improvement among school-aged children in South Africa.

OBJECTIVES: To evaluate different refractive cutoffs for spectacle provision with regards to their impact on visual improvement and spectacle compliance. DESIGN: Prospective study of visual improvement and spectacle compliance. PARTICIPANTS: South African school children aged 6-19 years receiving free spectacles in a programme supported by Helen Keller International. METHODS: Refractive error, age, gender, urban versus rural residence, presenting and best-corrected vision were recorded for participants. Spectacle wear was observed directly at an unannounced follow-up examination 4-11 months after initial provision of spectacles. The association between five proposed refractive cutoff protocols and visual improvement and spectacle compliance were examined in separate multivariate models. MAIN OUTCOMES: Refractive cutoffs for spectacle distribution which would effectively identify children with improved vision, and those more likely to comply with spectacle wear. RESULTS: Among 8520 children screened, 810 (9.5%) received spectacles, of whom 636 (79%) were aged 10-14 years, 530 (65%) were girls, 324 (40%) had vision improvement > or = 3 lines, and 483 (60%) were examined 6.4+/-1.5 (range 4.6 to 10.9) months after spectacle dispensing. Among examined children, 149 (31%) were wearing or carrying their glasses. Children meeting cutoffs < or = -0.75 D of myopia, > or = +1.00 D of hyperopia and > or = +0.75 D of astigmatism had significantly greater improvement in vision than children failing to meet these criteria, when adjusting for age, gender and urban versus rural residence. None of the proposed refractive protocols discriminated between children wearing and not wearing spectacles. Presenting vision and improvement in vision were unassociated with subsequent spectacle wear, but girls (p < or = 0.0006 for all models) were more likely to be wearing glasses than were boys. CONCLUSIONS: To the best of our knowledge, this is the first suggested refractive cutoff for glasses dispensing validated with respect to key programme outcomes. The lack of association between spectacle retention and either refractive error or vision may have been due to the relatively modest degree of refractive error in this African population.

Viewpoint: evaluating the impact of malaria control efforts on mortality in sub-Saharan Africa.

OBJECTIVE: To describe an approach for evaluating the impact of malaria control efforts on malaria-associated mortality in sub-Saharan Africa, where disease-specific mortality trends usually cannot be measured directly and most malaria deaths occur among young children. METHODS: Methods for evaluating changes in malaria-associated mortality are examined; advantages and disadvantages are presented. RESULTS: All methods require a plausibility argument-i.e., an assumption that mortality reductions can be attributed to programmatic efforts if improvements are found in steps of the causal pathway between intervention scale-up and mortality trends. As different methods provide complementary information, they can be used together. We recommend following trends in the coverage of malaria control interventions, other factors influencing childhood mortality, malaria-associated morbidity (especially anaemia), and all-cause childhood mortality. This approach reflects decreases in malaria's direct and indirect mortality burden and can be examined in nearly all countries. Adding other information can strengthen the plausibility argument: trends in indicators of malaria transmission, information from demographic surveillance systems and sentinel sites where malaria diagnostics are systematically used, and verbal autopsies linked to representative household surveys. Health facility data on malaria deaths have well-recognized limitations; however, in specific circumstances, they could produce reliable trends. Model-based predictions can help describe changes in malaria-specific burden and assist with program management and advocacy. CONCLUSIONS: Despite challenges, efforts to reduce malaria-associated mortality in Africa can be evaluated with trends in malaria intervention coverage and all-cause childhood mortality. Where there are resources and interest, complementary data on malaria morbidity and malaria-specific mortality could be added.

Factors associated with spectacle-wear compliance in school-aged Mexican children.

PURPOSE: To study the prevalence and determinants of compliance with spectacle wear among school-age children in Oaxaca, Mexico, who were provided spectacles free of charge. METHODS: A cohort of 493 children aged 5 to 18 years chosen by random cluster sampling from primary and secondary schools in Oaxaca, Mexico, all of whom had received free spectacles through a local program, underwent unannounced, direct examination to determine compliance with spectacle wear within 18 months after initial provision of spectacles. Potential determinants of spectacle wear including age, gender, urban versus rural residence, presenting visual acuity, refractive error, and time since dispensing of the spectacles were examined in univariate and multivariate regression models. Children not currently wearing their spectacles were asked to select the reason from a list of possibilities, and reasons for noncompliance were analyzed within different demographic groups. RESULTS: Among this sample of children with a mean age of 10.4 +/- 2.6 years, the majority (74.5%) of whom were myopic (spherical equivalent [SE] < or = -0.50 D), 13.4% (66/493) were wearing their spectacles at the time of examination. An additional 34% (169/493) had the spectacles with them but were not wearing them. In regression models, the odds of spectacle wear were significantly higher among younger (OR = 1.19 per year of age; 95% CI, 1.05-1.33) rural (OR = 10.6; 95% CI, 5.3-21.0) children and those with myopia < or = -1.25 D (OR = 3.97; 95% CI, 1.98-7.94). The oldest children and children in urban-suburban areas were significantly more likely to list concerns about the appearance of the glasses or about being teased than were younger, rurally resident children. CONCLUSIONS: Compliance with spectacle wear may be very low, even when spectacles are provided free of charge, particularly among older, urban children, who have been shown in many populations to have the highest prevalence of myopia. As screening programs for refractive error become increasingly common throughout the world, new strategies are needed to improve compliance if program resources are to be maximized.

The value of electromyography of the lumbar paraspinal muscles in discriminating between chronic-low-back-pain sufferers and normal subjects.

The authors studied the surface electromyographic (EMG) spectrum of the paraspinal muscles of 350 subjects. They were classified by their history as normal (n=175), chronic low back pain (n=145), or past history (n=30). They pulled upwards on a floor-mounted load cell at two-thirds of their maximum voluntary contraction for 30 s, while the EMG was measured from the paraspinal muscles at the L4/L5 level. From the EMG signal the root-mean-square (RMS) was calculated. Power spectrum analysis allowed calculation of the median frequency slope, the initial median frequency (IMF), modal frequency, peak amplitude and spectral width at half peak amplitude (half-width). All of the variables of the chronic group were significantly different from the normal group, except the median frequency slope, RMS slope and mode. Half-width, age and maximum voluntary contraction were shown to be independent predictors of back pain classification. Half-width classified the subjects with a sensitivity of 0.65 and a specificity of 0.75.

Role of flies and provision of latrines in trachoma control: cluster-randomised controlled trial.

BACKGROUND: Eye-seeking flies have received much attention as possible trachoma vectors, but this remains unproved. We aimed to assess the role of eye-seeking flies as vectors of trachoma and to test provision of simple pit latrines, without additional health education, as a sustainable method of fly control. METHODS: In a community-based, cluster-randomised controlled trial, we recruited seven sets of three village clusters and randomly assigned them to either an intervention group that received regular insecticide spraying or provision of pit latrines (without additional health education) to each household, or to a control group with no intervention. Our primary outcomes were fly-eye contact and prevalence of active trachoma. Frequency of child fly-eye contact was monitored fortnightly. Whole communities were screened for clinical signs of trachoma at baseline and after 6 months. Analysis was per protocol. FINDINGS: Of 7080 people recruited, 6087 (86%) were screened at follow-up. Baseline community prevalence of active trachoma was 6%. The number of Musca sorbens flies caught from children's eyes was reduced by 88% (95% CI 64-100; p<0.0001) by insecticide spraying and by 30% (7-52; p=0.04) by latrine provision by comparison with controls. Analysis of age-standardised trachoma prevalence rates at the cluster level (n=14) showed that spraying was associated with a mean reduction in trachoma prevalence of 56% (19-93; p=0.01) and 30% with latrines (-81 to 22; p=0.210) by comparison with the mean rate change in the controls. INTERPRETATION: Fly control with insecticide is effective at reducing the number of flies caught from children's eyes and is associated with substantially lower trachoma prevalence compared with controls. Such a finding is consistent with flies being important vectors of trachoma. Since latrine provision without health education was associated with a significant reduction in fly-eye contact by M sorbens, studies of their effect when combined with other trachoma control measures are warranted.

Herpes simplex 2 risk among women in a polygynous setting in rural West Africa.

OBJECTIVES: To determine risk factors for herpes simplex 2 (HSV2) infection in women in a polygynous rural Gambian population. METHODS: Data from women who participated in a cross-sectional survey of reproductive health were matched to their own and, for women who had been or were married (ever-married), their spouses' data collected in a cross-sectional survey of fertility interests, including information on marital histories. RESULTS: Data were available on 150 never-married and 525 ever-married women. HSV2 prevalence was 16% amongst never-married women and 36% amongst ever-married women. For ever-married women, their own personal characteristics (age, ethnicity and genital cutting status) and events from their husbands' marriage history were important determinants of HSV2 infection. Women whose husbands married for the first time over age 35 were at greater risk than women whose husbands married by age 24 [odds ratio (OR) 2.72, 95% confidence interval (CI) 1.20-6.10]. Women whose husband reported interest in a new marriage were more likely to be HSV2 positive (OR 1.91, 95% CI 1.18-3.09). Women whose husbands were currently monogamous but had had previous marriages (OR 2.76, 95% CI 1.30-5.88) and women in currently polygynous marriages (OR 2.88, 95% CI 1.66-5.01) were three times as likely to be HSV2 positive as women who were their husband's only wife ever. CONCLUSION: Much transmission of HSV2 in this setting occurs within marriage where opportunity for personal protection is limited. High levels of transmission within marriage may undermine the impact of sexual behaviour change programmes aiming to reduce HSV2 and HIV incidence and complicate their evaluation.

Reporting of fertility events by men and women in rural Gambia.

We conducted a survey of male and female fertility in rural villages in The Gambia and compared men and women's reports of recent pregnancy events in the aggregate and of children ever born for matched couples. Despite widespread polygyny and sex differences in fertility, men's and women's reports were similar. Small sex differences in reports of recent stillbirths and neonatal deaths were found. For matched couples, husbands reported 0.23 more children ever born than their wives on average, but discordant reporting had little effect on recent marital fertility rates. Modeling of discordant reports indicates that fertility reports are more likely to be underestimated by both men and women for their earliest marriages. Reliable fertility data can be collected from men in this population.