Are Farm‐to‐School Programs More Common in States With Farm‐to‐School‐Related Laws?

BACKGROUND: Farm‐to‐School programs (FTSPs) connect schools with locally grown food. This article examines whether FTSPs are more common in public elementary schools (ESs) in states with a formal, FTSP law or with a related, locally grown procurement law. METHODS: A pooled, cross‐sectional analysis linked nationally representative samples of 1872 public ESs (across 47 states) for the 2006–2007, 2007–2008, and 2008–2009 school years with state laws effective as of the beginning of September of each year that were collected and analyzed for all states. Multivariate logistic regression models examined the impact of state law on school FTSP participation, controlling for year and school‐level race/ethnicity, region, locale, free‐reduced lunch participation, and school size. RESULTS: The percentage of schools located in a state with a FTSP‐specific law increased from 7.3% to 20.4% over the 3‐year period, while the percentage of schools located in a state with a locally grown procurement law was approximately 30% across all years. The percentage of schools with FTSPs has more than tripled over the last 3 years (from 4.9% to 17.7%). After adjusting for all covariates except year, FTSPs were significantly more likely in states with a FTSP‐specific law (OR = 2.45, 95% CI = 1.28‐4.67); once adjusting for year, the results were marginally significant (OR = 1.72, 95% CI = .91‐3.25). School‐level FTSPs were not related to state locally grown procurement laws. CONCLUSION: Although the percentage of schools with FTSPs is relatively small, these programs are becoming more common, particularly in states with FTSP‐specific laws.     

Cost‐effectiveness analysis based on the number‐needed‐to‐treat: common sense or non‐sense?

This paper explores and critically discusses some of the methodological limitations of using the number-needed-to-treat (NNT) in economic evaluation. We argue that NNT may be a straightforward measure of benefit when the effect of an intervention is immediate, but that serious problems arise when the effect is delay rather than avoidance of an adverse event. In this case, NNT is not a robust or accurate measure of effect, but will vary considerably and inconsistently over time. This weakness will naturally spill over onto any CEA based on NNT. A literature review demonstrated that CEAs based on NNT were all published within the last five years, and that all studies suffered from important limitations. A major weakness of using NNT is the imposed restrictions on the outcome measure, which can only be strictly uni-dimensional and non-generic. Using NNT in economic evaluations is obtained at a cost in terms of both methodological shortcomings, and a reduced ability for such evaluations to serve as a useful tool in decision making processes. The use of NNT in economic evaluations might be better avoided.To every complicated question, there is a simple, straightforward, easy--and probably wrong answer (Occam's Sledgehammer).     

When does quality‐adjusting life‐years matter in cost‐effectiveness analysis?

PURPOSE: This paper investigates the impact of quality-of-life adjustment on cost-effectiveness analyses, by comparing ratios from published studies that have reported both incremental costs per (unadjusted) life-year and per quality-adjusted life-year for the same intervention. METHODS: A systematic literature search identified 228 original cost-utility analyses published prior to 1998. Sixty-three of these analyses (173 ratio pairs) reported both cost/LY and cost/QALY ratios for the same intervention, from which we calculated medians and means, the difference between ratios (cost/LY minus cost/QALY) and between reciprocals of the ratios, and cost/LY as a percentage of the corresponding cost/QALY ratio. We also compared the ratios using rank-order correlation, and assessed the frequency with which quality-adjustment resulted in a ratio crossing the widely used cost-effectiveness thresholds of 20, 000 US dollars, 50,000 US dollars, and 100,000 US dollars/QALY or LY. RESULTS: The mean ratios were 69,100 US dollars/LY and 103,100 US dollars/QALY, with corresponding medians of 24,600 US dollars/LY and 20,400 US dollars/QALY. The mean difference between ratios was approximately -34,300 US dollars (median difference: 1300 US dollars), with 60% of ratio pairs differing by 10,000 US dollars/year or less. Mean difference between reciprocals was 59 (QA)LYs per million dollars (median: 2.1). The Spearman rank-order correlation between ratio types was 0.86 (p<0.001). Quality-adjustment led to a ratio moving either above or below 50,000 US dollars/LY (or QALY) in 8% of ratio pairs, and across 100,000 US dollars in 6% of cases. CONCLUSIONS: In a sizable fraction of cost-utility analyses, quality adjusting did not substantially alter the estimated cost-effectiveness of an intervention, suggesting that sensitivity analyses using ad hoc adjustments or 'off-the-shelf' utility weights may be sufficient for many analyses. The collection of preference weight data should be subjected to the same scrutiny as other data inputs to cost-effectiveness analyses, and should only be under-taken if the value of this information is likely to be greater than the cost of obtaining it.     

Into a SEIFA SES cul‐de‐sac?

Use of the Australian Bureau of Statistics' SEIFA scores has almost become an automatic practice in area-based research on socio-economic status (SES) correlates of health inequalities in Australia in recent years. This article questions the wisdom of this emerging heavy, often singular, reliance on the SEIFA indexes for representing a real socio-economic condition. It is argued that improvements in our understanding of the social and economic processes that produce health inequalities will not occur unless we move beyond SEIFA's broad brush. Data for New England Area Health local government areas are used to support the argument.     

Meta‐analysis of Gaussian individual patient data: Two‐stage or not two‐stage?

Quantitative evidence synthesis through meta‐analysis is central to evidence‐based medicine. For well‐documented reasons, the meta‐analysis of individual patient data is held in higher regard than aggregate data. With access to individual patient data, the analysis is not restricted to a “two‐stage” approach (combining estimates and standard errors) but can estimate parameters of interest by fitting a single model to all of the data, a so‐called “one‐stage” analysis. There has been debate about the merits of one‐ and two‐stage analysis. Arguments for one‐stage analysis have typically noted that a wider range of models can be fitted and overall estimates may be more precise. The two‐stage side has emphasised that the models that can be fitted in two stages are sufficient to answer the relevant questions, with less scope for mistakes because there are fewer modelling choices to be made in the two‐stage approach. For Gaussian data, we consider the statistical arguments for flexibility and precision in small‐sample settings. Regarding flexibility, several of the models that can be fitted only in one stage may not be of serious interest to most meta‐analysis practitioners. Regarding precision, we consider fixed‐ and random‐effects meta‐analysis and see that, for a model making certain assumptions, the number of stages used to fit this model is irrelevant; the precision will be approximately equal. Meta‐analysts should choose modelling assumptions carefully. Sometimes relevant models can only be fitted in one stage. Otherwise, meta‐analysts are free to use whichever procedure is most convenient to fit the identified model.     

Embodiment without bodies? Analysis of embodiment in US‐based pro‐breastfeeding and anti‐male circumcision movements

This article uses the cases of pro‐breastfeeding and anti‐circumcision activism to complicate the prevailing conceptualisation of embodiment in research on embodied health movements (EHMs). Whereas most EHM activists draw on their own bodily experiences, in the breastfeeding and circumcision movements, embodiment by proxy is common. Activists use embodiment as a strategy but draw on physical sensations that they imagine for other people's bodies, rather than on those they experience themselves. Pro‐breastfeeding activists, who seldom disclose whether they were themselves breastfed, target mothers, encouraging them to breastfeed rather than to formula feed their children in order to reduce their child's risk of disease. Anti‐circumcision activists, only some of whom are circumcised men, urge parents to leave their sons' penises intact in order to avoid illness and disfigurement and to preserve the sons' rights to make their own informed decisions as adults. In both movements activists use embodiment as a persuasive strategy even though they themselves do not necessarily embody the risks of the negative health outcomes with which they are concerned. Future research on EHMs should reconceptualise EHMs to include embodiment by proxy and examine whether this important phenomenon systematically affects movement strategies and outcomes.     

First‐event or marginal estimation of cause‐specific hazards for analysing correlated multivariate failure‐time data?

In the analysis of multivariate failure-time data, the effect of a treatment or an exposure on the hazard of each failure type is sometimes evaluated using only the information on the first event that occurs in every individual, ignoring all events that follow. A Cox proportional hazards model may be fitted to such data, yielding a cause-specific hazard ratio (HR) estimate of the exposure for each failure type conditional on surviving all other failure types. However, such an estimate would not fully utilize all the available information on event times. Alternatively, a marginal approach may be implemented to model the time distribution of each failure type beyond the subject's first failure to (any) second and later failures. We investigate the performance of these two approaches by simulating positive and negative correlated event times from exponential distributions. Surprisingly, our results suggest that the first-event-only method (when multiple failures are possible) performs as well as the marginal method in most practical situations. Generally, for a modest sample size of 400, it is possible to achieve at least 85 per cent coverage of the true marginal HR with the first-event method. Although the coverage is poor for a correlation of 0.7 and beyond, such a high correlation between competing event times may be biologically rather implausible.     

Does the School Nurse‐to‐Student Ratio Make a Difference?

Public schools must provide an appropriate education for students with complex health needs. Chronic illnesses such as asthma and diabetes, social morbidities, injuries, and conditions that limit learning such as poor vision commonly affect school-aged children. School nurses often assume a leadership role in providing services for these children. However, although a national standard for school nurse/student ratio has been proposed, little research has examined the relationship between different school nurse-to-student ratios, level of health services provided in schools, and student outcomes. This study examined data in a 21-county region in eastern North Carolina served by a regional school nurse consultant. The school nurse-to-student ratio in these counties ranged from 1:451 to 1:7,440 based on full-time equivalencies. Two systems offered no school nursing services. Data from school years ending in 2000-2002 were considered in the analysis. A significant correlation was found between the increased presence of school nurses and services provided to children with diabetes (r = -.52, p = .000) and asthma (r = -.43, p = .002). Schools with better ratios provided more counseling services to children for social conditions such as depression and unintended pregnancy (r = -.38, p = .006), and more follow-up for school-related injuries (r = -.43, p = .003), and a higher percentage of children with vision problems received follow-up care (r = -.37, p = .007). A case analysis of one school district that experienced a significant improvement in nurse/student ratio over the study period provides further evidence that school nurses make a difference.     

Pairwise non‐parametric non‐inferiority tests in 3 × 3 cross‐over trials: should we adjust for period?

Before an active compound can be registered as a drug, it needs to be demonstrated that it does not cause particular unwanted compound specific side effects. Demonstrating non-inferiority of an active drug with placebo with respect to an unwanted side effect is often based on pharmacodynamic 3 x 3 cross-over trials, with the third treatment arm consisting of a positive control. The main comparison then is the pairwise comparison between active drug and placebo. In this paper, two different non-parametric methods with adjustment for period effect are compared with the non-parametric non-adjusted test and the parametric period-adjusted test with respect to size and power. The non-parametric test with period adjustment based on rank alignment has generally the largest power, but its size exceeds the nominal significance level. The non-parametric test with period adjustment based on stratification has low power for trials with a small number of subjects. The non-parametric test without period adjustment is a valid alternative in such cases, but its power decreases substantially in the presence of period effects. In the case of unbalanced designs, however, only the non-parametric test with period adjustment based on stratification can be used.     

Web‐ or paper‐based portfolios: is there a difference?

OBJECTIVE: To determine the differential effects of a paper-based versus a web-based portfolio in terms of portfolio quality, user-friendliness and student motivation. METHODS: An experimental design was used to compare Year 1 medical students' reflective portfolios. The portfolios differed in presentation medium only (i.e. web-based versus paper-based). Content analysis, a student questionnaire and mentor interviews were used to evaluate portfolio quality, user-friendliness and student motivation. A total of 92 portfolios were scored independently by 2 raters using a portfolio quality-rating instrument. RESULTS: Portfolio structure, quality of reflection and quality of evidence showed no significant effects of presentation medium. Multi-level analysis showed a significant effect for student motivation: web-based portfolios scored 0.39 more than paper-based portfolios (P < 0.05; effect size 0.76). The mentors reported no differences in portfolio quality, except that there were more visuals in web-based portfolios. Students spent significantly more time preparing the web-based than the paper-based portfolios (15.4 hours versus 12.2 hours; t = 2.1, P < 0.05; effect size 0.46). The 2 student groups did not differ significantly in terms of their satisfaction with the portfolio. The mentors perceived the web-based portfolios as more user-friendly. CONCLUSIONS: The web-based portfolios were found to enhance students' motivation, were more user-friendly for mentors, and delivered the same content quality compared with paper-based portfolios. This suggests that web-based presentation may promote acceptance of portfolios by students and teachers alike.     

How does cost matter in health‐care discrete‐choice experiments?

Willingness‐to‐pay (WTP) estimates derived from discrete‐choice experiments (DCEs) generally assume that the marginal utility of income is constant. This assumption is consistent with theoretical expectations when costs are a small fraction of total income. We analyze the results of five DCEs that allow direct tests of this assumption. Tests indicate that marginal utility often violates theoretical expectations. We suggest that this result is an artifact of a cognitive heuristic that recodes cost levels from a numerical scale to qualitative categories. Instead of evaluating nominal costs in the context of a budget constraint, subjects may recode costs into categories such as ‘low’, ‘medium’, and ‘high’ and choose as if the differences between categories were equal. This simplifies the choice task, but undermines the validity of WTP estimates as welfare measures. Recoding may be a common heuristic in health‐care applications when insurance coverage distorts subjects' perception of the nominal costs presented in the DCE instrument. Recoding may also distort estimates of marginal rates of substitution for other attributes with numeric levels. Incorporating ‘cheap talk’ or graphic representation of attribute levels may encourage subjects to be more attentive to absolute attribute levels. Copyright © 2010 John Wiley & Sons, Ltd.     

Evidence‐based practice?

In 1948, the world of health and medical research changed forever with the publication of the Medical Research Council's randomised controlled trial (RCT) on the use of streptomycin in the treatment of tuberculosis (TB).1,2 In that year, there were 341 papers on Streptomycin and TB published in the medical literature, but only one RCT, and this was the paper that defined a generation of TB treatment. The process of undertaking RCTs was quickly developed to include health promotion.     

What explains away the increased risk of histological chorioamnionitis in African‐American mothers of very‐low‐birthweight infants?

We sought explanations for African-American mothers' increased risk of chorioamnionitis by sequentially adjusting for confounder variables both individually and in groups. We searched for a subset of covariates that had the most influence on the chorioamnionitis odds ratio (OR) of these women. The sample consisted of 305 African-American and 520 White mothers who gave birth to a very-low-birthweight (< or = 1500 g) infant between 1991 and 1993, whose placenta was examined according to protocol and whose hospital chart was reviewed. Histologically proven chorioamnionitis was present in 43% of the placentas from African-American women and in 27% of those from Whites (crude OR 2.1, 95% confidence interval 1.5, 2.8). Singleton status appeared to be the most important effect modifier, with significant crude ORs of 1.5 among singletons and 3.4 among non-singletons. Using logistic regression models in the whole sample and in subgroups, we sought to 'explain away' this increased risk. Indeed, addition of information about confounder variables resulted in considerable reduction in the ORs to 1.1 among singletons and 1.9 among non-singletons. Particularly important among the confounders were singleton birth, Medicaid insurance, duration of ruptured membranes and gestational age. We discuss the possibility that this set of confounding variables conveys, in part, the same information as the variable African-American, and also perhaps information about the availability and/or utilisation of prenatal health care.