Reduce dimension or reduce weights? Comparing two approaches to multi‐arm studies in network meta‐analysis

Network meta‐analysis is a statistical method combining information from randomised trials that compare two or more treatments for a given medical condition. Consistent treatment effects are estimated for all possible treatment comparisons. For estimation, weighted least squares regression that in a natural way generalises standard pairwise meta‐analysis can be used. Typically, as part of the network, multi‐arm studies are found. In a multi‐arm study, observed pairwise comparisons are correlated, which must be accounted for. To this aim, two methods have been proposed, a standard regression approach and a new approach coming from graph theory and based on contrast‐based data (Rücker 2012). In the standard approach, the dimension of the design matrix is appropriately reduced until it is invertible (‘reduce dimension’). In the alternative approach, the weights of comparisons coming from multi‐arm studies are appropriately reduced (‘reduce weights’). As it was unclear, to date, how these approaches are related to each other, we give a mathematical proof that both approaches lead to identical estimates. The ‘reduce weights’ approach can be interpreted as the construction of a network of independent two‐arm studies, which is basically equivalent to the given network with multi‐arm studies. Thus, a simple random‐effects model is obtained, with one additional parameter for a common heterogeneity variance. This is applied to a systematic review in depression. Copyright © 2014 John Wiley & Sons, Ltd.     

Health and labour force participation of older people in Europe: What do objective health indicators add to the analysis?

This paper studies labour force participation of older individuals in 11 European countries. The data are drawn from the new Survey of Health, Ageing and Retirement in Europe (SHARE). We examine the value added of objective health indicators in relation to potentially endogenous self-reported health. We approach the endogeneity of self-reported health as an omitted variables problem. In line with the literature on the reliability of self-reported health ambiguous results are obtained. In some countries self-reported health does a fairly good job and controlling for objective health indicators does not add much to the analysis. In other countries, however, the results show that objective health indicators add significantly to the analysis and that self-reported health is endogenous due to omitted objective health indicators. These latter results illustrate the multi-dimensional nature of health and the need to control for objective health indicators when analysing the relation between health status and labour force participation. This makes an instrumental variables approach to deal with the endogeneity of self-reported health less appropriate.     

Do HIV prevention interventions in Asia lead to increase in condom utilization?: A meta‐analysis study

This meta‐analytic review synthesizes the findings of published independent studies that assessed the effectiveness of HIV prevention interventions in promoting condom usage in Asia. We focused on 20 studies that used randomized control studies to establish whether HIV prevention interventions lead to increased condom utilization. As compared with the average control group client, we found that 68% of the clients who participated in an intervention group increased their condom usage. The effectiveness of HIV prevention intervention is not related to country and rural‐urban contexts, to male‐female samples, to the type of outcome measured, or the timing of the latest follow‐up assessment. The implications of these findings are that policy‐makers, health administrators, and international donors should view behavioral HIV prevention interventions as an effective approach through which to increase condom usage in order to reduce the HIV infection rate in Asia.     

Socioeconomic factors, ethnicity and alcohol-related mortality in regions in Slovakia. What might a tree analysis add to our understanding?

Regional differences in alcohol-related mortality might reflect strong socioeconomic differences between regions. The present study examines the contribution of education, unemployment, income and minority proportion on regional differences in alcohol-related mortality for inhabitants aged 20-64 years. Linear regression analysis and a non-parametric regression tree analysis were used separately for males and females. The unemployment rate and low education appeared as important determinants of regional alcohol-related mortality, while the proportion of Roma and income were not significantly associated with alcohol-related mortality among males in Slovak districts. A district's unemployment rate was assumed to be the strongest predictor of the outcome measure.     

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.     

The place of death of cancer patients: can qualitative data add to known factors?

Research on the distribution of cancer deaths by setting-hospital, hospice, home, other--is longstanding, but has been given fresh impetus in the UK by policy commitments to increase the proportion of deaths occurring in patients' homes. Studies of factors associated with the location of cancer deaths fall into two main categories: geo-epidemiological interrogations of routinely collected death registration data, and prospective and retrospective cohort studies of terminally ill cancer patients. This paper summarises the findings of these studies and considers the place of death factors that are generated in semi-structured interviews with 15 palliative care service providers working in the Morecambe Bay area of north-west England. These qualitative data are found not only to confirm and considerably enrich understanding of known factors, but also to bring new factors into view. New factors can be grouped under the headings: service infrastructure, patient and carer attitudes, and cultures of practice. Such an approach provides useful information for policy makers and practitioners in palliative care.     

Mountain Dew® or Mountain Don't?: A Pilot Investigation of Caffeine Use Parameters and Relations to Depression and Anxiety Symptoms in 5th‐ and 10th‐Grade Students

BACKGROUND: Caffeine, the only licit psychoactive drug available to minors, may have a harmful impact on students' health and adjustment, yet little is known about its use or effects on students, especially from a developmental perspective. Caffeine use in 5th- and 10th-grade students was examined in a cross-sectional design, and relations and potential mediators of caffeine use to depression and anxiety symptoms were investigated.
METHODS: Children ( n = 135) and adolescents ( n = 79) completed a measure of naturalistic use of caffeinated and noncaffeinated beverages. Furthermore, daily availability, perceived benefits, and stimulating, psychological, and withdrawal effects of caffeinated and noncaffeinated beverages were assessed. Measures of depression and anxiety were also administered.
RESULTS: Fifth and 10th graders used caffeine frequently. Depression was positively related to caffeine use for both cohorts, though mediated by caffeine withdrawal effects. Surprisingly, anxiety was unrelated to use. Fifth graders reported less daily access to caffeine, but more psychological and stimulating effects of caffeine than 10th graders.
CONCLUSIONS: Although both children and adolescents experience negative caffeine-related outcomes, intake is seemingly not greatly limited in either cohort. In particular, youth appear vulnerable to increased depressive symptoms with increasing caffeine consumption. Implications for school policy regarding students' caffeine use are discussed.     

The death of cost‐minimization analysis?

Four different types of evaluation methods, cost-benefit analysis (CBA), cost-utility analysis (CUA), cost-effectiveness analysis (CEA) and cost-minimization analysis (CMA), are usually distinguished. In this note, we pronounce the (near) death of CMA by showing the rare circumstances under which CMA is an appropriate method of analysis. We argue that it is inappropriate for separate and sequential hypothesis tests on differences in effects and costs to determine whether incremental cost-effectiveness (or cost-utility) should be estimated. We further argue that the analytic focus should be on the estimation of the joint density of cost and effect differences, the quantification of uncertainty surrounding the incremental cost-effectiveness ratio and the presentation of such data as cost-effectiveness acceptability curves. Two examples from recently published CEA are employed to illustrate the issues. The first shows a situation where analysts might be tempted (inappropriately) to employ CMA rather than CEA. The second illustrates one of the rare circumstances in which CMA may be justified as a legitimate form of analysis.     

Income‐rich and wealth‐poor? The impact of measures of socio‐economic status in the analysis of the distribution of long‐term care use among older people

This article aims to investigate the impact of using 2 measures of socio‐economic status on the analysis of how informal care and home care use are distributed among older people living in the community. Using data from the Survey of Health, Ageing and Retirement in Europe for 14 European countries, we estimate differences in corrected concentration indices for use of informal care and home care, using equivalised household net income and equivalised net worth (as a proxy for wealth). We also calculate horizontal inequity indices using both measures of socio‐economic status and accounting for differences in need. The findings show that using wealth as a ranking variable results, as a rule, in a less pro‐poor inequality of use for both informal and home care. Once differences in need are controlled for (horizontal inequity), wealth still results in a less pro‐poor distribution for informal care, in comparison with income, whereas the opposite is observed for home care. Possible explanations for these differences and research and policy implications are discussed.     

The change in lifestyle data during 9?years: the reliability and continuity of baseline health practices

Objectives

To reveal change patterns in self-reported lifestyle data for 9 years, and examine the characteristics of changes by type of lifestyle and ageing.

Methods

The authors used the lifestyle data of 7,080 male workers aged 20–59 who received checkups for 9 years. The proportions of change patterns during the 9 years were determined in seven health practices; smoking, eating breakfast, sleeping hours, working hours, physical exercise, eating nutritional balanced diets, and mental stress.

Results

Among seven health practices, the keep rate of good health practice was highest for the non-smoking (90.8 %), followed by eating breakfast (69.0 %);and the lowest was physical exercise (13.7 %). The keep rate of poor health practice was highest for smoking (73.8 %), followed by non-physical exercise (67.1 %). The lowest rate of multiple changes during 10 years was smoking (7.1 %); the highest was mental stress (68.5 %).

Conclusions

As for the life style on smoking and eating breakfast seemed to be stable, using the data obtained at a specific point in time wouldn’t much affect the results. On the contrary, for other life styles, they showed poor continuity during 9 years, so it would be necessary to take into consideration the time point of data collection.     

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.     

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.     

What do medical students read and why? A survey of medical students in Newcastle‐upon‐Tyne,England

OBJECTIVES: There is increasing interest in the role of medical humanities within the undergraduate curriculum, but we know little about medical students' views on this or about their reading habits. Our study explored the reading habits of medical students, and their attitudes towards literature and the introduction of humanities into the curriculum. DESIGN: Self-completion questionnaire survey. SETTING: Newcastle University and Medical School. SUBJECTS: All first-, second- and third-year undergraduate medical students (384), biology students (151) and a random sample of law students (137) were sent a self-completion questionnaire to assess reading levels, attitudes towards literature and the medical humanities (medical students) and the perceived benefits of reading. RESULTS: Medical students read widely beyond their course and articulate a range of benefits from this, including: increasing awareness of life outside their experience; introspection or inspiration; emotional responses; and stimulation of an interest in reading or literature. Of the medical students, 40% (103/258) read one or more fiction books per month, but 75% (193) read fewer non-curricular books since starting university, largely because of time pressures, work, study or academic pressures and restricted access to books. A total of 77% (194) thought that medical humanities should definitely or possibly be offered in the curriculum, but of these 73% (141) thought it should be optional and 89% (172) that it should not be examined. CONCLUSIONS: Medical students read literature for a variety of very positive and valued reasons, but have found leisure reading harder to maintain since starting university. They support inclusion of the humanities in medical education, but have mixed views on how this should be done.     

What Works? Process Evaluation of a School‐Based Fruit and Vegetable Distribution Program in Mississippi

BACKGROUND: During the 2004–2005 school year, the Mississippi Department of Education, Office of Child Nutrition, initiated a pilot program to distribute free fruit and vegetable snacks to students during the school day. This article describes the first‐year implementation of the Mississippi Fruit and Vegetable Pilot Program. METHODS: The process evaluation addressed where, when, and how produce was distributed; what was distributed; challenges and successes; and recommended modifications. Five of the 25 program schools were selected to participate in the evaluation; selection was based on grade levels served and demographic characteristics. Data were collected from program staff (N = 11) and administrators (N = 6) via interviews and logs; student (N = 42) and parent (N = 19) focus groups; student questionnaires (N = 660); and school staff questionnaires (N = 207). RESULTS: Distributing fresh fruit and vegetable snacks at school was well received by staff and students. Most schools distributed the fresh fruit and vegetable snacks at morning break in classrooms or a central courtyard. Twenty‐two types of fresh fruit, 4 types of dried fruit, and 7 types of vegetables were served to students during the program year. Commonly distributed fruit included apples, oranges, pears, bananas, and tangerines. Carrots were the staple vegetable, followed by celery. Key challenges included getting students to try new foods and receiving the produce in a timely manner without spoiling. Main successes included seeing students try new fruit and vegetable snacks, having the program run smoothly, and teacher support. CONCLUSIONS: The program fit well within the school structure and could be an effective component of a multifaceted approach to enhancing child nutrition.