Model based estimation of population total in presence of non-ignorable non-response.

The problem of handling non-ignorable non-response has been typically addressed under the design-based approach using the well-known sub-sampling technique introduced by Hansen and Hurwitz [1946, Journal of the American Statistical Association, Vol 41(236), Page 517- 529]. Alternatively, the model-based paradigm emphasizes on utilizing the underlying model relationship between the outcome variable and one or more covariate(s) whose population values are known prior to the survey. This article utilizes the model relationship between the study variable and covariate(s) for handling non-ignorable non-response and obtaining an unbiased estimator for the population total under the sub-sampling technique. The main idea is to combine the estimates obtained from the sample on first call and the sub-sample from second call using separate model relationships. The contribution of this paper helps us in providing unbiased estimates with an improved efficiency under model-based paradigm in presence of non-ignorable non-response. The provided method is more economical than the available estimators under callback methods as we are working sub-sampling and also increase response rate as a stronger mode of interview is employed for data collection. A numerical study using Monte Carlo is presented to illustrate the behavior of the proposed and the efficiency comparison.

A critical issue of using the variance of the total in the linearization method - In the context of unequal probability sampling.

Publicly available national survey data are useful for the evidence-based research to advance our understanding of important questions in the health and biomedical sciences. Appropriate variance estimation is a crucial step to evaluate the strength of evidence in the data analysis. In survey data analysis, the conventional linearization method for estimating the variance of a statistic of interest uses the variance estimator of the total based on linearized variables. We warn that this common practice may result in undesirable consequences such as susceptibility to data shift and severely inflated variance estimates, when unequal weights are incorporated into variance estimation. We propose to use the variance estimator of the mean (mean-approach) instead of the variance estimator of the total (total-approach). We show a superiority of the mean-approach through analytical investigations. A real data example (the National Comorbidity Survey Replication) and simulation-based studies strongly support our conclusion.

Evaluation of the Bifactor Nominal Response Model Analysis of a Health Efficacy Measure.

The bifactor nominal response item response theory (IRT) model, proposed by Cai, Yang and Hansen (2011), provides an extension of Bock's (1972, 1997) unidimensional nominal response model to multidimensional IRT. This model has not been utilized in any published studies since its original development. In this study, the model was applied to data from a sample of college students (N = 799) to evaluate the psychometric properties of a health efficacy measure. The nominal response model has the unique capability to estimate the functioning of each single response category, and higher response categories were found to have better functioning in this study. Poor-functioning categories were identified and combined into their adjacent categories. Items with revised response format showed improved functioning. The bifactor nominal response model is a useful tool for evaluation of bifactor scales with ordered while non-equivalently functioning categories.

Mediation analysis for survival data using semiparametric probit models.

Causal mediation modeling has become a popular approach for studying the effect of an exposure on an outcome through mediators. Currently, the literature on mediation analyses with survival outcomes largely focused on settings with a single mediator and quantified the mediation effects on the hazard, log hazard and log survival time (Lange and Hansen 2011; VanderWeele 2011). In this article, we propose a multi-mediator model for survival data by employing a flexible semiparametric probit model. We characterize path-specific effects (PSEs) of the exposure on the outcome mediated through specific mediators. We derive closed form expressions for PSEs on a transformed survival time and the survival probabilities. Statistical inference on the PSEs is developed using a nonparametric maximum likelihood estimator under the semiparametric probit model and the functional Delta method. Results from simulation studies suggest that our proposed methods perform well in finite sample. We illustrate the utility of our method in a genomic study of glioblastoma multiforme survival.

The pits and falls of graphical presentation.

Graphics are powerful tools to communicate research results and to gain information from data. However, researchers should be careful when deciding which data to plot and the type of graphic to use, as well as other details. The consequence of bad decisions in these features varies from making research results unclear to distortions of these results, through the creation of "chartjunk" with useless information. This paper is not another tutorial about "good graphics" and "bad graphics". Instead, it presents guidelines for graphic presentation of research results and some uncommon, but useful examples to communicate basic and complex data types, especially multivariate model results, which are commonly presented only by tables. By the end, there are no answers here, just ideas meant to inspire others on how to create their own graphics.

Reportable neurologic diseases in refugee camps in 19 countries.

BACKGROUND: Approximately one-third of refugees worldwide live in refugee camps. Selected neurologic diseases are actively reported in some refugee camps. METHODS: The United Nations High Commissioner for Refugees monitors health visits in refugee camps with the assistance of more than 25 partner organizations using standardized case definitions. Neurologic diseases were selected and searched for the years 2008 to 2011. The number of health care visits for a neurologic disease was calculated and divided by the aggregated number of reporting months available for each refugee camp ("visits per camp-month"). RESULTS: Five neurologic diseases were reported from 127 refugee camps in 19 countries. Visits for chronic, noncommunicable diseases including epilepsy (53,941 visits in 1,426 camp-months, 48% female) and cerebrovascular disease (4,028 visits in 1,333 camp-months, 51% female) far exceeded those for neurologic infectious diseases (acute flaccid paralysis/poliomyelitis, 78 visits in 3,816 camp-months, 42% female; leprosy, 74 visits in 3,816 camp-months, 66% female; meningitis, 477 visits in 3,816 camp-months, 51% female). In 2011, these diseases accounted for 31,349 visits globally with 91% of visits for epilepsy. CONCLUSIONS: Targeted programs addressing epilepsy and stroke among refugees in camps should become a priority and indicate that other chronic neurologic diseases that may be under- or misdiagnosed may also be common in refugee camps. Given that significant under-reporting is likely, our findings demonstrate the pressing need for coordinated preventive and interventional measures for epilepsy and stroke in refugee camps.

Estatística espacial na investigação epidemiológica de fatores associados à detecção de casos de hanseníase no Rio de Janeiro / Spatial statistics in epidemiological investigation of factors associated with the detection of leprosy cases in Rio de Janeiro

A hanseníase é uma doença infecciosa crônica, que apresenta uma meta deeliminação como problema de saúde pública recomendada pela OrganizaçãoMundial da Saúde, baseada apenas na taxa de prevalência. Nas últimas duasdécadas, observou-se uma redução progressiva dessa taxa em muitos países, muitos dos quais conseguiram alcançar a meta de eliminação, mas alguns permanecem com taxas de detecção elevadas em pequenas áreas, entre eles o Brasil e a Índia. Uma questão sem resposta até o momento é o quanto a atual estratégia de controle da transmissão é capaz de prevenir e controlar a doença. Para alcançar respostas, é preciso considerar o problema a partir dos múltiplosfatores de risco envolvidos na dinâmica do fenômeno, com suas diversidades notempo e no espaço. Estudos sobre a epidemiologia da doença e seu ciclo detransmissão, considerando o espaço na análise, podem ser tentativas de abranger essa complexidade fenomenológica. Diante deste cenário, propõe-se um estudo ecológico de base populacional, pautado em casos novos de hanseníase residentes no município de Duque de Caxias, Estado do Rio de Janeiro, entre 1998 e 2006, e tendo como unidade de análise os bairros. Foram realizados três métodos de análise na busca de melhores explicações dos fatores envolvidos. Primeiro, uma análise espacial descritiva da distribuição das taxas de detecção da doença através doÍndice de Moran Local; segundo, uma análise multivariada não espacial usandomodelo de Regressão de Poisson log-normal e; terceiro, uma análise multivariada espacial local utilizando modelo de Regressão Ponderada Geograficamente (GWR). Na análise com índice de Moran local identificou-se cluster na faixa sul-noroestemunicipal, não relacionado diretamente às ações de campanhas ou descentralização da assistência no município.

A análise multivariada não espacial identificou fatores associados à detecção de casos novos, tais como a descentralização da assistência e casos multibacilares. A análise com método GWR descreveu, através da construção de mapas, os fatores associado à taxa de detecção, com suas variações ao longo do município. Variáveis como proporção de domicílios com rede de água mostrou maior proteção nos bairros localizados mais no nordeste do município, e forma clínica indeterminada associada com maior detecção na região sul. Verificouse um quadro epidemiológico favorável e otimista da doença no município no período analisado, associada à detecção mais precoce dos casos e às açõesrealizadas na região de combate à doença. Os dados desse estudo mostramheterogeneidade da endemia dentro do município, reforçando a importância daanálise nas microrregiões e não só nas macrorregiões. Os achados a partir dosdiferentes métodos se complementaram, e podem dar indícios do início da ocorrência de mudanças na dinâmica do processo saúde-doença, que possam estar sendo influenciadas por ações estratégicas recentes de combate à doença, no município de Duque de Caxias, no período da análise.

Special considerations in the design, conduct and analysis of prophylaxis trials.

Methodological considerations in clinical trials apply to prophylaxis studies also. In addition, there are certain aspects that need special attention. These are the identification of a valid group of controls, the choice of the unit of randomization and its impact on subsequent analyses, the specificity and the sensitivity of case diagnosis and their impact on estimated efficacy and its reliability. The ethical aspects of the trial also need special consideration, bearing in mind that the intervention is on healthy individuals, and not patients with disease. These are discussed in the context of community prophylaxis trials of tuberculosis and leprosy undertaken in south India.

Measurement of ERP latency differences: a comparison of single-participant and jackknife-based scoring methods.

We used computer simulations to evaluate different procedures for measuring changes in the onset latency of a representative range of event-related components (the auditory and visual N1, P3, N2pc, and the frequency-related P3 difference wave). These procedures included several techniques to determine onset latencies combined with approaches using both single-participant average waveforms and jackknife-subsample average waveforms. In general, the jackknife-based approach combined with the relative criterion technique or combined with the fractional area technique (J.C. Hansen & S.A. Hillyard, 1980; S.J. Luck, 2005) provided the most accurate method and the greatest statistical power, with no inflation of Type I error rate.

A note on testing for intervention effects on binary responses.

OBJECTIVES: In some circumstances controlled trials are not feasible and treatments can only be evaluated using clinical databases. Here we consider the situation where treatment is introduced at a particular calendar time and can only be evaluated by comparison with historical controls. In these circumstances Heuer and Abel recommended using change-point methods to search for change in characteristics over the whole study period rather than simply comparing treated and untreated patients. Their recommendation is to only conclude that the intervention had an effect if a change-point could be demonstrated close in time to the introduction of the new treatment. This reduces the risk of false positives caused by confounding changes in population characteristics or changes in patient management. For binary data we develop a method that follows their philosophy and apply it to an observational study in the treatment of pin sites after orthopaedic surgery. METHODS: Tests for change in binomial probabilities based on Brownian bridge and Hansen's approximation for maximally selected chi(2) statistics are compared to an exact test by Worsley. The approximate method is generalized to logistic regression models allowing for covariates. RESULTS: The agreement of the exact and approximate method is good for sample sizes of 100 or more. The actual test size of the Hansen approximate test allowing for covariates is close to the nominal level, whereas the Brownian bridge approximation is slightly conservative. The change in pin site treatment significantly reduces the risk of infection for both adults and children. CONCLUSIONS: We consider the Hansen approximation to provide a very good and very simple method for obtaining the p-value when testing for a change in binary data event probabilities, with or without covariates.

Reliability and validity of the Bangla version of WHOQOL-BREF in an adult population in Dhaka, Bangladesh.

The aim of the present study was to evaluate the validity and reliability of the Bangla version of the World Health Organization Quality of Life questionnaire (WHOQOL-BREF) in an adult population in Bangladesh. Approximately 200 adults in the Dhaka district were interviewed using a questionnaire containing the Bangla version of the WHOQOL-BREF, as well as questions related to sociodemographic data. To assess the reliability of WHOQOL-BREF, Cronbach's alpha was calculated, and test-retest reliability was evaluated using intraclass correlation coefficient (ICC) of the first and second administrations. For comparison, approximately 200 leprosy patients were also interviewed with the questionnaire to examine the discriminant validity between groups. On the whole, sufficient validity was observed, and the Bangla version of the WHOQOL-BREF was deemed to be valid and reliable in assessing the quality of life of an adult population in Bangladesh.

Summarizing data through a piecewise linear growth curve model.

Most of the research in clinical trials is based on longitudinal designs, which involve repeated measurements of a variable of interest. Such designs are very powerful, both statistically and scientifically. Recent advances in statistical theory and software development incorporate the covariance structures such as unstructured, compound symmetry, auto-regressive and random effects, etc., for analysing longitudinal data. Hathaway et al. propose a technique for summarizing longitudinal data using linear growth curve model and establish that the number of summary statistics is fixed as four irrespective of the length of study. In this paper, we develop a procedure for analysing the longitudinal data through a piecewise linear growth curve model on the lines of Hathaway et al. Under different covariance structures, the linear model is fitted for Leprosy data and the residual sum of squares computed. Goodness of fit has also been considered for various models. In order to prove that the proposed method is robust and better than the others in terms of goodness of fit, simulation studies are carried out and the results presented.

Epidemiology of leprosy on five isolated islands in the Flores Sea, Indonesia

We conducted a population-based survey on 5 small island in South Sulawesi Province (Indonesia) to collect baseline data previous to a chemoprophylactic intervention study aiming at interrrupting the transmission of Mycobacterium leprae. Here we describe the present leprosy epidemiology on these geographically isolated islands. Of the 4774 inhabitants living in the study area 4140 were screened for leprosy (coverage: 87%). We identified 96 leprosy patients (85 new and 11 old patients), representing a new case detection rate (CDR) of 205/10,000 and prevelence rate 195/10,000. CDRs were similar for males and females. Male patients were more often classified as multibacillary (MB) than women. Of the new patients, 33 (39%) were classified as MB, 16 (19%) as paucibacillary (PB) 2-5 lesions and 36 (42%) as PB single lesion. In this area of high leprosy endemicity leprosy patients were extensively clustered, i.e. not equally distributed among the islands and within the islands among the houses

Risk factores for developing leprosy -A population-based cohort study in Indonesia

This study identified risk factors for developing leprosy through yearly incidence rates in five island populations. Personal factors, like age, sex, household size and the presence of M.leprae-specific antibodies as well as contact were studied. Of the 94 index patients (patients diagnosed in 2000) 43 (46%) were classified as multibacillary (MB), 17 (19%) were seropositive and 6 (7%) presented M.leprae DNA in nasal swabs as determined by polumerase chain reaction (PCR). All PCR positive patients were also seropositive. Forty-four of the 4903 persons initially without symptoms of leprosy developed leprosy in almost four years follow-up, giving an incidence rate of 2.98 per 1000 person-years. Men had a 2.2 times higher risk (95% Confidence Interval [CI]: 1.2-4.1) to developd leprosy than women. Persons living in households of more than 7 household members. Persons who were seropositive in 2000 had a 3.7 times higher risk (95% CI:1.1-12.4) than seronegative persons. Household contacts of MB patients had an adjusted hazard ratio (aHR) of 4.6 (95% CI:1.6-12.9) and household contacts of PCR positive patients an aHR of 9.36 (95% CI: 2.5-34.9) compared to non-contacts. Patients with PCR positive nasal swabs, suggesting nasal excretion of M.leprae, are probably the patients with the highest transmission patential. Since all index patients who were PCR positive were also seropositive, serology semms an adequate tool to identify these patients. Preventing seropositive persons to become seropositive patients and thus the main source of infection may break the chain of transmission

Espacializacao da endemia hansenica no Estado de Sao Paulo, 1991 a 2002 / ?

Objetivo: analisar o padrao espacial da ocorrencia dos caso de hanseniase detectados no Estado de Sao Paulo no periodo de 1991 a 2002. Metodo: estudo ecologico, tendo como unidade de analise os municipios do Estado de Sao Paulo georreferenciados em seus centroides, utilizando-se a geoestatistica para deteccao das areas de probabilidade de risco para hanseniase e quantificacao da dependencia espacial dos casos. A fonte de dados utilizada foi o banco informatizado dos casos de hanseniase notificados do Centro de Vigilancia Epidemiologica do Estado de Sao Paulo. Resultados: a dependencia espacial detectada foi de 0,55 graus de coordenadas georreferenciadas que corresponde aproximadamente a 55 km. As principais areas de probabilidade de risco encontradas foram as regioes nordeste, norte e noroeste do estado. Conclusoes: a verificacao de areas com probabilidade de riscos, utilizando-se a analise de dependencia espacial de casos de hanseniase, pode ser ferramenta util para avaliar a situacao de saude e planejar alocacao de recursos

Pacientes com hanseniase paucibacilar lesao unica tratados com terapia de dose unica ROM: avaliacao de progressao da doenca em coorte multicentrica (1997 - 2001) / ?

Este projeto de investigacao teve por objetivo definir o perfil clinico, epidemiologico, histopatologico e imunologico ao diagnostico e tambem avaliar os fatores prognosticos associados com a evolucao clinica dos pacientes de hanseniase PB/lesao unica tratados com ROM/dose unica em condicoes do campo. Trata-se de estudo de coorte multicentrica com seguimento de tres anos, conduzido em diferentes regioes do Brasil: Norte, Centro-Oeste e Sudeste. Os casos recem-detectados de hanseniase PB/lesao unica (n=259) foram recrutados e tratados de outubro de 1997 a dezembro de 1998. Na linha de base, a media de idade foi 32,4 (dp=16,0). 61,8% dos pacientes eram mulheres e 33,6% dos pacientes apresentaram histomorfologia BT, 29,7% I e33,6% TT. 74,4% tiveram Mitsuda positivo e 17,7% anti-PGL1 positivo. A avaliação clinica dos pacientes recrutados foi mensal no primeiro semestre e de 6 em 6 meses de outubro de 1997 a dezembro de 2001. O completo desaparecimento ou reducao do tamanho da lesao foi categorizado como evolucao favoravel. Evolucao desfavoravel foi definida como aumento no numero das lesoes, episodios de reacao reversa e/ou neurites perifericas isoladas, alem de mudanca de forma clinica para MB. A densidade de incidencia (DI) de eventos desfavoraveis foi calculada por pessoas-mes. Para calcular a probalidade cumulativa de intervalos livres de eventos foi usado de Metodo de Kaplan-Meier. Em 03 anos de seguimento, 88,1% dos pacientes tratados apresentaram completo desaparecimento da lesao. A incidencia de evolucao desfavoravel foi 17,8% (DI= 6,9/1000 pessoas-mes). A reacao reversa (RR) foi o evento desfavoravel predominante, principalmente entre pacientes BT. Houve baixa taxa de incidencia de progressao de PB para MB (7/10 000 pessoas-mes). Para a coorte como um todo, a probabilidade de permanecer livre de eventos a partir do diagnostico foi 92% (+/- 1,7%) no primeiro ano e 80,6% (+/- 2,55) no terceiro ano, estimado pelo metodo de Kaplan-Meier. A probalidade de permanecer livre de eventos foi mais alta entre pacientes com menos de 40 anos (92,4% no primeiro ano e 85,7% no terceiro ano de acompanhamento). Os pacientes BT tiveram mais probabilidade cumulativa de apresentar evolucao desfavoravel durante os intervalos seguimento, quanto comparados com os pacientes I e TT (p=0,03). Em conclusao, nossos resultados evidenriaram a predominancia de evolucao favoravel e baixa densidade de incidencia de progressao para formas MB nos