Account Missing Data Research Articles

Dealing with misclassification and missing data when estimating prevalence and incidence of caries experience

The aim of this research was to estimate the prevalence and incidence of caries experience (CE) in first permanent molars while dealing with misclassification and missing of data. CE was modeled as a Hidden Markov Model in which the response variable is subject to misclassification and missingness. The proposed analysis extends that of García-Zattera et al. (Stat Med 2010;29:3103) by allowing for various patterns of missing data. Findings were illustrated using data from the Signal Tandmobiel study that is a longitudinal oral health intervention study. Differences in the parameter estimates were noted between models that take into account misclassification and missing data and those that do not. Unbiased parameter estimates of prevalence and incidence were obtained without the use of validation data. Models that include subjects with missing data have smaller standard deviations than models that do not. It is important to account for misclassification to obtain less biased estimates of prevalence and incidence. For a proper estimation of prevalence and incidence in a longitudinal study subject to misclassification, validation data are not needed but when internal they can increase the efficiency in estimating the model. Also, including subjects with missing data increases the efficiency of estimating the parameters.

Relationship between PM2.5 Collected at Residential Outdoor Locations and a Central Site

Regression models are developed to describe the relationship between ambient PM2.5 (particulate matter [PM] ≤2.5 μm in aerodynamic diameter) mass concentrations measured at a central-site monitor with those at residential outdoor monitors. Understanding the determinants and magnitude of variability and uncertainty in this relationship is critical for understanding personal exposures in the evaluation of epidemiological data. The repeated measures regression models presented here address temporal and spatial characteristics of data measured in the 2004–2007 Detroit Exposure and Aerosol Research Study, and they take into account missing data and other data features. The models incorporate turbulence kinetic energy and planetary boundary layer height, meteorological data that are not routinely considered in models that relate central-site concentrations to exposure to health effects. It was found that turbulence kinetic energy was highly statistically significant in explaining the relationship of PM2.5 measured at a particular stationary outdoor air monitoring site with PM2.5 measured outside nearby residences for the temporal coverage of the data.

Does confined placental mosaicism account for adverse perinatal outcomes in IVF pregnancies?

IVF singletons have poorer perinatal outcomes than singletons from spontaneous conceptions. This may be due to the influence of ovarian stimulation on the chromosomal constitution of the embryos which could be translated into localized chromosomal anomalies in the placenta. The aim of this study was to compare the incidence of confined placental mosaicism (CPM) in IVF/ICSI pregnancies and spontaneous conceptions. We conducted a multi-centre retrospective analysis of karyotype results obtained by chorionic villus sampling (CVS), performed due to advanced maternal age (>or=36 years at 18 weeks of gestation), in the Netherlands between 1995 and 2005. From a total of 322 246 pregnancies, 20 885 CVS results were analysed: 235 in the IVF/ICSI group and 20 650 in the control group. The mean age of women in both groups was 38.4 years (mean difference -0.08, 95% CI -0.35 to 0.18). Data relating to the fetal karyotype were missing in 143 cases in the control group. When taking into account missing data, the incidence of CPM was lower in the IVF-ICSI group than in the control group, 1.3% versus 2.2% (odds ratio 0.59, 95% CI 0.19-1.85), whereas the incidence of fetal chromosomal anomalies was increased 4.3% versus 2.4% (odds ratio 1.81, 95% CI 0.95-3.42). Neither differences were statistically significant. The incidence of CPM is not increased in IVF/ICSI pregnancies compared with spontaneous conceptions. CPM probably does not account for the adverse perinatal outcomes following IVF/ICSI.

Client initiated termination of therapy at NHS primary care counselling services

AbstractThe main aim of this paper is to provide estimated benchmarks for unplanned endings for primary care counselling services using data drawn from the CORE National Research Database. Due to the problem of missing data, benchmarks are provided for ‘declared’ therapy endings and ‘estimated’ therapy endings. ‘Declared’ therapy endings are where the practitioner has provided data; ‘estimated’ therapy endings take into account missing data where clients are more likely to have had an unplanned rather than a planned ending to therapy. The average ‘declared’ rate of unplanned endings was calculated at 32% with a high rate determined as 40% or higher and a low rate as 21% or lower. The average ‘estimated’ rate of unplanned endings was calculated at 50% with a high rate determined as 58% or higher and a low rate as 38% or lower. It is likely that the true average, high and low rates of unplanned endings fall between the two sets of figures.

Bayesian reliability analysis of complex repairable systems

AbstractStemming from a consulting project about a gas distribution network, a new, Bayesian model is proposed to describe failures in a complex, expanding over time, repairable system, which is split into components installed over different years. Both exchangeable and independent Poisson processes, homogeneous in space but not in time, are used to model the components. The model takes also into account missing data, due either to unrecorded early failures or unknown installation dates of failed parts.Actual escape data from a gas distribution network illustrate the implementation of the model, which relies on the use of Markov chain Monte Carlo methods. Copyright © 2004 John Wiley & Sons, Ltd.

Modeling and Testing Change: An Introduction to the Latent Growth Curve Model

The purposes of this article are threefold: (a) to outline the basic concepts associated with latent growth curve (LGC) modeling; (b) to demonstrate the modeling and testing of LGC models based on three relatively simple, albeit increasingly complex, examples; and (c) to illustrate the modeling mechanism used in testing for the tenability of key statistical assumptions associated with LGC modeling. Based on 3-wave data comprising an original sample of 601 adolescents (Grades 8, 9, and 10) and using a multiple-sample approach that takes into account missing data resulting from time-related attrition, we "walk" the reader through the various stages of the model specification and testing processes. Based on self-rating scores of perceived ability as the outcome variable, we begin with a single-domain LGC model of perceived math ability and then follow with a more complex multiple-domain model that includes perceived ability in math, language, and science. Our final application extends the multiple-domain model to include the predictor variable of gender. We conclude by summarizing several advantages of LGC modeling over the more traditional methods used in the measurement of change.

Ambient air quality levels in Tehran, Iran, from 1988 to 1993

The objectives of this study were to describe trends in ambient air quality in Tehran between 1988 and 1993, to determine if these levels exceeded the World Health Organization (WHO) guidelines, and to discuss possible health effects related to exposure for these particular pollutants. Data were acquired from Iran's Environmental Protection Agency (IEPA) and the Ministry of Health (MH). These agencies operate five automated ambient air monitoring stations located in areas with heavy traffic. Daily samples of SO2, NO2, CO, total suspended particulate matter (TSM), and hydrocarbons (HC) were collected to provide 24 hour averages for each pollutant. Every three months, mean concentrations were reported to IEPA. Composite samples from all five stations were stored in a databank operated by IEPA. The ambient air quality guidelines were obtained from WHO reports. Statistical analysis was carried out using a regression model, which was designed to fit the air pollution data and take into account missing data. The results showed that there was a statistically significant upward trend in air pollution levels for all of the measured pollutants, except NO2, during the years 1988 to 1993. WHO guidelines were routinely and substantially exceeded by all pollutants except TSM. These findings suggest that as the population continues to grow, and increasing numbers of motor vehicles are driven in Tehran, there is concern for the health effects that may result from exposure to these pollutants.