Multiple imputation for missing continuous and dichotomous data: The effects of patterns of missingness and variable correlations on type I error with small samples

The purpose of this study is to investigate the effects of predictor variable correlations and patterns of missingness with dichotomous and/or continuous data in small samples when missing data is multiply imputed. Missing data of predictor variables is multiply imputed under three different multivariate models: the multivariate normal model for continuous data, the multinomial model for dichotomous data and the general location model for mixed dichotomous and continuous data. Subsequent to the multiple imputation process, Type I error rates of the regression coefficients obtained with logistic regression analysis are estimated under various conditions of correlation structure, sample size, type of data and patterns of missing data. The distributional properties of average mean, variance and correlations among the predictor variables are assessed after the multiple imputation process. ^ For continuous predictor data under the multivariate normal model, Type I error rates are generally within the nominal values with samples of size n = 100. Smaller samples of size n = 50 resulted in more conservative estimates (i.e., lower than the nominal value). Correlation and variance estimates of the original data are retained after multiple imputation with less than 50% missing continuous predictor data. For dichotomous predictor data under the multinomial model, Type I error rates are generally conservative, which in part is due to the sparseness of the data. The correlation structure for the predictor variables is not well retained on multiply-imputed data from small samples with more than 50% missing data with this model. For mixed continuous and dichotomous predictor data, the results are similar to those found under the multivariate normal model for continuous data and under the multinomial model for dichotomous data. With all data types, a fully-observed variable included with variables subject to missingness in the multiple imputation process and subsequent statistical analysis provided liberal (larger than nominal values) Type I error rates under a specific pattern of missing data. It is suggested that future studies focus on the effects of multiple imputation in multivariate settings with more realistic data characteristics and a variety of multivariate analyses, assessing both Type I error and power. ^

The association of sexual behavior with oropharyngeal cancer and correlations with HPV-16 serologic status

Objectives. This hospital-based case-case study compared the characteristics of sexual behavior in patients with cancer of the oropharynx to patients with cancers of other head and neck sites. Additionally, the prevalence of certain sexual behaviors of HPV-16 seropositive head and neck cancer patients was compared to that of seronegative patients. ^ Methods. One hundred sixty five oropharyngeal cancer patients and 86 patients with cancers of other head and neck sites completed a sexual history questionnaire. ^ Results. Oropharyngeal cancer patients were significantly more likely to have had a greater number of lifetime sex partners, to have engaged in oral-genital sex, and to have had a greater number of oral-genital sex partners than non-oropharyngeal cancer patients. Oral-genital sex was significantly more common in the HPV-16 seropositive group. ^ Conclusion. These findings add to the evidence that HPV-16 is sexually transmitted to the upper aerodigestive tract and that certain sexual behaviors increase the risk for HPV-associated oropharyngeal cancer. ^

The characterization of residential fungal spores and the relationship with housing characteristics in the Houston area

A number of indoor environmental factors, including bioaerosol or aeroallergen concentrations have been identified as exacerbators for asthma and allergenic conditions of the respiratory system. People generally spend 90% to 95% of their time indoors. Therefore, understanding the environmental factors that affect the presence of aeroallergens indoors as well as outdoors is important in determining their health impact, and in identifying potential intervention methods. This study aimed to assess the relationship between indoor airborne fungal spore concentrations and indoor surface mold levels, indoor versus outdoor airborne fungal spore concentrations and the effect of previous as well as current water intrusion. Also, the association between airborne concentration of indoor fungal spores and surface mold levels and the age of the housing structure were examined. Further, the correlation between indoor concentrations of certain species was determined as well. ^ Air and surface fungal measurements and related information were obtained from a Houston-area data set compiled from visits to homes filing insurance claims. During the sampling visit these complaint homes exhibited either visible mold or a combination of visible mold and water intrusion problems. These data were examined to assess the relationships between the independent and dependent variables using simple linear regression analysis, and independent t-tests. To examine the correlation between indoor concentrations of certain species, Spearman correlation coefficients were used. ^ There were 126 houses sampled, with spring, n=43 (34.1%), and winter, n=42 (33.3%), representing the seasons with the most samples. The summer sample illustrated the highest geometric mean concentration of fungal spores, GM=5,816.5 relative to winter, fall and spring (GM=1,743.4, GM=3,683.5 and GM=2,507.4, respectively). In all seasons, greater concentrations of fungal spores were observed during the cloudy weather conditions. ^ The results indicated no statistically significant association between outdoor total airborne fungal spore concentration and total living room airborne fungal spore concentration (β = 0.095, p = 0.491). Second, living room surface mold levels were not associated with living room airborne fungal spore concentration, (β= 0.011, p = 0.669). Third, houses with and without previous water intrusion did not differ significantly with respect to either living room (t(111) = 0.710, p = 0.528) or bedroom (t(111) =1.673, p = 0.162) airborne fungal spore concentrations. Likewise houses with and without current water intrusion did not differ significantly with respect to living room (t(109)=0.716, p = 0.476) or bedroom (t(109) = 1.035, p = 0.304) airborne fungal spore concentration. Fourth, houses with and without current water intrusion did not differ significantly with respect to living room (χ 2 (5) = 5.61, p = 0.346), or bedroom (χ 2 (5) = 1.80, p = 0.875) surface mold levels. Fifth, the age of the house structure did not predict living room (β = 0.023, p = 0.102) and bedroom (β = 0.023, p = 0.065) surface mold levels nor living room (β = 0.002, p = 0.131) and bedroom (β = 0.001, p = 0.650) fungal spore airborne concentration. Sixth, in houses with visually observed mold growth there was statistically significant differences between the mean living room concentrations and mean outdoor concentrations for Cladosporium (t (107) = 11.73, p < 0.0001), Stachybotrys (t (106)=2.288, p = 0.024, and Nigrosporia (t (102) = 2.267, p = 0.025). Finally, there was a significant correlation between several living room fungal species pairs, namely, Cladosporium and Stachybotrys (r = 0.373, p <0.01, n=65), Curvularia and Aspergillus/Penicillium (r = 0.205, p < 0.05, n= 111)), Curvularia and Stachybotrys (r = 0.205, p < 0.05, n=111), Nigrospora and Chaetomium (r = 0.254, p < 0.01, n=105) and Stachybotrys and Nigrospora (r = 0.269, p < 0.01, n=105). ^ This study has demonstrated several positive findings, i.e., significant pairwise correlations of concentrations of several fungal species in living room air, and significant differences between indoor and outdoor concentrations of three fungal species in homes with visible mold. No association was observed between indoor and outdoor fungal spore concentrations. Neither living room nor bedroom airborne spore concentrations and surface mold levels were related to the age of the house or to water intrusion, either previous or current. Therefore, these findings suggest the need for evaluating additional parameters, as well as combinations of factors such as humidity, temperature, age of structure, ventilation, and room size to better understand the determinants of airborne fungal spore concentrations and surface mold levels in homes. ^

Nosocomial infections and their relationship to non-traditional risk factors in a medical intensive care unit

Background. Nosocomial infections are a source of concern for many hospitals in the United States and worldwide. These infections are associated with increased morbidity, mortality and hospital costs. Nosocomial infections occur in ICUs at a rate which is five times greater than those in general wards. Understanding the reasons for the higher rates can ultimately help reduce these infections. The literature has been weak in documenting a direct relationship between nosocomial infections and non-traditional risk factors, such as unit staffing or patient acuity.^ Objective. To examine the relationship, if any, between nosocomial infections and non-traditional risk factors. The potential non-traditional risk factors we studied were the patient acuity (which comprised of the mortality and illness rating of the patient), patient days for patients hospitalized in the ICU, and the patient to nurse ratio.^ Method. We conducted a secondary data analysis on patients hospitalized in the Medical Intensive Care Unit (MICU) of the Memorial Hermann- Texas Medical Center in Houston during the months of March 2008- May 2009. The average monthly values for the patient acuity (mortality and illness Diagnostic Related Group (DRG) scores), patient days for patients hospitalized in the ICU and average patient to nurse ratio were calculated during this time period. Active surveillance of Bloodstream Infections (BSIs), Urinary Tract Infections (UTIs) and Ventilator Associated Pneumonias (VAPs) was performed by Infection Control practitioners, who visited the MICU and performed a personal infection record for each patient. Spearman's rank correlation was performed to determine the relationship between these nosocomial infections and the non-traditional risk factors.^ Results. We found weak negative correlations between BSIs and two measures (illness and mortality DRG). We also found a weak negative correlation between UTI and unit staffing (patient to nurse ratio). The strongest positive correlation was found between illness DRG and mortality DRG, validating our methodology.^ Conclusion. From this analysis, we were able to infer that non-traditional risk factors do not appear to play a significant role in transmission of infection in the units we evaluated.^

Cardiovascular diseases risk factors during the first year of life

The pattern of change in cardiovascular risk factors, blood pressure (SBP and DBP) and plasma total cholesterol (TC), over time, their tracking and their relation to anthropometric measurements during the first year of life were investigated. Also, the effect of breast feeding on TC and the relationship of blood pressure measurements and family history of CVD risk factors were examined. One hundred five newborn term, healthy infants who were seen at a pediatric clinic in The Woodlands, Texas were followed longitudinally from 2 weeks to 1 year of age. TC, blood pressure, weight and length of the infants were measured at age 2 weeks, and again at 2, 4, 6, 9 and 12 months. In addition, family history, maternal and paternal, of CVD risk factors was obtained. Data analyses included only 40 infants who completed one year of follow up.^ At 2 weeks of age, the median value for TC was 23 mg/dl higher for females than for males. This difference disappeared as infants got older. For males, most of the increase in TC median levels, from 114 to 137 mg/dl, occurred between the ages of 2 weeks and 2 months, whereas for the female group, TC levels increased moderately, about 10 mg/dl, between 9 and 12 months of age. Tracking of TC was examined by using Spearman's correlation analysis. There were strong correlations between measurements taken as early as 2 weeks of age with later measurements. These correlations were stronger and more significant for males than for females (for males, r varied between 0.51 to 0.70, whereas for females, r varied between 0.11 to 0.70). The association of body measurements with TC is no more than modest and is closer for female infants than for male infants. Analysis, also, showed that infants who received breast milk had a TC mean value 47 mg/dl higher than that for infants who received formula milk only during the period of breast feeding and this difference disappeared by age 12 months.^ In both genders, most of the increase in blood pressure (about 10-15 mmHg in both SBP and DBP) occurred during the first 4 months of life. Most of the increase for male infants occurred during the first 2 months of life, while for females, the increase in SBP and DBP was between the age of 2 and 4 months. Neither SBP nor DBP track well during the first year of life and most of the correlations between measurements at different ages were not significant for either gender. The cross-sectional relationship of blood pressure measurements and selected body measurements was assessed. For females, only at age of 12 months did DBP have positive and significant correlations with weight, length and Quetelet index (r = 0.57, 0.60 and 0.57, respectively). There were no significant correlations between blood pressure and body measurements for males. Finally, analysis showed that maternal history of CV risk factors was significantly related to SBP in the female infant group, but not for males. For DBP, neither maternal nor paternal history was related. ^

Behavioral responses to methylphenidate: correlations with neuronal activity in the caudate nucleus

Methylphenidate is currently a drug of abuse and readily prescribed to both adolescents and adults. Chronic methylphenidate (MPH) exposure results in an increase in DA in the motive circuit, including the caudate nucleus (CN), similar to other drugs of abuse. This study focuses on research aimed to elucidate if there are intrinsic underlying differences in the CN electrophysiological activity of animals exhibiting different chronic responses to the same dose of MPH. Behavioral and caudate nucleus (CN) neuronal activity following acute and chronic doses of MPH was assessed by simultaneously recording the behavioral and neuronal activity. The experimental protocol lasted for 10 days using four groups; saline, 0.6, 2.5 and 10.0mg/kg MPH. Initially, the study determined that animals exposed to the same dose of MPH exhibited either behavioral sensitization or behavioral tolerance. Therefore animals were classified into two groups (behaviorally sensitized/tolerant) and their neuronal activity was evaluated. Four hundred and fifty one units were evaluated. Overall, a mixture of increases and decreases in CN neuronal populations was observed at initial MPH exposure, and at ED10 baseline and ED10 rechallenge. When separated based on their behavioral response (sensitized/tolerant), significant differences in neuronal response patterns was revealed. Animals exhibiting sensitization were more likely to increase their neuronal activity at ED1 and ED10 baseline, expressing the opposite response at ED10 rechallenge. Furthermore, when neuronal populations recorded from those animals exhibiting behavioral sensitization were statistically compared to those from animals exhibiting behavioral tolerance significant differences were observed. Collectively, these findings tell us that animals exposed to the same dose of MPH can respond oppositely and moreover that there is in fact some intrinsic difference in the two population’s neuronal activity. This study offers new insight into the electrophysiological differences between sensitized and tolerant animals.