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. ^

Comparison of financial trends in small rural and urban hospitals in Texas, 1983--1985

The purpose of this study was to compare the financial performance of small rural hospitals to that of small urban hospitals in Texas. Hospital-specific and environmental factors were studied as control variables.^ Small rural hospitals were found to be financially stronger on measures of liquidity but weaker on measures of profitability. Small urban hospitals performed better on measures of profitability and long-range solvency. When all measures in the five dimensions of financial performance were analyzed, no significant difference was found between the two groups of hospitals. None of the control variables included in the study was significantly associated with financial performance both for rural and urban hospitals. Conclusions were that small rural hospitals in Texas are experiencing a deterioration in financial condition but small, rural hospitals are not doing any worse than small urban hospitals; and that the financial hardship which rural hospitals suffer may be inherent in the nature of the institutions themselves, and not as a result of their smallness nor their rural settings. ^

Community health workers: The critical link to sustainable, healthy societies in low-resource settings---A model for the Baylor Shoulder to Shoulder Foundation

Community health workers (CHWs) can serve as a bridge between healthcare providers and communities to positively impact social determinants of health and, thus, the overall health of the population. The potential to effect lasting change is particularly significant within resource-poor settings with limited access to formally trained health care providers such as the small, rural village of Santa Ana Intibucá, Honduras and surrounding areas—located on the geographically and politically isolated border of Honduras and El Salvador. The Baylor Shoulder to Shoulder Foundation (BSTS) works in conjunction with Santa Ana's volunteer health committee to bring a health brigade that has provided health care and public health projects to the area at least twice a year since 2001. They have also hired a full-time Honduran physician, a Honduran in-country administrative director, and built a clinic; yet, no community health worker program exists. This CHW program model is the response to a clear need for a CHW program within the area served by BSTS and presents a CHW program model specific to Santa Ana Intibucá and surrounding areas to be implemented by BSTS. Methods used to develop this model include reviewing the literature for recommendations from leading authorities as well as successfully implemented CHW programs in comparable regions. This information was incorporated into existing knowledge and materials currently being used in the area. Using the CHW model proposed here, each brigade, in conjunction with the communities served, can help develop new modules to respond to the specific health priorities of the region at that time, incorporating consistent modes of contact with the local physician and the CHWs to provide refresher courses, training in new topics of interest, and to be reminded of the importance of community health workers' role as the critical link to healthy societies. With cooperation, effort, and support, the brigade can continue to help integrate a sustainable CHW system in which communities may be able to maximize the care they receive while also learning to care for their own health and the future of their communities.^