A MULTIPLE IMPUTATION PROCEDURE FOR RECORD LINKAGE AND CAUSAL INFERENCE TO ESTIMATE THE EFFECTS OF HOME-DELIVERED MEALS

Causal analysis of observational studies requires data that comprise of a set of covariates, a treatment assignment indicator, and the observed outcomes. However, data confidentiality restrictions or the nature of data collection may distribute these variables across two or more datasets. In the absence of unique identifiers to link records across files, probabilistic record linkage algorithms can be leveraged to merge the datasets. Current applications of record linkage are concerned with estimation of associations between variables that are exclusive to one file and not causal relationships. We propose a Bayesian framework for record linkage and causal inference where one file comprises all the covariate and observed outcome information, and the second file consists of a list of all individuals who receive the active treatment. Under certain ignorability assumptions, the procedure properly propagates the error in the record linkage process, resulting in valid statistical inferences. To estimate the causal effects, we devise a two-stage procedure. The first stage of the procedure performs Bayesian record linkage to multiply impute the treatment assignment for all individuals in the first file, while adjustments for covariates' imbalance and imputation of missing potential outcomes are performed in the second stage. This procedure is used to evaluate the effect of Meals on Wheels services on mortality and healthcare utilization among homebound older adults in Rhode Island. In addition, an interpretable sensitivity analysis is developed to assess potential violations of the ignorability assumptions.

Diffusion Dynamics of Radiology IT - Systems in German Hospitals - A Bayesian Bass Model

Radiology has a reputation for having a high affinity to innovation - particularly with regard to information technologies. Designed for supporting the peculiarities of radiological diagnostic workflows, Radiology Information Systems (RIS) and Picture Archiving and Communication Systems (PACS) developed into widely used information systems in hospitals and form the basis for advancing the field towards automated image diagnostics. RIS and PACS can thus serve as meaningful indicators of how quickly IT innovations diffuse in secondary care settings - an issue that requires increased attention in research and health policy in the light of increasingly fast innovation cycles. We therefore conducted a retrospective longitudinal observational study to research the diffusion dynamics of RIS and PACS in German hospitals between 2005 and 2017. Based upon data points collected within the "IT Report Healthcare" and building on Rogers' Diffusion of Innovation (DOI) theory, we applied a novel methodological technique by fitting Bayesian Bass Diffusion Models on past adoption rates. The Bass models showed acceptable goodness of fit to the data and the results indicated similar growth rates of RIS and PACS implementations and suggest that market saturation is almost reached. Adoption rates of PACS showed a slightly higher coefficient of imitation (q = 0.25) compared to RIS (q = 0.11). However, the diffusion process expands over approximately two decades for both systems which points at the need for further research into how innovation diffusion can be accelerated effectively. Furthermore, the Bayesian approach to Bass modelling showed to have several advantages over the classical frequentists approaches and should encourage adoption and diffusion research to adapt similar techniques.