Data Accuracy in Medical Record Abstraction

Clinical Research Data Quality Literature Review and Pooled Analysis We present a literature review and secondary analysis of data accuracy in clinical research and related secondary data uses. A total of 93 papers meeting our inclusion criteria were categorized according to the data processing methods. Quantitative data accuracy information was abstracted from the articles and pooled. Our analysis demonstrates that the accuracy associated with data processing methods varies widely, with error rates ranging from 2 errors per 10,000 files to 5019 errors per 10,000 fields. Medical record abstraction was associated with the highest error rates (70–5019 errors per 10,000 fields). Data entered and processed at healthcare facilities had comparable error rates to data processed at central data processing centers. Error rates for data processed with single entry in the presence of on-screen checks were comparable to double entered data. While data processing and cleaning methods may explain a significant amount of the variability in data accuracy, additional factors not resolvable here likely exist. Defining Data Quality for Clinical Research: A Concept Analysis Despite notable previous attempts by experts to define data quality, the concept remains ambiguous and subject to the vagaries of natural language. This current lack of clarity continues to hamper research related to data quality issues. We present a formal concept analysis of data quality, which builds on and synthesizes previously published work. We further posit that discipline-level specificity may be required to achieve the desired definitional clarity. To this end, we combine work from the clinical research domain with findings from the general data quality literature to produce a discipline-specific definition and operationalization for data quality in clinical research. While the results are helpful to clinical research, the methodology of concept analysis may be useful in other fields to clarify data quality attributes and to achieve operational definitions. Medical Record Abstractor’s Perceptions of Factors Impacting the Accuracy of Abstracted Data Medical record abstraction (MRA) is known to be a significant source of data errors in secondary data uses. Factors impacting the accuracy of abstracted data are not reported consistently in the literature. Two Delphi processes were conducted with experienced medical record abstractors to assess abstractor’s perceptions about the factors. The Delphi process identified 9 factors that were not found in the literature, and differed with the literature by 5 factors in the top 25%. The Delphi results refuted seven factors reported in the literature as impacting the quality of abstracted data. The results provide insight into and indicate content validity of a significant number of the factors reported in the literature. Further, the results indicate general consistency between the perceptions of clinical research medical record abstractors and registry and quality improvement abstractors. Distributed Cognition Artifacts on Clinical Research Data Collection Forms Medical record abstraction, a primary mode of data collection in secondary data use, is associated with high error rates. Distributed cognition in medical record abstraction has not been studied as a possible explanation for abstraction errors. We employed the theory of distributed representation and representational analysis to systematically evaluate cognitive demands in medical record abstraction and the extent of external cognitive support employed in a sample of clinical research data collection forms. We show that the cognitive load required for abstraction in 61% of the sampled data elements was high, exceedingly so in 9%. Further, the data collection forms did not support external cognition for the most complex data elements. High working memory demands are a possible explanation for the association of data errors with data elements requiring abstractor interpretation, comparison, mapping or calculation. The representational analysis used here can be used to identify data elements with high cognitive demands.

Automated soldier medical readiness: Prevalence of " medically not ready" using data from the electronic soldier readiness processing system (ESRP)

Identifying accurate numbers of soldiers determined to be medically not ready after completing soldier readiness processing may help inform Army leadership about ongoing pressures on the military involved in long conflict with regular deployment. In Army soldiers screened using the SRP checklist for deployment, what is the prevalence of soldiers determined to be medically not ready? Study group. 15,289 soldiers screened at all 25 Army deployment platform sites with the eSRP checklist over a 4-month period (June 20, 2009 to October 20, 2009). The data included for analysis included age, rank, component, gender and final deployment medical readiness status from MEDPROS database. Methods.^ This information was compiled and univariate analysis using chi-square was conducted for each of the key variables by medical readiness status. Results. Descriptive epidemiology Of the total sample 1548 (9.7%) were female and 14319 (90.2%) were male. Enlisted soldiers made up 13,543 (88.6%) of the sample and officers 1,746 (11.4%). In the sample, 1533 (10.0%) were soldiers over the age of 40 and 13756 (90.0%) were age 18-40. Reserve, National Guard and Active Duty made up 1,931 (12.6%), 2,942 (19.2%) and 10,416 (68.1%) respectively. Univariate analysis. Overall 1226 (8.0%) of the soldiers screened were determined to be medically not ready for deployment. Biggest predictive factor was female gender OR (2.8; 2.57-3.28) p<0.001. Followed by enlisted rank OR (2.01; 1.60-2.53) p<0.001. Reserve component OR (1.33; 1.16-1.53) p<0.001 and Guard OR (0.37; 0.30-0.46) p<0.001. For age > 40 demonstrated OR (1.2; 1.09-1.50) p<0.003. Overall the results underscore there may be key demographic groups relating to medical readiness that can be targeted with programs and funding to improve overall military medical readiness.^

NOVEL PHANTOMS AND POST-PROCESSING FOR DIFFUSION SPECTRUM IMAGING

High Angular Resolution Diffusion Imaging (HARDI) techniques, including Diffusion Spectrum Imaging (DSI), have been proposed to resolve crossing and other complex fiber architecture in the human brain white matter. In these methods, directional information of diffusion is inferred from the peaks in the orientation distribution function (ODF). Extensive studies using histology on macaque brain, cat cerebellum, rat hippocampus and optic tracts, and bovine tongue are qualitatively in agreement with the DSI-derived ODFs and tractography. However, there are only two studies in the literature which validated the DSI results using physical phantoms and both these studies were not performed on a clinical MRI scanner. Also, the limited studies which optimized DSI in a clinical setting, did not involve a comparison against physical phantoms. Finally, there is lack of consensus on the necessary pre- and post-processing steps in DSI; and ground truth diffusion fiber phantoms are not yet standardized. Therefore, the aims of this dissertation were to design and construct novel diffusion phantoms, employ post-processing techniques in order to systematically validate and optimize (DSI)-derived fiber ODFs in the crossing regions on a clinical 3T MR scanner, and develop user-friendly software for DSI data reconstruction and analysis. Phantoms with a fixed crossing fiber configuration of two crossing fibers at 90° and 45° respectively along with a phantom with three crossing fibers at 60°, using novel hollow plastic capillaries and novel placeholders, were constructed. T2-weighted MRI results on these phantoms demonstrated high SNR, homogeneous signal, and absence of air bubbles. Also, a technique to deconvolve the response function of an individual peak from the overall ODF was implemented, in addition to other DSI post-processing steps. This technique greatly improved the angular resolution of the otherwise unresolvable peaks in a crossing fiber ODF. The effects of DSI acquisition parameters and SNR on the resultant angular accuracy of DSI on the clinical scanner were studied and quantified using the developed phantoms. With a high angular direction sampling and reasonable levels of SNR, quantification of a crossing region in the 90°, 45° and 60° phantoms resulted in a successful detection of angular information with mean ± SD of 86.93°±2.65°, 44.61°±1.6° and 60.03°±2.21° respectively, while simultaneously enhancing the ODFs in regions containing single fibers. For the applicability of these validated methodologies in DSI, improvement in ODFs and fiber tracking from known crossing fiber regions in normal human subjects were demonstrated; and an in-house software package in MATLAB which streamlines the data reconstruction and post-processing for DSI, with easy to use graphical user interface was developed. In conclusion, the phantoms developed in this dissertation offer a means of providing ground truth for validation of reconstruction and tractography algorithms of various diffusion models (including DSI). Also, the deconvolution methodology (when applied as an additional DSI post-processing step) significantly improved the angular accuracy of the ODFs obtained from DSI, and should be applicable to ODFs obtained from the other high angular resolution diffusion imaging techniques.