Dataset Information

Demonstration of a software design and statistical analysis methodology with application to patient outcomes data sets.

ABSTRACT: With emergence of clinical outcomes databases as tools utilized routinely within institutions, comes need for software tools to support automated statistical analysis of these large data sets and intrainstitutional exchange from independent federated databases to support data pooling. In this paper, the authors present a design approach and analysis methodology that addresses both issues.A software application was constructed to automate analysis of patient outcomes data using a wide range of statistical metrics, by combining use of C#.Net and R code. The accuracy and speed of the code was evaluated using benchmark data sets.The approach provides data needed to evaluate combinations of statistical measurements for ability to identify patterns of interest in the data. Through application of the tools to a benchmark data set for dose-response threshold and to SBRT lung data sets, an algorithm was developed that uses receiver operator characteristic curves to identify a threshold value and combines use of contingency tables, Fisher exact tests, Welch t-tests, and Kolmogorov-Smirnov tests to filter the large data set to identify values demonstrating dose-response. Kullback-Leibler divergences were used to provide additional confirmation.The work demonstrates the viability of the design approach and the software tool for analysis of large data sets.

SUBMITTER: Mayo C

PROVIDER: S-EPMC3815052 | biostudies-other | 2013 Nov

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:BackgroundThe consideration of health-related quality of life (HRQL) is a hallmark of best practice in HIV care. Information technology offers an opportunity to more closely engage patients with chronic HIV infection in their long-term management and support a focus on HRQL. However, the implementation of patient-reported outcome (PRO) measures, such as HRQL in routine care, is challenged by the need to synthesize data generated by questionnaires, the complexity of collecting data between patient visits, and the integration of results into clinical decision-making processes.ObjectiveOur aim is to design and pilot-test a multimedia software platform to overcome these challenges and provide a vehicle to increase focus on HRQL issues in HIV management.MethodsA multidisciplinary team in France and Australia conducted the study with 120 patients and 16 doctors contributing to the design and development of the software. We used agile development principles, user-centered design, and qualitative research methods to develop and pilot the software platform. We developed a prototype application to determine the acceptability of the software and piloted the final version with 41 Australian and 19 French residents using 2 validated electronic questionnaires, the Depression, Anxiety and Stress Scale-21 Items, and the Patient Reported Outcomes Quality of Life-HIV.ResultsTesting of the prototype demonstrated that patients wanted an application that was intuitive and without excessive instruction, so it felt effortless to use, as well as secure and discreet. Clinicians wanted the PRO data synthesized, presented clearly and succinctly, and clinically actionable. Safety concerns for patients and clinicians included confidentiality, and the potential for breakdown in communication if insufficient user training was not provided. The final product, piloted with patients from both countries, showed that most respondents found the application easy to use and comprehend. The usability testing survey administered found that older Australians had reduced scores for understanding the visual interface (P=.004) and finding the buttons organized (P=.02). Three-fourths of the respondents were concerned with confidentiality (P=.007), and this result was more prevalent in participants with higher anxiety and stress scores (P=.01), as measured by the Depression, Anxiety and Stress Scale-21 Items. These statistical associations were not observed in 15 French patients who completed the same questionnaire.ConclusionsDigital applications in health care should be safe and fit for purpose. Our software was acceptable to patients and shows potential to overcome some barriers to the implementation of PROs in routine care. The design of the clinicians' interface presents a solution to the problem of voluminous data, both synthesizing and providing a snapshot of longitudinal data. The next stage is to conduct a randomized controlled trial to determine whether patients experience increased satisfaction with care and whether doctors perceive that they deliver better clinical care without compromising efficiency.

Dataset Information

Demonstration of a software design and statistical analysis methodology with application to patient outcomes data sets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets