Gunes Koru

Testing the theory of relative defect proneness for closed-source software

Recent studies on open-source software (OSS) products report that smaller modules are proportionally more defect prone compared to larger ones. This phenomenon, referred to as the Theory of Relative Defect Proneness (RDP), challenges the traditional QA approaches that give a higher priority to larger modules, and it attracts growing interest from closed-source software (CSS) practitioners. In this paper, we report the findings of a study where we tested the theory of RDP using ten CSS products. The results clearly confirm the theory of RDP. We also demonstrate the useful practical implications of this theory in terms of defect-detection effectiveness. Therefore, this study does not only make research contributions by rigorously testing a scientific theory for a different category of software products, but also provides useful insights and evidence to practitioners for revising their existing QA practices.

De-identifying a public use microdata file from the Canadian national discharge abstract database

BackgroundThe Canadian Institute for Health Information (CIHI) collects hospital discharge abstract data (DAD) from Canadian provinces and territories. There are many demands for the disclosure of this data for research and analysis to inform policy making. To expedite the disclosure of data for some of these purposes, the construction of a DAD public use microdata file (PUMF) was considered. Such purposes include: confirming some published results, providing broader feedback to CIHI to improve data quality, training students and fellows, providing an easily accessible data set for researchers to prepare for analyses on the full DAD data set, and serve as a large health data set for computer scientists and statisticians to evaluate analysis and data mining techniques. The objective of this study was to measure the probability of re-identification for records in a PUMF, and to de-identify a national DAD PUMF consisting of 10% of records.MethodsPlausible attacks on a PUMF were evaluated. Based on these attacks, the 2008-2009 national DAD was de-identified. A new algorithm was developed to minimize the amount of suppression while maximizing the precision of the data. The acceptable threshold for the probability of correct re-identification of a record was set at between 0.04 and 0.05. Information loss was measured in terms of the extent of suppression and entropy.ResultsTwo different PUMF files were produced, one with geographic information, and one with no geographic information but more clinical information. At a threshold of 0.05, the maximum proportion of records with the diagnosis code suppressed was 20%, but these suppressions represented only 8-9% of all values in the DAD. Our suppression algorithm has less information loss than a more traditional approach to suppression. Smaller regions, patients with longer stays, and age groups that are infrequently admitted to hospitals tend to be the ones with the highest rates of suppression.ConclusionsThe strategies we used to maximize data utility and minimize information loss can result in a PUMF that would be useful for the specific purposes noted earlier. However, to create a more detailed file with less information loss suitable for more complex health services research, the risk would need to be mitigated by requiring the data recipient to commit to a data sharing agreement.

A privacy protection technique for publishing data mining models and research data

Data mining techniques have been widely used in many research disciplines such as medicine, life sciences, and social sciences to extract useful knowledge (such as mining models) from research data. Research data often needs to be published along with the data mining model for verification or reanalysis. However, the privacy of the published data needs to be protected because otherwise the published data is subject to misuse such as linking attacks. Therefore, employing various privacy protection methods becomes necessary. However, these methods only consider privacy protection and do not guarantee that the same mining models can be built from sanitized data. Thus the published models cannot be verified using the sanitized data. This article proposes a technique that not only protects privacy, but also guarantees that the same model, in the form of decision trees or regression trees, can be built from the sanitized data. We have also experimentally shown that other mining techniques can be used to reanalyze the sanitized data. This technique can be used to promote sharing of research data.

An Exploratory Study of Home Healthcare Robots Adoption Applying the UTAUT Model

The home healthcare initiative is aimed to reduce readmission costs, transportation costs, and hospital medical errors, and to improve post hospitalization healthcare quality, and enhance patient home independency. Today, it is almost unimaginable to consider this initiative without information technology. Home healthcare robots are one of such emerging technologies. Several robots have been developed to facilitate home healthcare such as remote presence robots e.g., RP2 and Paro. Most previous research in this area has focused on technology and implementation issues of home healthcare robots, but ignored the factors that influence their adoption. To address the limitation, the current research applied and extended the UTAUT model to the home healthcare domain. The model was tested using survey questionnaire. The empirical results not only confirmed the effects of some constructs from the original UTAUT model but also identified perceived security as a new factor that directly affects usage intention of home healthcare robots. In addition, effort expectancy did not show a direct effect but an indirect effect through performance expectancy on usage intention. Several practical and theoretical implications are also discussed.

A longitudinal analysis of the dependency concentration in smaller modules for open-source software products

Our recent studies on single releases of multiple open-source software (OSS) products showed a higher concentration of dependencies in smaller modules. For one of the products, it was observed that an isolatable and observable refactoring initiative exacerbated this concentration inequality. In this paper, we empirically investigate the dependency concentration in smaller modules from a longitudinal perspective: (1) whether this concentration inequality always exists over product life time; (2) how it changes. We hypothesize that the concentration inequality should either remain at same levels or increase over time. This is because large-scale and long-lived software products usually go through some degree of continuous and intermittent refactoring. Our results show that dependencies concentrate in smaller classes in all releases, and this concentration inequality generally increases over successive releases. We suggest that software practitioners continuously pay a higher QA attention to smaller modules. We also recommend increasing such QA focus as a product matures and goes through refactoring activities.

Testing the theory of relative dependency from an evolutionary perspective: higher dependencies concentration in smaller modules over the lifetime of software products

Recent studies conducted on the single releases of multiple software products showed that dependencies concentrate on smaller modules, that is, smaller modules have more dependencies per source line of code. This phenomenon, called the Theory of Relative Dependency, explains why some earlier studies reported that smaller modules were proportionally more defect prone. It is important to test the Theory of Relative Dependency from multiple perspectives so that it can be used as an explanatory argument when garnering organizational support to give a higher quality assurance (QA) priority to smaller modules. In this study, we test the validity of this theory from an evolutionary perspective by examining the consecutive releases of a number of software products. Dependencies do concentrate over smaller modules regardless of the product age. Furthermore, continuous refactoring efforts are associated with increasing concentration of dependencies on smaller modules over product lifetime. Based on the consistent results, software managers and developers should consider giving a higher QA priority to smaller modules over the lifetime of a software product. In the projects where refactoring is adopted continuously, the QA priority on smaller modules should be further increased as the software product ages. Copyright

Leveraging Health Information Technology for Fall-Risk Management in Home Care A Qualitative Exploration of Clinicians’ Perspectives

Falls have a detrimental impact on home care outcomes. This study investigated home care clinicians’ perspectives on (1) how the currently adopted health information technology (IT) solutions supported fall-risk management activities, (2) various shortcomings of these solutions, and (3) ways to improve upon them. Within a case study design, qualitative data were collected through interviews and focus groups at one suburban home health agency in the United States. The clinicians mainly used the electronic health record but experienced substantial difficulties with documenting and accessing fall-related information. Improvement opportunities include providing better workflow support and usability, implementing medication alerts, facilitating fall-related information exchange, and leveraging the IT capabilities at patients’ homes. Additional IT solutions can improve the awareness, assessment, and intervention dimensions of fall-risk management.

Identifying the key performance improvement domains for home health agencies

Objective: The objective of this study was to support home health agencies (HHAs) in the United States (US) in their individualized quality assessment and performance improvement (QAPI) initiatives by identifying their key performance improvement domains (KPIDs). Methods: Qualitative research was conducted by following the Framework method. Rich contextual data were obtained through focus group meetings participated by domain experts. The analysis results were further refined in an online forum and validated at a final meeting. Results: Four focus groups involving a total of 20 participants resulted in useful discussions during which various perspectives were expressed by the expert participants. A well-defined set of 17 KPIDs emerged under four categories, namely, economical value, sociocultural sensitivity, interpersonal relationships, and clinical capabilities. Conclusions: The feedback we received from the focus groups indicates that performance improvement in HHAs is a lot more complicated than simply assessing whether certain clinical tasks are performed. The KPIDs identified in this study can help HHAs in their focused and individualized QAPI initiatives. Therefore, the results should be immediately relevant, interesting, and useful to the home care industry and policy makers in the US.

Technology Innovations for Better Fall Risk Management in Home Care

Achieving better fall risk management is an integral component of quality home care. The current qualitative study uncovers the challenges and opportunities of home health agencies (HHAs) in achieving better fall risk management. A secondary document analysis was adopted to learn from rich contextual information in fall incident reports recorded in a HHA. Poor engagement of patients and caregivers was a contributing factor in many fall incidents. Patients often fell as a result of not understanding or accepting their physical limitations. For better fall risk management, many incidents highlighted a need for providing complete and thorough care, better coordination of care, higher levels of sociocultural awareness, patient engagement, and caregiver involvement. The results provide evidence regarding the challenges and opportunities for improving fall risk management in home care along with insight about how information technology solutions can support improvement initiatives. [Journal of Gerontological Nursing, 44(7), 15-20.].

Facilitating Focused Process Improvement Efforts in Home Health Agencies to Improve Utilization Outcomes Effectively and Efficiently

The aim of this study was to identify a smaller set of clinical practices most associated with the utilization outcomes of the home health agencies. A secondary data analysis approach was adopted using the publicly accessible Home Health Compare repository as the main data source; the control variables such as rurality, agency size, and median income levels were obtained from other public data repositories. Checking for fall risks and starting care in a timely manner were most associated with hospital readmissions. These two practices and treating patients for pain are most associated with the emergency room visit rates. The results provide additional guidance to home health agencies in their prioritized and focused performance improvement initiatives to improve their utilization outcomes.