SBIR-STTR Award

This proposal addresses the need for objective hospital quality/ cost ranking tools that are relevant to patients' experiences and customizable to their individual conditions. This need is created by market changes where patients are increasingly being made aware of differences in quality between providers and pay higher portions of the total healthcare bill out of their pocket. Medicare, state governments, insurers and several private companies offer information on hospital quality and cost across many conditions and incorporating a variety of data, process and outcomes quality indicators. However, according to investors, bankers, industry analysts and entrepreneurs, comparison tools are still at an embryonic stage. They suffer from limited specificity, low resolution and conflicts of interest which in turn lead to patient confusion and perceptions that information is meaningless for decision making. PickCare(tm) is a hospital comparison tool that will address gaps that exist in current hospital ranking systems using advanced statistical and econometric models that have never been applied in this space. These models will: 1) Improve specificity through customization to patient factors that our analysis unearths are most important such as age and comorbidities; 2) Improve resolution by using an expanded history of risk factors; 3) Begin to build patient trust by detecting selection bias; 4) Present the results to patients in ways that are more meaningful and relevant to experiences they have had. The deliverable of Phase 1 will be a vertical prototype on kidney transplant programs that will integrate the analytical models with a web-based interface which will incorporate user feedback. Phase 2 will expand the functionality and conditions covered by the system in order to explore different paths to commercialization

This proposal addresses the need for objective hospital quality ranking tools that are relevant to patients' experiences and customizable to their individual conditions. This need is created by market changes where patients are increasingly being made aware of differences in quality between providers who will improve their odds at success. Medicare, state governments, insurers and several private companies offer information on hospital quality and cost across many conditions and incorporating a variety of data, process and outcomes quality indicators. However, according to investors, bankers, industry analysts and entrepreneurs, comparison tools are still at an embryonic stage. They suffer from limited specificity, low resolution, conflicts of interest and misfit in the patient information finding process. This leads to patient confusion and perceptions that information is meaningless for decision making. PickCare (tm) is a hospital comparison tool that will address gaps that exist in current hospital ranking systems using advanced statistical and econometric models that have never been applied in this space. In phase 1 we developed Pickcare v2.0 in kidney transplantation and showed that: 1) these models improve specificity through customization to patient factors that our analysis unearths are most important; 2) our models improve resolution by using an expanded history of risk factors; 3) the user interface tested with patients/ family communicates the system's improved resolution as differences across hospitals that patients can understand, and its specificity as a message of hope and alternatives. In phase 2 we will expand to Pickcare v3.0 to 1) accommodate user feedback from phase 1, 2) scale to accommodate more treatments beginning with dialysis and 3) further imbed the system in patients' information finding process. The deliverable of Phase 2 will be version 3 of Pickcare to be released on-line for kidney transplantation and tested internally with recruited patients for dialysis. A manuscript describing how the system influences patient choice in kidney transplantation based on analysis of on-line user data will also be submitted for publication.

Public Health Relevance:
This proposal addresses the need for user-friendly web-based tools to help patients and their families choose between hospitals based on the quality of care they will receive. The tool is unique in that it is customized to patient conditions, it has high resolution to separate excellent, average and poor hospitals, it addresses patients' motivations to search for hospital information and builds trust by answering their questions independently of insurers or hospitals.

Thesaurus Terms:
Address; Advertising; Affect; Age; Algorithms; Architecture; Care, Health; Case Mixes; Cessation Of Life; Clinic; Clinical; Clinics And Hospitals; Clinics Or Hospitals; Cohort Studies; Concurrent Studies; Confidential Information; Conflict Of Interest; Confusion; Confusional State; Data; Data Banks; Data Bases; Databank, Electronic; Databanks; Database, Electronic; Databases; Death; Decision Making; Dialysis; Dialysis Patients; Dialysis Procedure; Econometric Models; Embryo; Embryonic; Engineering / Architecture; Family; Family Member; Feedback; Goals; Grafting, Kidney; Hosp; Health; Health Insurance For Aged And Disabled, Title 18; Health Insurance For Aged, Title 18; Health Insurance For Disabled Title 18; Healthcare; History; Hospitals; Individual; Industry; Information, Confidential; Insurance; Insurance Carriers; Insurers; Internet; Kidney Transplantation; Kidney Transplants; Learning, Machine; Link; Machine Learning; Manuscripts; Marketing; Measures; Medicare; Mental Confusion; Modeling; Models, Econometric; Motivation; On-Line Systems; Online Systems; Outcome; Prov; Patients; Perception; Phase; Physicians; Process; Programs (Pt); Programs [publication Type]; Provider; Provincial Government; Publications; Qoc; Quality Indicator; Quality Of Care; Recommendation; Recording Of Previous Events; Recruitment Activity; Recurrence; Recurrent; Renal Transplantation; Renal Transplants; Research; Resolution; Risk; Risk Adjustment; Risk Factors; Science Of Statistics; Scientific Publication; Specificity; Staging; State Government; Statistics; System; System, Loinc Axis 4; Testing; Title 18; Transplantation; Trust; Www; Base; Case Control; Clinical Data Repository; Clinical Data Warehouse; Computerized Data Processing; Cost; Data Processing; Data Repository; Dialysis Therapy; Experience; Frailty; Health Insurance For Disabled; Improved; Innovate; Innovation; Innovative; Kernel Methods; Online Computer; Preference; Programs; Prospective; Public Health Relevance; Recruit; Relational Database; Signal Processing; Statistical Learning; Statistics; Success; Support Vector Machine; Tool; Transplant; User-Friendly; Web; Web Based; World Wide Web