A patient registry is a structured system that uses observational study methods to collect uniform data for the scientific assessment of patient or population outcomes.1 Patient registries create and track patient care data specific to a disease or common comorbidities that may make the disease worse if not managed. The data can be specific to patients in your practice, or disease management processes at the macrolevel through population health management. The most important functions of a patient care registry are to improve care, identify gaps in care, and to evaluate or support new models of care.1 A given data management system allows the provision of evidence derived from quality measures (QMs), such as access, safety, quality, performance improvement, and resource allocation (cost). The use of "big data" or electronic management through the use of health information systems and electronic medical records allows robust queries utilizing individual data or aggregate data, by patient and practice. Registries are the systematization of data related to patient care, the process of care delivery, outcomes, and quality improvements.1
Disease registries usually include global approaches to chronic disease management, such as practitioner feedback, and case management through the use of clinical information and claims data systems. At the patient level, they incorporate patient self-management skills (clinician teaching and patient learning outcomes), which can be measured as part of an educational continuous quality improvement cycle. At the specialty level, they may incorporate adherence to clinical practice guidelines. The national strategy then is to better serve patients by developing data systems that are transparent and accessible to interprofessional teams and payers to provide better care.
The initial step in the process is to plan and design the registry. This usually begins with a statement of purpose, for example, to evaluate adherence to a particular element of a clinical practice guideline, such as daily skin inspections. Developing a registry team is similar to assembling any other mission-specific group: identifying stakeholders, clarifying the central parameters of interest, and defining the target population setting the scope of the project. Given the ethical considerations related to registries, the collection of patient data portends institutional risks; therefore, checks and balances should be instituted. The collection of data through registries is considered research; therefore, the institutional review board should approve the project. At the national level, there exists concern about the coordination and oversight of patient registries.1,2
A recent article published in the Journal for Healthcare Quality3 revealed some startling statistics regarding clinical registries. According to the authors who conducted a study of 153 clinical registries, most lacked a governing body, and there are no uniform standards for what constitutes a standard patient registry. Moreover, there is no evidence of a central or national clearinghouse for such registries. The salient points of the article find that "Among the 117 AMA specialty societies, just 16.2% were affiliated with a registry, government funding was associated with only 26.1% of registries, only 23.5% of registries risk adjusted outcomes, and only 18.3% of registries audited their data. Mandatory public reporting of hospital outcomes for all participating hospitals was associated with just 2% of registries."4,5
The chain of development of QMs is not always rooted in evidence-based medicine resulting from large randomized clinical trials. More commonly, QMs are generated from consensus and find their way into clinical practice guidelines.
Quality measures are now an integral part of pay for performance promulgated by the Affordable Care Act and the Centers for Medicare & Medicaid Services. Some QMs, such as those for pressure ulcer surveillance, suffer from a lack of consensus. Differences in taxonomy, such as "administrative versus surveillance incidence," also exist. The problem brings us back to the "old saw"--the definitions of incidence and prevalence. A recent retrospective analysis by Meddings et al5 included "2 million all-payer administrative records for 448 California hospitals and quarterly hospital-wide surveillance data for 213 hospitals, from the California Nursing Outcomes and Prevalence Study (as publicly reported on Cal Hospital Compare) for hospital-acquired pressure ulcers (HAPUs)." The data used to measure the outcomes were conflated by the use of "administrative data (as an incidence rate) or surveillance data (as a point-prevalence rate)." According to the authors, there was little correlation between each hospital's HAPUs rates from either administrative data or surveillance data. The authors raise legitimate concerns about the use of administrative data alone for comparing hospitals by HAPU rates and for public reporting or financial penalty or payment.5 It is clear policies, procedures, and politics that promulgate incentives for meaningful use of medical records and comprehensive registries are transitional but not yet transformative.3
REFERENCES