Are registry data granular enough to make patient care-related decisions?
Click Here to Manage Email Alerts
Click here to read the Cover Story, "To assist, alert: Use of registries in orthopedics."
Granular capabilities
Orthopedic registry studies have long served as instrumental tools for guiding the quality of patient care.
National registries, such as the Swedish or American Academy of Orthopaedic Surgeons registries, gather large datasets from hospitals, hospital systems, surgical centers and private practices to create a comprehensive view of population-level trends. Institutional registries at Kaiser Permanente, Hospital for Special Surgery and the Mayo Clinic offer more granular insight into the individual patient via their relatively closed systems of care and leverage of electronic medical record data. The utilization of both types of registries provides clinicians and researchers a unique opportunity to leverage data to influence care-related decisions in a way that ultimately improves the quality of patient care.
The national registries have a distinct advantage in their ability to collect massive amounts of high-level data from a diverse population. The information gathered from various settings and regions of the country offers researchers insight into national trends and outcomes that become normalized over enormous volumes. For example, in the Australian Arthroplasty Registry, the high revision risk associated with metal-on-metal articular surface replacement resurfacing implants (DePuy) and articular surface replacement XL acetabular implants (DePuy) was later confirmed by several other international arthroplasty registries, including ones from England, Wales, Northern Ireland and New Zealand. This resulted in a worldwide voluntary recall of the implants and subsequently safer arthroplasties for patients.
In contrast, institutional registries such as those found at Kaiser Permanente, Hospital for Special Surgery and the Mayo Clinic present their own distinct advantages. What they may lack in magnitude of data compared to the national registries, they excel in their unique capacity to meticulously track individual patient data and outcomes. This offers decision-makers at these institutions the ability to highlight specific questions, drive important metrics within their systems and track changes in practice. This has led to reductions in opioid prescribing by clinicians, acceptance of same-day discharge in shoulder arthroplasty and increases in cemented hip hemiarthroplasty after fracture to name a few institutional changes after registry data supported efforts.
National and institutional registries complement each other in terms of the type of data that can be extracted and the questions that can be answered. Limitations in registry studies do exist, which magnifies the importance that clinicians and researchers target specific questions and understand the clinical relevance of their findings. Prior studies of national level registries and the success of various institutional registries suggest that registries have the capability to be granular enough to influence how we make patient care decisions.
- References:
- Hailer NP, et al. Acta Orthop. 2015;doi:10.3109/17453674.2014.1002183.
- Hughes RE, et al. Curr Rev Musculoskelet Medi. 2017;doi:10.1007/s12178-017-9408-5.
- Kramer JD, et al. J Shoulder Elbow Surg. 2020;doi:10.1016/j.jse.2019.09.037.
- Navarro RA, et al. Semin Arthroplasty. 2022;doi:10.1053/j.sart.2021.12.001.
- Okike K, et al. JAMA. 2020;doi:10.1001/jama.2020.1067.
Ronald A. Navarro, MD, is of Kaiser Permanente School of Medicine in Pasadena, California. Randall Arroyo is an MS2 from the University of California, Riverside School of Medicine in Riverside, California.
Lack of granularity
Orthopedic registries collect important data that, in their present state, are most frequently used to report trends in treatments and patient demographics. Individual surgeons and hospitals can also use registry data to benchmark their aggregated patient population data and outcomes against those from a registry to identify potential opportunities for quality improvement. Registries also provide crucial surveillance functions, signaling possible problematic treatments or implants early on. Further, they can inform treatment selection for specific patient groups; for example, identifying joint arthroplasty implants with either superior or inferior performance in given demographic cohorts. However, despite the number of patients in some registries, they may not possess the granularity necessary to prescribe point-of-care decisions for an individual patient.
We know as practicing surgeons that individual patients exhibit significant variability that may not be captured or reflected in registry data. Even if a registry collects robust demographic, comorbidity and condition-specific data, the way in which registry data are publicly accessible tends to represent averages or aggregates. This lack of granularity can lead to generalizations that overlook unique patient characteristics that might impact patient outcomes with a specific treatment or implant.
There are other limitations of registry data. The quality and completeness of its data can vary, potentially influencing the reliability of conclusions drawn from them. Missing data points or inconsistencies in data entry can compromise its accuracy, making it difficult to draw accurate inferences about patient outcomes. In addition, registry data may not reflect the most up-to-date information on surgical techniques, implant designs or postoperative care. The latency between real-world practice changes and registry reporting can limit its applicability in making real-time patient care decisions, particularly for new techniques and implants.
Registries often focus on objective clinical outcomes, such as implant survival or revision rates, and it often lacks patient-reported outcome measures (PROMs). PROMs capture patient perspectives on pain, function and satisfaction, providing insights that complement objective measures and offer a more comprehensive view of patient health. Certainly, PROMs are more a stated priority for registries, but the collection rates and completeness of PROMs data even in registries reporting PROMs lag those of more objective data elements. Increasingly, we also recognize that patient preferences, lifestyle factors and social determinants of health impact patient outcomes after orthopedic procedures, and these are not frequently captured in registries. Overreliance on registry data for clinical care decisions can overlook these critical aspects of patient care, potentially leading to treatment plans that do not align with patient values or circumstances.
In conclusion, registries provide valuable data but have limitations vis-à-vis point-of-care clinical decision-making for individual patients. Application of registry data must be complemented by a surgeon’s clinical expertise, patient engagement and a comprehensive understanding of individual patient needs and circumstances.
Alexander S. McLawhorn, MD, MBA, is a hip and knee surgeon at Hospital for Special Surgery in New York.
Collapse
Read more about
Click Here to Manage Email Alerts