Data from both electronic health records (EHRs) and longitudinal patient registries are central to implementing, managing, and monitoring health care in emerging value-based models. Medicare’s new Comprehensive Care for Joint Replacement (CJR) program is a good illustration of the shift from fee-for-service payment to bundled payment programs with fixed global reimbursements linked to the quality of care. Under the CJR program, postoperative adverse events, as well as patient-reported satisfaction and functional outcomes, are used to calculate bundle quality incentives.
While EHR adoption is revolutionizing the capture and storage of health care data during the acute episode (e.g., during total joint replacement surgery), institutional EHRs include limited postoperative outcome data because (1) patients seek care across multiple health systems and (2) regional health information exchanges are limited. This “siloing” of data can limit the ability to evaluate longitudinal clinical care, monitor quality, and conduct research to guide policy and best practices.
Recent reports on several national joint replacement databases, however, have illustrated how high-quality longitudinal patient-reported and clinical data can be aggregated to complement EHR data, engage patients, and simultaneously meet new value-based payment regulations. We propose that smartly designed registries such as the Function and Outcomes Research for Comparative Effectiveness in Total Joint Replacement (FORCE-TJR) database can overcome the shortcomings of institutional EHRs and complement EHR data for the purposes of quality and value monitoring and comparative outcomes research.
Capturing Longitudinal Outcomes
Established in 2010 with funding from the Agency for Healthcare Research and Quality, the FORCE-TJR web-based data infrastructure captures and analyzes comprehensive total joint replacement outcomes for the purposes of quality and value monitoring. To date, the FORCE-TJR database has been used to collect information on >50,000 patients from a diverse group of >200 orthopedic surgeons representing practices in 28 states.
Web-based portals such as those deployed by FORCE-TJR are designed to facilitate the collection of data on pre- and post-hospital care and on patient-reported outcomes across time and place. The FORCE-TJR system facilitates the integration of patient-reported, EHR, and claims data as described below:
- Patient-Reported Data: The FORCE-TJR system sends automated emails directly to patients at home both preoperatively and postoperatively. The emails include a unique, secure link that allows the patient to access a survey based on his or her stage of treatment or recovery. These surveys are used to collect information on preoperative risk factors, postoperative outcomes (i.e., pain and function as rated with use of standardized patient-reported outcome measures, or PROMs), and adverse events for all enrolled patients throughout the total joint replacement care cycle (before surgery, 30–90 days and 6–12 months after surgery, and yearly thereafter).
- Preoperatively, patients report on risk factors, including demographic characteristics, body mass index, medical and emotional comorbidities, smoking status, and arthritis symptoms in the knee, hip, and low back.
- Postoperatively, patients report any visits to the emergency room, any returns to the operating room, and any hospitalizations, as well as the use of any rehabilitation services. For example, at 30 days after surgery, patients report pain status to assess joint recovery and are asked about any visits to an emergency room for the evaluation of the involved knee or hip. While patients may not know the precise diagnosis, they typically are able to accurately report the visit, and, if the visit uncovers a serious knee or hip issue, the surgeon is notified.
- EHR and Claims Data: Hospitals are often unaware of patient visits that occur outside the health system where the surgery was performed and therefore undercount the costs of total health delivery for these patients. To address this void, the FORCE-TJR system supplements the patient-reported postoperative data with EHR and claims data. Specifically, FORCE-TJR staff members review each patient’s EHR to understand the clinical details of postoperative events. For patients over 65 years of age, the staff then compare the EHR data with claims data from the Center for Medicare and Medicaid Services (CMS). The EHR and claims data provide additional clinical information beyond that reported by the patient and are used to verify the timing of adverse events, diagnoses, and procedures.
- Combining and Using the Data: FORCE-TJR analysts combine patient-reported, EHR, and claims data to generate comparative post-TJR outcome data for member hospitals. The analysts provide lists of patients with postoperative hospitalizations or emergency room visits within 30–90 days after surgery so that physicians and hospital administrators can evaluate the reasons and impact on cost and quality. In addition, hospitals and surgeons use the FORCE-TJR web-reporting system to obtain risk-adjusted, comparative outcome profiles for postoperative outcomes. These comparative reports enable surgeons and hospitals to understand how the outcomes for their patients (e.g., pain relief, functional improvement, rate of readmissions, etc.) compare with those at other hospitals.
Involving Registry Staff and Engaging Patients to Capture Complete Data
While the FORCE-TJR web-based, direct-to-patient data-capture system facilitates longitudinal data capture, ensuring that patients actually report complete data requires more than technological tools. To that end, registry staff initiate and maintain a longitudinal relationship with the patient, obtain patient consent (to eliminate regulatory challenges for the release of EHR data across multiple hospitals and doctor offices), educate the patient about the value of this information for ongoing quality monitoring and future research efforts, and monitor the completeness of data reporting. Centralizing these registry-related functions to the registry staff frees the clinical staff to focus on acute care.
Patient engagement is assessed by monitoring self-reported outcomes as demonstrated by PROM completion rates. In the FORCE-TJR registry, >80% of patients complete postoperative assessments: 67% respond to the initial request at 6 to 12 months postoperatively, and an additional 15% to 18% respond following automated email reminders and, if needed, staff calls. Surgeons and hospitals alike benefit from the consistent, complete reporting of preoperative and postoperative information (e.g., risk factors, adverse events, patient-reported outcomes, etc.) that can be collected independent of office visits and patient geographic location.
Adding National Benchmarks to Patient Data
Monthly, the FORCE-TJR web-based reporting system returns outcome data to the treating physician and hospital administrators, with site-specific outcomes compared with risk-adjusted national norms, so that the information can be used for quality monitoring. Thus, the registry provides national benchmarks against which to evaluate institutional outcomes and ensures complete reporting of events and outcomes beyond the treating hospital.
For regulatory reporting, FORCE-TJR is a CMS-certified Quality Clinical Data Registry and meets the requirements for the new CJR value-based payment pilot program. Hospitals can use FORCE-TJR data to guide their internal quality and utilization management and can also submit these data to meet external mandates such as the CMS bundled payment program.
Real-Time, Personalized Patient Summaries and Outcome Predictions
National benchmarks can also be valuable when interpreting individual patient data. Today, with research funds from the Patient-Centered Outcomes Research Institute, the FORCE-TJR web-based system can be used to collect patient data at the time of an office visit and return a real-time, personalized patient summary to the patient and surgeon. This individual patient summary can then be used to compare the patient’s symptom and risk profiles with national norms in the FORCE-TJR database. In addition, statistical algorithms within the FORCE-TJR system can be used to generate predictions of likely surgical outcomes for individual patients based on the national data.
These data are formatted in easy-to-read, color-coded, two-page reports to support shared decisions between surgeons and patients. For example, after the patient has completed the FORCE-TJR survey, the office staff can print out the report so that the patient can review it with his or her surgeon. The patient and surgeon can then discuss likely estimates of postoperative improvement in pain and function scores based on the experience of previous patients with similar comorbidity profiles. In addition, the patient can review his or her individual risk factors (e.g., smoking or diabetes) that may influence the timing and outcome of total joint replacement.
Using these data, the patient and physician can determine whether it is likely that the patient will achieve the desired outcome following surgery or might prefer an alternative therapy. After the data populate the patient’s report to guide treatment decisions, the patient’s data are stored in the national FORCE-TJR database to be included (anonymously) in future calculations of national norms.
Patients: A Key Resource for Information
Patients are a key resource for information that is traditionally missing from EHR systems and should be central to a registry’s national efforts. Specifically:
- Patients can provide comprehensive information on risk factors, including behaviors and historical medical events that are inconsistently captured in the hospital’s EHR at the time of surgery.
- Patients can report any rehospitalizations or emergency room visits that occur after the index procedure.
- Our preliminary research indicates that, on average, 25% of readmissions that occur within 30 days after discharge are to a hospital that is different from the one where the total joint replacement surgery was performed. Moreover, in urban areas, almost 40% of emergency room visits and readmissions are to a different hospital than the one where the surgery was performed.
- Hospital-based data warehouses store information only on care that is provided within that hospital system and do not include postoperative care provided in other health systems. Regional data-sharing is one solution, but regional data systems are not the current norm. Alternately, a registry can capture the total utilization data through the combination of health care claims and patient-reported data. Specifically, when patients report receiving postoperative care at a different health system, FORCE-TJR verifies this care by reviewing the EHR or claims.
- Over 80% of patient-reported events can be validated by one of these two methods.
- Patients and their caregivers are the only source of information on symptoms (e.g., pain) and behaviors (e.g., physical activity and function). In an era of chronic disease, symptom progression is a key data element for guiding treatment and assessing outcomes.
- To capture patient-generated information, most EHR portals now disseminate patient surveys. However, in today’s disjointed health system, patients must maintain portal access with each hospital or health system where they receive care — a cumbersome situation.
- In contrast, the FORCE-TJR registry maintains a single web-based system to capture longitudinal data and can return this information to multiple EHRs with patient permission. Thus, the registry serves the function of integrating data across time and place and simplifies patient reporting.
Patient-Centered Registries: Providing Augmentation, Not Duplication
As value-based purchasing and quality measurement policies expand, policymakers, clinicians, and researchers need methods to ensure that data collected to serve policy mandates mesh with ongoing clinical care and research efforts. The lessons learned from FORCE-TJR can be extrapolated to other chronic conditions such as diabetes or chronic pulmonary disease to capture comprehensive pre-treatment risk factors, post-treatment outcomes, and risk-adjusted comparative analyses for outcome-improvement activities in a value-based payment system.
Because registries such as FORCE-TJR transcend individual health care delivery systems, creative funding mechanisms are required to maintain and support data capture and analysis. However, the longitudinal outcome data benefit diverse partners, including hospitals, payers, and implant manufacturers, affording broad opportunities to define fiscal support.
Well-designed registries engage patients as partners to ensure comprehensive, consistent data to inform quality and outcome monitoring beyond the EHR of a single hospital or health system. Smartly designed registries do not duplicate the EHR’s clinical data but rather augment this information to serve ongoing quality improvement, policy, and research efforts.
Disclosures and Disclaimer: FORCE-TJR was funded in part by grants from the Agency for Healthcare Research and Quality (AHRQ grant #P50HS018910), and new patient reports were funded in part by the Patient-Centered Outcomes Research Institute (PCORI grant #1507-31714). The content of this paper represents the opinions of the authors, who are responsible for its content, and do not necessarily represent the views of the Agency for Healthcare Research and Quality (AHRQ) or the Patient-Centered Outcomes Research Institute (PCORI).
Acknowledgement: The authors would like to thank Dr. Sylvie Puig for her editorial assistance with this manuscript.