User’s Guide for Patient-Reported Outcomes in the Electronic Health Record was released by PCORI in May 2017. It provides a thoughtful evaluation of various approaches for integrating PROs into EHRs.
A summary, “Oh, the Places We’ll Go: Patient-Reported Outcomes and Electronic Health Records” is also available.
Integration in Clinical Settings
HealthMeasures can be integrated in clinical settings. When doing so, consider the following factors:
Identify Your Clinical Practice Needs
Use of patient-reported outcome measures (PROs) in clinical practice requires knowing your assessment goals, appropriate measures for those goals, which patients will complete measures, when patients will provide that information, and what you will do with the results. The International Society of Quality of Life Research (ISOQOL) published a User’s Guide for implementing PROs in clinical practice. Though not specific to HealthMeasures, the insights and suggestions apply to HealthMeasures.
- Review our recommended HealthMeasures
- Review our guidance on selecting the appropriate HealthMeasure
- Evaluate available data collection tools for the features you need including the advantages and disadvantages of collecting PROs in the electronic health record (EHR)
- Read examples of implementations with HealthMeasures
- If collecting PROs in your EHR, assemble the information that your local EHR team will need. This 3-page guide identifies some of the questions you will need to answer: What you need to know before requesting PROs in your EHR>>
Minimize Workflow Disruption
Collecting patient-reported outcome measures (PROs) in clinical settings requires time at multiple points in the clinic workflow. Ordering an assessment, completing an assessment, following-up on an incomplete assessment, reviewing results of an assessment, and addressing the results within a care delivery take time from healthcare providers, staff, and patients.
- Assemble a multi-disciplinary team to identify a strategy that minimize disruptions in your setting
- Utilize existing wait time for patients to complete an assessment
- Create time for a patient to complete an assessment prior to interacting with his or her provider (e.g., schedule assessment appointment 20 minutes before visit)
- Have a plan for patients that did not complete an assessment
- Clearly define roles (e.g., who checks that assessment was completed, who answers technical questions, who cleans and charges devices used for data collection)
- After implementation, communicate across all team members to address workflow issues
For more information on minimizing workflow disruptions, watch this 6 minute video.
Enable Multiple Data Capture Options
PROs are increasingly available within electronic health records. Using the EHR's patient portal can aid in collecting information from patients outside of the clinical setting. However, not all patients will complete an assessment through the EHR's patient portal for a variety of reasons.
- Implement departmental or organizational efforts to encourage enrollment in the patient portal
- Provide information about privacy
- Ensure there are benefits to the patient for enrollment
- Improve ease of registering and using the patient portal
Facilitate Use of Assessment Results in Clinical Care
If patients do not see their assessment results being used, they may not see a benefit to continuing to complete assessments. If clinicians do not know how to interpret and respond to a PRO score, they won’t use the information the patient provides.
- Enable patients to track their scores over time
- Graphs are often easier to interpret than tables of numbers
- Select PRO measures that provide information that care providers and patients can use
- Engage care providers and patients before implementation to identify what scores should trigger specific clinical actions (e.g., further assessment, education on symptom management, referral)
Additional Factors to Consider
- Number of assessments: Frequent assessments provide a more complete picture of a person’s health over time but can be burdensome for patients (to complete) and clinicians (to review).
- Measures can be completed at a healthcare visit. This information is then available to the care provider. Alternately, measures can be completed between clinic visits. This allows more frequent assessment and less disruption to the workflow. However, it may not reflect how a person is doing on the day of a clinic visit and may require a process for providers to follow-up with people who are having problems or concerns.
- Reporting of data: Data should be presented in an understandable format. This format may be different for the person who completed the assessment than for the care provider. This could involve terminology (e.g., “dyspnea” versus “trouble breathing”).
When to Assess
For information on when and how often to assess PROs, watch this 4 minute video.
Recommendations from Successful System-wide Integrations
The Cincinnati Children’s Hospital Medical Center (CCHMC) implemented system-wide assessment with PROMIS and other PRO measures. Across clinics, their completion rate is 75%. Their experience resulted in a publication (Gerhardt et al., 2018) describing recommendations for successful implementation including identification of a physician or nurse champion, matching measures with clinical need, knowing the meaning of scores including when and how to intervene, training providers and staff, and monitoring completion rates. Further discussion by (Lavallee et al, 2018) highlighted the need for such case studies.
University of Utah Health also implemented system-wide PRO collection that includes PROMIS CATs (Biber et al 2018). Their recommendations include engaging in pre-implementation planning, creating a standardized process for implementation, having organizational and clinical leadership, and collaboration across teams.
Although implemented as part of clinical research studies, Stover et al (2019) describe strategies to facilitate implementation of PROs in clinical care through use of stakeholder engagement. Specifically, clinician and patient input was used to establish thresholds for clinical action and create decision support tools for how to respond to concerning scores. Tracking and reporting the impact of those thresholds (e.g., frequency of concerning scores) was also useful in understanding the impact of PRO implementation on clinician workload and evaluate the frequency of false positives. Learn more>>
Clinical Integration Resources
Materials to facilitate integration of PROMIS measures in Epic, Cerner, OpenEMR, and other electronic health records were developed by the Electronic Health Record Access to Seamless Integration of PROMIS (EASIPRO) grant team. These include a Planning Guide, Decision Log, and Implementation Plan Template. Learn more>>
The ePRO Toolkit provides a set of guidelines for governance, integration, and score reporting for implementing patient-reported outcome measures in clinical practice settings. They provide interactive tools including checklists and sample plans. Learn more>>
A McGill University research team created a workshop to address barriers in using patient-reported outcome scores in care for patients with low back pain. Empirically supported behavior change techniques (e.g., modeling, rehearsal of skills, reward/incentives) were matched with identified barriers (e.g., skills to interpret PRO scores). An outline of the intervention including sample graphical reports of PRO score change over time and treatment recommendations by PRO score is available as supplementary information to their 2020 manuscript. Learn more>>
Using the Theoretical Domain Framework, a research team at McGill University identified both enablers and barriers for healthcare providers to use patient-reported outcome scores in clinical care for patients with low back pain. Learn more>>
- Knowledge (e.g., objective in collecting PROs)
- Skills (e.g., how to interpret a PRO score)
- Professional role (e.g., perceive responsibility for using PRO in care)
- Optimism (e.g., expect PROs will improve patients’ outcomes)
- Beliefs about consequences (e.g., believe in benefits of using PROs)
- Reinforcement (e.g., having improved patient outcomes after using PROs reinforces use of PROs
- Intentions (e.g., plan to use PROs in clinical care delivery)
- Skills (e.g., need new skills to use PROs)
- Professional role (e.g., unsure of role with using PROs)
- Beliefs about consequences (e.g., don’t need PROs to achieve improved outcomes)
- Goals (e.g., PROs are not more important than other sources of information)
- Decision processes (e.g., unclear how to use PROs in treatment decisions)
- Environmental context and resources (e.g., lack of time)
- Social influences (e.g., don’t consult with others on need to use PROs)
- Behavioral regulation (e.g., don’t assess patients’ motivation to complete PROs)
A 2019 publication provides an overview of challenges and opportunities in using patient-reported outcomes in health care delivery. Hsiao and colleagues summarized two technical expert panel meetings held in 2016 by the Agency for Healthcare Research and Quality. They identified challenges at pre-implementation (e.g., lack of a business case to collect PROs), data collection (e.g., liability and privacy concerns), and post-data collection (e.g., data display is not patient-friendly). They then identified strategies to address these challenges today (e.g., provide training on benefits of PRO collection; provide multiple data collection modalities; provide guidance on PRO score interpretation) and research opportunities (e.g., identify what PRO measures in what context provide what specific benefits; identify how PRO information can be integrated into care plans). Learn more>>
The National Quality Forum (NQF) published Patient-Reported Outcomes: Approaches and Challenges in Selection and Data Collection – An Environmental Scan in December 2019. Approaches to measure selection include psychometric soundness, person-centeredness, meaningfulness, amenability to change, and implementability. Methodological options (e.g., self-report versus proxy-report, self-administration versus interviewer administration, paper versus electronic, clinic setting versus home setting versus other settings, classical test theory versus item response theory) are also explored. Challenges include data collection burden, difficulties incorporating into the clinician workflow, difficulties incorporating data into electronic medical records, lack of clear interpretation of scores, both to patients and clinicians, and achieving stakeholder buy-in. Learn more>>