Common Patient-Reported Outcomes Within the Food and Drug Administration Voice of the Patient Reports.

OBJECTIVES: Proponents of disease-specific patient-reported outcome measurements (PROMs) often argue disease-agnostic measures do not adequately capture their patient population's experience. Patient-Reported Outcomes Measurement Information System (PROMIS) provides a disease-agnostic domain set that may adequately cover many diseases. This study seeks to investigate whether PROMIS's quality of life domain coverage can span patient-reported outcomes (PROs) elicited from patients across unrelated diseases. METHODS: The Food and Drug Administration Voice of the Patient reports were an initiative to elevate patient voices regarding their condition and associated treatments. Two reviewers extracted patient-reported health-related (quality of life) domains from the reports and categorized them into PROMIS domains or non-PROMIS domains. Domain coverage was summarized for each report. Any extracted PROs not covered by PROMIS domains were placed in an "other" category and analyzed for common themes. RESULTS: Across 26 reports, PROMIS covered 216 of 374 (70%) of the reports' PRO domains. The heritable bleeding disorders report had the highest coverage (82%). Human immunodeficiency virus had the lowest coverage (50%). The most common PROMIS domain, "ability to participate in social roles," appeared in 25 reports (96%). The most common domains not included in PROMIS were stigma, sensitivities, and sensory deficits as evident in 19 (73%), 18 (69%), and 18 reports (69%), respectively. If the top 3 unincluded domains were amended into PROMIS, the total domain coverage would increase to 84%. CONCLUSIONS: PRO domains elicited in the Food and Drug Administration Voice of the Patient reports were widely captured by PROMIS, suggesting domains patients experience contain enough overlap to be recorded by appropriate PROMIS domains. PROMIS could increase its coverage by adding domains.

Do patients want clinicians to ask about social needs and include this information in their medical record?

BACKGROUND: Social needs screening in primary care may be valuable for addressing non-medical health-related factors, such as housing insecurity, that interfere with optimal medical care. Yet it is unclear if patients welcome such screening and how comfortable they are having this information included in electronic health records (EHR). OBJECTIVE: To assess patient attitudes toward inclusion of social needs information in the EHR and key correlates, such as sociodemographic status, self-rated health, and trust in health care. DESIGN, PARTICIPANTS, AND MAIN MEASURES: In a cross-sectional survey of patients attending a primary care clinic for annual or employment exams, 218/560 (38%) consented and completed a web survey or personal interview between 8/20/20-8/23/21. Patients provided social needs information using the Accountable Care Communities Screening Tool. For the primary outcome, patients were asked, "Would you be comfortable having these kinds of needs included in your health record (also known as your medical record or chart)?" ANALYSES: Regression models were estimated to assess correlates of patient comfort with including social needs information in medical records. KEY RESULTS: The median age was 45, 68.8% were female, and 78% were white. Median income was $75,000 and 84% reported education beyond high school. 85% of patients reported they were very or somewhat comfortable with questions about social needs, including patients reporting social needs. Social need ranged from 5.5% (utilities) to 26.6% (housing), and nonwhite and gender-nonconforming patients reported greater need. 20% reported "some" or "complete" discomfort with social needs information included in the EHR. Adjusting for age, gender, race, education, trust, and self-rated health, each additional reported social need significantly increased discomfort with the EHR for documenting social needs. CONCLUSIONS: People with greater social needs were more wary of having this information placed in the EHR. This is a concerning finding, since one rationale for collecting social need data is to use this information (presumably in the EHR) for addressing needs.

Predicting Unplanned Health Care Utilization and Cost: Comparing Patient-reported Outcomes Measurement Information System and Claims.

OBJECTIVES: There is little literature describing if and how payers are utilizing patient-reported outcomes to predict future costs. This study assessed if Patient-reported Outcomes Measurement Information System (PROMIS) domain scores, collected in routine practice at neurology clinics, improved payer predictive models for unplanned care utilization and cost. STUDY DESIGN: Retrospective cohort analysis of private Health Plan-insured patients with visits at 18 Health Plan-affiliated neurology clinics. METHODS: PROMIS domains (Anxiety v1.0, Cognitive Function Abilities v2.0, Depression v1.0, Fatigue v1.0, Pain Interference v1.0, Physical Function v2.0, Sleep Disturbance v1.0, and Ability to Participate in Social Roles and Activities v2.0) are collected as part of routine care. Data from patients' first PROMIS measures between June 27, 2018 and April 16, 2019 were extracted and combined with claims data. Using (1) claims data alone and (2) PROMIS and claims data, we examined the association of covariates to utilization (using a logit model) and cost (using a generalized linear model). We evaluated model fit using area under the receiver operating characteristic curve (for unplanned care utilization), akaike information criterion (for unplanned care costs), and sensitivity and specificity in predicting top 15% of unplanned care costs. RESULTS: Area under the receiver operating curve values were slightly higher, and akaike information criterion values were similar, for PROMIS plus claims covariates compared with claims alone. The PROMIS plus claims model had slightly higher sensitivity and equivalent specificity compared with claims-only models. CONCLUSION: One-time PROMIS measure data combined with claims data slightly improved predictive model performance compared with claims alone, but likely not to an extent that indicates improved practical utility for payers.

Linking AM-PAC Cognition to PROMIS Cognitive Function.

OBJECTIVE: To link the Activity Measure for Post-Acute Care (AM-PAC) Applied Cognition to the Patient-Reported Outcomes Measurement Information System (PROMIS) Cognitive Function, allowing for a common metric across scales. DESIGN: Cross-sectional survey study. SETTING: Outpatient rehabilitation clinics. PARTICIPANTS: Consecutive sample of 500 participants (N=500) aged ≥18 years presenting for outpatient therapy (physical, occupation, speech). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: AM-PAC Medicare and Generic Cognition short forms and PROMIS Cognitive Function items representing the PROMIS Cognitive Function item bank. RESULTS: The calibration of 25 AM-PAC cognition items with 11 fixed PROMIS cognitive function item parameters using item-response theory indicated that items were measuring the same underlying construct (cognition). Both scales measured a wide range of functioning. The AM-PAC Generic Cognitive assessment showed more reliability with lower levels of cognition, whereas the PROMIS Cognitive Function full-item bank was more reliable across a larger distribution of scores. Data were appropriate for a fixed-anchor item response theory-based crosswalk and AM-PAC Cognition raw scores were mapped onto the PROMIS metric. CONCLUSIONS: The crosswalk developed in this study allows for converting scores from the AM-PAC Applied Cognition to the PROMIS Cognitive Function scale.

A scoping review of US insurers' use of patient-reported outcomes.

OBJECTIVES: To complete a scoping review of US health insurers' use of health-related quality of life (HRQOL) patient-reported outcome measures (PROMs). STUDY DESIGN: Literature review. METHODS: A literature search was constructed for articles that contained an insurer-related term and an HRQOL-related term between 1999 and 2019 using the MEDLINE, Embase, Web of Science, Cochrane Database of Systematic Reviews, EconLit, and Business Source Complete databases. RESULTS: The search identified 14,253 unduplicated records, of which 2340 passed abstract screening and 350 were included in the review. The populations addressed in these studies included both populations with specific health conditions (eg, diabetes) and an entire member population. The most common purpose of the article was to evaluate a policy or program (n = 255; 72.9%); the range of interventions evaluated included federal policy, subgroup management strategies, and identification of individual patients. The most common insurance mechanism was Medicare (n = 205; 58.6%). The most common source of data was collected specifically for a research project (n = 172; 49.1%), and the least common source of data was collected by providers at the point of care (n = 34; 9.7%). The most commonly addressed age group was 65 years and older (n = 262; 74.9%), and the least commonly addressed was younger than 18 years (n = 36; 10.3%). The most commonly used PROMs were single-item self-rated health (n = 138; 47.1%) and activities of daily living (n = 88; 30.0%), with validated depression questionnaires (n = 56; 19.1%) being the most common disease-focused questionnaire. CONCLUSIONS: This review found a wide variety of articles across insurance providers, health conditions, and uses of PROMs. There is a noted paucity of data in pediatric populations and little information about the use of data collected within health care settings that is transmitted to health insurers.

Linking AM-PAC Mobility and Daily Activity to the PROMIS Physical Function Metric.

OBJECTIVE: The purpose of this study was to link Activity Measure for Post-Acute Care (AM-PAC) Mobility and Daily Activity scales to the PROMIS Physical Function (PF) allowing for a common metric across scales. METHODS: Cross-sectional study of patients 18 years and older presenting to 1 of 8 outpatient rehabilitation clinics in southwestern Pennsylvania. Patients completed one survey with questions from the AM-PAC Daily Activity and Mobility short forms, and the PROMIS PF item bank. Using item response theory, 2 rounds of fixed-parameter calibration were performed. In the first, the AM-PAC Daily Activity and Mobility items were calibrated with 27 fixed item parameters from the PROMIS PF. Second, the AM-PAC Daily Activity items were calibrated with 11 PROMIS Upper Extremity fixed item parameters. This process uses the construct of physical function and equates AM-PAC items on the same underlying measurement scale for the PROMIS PF. RESULTS: Both scales measured a wide range of functioning and demonstrated good calibration. Data were appropriate for a fixed-parameter item response theory-based crosswalk. AM-PAC Daily Activity and Mobility raw scores were mapped onto the PROMIS PF metric. AM-PAC Daily Activity scores were also mapped onto the PROMIS PF Upper Extremity metric. CONCLUSION: Question items from the AM-PAC Daily Activity, AM-PAC Mobility, and PROMIS PF similarly measure the construct of physical function. This consistency allows for a crosswalk of AM-PAC scores onto the PROMIS PF metric. IMPACT: Crosswalk tables developed in this study allow for converting scores from the AM-PAC Daily Activity and Mobility scales to the PROMIS PF metric. This will facilitate monitoring of longitudinal change in function over time and across settings.

Uptake of an Integrated Electronic Questionnaire System in Community Pediatric Clinics.

OBJECTIVE: The study aimed to evaluate an integrated electronic questionnaire system implementation in outpatient community pediatric practices on workflow, completion rates, and recorded scores. METHODS: We evaluated the implementation and outcomes of an integrated electronic questionnaire system at 45 community pediatric practices that used standardized questionnaires to screen for autism, depression, and substance use and to measure asthma control. Electronic health record (EHR) data for all well child visits were extracted for the 3 months before and after implementation. We used statistical process control charts to evaluate questionnaire completion rates and Chi-square tests to evaluate screening completion and positive screening rates. The collection and entry of questionnaire information was observed and timed. RESULTS: EHR data included 107,120 encounters across 45 practices that showed significant and sustained improvement in completion rates for all questionnaires. The rate of recorded concerning questionnaires decreased for asthma control (19.3 vs. 12.8%, p < 0.001), stayed the same for autism (96.6 vs. 96.2%, p = 0.38), decreased for depression (9.5 vs. 6.7%, p ≤ 0.001), and increased for any substance use (9.8 vs. 12.8%, p < 0.001). Twelve practices were observed, and patient time and staff time managing questionnaires were decreased after implementation. DISCUSSION: Electronic questionnaire administration saved staff time and patient time. We report overall improvement in questionnaire completion rates, with notable variation in improvement in completion across practices and in change in concerning recorded result rates across measures. CONCLUSION: Conversion of four standard paper questionnaires to an integrated electronic system reduces patient and staff time while increasing completion rates when well integrated into routine care.