e23192 Background: Patient-reported outcome measures (PROMs) are critical for patient-centered care in oncology, as they allow patients to communicate with their care team about a range of symptoms using well-validated tools. However, successful use of PROMs in routine care depends on the clinical user experience: the opinions and experiences of the clinicians and staff who administer and make treatment decisions using PROMs. This study characterizes the use of the NIH-sponsored Patient Reported Outcomes Measurement Information System (PROMIS) in routine oncological care through a literature review and user survey. Methods: First, we conducted a narrative review of papers published from 2010 to June 2023 to identify breadth of PROMIS use in routine care. We screened 262 abstracts and identified 10 articles describing routine care in patients with cancer. We coded for institution type, PROM administration methods, and types of PROMIS measures. Next, we conducted a survey to solicit the opinions and experiences of diverse stakeholders implementing PROMIS in routine care. We contacted 209 known users of PROMIS (published authors, PROMIS API clients, and respondent referrals). Item frequencies were used to characterize survey responses. Chi-square tests were used to test differences between users in oncology and other fields. Results: Review. Nine of the 10 institutions using PROMIS in routine cancer care were academic medical centers. Articles reported on data from 2011-2021. The most popular PROMIS measure was PROMIS Global Health which provides summary scores of physical and mental health. Other popular domains included fatigue (4), physical function (3), depression (3), and anxiety (3), all measured by PROMIS computerized adaptive tests (CATs). They were most often administered in the clinic waiting room (7) or online before appointments (6). Most institutions (6) had PROMIS results automatically integrated into the electronic health record. Survey. We had 89 respondents (6 from oncology). PROMIS users in oncology were more likely to have structured training on the use of PROMIS than other users (χ2 (1,79) = 6.0, p = 0.01) and less likely to give patients direct access to their results (χ2 (1,72) = 6.0, p = 0.01). Users in oncology did not differ from the whole sample in opinions of PROMIS (% agree): users viewed PROMIS as worthwhile (86%), clinically relevant (85%), easy to administer (76%), statistically reliable (73%), and beneficial to treatment monitoring (68%), but were less confident that PROMIS improves doctor-patient communication (57%), treatment decision-making (54%), or patient satisfaction (35%). Conclusions: Conclusions PROMIS has been successfully integrated into routine oncological care in several academic medical centers. While adoption is not widespread, these initial results suggest that PROMIS is a useful tool for symptom monitoring in cancer care.
Read full abstract