Using Patient-Reported Outcome Measures during Routine Care of Patients with Type 2 Diabetes [Internet]

Excerpt

Background: The use of patient-reported outcome measures (PROMs) in clinical care has been shown to improve the process and outcomes of care, but many challenges to their routine implementation have been noted and their use is limited. Traditional measures and existing models do not address broad functional status outcomes. PROMs can improve outcomes, particularly when used to manage treatment of specific conditions and can improve communication between patients and providers. Demonstrating models for using PROMs that are meaningful to patients and the clinical care team is a high priority and supports efforts to monitor quality of care.

Objective: This project seeks to assess the feasibility of using PROMs in routine care for people with diabetes.

Methods: At 2 clinics serving diverse populations, teams involving clinicians and patients developed a workflow to collect the PROMIS®-29 survey assessing 6 domains of physical, mental, and social health to guide goal setting at baseline and at 3-month follow-up. The study population was patients with type 2 diabetes who had a primary care visit and an HbA_1c ≥6.5%. We used chi-square and multivariable regression (controlling for site) to examine patient characteristics associated with participation in PROM data collection and goal setting. We followed PROMIS scoring guidelines; all PROMIS measures have been calibrated using the T score metric, in which scores have a mean of 50 and a standard deviation of 10, compared with the general population. We used bayesian methods to capture likelihood of effect sizes for various patient factors. We interviewed 36 patients and 13 clinicians and care managers to determine the acceptability and meaningfulness of gathering and using PROM data in patient care.

Results: Of 1864 patients in the target group, 26% (n = 490) of targeted patients participated; of these, 356 (73%) provided follow-up data. At site 1, 40% of patients set a health goal; at site 2, 90% of patients set a goal. Based on multivariable analyses controlling for site, participation in data collection differed by age, race/ethnicity, language preference, and presence of multiple chronic conditions. Patients aged 65 and older were less likely to participate in the initial PROM data collection (OR, 0.72; P = .03) and to complete all items (OR, 0.59; P = .09). Non-Hispanic Black participants were more likely to participate (OR, 1.24; P = .06), provide complete responses (OR, 1.71; P = .03), and set a goal (OR, 2.26; P = .04) than non-Hispanic White participants. For Spanish-preferred Hispanics, the opposite was true for providing complete responses (OR, 0.36; P = .00) and goal setting (OR, 0.71; P = .05). Patients with 2 to 4 chronic conditions were twice as likely than patients with diabetes only to participate at baseline (OR, 2.19; P = .01) and times as likely to set a goal (OR, 3.54; P = .08). The only factor significantly associated with participation at follow-up was presence of multiple chronic conditions: Patients with 5 or more conditions were 4 times as likely to participate in the follow-up PROM data collection (OR, 4.10; P = .03). Compared with national norms, our study population's worst functional areas were physical functioning and pain interference (ie, how pain interferes with a person's ability to participate in daily activities). For example, the average score for pain interference was 58.0 at site 1 and 61.1 at site 2 (higher score indicates worse functioning). At least 33% of patients had a clinically meaningful change between baseline and follow-up assessments (half an SD, or 5 points) on each domain. Nearly all patients at the site with dedicated care managers set a goal (90%), compared with 40% of patients at the site that incorporated data collection in primary care and other visits.

Results of bayesian analyses to examine the relationship of patient clinical and demographic characteristics with change scores on PROMIS scales suggest that having more chronic conditions, public insurance, and Hispanic ethnicity are likely to be associated with decreased functioning over time. Our exploratory analysis did not suggest that setting a goal related to mental health was related to improvement in mental health functioning.

Patients and care team members generally felt that PROM data can improve communication and lead to better engagement and specific care changes, but reported the need for better explanations of PROMs, and care team members noted the need for training and resources to conduct data collection and goal-setting conversations.

Limitations: The generalizability of these findings is limited because we conducted this study at only 2 sites, included only patients with type 2 diabetes, and used a single PROM. Low participation rates and differences in the timing of follow-up data collection at the 2 sites may have affected the PROM results.

Conclusions: Collecting PROM data in routine diabetes care is valued by patients and clinical care teams but is challenging to implement. The use of the PROMIS-29 in routine clinical care identified areas of functional limitations among people with diabetes, particularly in pain interference and physical functioning. Patients and care team members gave examples of how data influenced care.

Based on the qualitative interviews and site debriefings, we conclude the following: Careful implementation of PROMs into the clinical workflow should consider the target population; preparation and support of care teams and patients; better methods for collecting PROMs, particularly in diverse populations; better understanding of the human-factor issues involved in interpreting and using PROMs; and data on the patterns of change in PROMs over time and in relation to goal-setting interventions. In particular, research is needed to address the needs of diverse populations, taking into consideration race/ethnicity, language, and other factors. Broader implementation will likely require a clear business model that supports the investment in time and resources needed for successful implementation.