Proxy-reported outcomes accurately reflect objective patient-reported outcomes in older adult patients with traumatic orthopaedic injuries

Abstract

Background

Older adults make up an increasing portion of orthopedic trauma care. Proxy reports are particularly valuable when patients face difficulties formulating answers due to pre-existing or temporary cognitive impairment, and provide critical insights into patient well-being.

Questions/purposes

This study examines the agreement between patient- and proxy-reported outcome measures across various health domains of older adult orthopedic trauma patients, including those with mild cognitive impairment.

Patients and methods

A prospective cohort study was conducted in the Orthopedic trauma clinic of two Level 1 trauma centers, involving 108 patients aged 70 years or older, with or without mild cognitive impairment, and a self-identified discernible proxy. Participants were evaluated using PROMIS measures for physical function, pain intensity, anxiety, depression, fatigue, and social roles and (instrumental) activities of daily living questionnaires. Agreement between patient and proxy assessments was analyzed using Intraclass Correlation Coefficient and Bland-Altman analyses. Subgroup comparison was made using confidence intervals.

Results

Most patients were married women with a mean age of 78 years (SD 6.2), 61 % had higher education, 67 % were walking independently, and 53 % had fractures of the pelvic ring or femur. Proxies were younger (mean age 64 years, SD 15), with 41 % being children and 40 % spouses. Significant correlations were found between patient and proxy assessments in physical function (ICC 0.74, 95 % CI 0.61-0.82), pain intensity (ICC 0.83, 95 % CI 0.74-0.89), (instrumental) activities of daily living (both ICC 0.86, 95 % CI 0.80-0.91). Moderate correlations were observed in anxiety, depression, fatigue, and participation in social roles (ICCs ranging from 0.56 to 0.66). Bland-Altman analyses confirmed good agreement with none to minimal systematic bias across all these domains.

Conclusion

Proxy-reported outcomes are valuable for evaluating health domains in older adult orthopedic trauma patients, including those with cognitive impairments. While proxies reliably assess physical and pain-related domains, additional strategies are needed to improve accuracy in more subjective domains. Future research should explore longitudinal agreements to better understand recovery perceptions over time and mitigate bias proxy-reporting, ultimately enhancing both patient care and outcomes research in geriatric orthopedic trauma patients.