1 Introduction

Patient-reported outcome measures (PROMs) are important questionnaires that enable the assessment of a patient's perspective regarding their condition, symptoms, and health status.^1,2 Historically, in orthopaedic surgery, objective outcome measurements reported by a physician (e.g., grip strength, range of motion, fracture healing) were used to determine patient outcomes and used as measures of success.^3,4 As the healthcare system shifts toward one that values patient-centered care, PROMs are valuable tools to capture the patient's perspective.⁵ For instance, a patient with a healed distal radius fracture (DRF) may continue to experience pain or difficulty with activities of daily living, which may not be captured by objective measures. By incorporating PROMs into clinical practice, physicians can improve patient-physician communication, monitor response to treatment, and/or identify unrecognized clinical issues, ultimately improving patient outcomes and experiences.^1,6,7

PROMs play a critical role in the management of patients with DRFs. For example, they can be utilized to understand post-operative outcomes clinically and in research settings.^8,9 PROMs can also be utilized to identify patients who may benefit from ancillary services (e.g. hand therapy, social work)¹⁰ and help guide physicians on how quickly patients can transition back to work.¹¹ Furthermore, PROMs can be utilized in shared decision-making processes to help align patient values and preferences with treatment plans.⁶ Several PROMs have been developed and are utilized for patients with DRFs; the most commonly utilized include Disabilities of the Arm, Shoulder, and Hand (DASH); Patient-Rated Wrist Evaluation (PRWE); and Medical Outcomes Study: 36-Item Short Form Survey (SF-36).^12–14 The development of PROMs involves several key steps, including the identification and definition of the construct (e.g., physical function) being evaluated, the creation and iteration of items/questions, and psychometric testing (e.g., to evaluate reliability, the presence of ceiling/floor effects).¹⁵ After development, PROMs undergo continuous instrument improvement and validation testing to ensure the PROM functions effectively in particular patient populations for a particular purpose.^1,2,16 However, the population in whom PROMs are developed and validated is foundational, as the PROM scores from one population may be different from those of another population.¹⁷ For instance, if a PROM was developed and validated primarily in a predominantly White, high-income patient population, its results may not accurately capture the experiences of racially and ethnically diverse patients from lower-income backgrounds. Therefore, if the patient population in whom PROMs are developed varies from that in whom PROMs are measured, the results may not be valid.

As such, we aim to systematically evaluate the demographic representation of participants in the development studies of the most commonly used PROMs for DRFs. We sought to assess the extent to which key demographic variables such as race/ethnicity, sex, socioeconomic status, and language are reported in the original development studies of these PROMs and compare the reported research population demographics to the United States 2020 census, the most recent comprehensive U.S. population data. Our null hypothesis is that the participant demographics reported in the studies represent a heterogeneous group that reflects the United States population in whom these PROMs are utilized.

2 Methods

2.3

2.3 Selection criteria

The search was limited to studies written in English with no timeframe restriction. Studies that met the following criteria were included in the review: (1) included a PROM used for the evaluation of patients with DRF, both generic and specific; (2) original development studies; (3) conducted with adults (age 18 or older); (4) with available full-text. We defined the development of a tool to include (1) generation/creation/opinions of survey questions and (2) item reduction where participants gave their own input. We considered non-development studies to include (1) validity, reliability, and responsiveness testing and (2) item reduction based on statistical analysis.

Common DRF PROMs were included based upon the review from the Distal Radius Outcomes Consortium, which includes: Disabilities of the Arm, Shoulder and Hand (DASH); QuickDASH (QDASH); Michigan Hand Outcomes Questionnaire (MHQ); Brief MHQ; Patient-Rated Wrist Evaluation (PRWE); Patient-Reported Outcomes Measurement Information System (PROMIS) – Upper Extremity; PROMIS – Physical Function; Medical Outcomes Study: 36-Item Short Form Survey (SF-36); EuroQol- 5 Dimension (EQ-5D).

Studies were imported into Rayyan, a web application designed for performing systematic reviews,¹⁹ and duplicates were removed. Two reviewers (ZR and SU) independently screened the title and abstracts for initial eligibility. The full texts of initially eligible records were then retrieved and reviewed by two independent reviewers based on specified inclusion and exclusion criteria. Studies were included if they evaluated PROMs for DRF, including both generic and condition-specific measures. Only original development studies were considered for inclusion, and all studies were required to involve adult participants (≥18 years old). Studies were excluded if they focused on validating the conclusions of an original development study or if they assessed PROMs in specific populations outside the original study, such as individuals with stroke, depression, or HIV/AIDS. Additionally, articles classified as comments, letters, editorial guidelines, conference reports, or reviews were excluded. Studies involving pediatric populations were not considered, and any study for which the full text was unavailable was also excluded from the review. In instances where consensus was not achieved, a third reviewer was consulted to reconcile conflicts (LMS). The PRISMA flow diagram summarizes the literature search, screening, and review as illustrated in Fig. 1.

Close

Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart of study selection.

Export to PPT

3 Results

The initial search resulted in 9323 records. After duplicates were removed, 5517 records remained. Twelve full-text studies fit the inclusion criteria for the development of ten commonly used PROMs in DRFs. Two studies were associated with the DASH and were aggregated. One article related to the QuickDASH provided similar information to the DASH. EQ (5)DL actually is informed by four publications, while the remaining PROMs (PRWE, PROMIS–UE, PROMIS–PF, MOS SF-36) were each represented by a single study. The PROMs and the relevant demographic information provided are illustrated in Table 2. Studies describing the development of the DASH reported the highest number of variables (9 in total), followed by PROMIS-PF with 7 variables. The PRWE reported the fewest number of variables.

The percentage of studies reporting each variable is shown in Table 3. The most common demographics reported were age (83 %), sex (75 %), and education (67 %), followed by race/ethnicity (34 %), work status (34 %) language spoken (25 %), income (17 %), and worker's compensation status (17 %).

Among the demographics reported across all the PROMs, 81 % of participants were white, 54 % were female, 83 % spoke English, and 74 % had some college education or degree. Compared to the U.S. 2020 Census, the study participants were most likely to be white (81 % vs. 62 %, p < 0.001) and female (54 % vs. 51 %, p < 0.001), less likely to be English speaking (83 % vs. 84 %, p = 0.006), and had above a high school education (74 % vs. 61 %, p < 0.001) (Table 4).

4 Discussion

The findings of this systematic review highlight a gap in the representation of participants in the development studies of the PROMs utilized for patients with DRFs. Demographic variables were inconsistently reported, with a minority of these studies providing demographic data on race/ethnicity and language spoken. The groups most represented were those who were white, English-speaking, and female with some college or higher-level education, which did not match the observed distribution of the U.S. 2020 census. Although we found a statistically significant difference in English-speaking participants, the majority of participants were still English-speaking, and only 25 % of studies reported language as a variable. While statistically significant, the difference may not be clinically relevant. Thus, despite the widespread development and use of PROMs, the populations in whom these tools were developed may not represent the diverse populations in which they are applied. Critically assessing the populations in which these measures were originally developed, along with their validation and adaptation for specific groups, is essential to ensure appropriate application across diverse clinical populations.

Patient-reported outcome measures (PROMs) are valuable tools in clinical practice, offering insights into patients' health status, recovery, and quality of life, nuanced clinical issues, and guiding clinical management or patient behavior.²¹ However, the omission of representation in the development of PROMs, such as education level, sex, race/ethnicity, and language, can lead to possible biases and misinterpretations. For example, Greene et al., found that patients with higher educational attainment reported better health-related quality of life (HRQoL) and experienced less postoperative pain compared to those with lower education levels,²² suggesting that education can shape a patient's perception of their health status and recovery status. Similarly, sex-related differences in PROMs have been documented, with female patients undergoing reverse total shoulder arthroplasty reporting significantly lower pre-operative PROMIS Physical Function scores compared to their male counterparts,²³ highlighting the risk of misrepresenting preoperative health status and recovery for female patients (and other patients' less commonly represented) if such differences are not accounted for. Racial, ethnic, and language disparities in PROM outcomes are also well documented. In a study by Kim et al., minority patients scored lower on the BREAST-Q, a PROM in implant-based breast reconstruction.²⁴ Izadi et al. demonstrated that the PROMIS Physical Function Short Form 10a showed weaker correlations with clinical measures among Chinese speakers and Hispanic patients and was less responsive to clinical improvements among Medicaid recipients and Spanish speakers, indicating that language proficiency affects PROM validity and responsiveness.²⁵ An example of how PROMs can reflect narrow cultural assumptions is seen in the original PRWE tool, which initially included a question about cutting meat, assuming dietary habits that may not apply to all individuals, particularly those who abstain from meat for cultural or religious reasons. Recognizing this limitation, the PRWE was recently updated to remove culturally specific language and improve its inclusivity.²⁶ As appropriate PROM use may lead to more accurate scores, the underrepresentation of diverse populations in PROM development risks perpetuating disparities, ultimately leading to inequitable outcomes and missed opportunities for shared decision-making, improved patient-physician communication, and other interventions.²⁷

For the purposes of research, the lack of demographic representation poses additional challenges. When PROMs do not reflect the populations to which they are applied, the data that is retrieved may be skewed and/or not applicable to all. Missing data in PROMs —especially when missingness is related to the outcome itself (i.e., not missing at random) —can introduce bias, reduce statistical power, and result in misleading conclusions, affecting the validity of research findings.²⁸ When certain patient subgroups are underrepresented or excluded due to incomplete data, the resulting PROM scores may not accurately reflect the broader population, ultimately compromising the validity and applicability of research findings. This presents challenges as health conditions can differ by demographic factors; individuals from different backgrounds experience diseases at different rates, navigate the healthcare system differently, and may have distinct perceptions of diagnoses and treatment outcomes. For example, sex-, race-, and literacy-based differences in DRF care and outcomes highlight the importance of demographic factors in outcomes research. Older women have a higher risk for DRF than older men, while Black, Latino, and lower literacy patients have worse outcomes and/or PROM scores than White patients and those with high literacy.^29–31 If PROMs that are primarily developed and validated in male, White, and high literacy populations are utilized in shared decision making processes or to assess post-operative recovery, they may fail to capture sex-, race-, and literacy-specific risk profiles, recovery patterns, and care needs. Without including these factors in PROM development, research may overlook key disparities that impact patient care. Although age and sex/gender are often included in development studies, incorporating additional factors such as education level, race/ethnicity, work status, and language is essential as these variables are associated with health outcomes.^32–34 Without such inclusion, underrepresented groups may be excluded, and PROMs designed primarily white, well-educated, English-speaking women, as in this study, may fail to capture the unique health experiences of other demographic groups.

Furthermore, the changing healthcare policy landscape underscores the importance of optimizing PROMs for patient care and reimbursement. For example, the Centers for Medicare & Medicaid Services (CMS) introduced the Total Hip Arthroplasty/Total Knee Arthroplasty (THA/TKA) Patient-reported Outcome-based Performance measure (PRO-PM), that requires the collection of PROM data for eligible patients.^35,36 CMS has also detailed “substantial clinical benefit” (SCB) thresholds for PROMs that, although not formally announced, may soon be utilized in reimbursement models. As patient demographics have known associations with the measurement of PROMs and changes in PROM scores, the collection and incorporation of such demographics is foundational to ensure such policies do not result in biased reimbursements, potentially favoring institutions that serve well-represented populations while disadvantaging those caring for marginalized groups, further perpetuating health disparities.

Although the collection of every variable and the inclusion of patients from all demographics pose challenges, we advocate for transparent collection and reporting of variables known to influence outcomes, and additionally, to include patients representative of the populations treated. For example, the Hand Surgery Quality Consortium, informed by a systematic review, conducted a multi-stakeholder Delphi process to set a national agenda for data collection in hand surgery that can be utilized to collect and report important demographic data.³⁷ To enhance representation, researchers can engage in community partnerships to recruit diverse populations and develop culturally translated materials to ensure broader applicability.^38,39 Collaborations with diverse research partners and those employing strategies like weighted sampling, as seen in initiatives like PROMIS, offer promising opportunities to further improve representation.⁴⁰ Additionally, journals should continue to encourage the reporting of demographic variables such as age, sex, gender, race, and ethnicity to increase transparency and improve diversity in patient sampling.^41,42 This transparency helps to identify potential disparities and improves the generalizability of PROMs in clinical and research settings.

Limitations of this review should be noted. Firstly, the demographic information is drawn from the publications detailing tool development, limiting the results to the details disclosed by the authors. As such, the populations in which these PROMs were developed may be more diverse than demonstrated in the results and discussion. The authors were not contacted for further information. Additionally, while we conducted multiple searches using predefined criteria to maximize the comprehensiveness of our review, it is possible that some relevant studies were missed. Finally, this review was limited to development studies and did not include validation studies which are designed to further investigate the nuances of the PROM utilization. For example, validation studies may aim to specifically evaluate a PROM's performance in a specific patient population. A recent systematic review evaluating the validation of PROMs for DRFs in Spanish-speaking populations found minimal validation and development of Spanish-language adaptations, highlighting a broader opportunity in PROMs research. These findings underscore the need for further research to ensure that PROMs are not only developed but also thoroughly tested and adapted for use in diverse populations.⁴³

5 Conclusion

PROMs are increasingly used in clinical practice and research to capture the patient's perspective; however, despite their widespread adoption, their applicability across diverse populations is limited. Prioritizing diversity in the initial development studies of these tools may help to ensure adequate representation of important patient variables such as race/ethnicity, language, work status, and education level. PROMs that are developed with biased findings may produce biased results and ultimately fail to capture the experiences of underrepresented patient groups. If developing a new PROM, transparent reporting and diverse representation is needed. If using an existing PROM, it may need to be adapted to be applicable to all populations. Further efforts are warranted to expand the collection of demographic variables and the inclusion of diverse patients to ensure that PROMs are both inclusive of and applicable to the patients we treat.

Authors contribution

oSciaska N. Ulysse – Investigation, Formal analysis, Roles/Writing – original draft, Writing – review & editingoZuivanna Rivas – Conceptualization, Investigation, Formal analysis, Roles/Writing – original draftoBrocha Z. Stern – Writing – review & editingoRobin N. Kamal – Writing – review & editing, SupervisionoJessica L. Abrolat – Investigation, Writing – review & editingoChristopher F. Orozco – Investigation, Writing – review & editingoGopal Lalchandani – Conceptualization, Writing – review & editingoLauren M. Shapiro – Conceptualization, Funding acquisition, Writing – review & editing, Supervision

Institutional ethical committee approval (for all human studies)

N/A.

Informed consent (patient/guardian), mandatory only for case reports/clinical images

No human subjects were involved in this study. Therefore, informed consent was not required.

Funding/sponsorship

Some of the authors have received funding to support this work. This work was supported by a National Institutes of Health (Grant #K23AR082960; Principal Investigator, LMS) award and the Orthopaedic Research and Education Foundation (Principal Investigator, LMS). The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of these organizations.