Psychometric properties of the Pediatric Evaluation of Disability Inventory - Patient Reported Outcome: A cognitively accessible measure of functional performance.

Psychometric properties of the Pediatric Evaluation of Disability Inventory – Patient Reported Outcome: A cognitively accessible measure of functional performance - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Dev Med Child Neurol . Author manuscript; available in PMC: 2026 Jun 1. Published in final edited form as: Dev Med Child Neurol. 2024 Nov 4;67(6):770–778. doi: 10.1111/dmcn.16117 Psychometric properties of the Pediatric Evaluation of Disability Inventory – Patient Reported Outcome: A cognitively accessible measure of functional performance Jessica M Kramer 1 , Pengsheng Ni 2 , Beth Pfeiffer 3 , Andrew Persch 4 , Fiorella Guerrero Calle 1 , Ariel Schwartz 5 , Elizabeth Barbour 1 , Daniel K Davies 6 Author information Article notes Copyright and License information 1 Department of Occupational Therapy, University of Florida, Gainesville, FL, USA 2 Biostatitics & Epidemiology Data Analytic Center, Boston University School of Public Health, Boston, MA, USA 3 Department of Health and Rehabilitation Sciences, Temple University, Philadelphia, PA, USA 4 Department of Occupational Therapy, Colorado State University, Fort Collins, CO, USA 5 Institute on Disability, University of New Hampshire, Durham, NH, USA 6 AbleLink Smart Living Technologies, LLC, Colorado Springs, CO, USA ✉ Correspondence: Jessica M. Kramer, Jessica.kramer@phhp.ufl.edu , Department of Occupational Therapy, College of Public Health and Health Professions, University of Florida, PO BOX 100164, Gainesville, FL 32610-0164, USA. Issue date 2025 Jun. PMC Copyright notice PMCID: PMC12049558  NIHMSID: NIHMS2028506  PMID: 39497492 The publisher's version of this article is available at Dev Med Child Neurol Abstract Aim: To evaluate the structural validity, internal reliability, and measurement invariance of the Pediatric Evaluation of Disability Inventory – Patient Reported Outcome (PEDI-PRO), a measure of functional performance of discrete tasks required to participate in everyday life situations important for adulthood. Method: This was a cross-sectional study with 306 young people aged 14 to 22 years (mean 18 years 10 months, SD 2 years 5 months) with developmental disabilities (43.1% autism spectrum disorder only, 23.9% intellectual disability, 17.6% other disability, 11.4% both autism spectrum disorder and intellectual disability, 3.9% missing) completed the PEDI-PRO. Following COnsensus-based Standards for selection of health Measurement INstruments (COSMIN) criteria, we conducted a confirmatory factor analysis, applied a Rasch rating-scale model, examined Cronbach’s alpha, Rasch person reliability and separation coefficients, and differential item functioning (DIF). Results: Structural validity was good for the daily activities and mobility domains, and acceptable for the social/cognitive domain. The 3-point Likert response scale functioned as intended. All domains demonstrated acceptable internal consistency on all criteria. One or two items in each domain demonstrated DIF, but the impact on all domain scores was less than 1.0 threshold. Interpretation: The cognitively accessible design and innovative conceptual measurement framework probably contributed to these promising findings. The PEDI-PRO addresses a gap in high-quality patient-reported outcome measures that assess priority outcomes for young people with developmental disabilities. Young people with developmental disabilities, including intellectual disability and autism, have the right to be involved in discussions about their strengths and goals for their health and wellbeing. 1 – 5 Patient-reported outcome measures (PROMs) are one way to gather direct reports and facilitate the engagement of young people in their rehabilitation. 6 , 7 Professionals have a responsibility to adopt rigorous, high-quality PROMs that are accessible and appropriate for the specific characteristics and needs of each young person. 8 , 9 Although the number of available PROMs is increasing, most have been developed for the general population and research suggests the measures may not have the same relevance or psychometric properties for young people with developmental disabilities. 10 , 11 A review of self-reported outcome measures for young people with developmental disabilities indicates there is limited psychometric evidence for PROMs for this respondent group, particularly for PROMs addressing their priorities. 12 These priority outcomes span the activity and participation domains of the International Classification of Functioning, Disability and Health including communication, self-care, mobility, community and social life, independent living, leisure, and physical activity. 13 – 15 To ensure young people can report on their readiness for success in these areas, there is a need for high-quality PROMs that evaluate functional abilities in the context of participation in everyday life. 16 The Pediatric Evaluation of Disability Inventory – Patient Reported Outcome (PEDI-PRO) measures the functional performance of discrete daily activity, mobility, and social/cognitive tasks required to participate in everyday life situations important for success in adulthood. The PEDI-PRO is administered on a computer or mobile device, and incorporates cognitive accessibility design features to enhance comprehension, independent use, and quality reporting of young people aged 14 to 22 years. 9 Previous research, including collaborative design with young people, established the content validity of the PEDI-PRO. 5 , 17 , 18 Occupational and physical therapists piloting a beta-version of the PEDI-PRO software reported excellent usability, and test–retest reliability was acceptable. 19 In this study, we evaluated the psychometric properties of the PEDI-PRO software following standards put forth by COnsensus-based Standards for selection of health Measurement INstruments (COSMIN). 12 , 20 Research questions included the following. What is the structural validity of the PEDI-PRO domains and response scale? What is the internal reliability of the PEDI-PRO domains? Is there evidence of measurement invariance, as indicated by stable item difficulties across gender, age, and racial/ethnic identity groups? METHOD Participants Following approval by the institutional review board from the University of Florida, young people were invited to participate in the study if they (1) met criteria for developmental disability as defined by the US Developmental Disabilities Act, section 102, 8 (2) were aged 14 to 22 years, (3) able to understand and express basic feelings, behaviors, and actions in English or using augmentative and alternative communication, and (4) were able work on a task for 15 minutes with support. Participants were recruited across the USA from August 2021 to March 2023 through professional contacts and organizations serving people with disabilities (e.g. schools, community programs), advertisements on Facebook, and research databases that provide verified disability information. Informed written consent was obtained for all participants. Young people with guardians provided informed assent using an accessible process and had the final decision to participate. Instruments: the PEDI-PRO The PEDI-PRO produces criterion-scaled scores in three domains: daily activities, social/cognitive, and mobility. These align with the three domains of the parallel parent-reported measure, the Pediatric Evaluation of Disability Inventory - Computer Adaptive Test (PEDI-CAT). 21 A 3-point response scale is used for all domains: a little hard, a little easy, very easy. In addition, the PEDI-PRO includes a ‘skip’ response option for specific functional tasks that are not relevant for a young person (e.g. not all young people may ‘put on a bra’). The ‘Learn how to use the PEDI-PRO’ module provides instruction, examples, and practice items with direct feedback to support appropriate use of the response categories. The PEDI-PRO software uses an innovative measurement framework, in which items aligned with the three domains are embedded in six everyday life situations: 5 working at a job, getting ready in the morning, going to a restaurant, cooking a meal, sports and exercise, and going to the doctor. For example, the everyday life situation of going to the doctor includes ‘peel the back off a medicine packet’ (daily activities, 36 items tested), ‘politely ask my doctor if I can change my medication or care’ (social/cognitive, 32 items tested), and ‘roll on my side on the doctor’s table’ (mobility, 28 items tested). Items selected for testing in this study were developed in previous research 18 and rated as important by a team of young adult researchers with developmental disabilities, and an advisory board that included occupational and physical therapists working in school, hospital, and outpatient/community rehabilitation settings. Procedures For this cross-sectional study, trained research staff met with young people virtually, or in a research laboratory, or in a community-based setting. Regardless of the context, all participants viewed and directly controlled the PEDI-PRO software. Participants completed the ‘Learn how to use the PEDI-PRO’ module and were required to demonstrate appropriate use of the response scale (e.g. selecting response categories that correspond with the demonstrated ability of standardized examples shown in videos). Then staff administered everyday life situations in counterbalanced order. Staff provided think-out-loud prompts as young people answered each item (e.g. ‘why is it a little easy to do this?’) but faded prompts as young people spontaneously provided information. Previous research indicates that this think-out-loud approach supports reflection and improves quality of responses 18 (see Appendix S1 for the PEDI-PRO administration flowchart and prompting structure). Participants who were unable to demonstrate appropriate understanding of the PEDI-PRO response scale or items in the ‘Learn how to use the PEDI-PRO’ module after two attempts and/or in their think-out-loud prompts were determined ineligible and dropped from the study ( n = 3, not included in final reported sample size). To prevent missing data, the PEDI-PRO software forces a response choice to continue to the next screen (a little hard, a little easy, very easy, or skip). Participants still had the autonomy to stop at any time; this happened rarely, as the frequency of unadministered questions was low (daily activities, mean 0.58%; social/cognitive, mean 0.62%; mobility, mean 0.61%). When needed to sustain attention and quality self-reporting, 28 young people (9.2%) completed the PEDI-PRO over two sessions, within 2 weeks. 19 Following completion of the PEDI-PRO, young people and their parents/guardians completed a brief demographics questionnaire. Self-reported disability category and IQ scores were verified using participants’ individualized education plans and educational records when available and provided by participants/guardians (verified disability category = 60.5%, verified IQ = 43.8%). Analysis Sample sizes met COSMIN criteria for very good for all analyses. 20 All analyses were conducted with available data without imputation. If necessary demographic variables were missing, the participant was dropped from measurement invariance analyses. Structural validity Structural validity was evaluated in four ways. First, a confirmatory factor analysis (CFA) based on a polychoric correlation matrix using Mplus 8.10 determined whether each domain’s item set measured a common construct. 22 Fit indices of the comparative fit index greater than 0.95, Tucker–Lewis index greater than 0.95, and root mean square error of approximation less than 0.08 are good, 23 , 24 although comparative fit index and Tucker–Lewis index greater than 0.90 may be acceptable. 25 If CFA criteria are not met, exploratory factor analysis can provide evidence for unidimensionality when the percentage of variance explained by the first factor is greater than 20% and when the ratio of the first and second eigenvalues is greater than 4. 23 Second, absolute residual correlations less than 0.2 between domain items indicated local independence. 23 To determine the percentage of absolute residual correlations greater than 0.2 likely to occur in a unidimensional scale, we generated 1000 data sets in Mplus for each domain and fitted each simulated data set to a unidimensional CFA model. 26 The percentage of simulated data sets with at least one absolute residual correlation greater than 0.2 varied by domain: 92.27% of social/cognitive datasets, followed by 63.9% of mobility data sets, and 29.5% of daily activities data sets. Therefore, we maintained one or two item pairs with absolute residual correlations greater than 0.20 in each domain to maximize item difficulty and maintain content validity. Content validity was determined by our research team which has expertise in rehabilitation, by our young people collaborators, and informed by past content validity research. 5 Third, we evaluated each domain for unidimensionality using the Rasch rating-scale model (RSM), with poor fit to the model indicated by infit and outfit mean squares between 0.6 and 1.4. 27 We conducted the RSM calibration in Winsteps, 28 which uses joint maximum likelihood estimation, allowing the inclusion of all available responses using the response scale (‘very easy’, ‘little easy,’ ‘little hard’), and treats ‘skip’ responses as missing. Fourth, the structural validity of the response scale was examined using the RSM, with monotonically increasing Rasch-Andrich threshold parameters and observed average measures indicating appropriate interpretation of the response scale. 29 Internal consistency Internal consistency reliability was examined using several approaches. First, Cronbach’s alpha was calculated on the basis of a polychoric correlation matrix. 30 Second, the omega coefficient was calculated on the basis of the parameters generated from the unidimensional CFA model. 31 For both, values greater than 0.70 were considered acceptable. 32 Internal consistency was also evaluated using the Rasch person reliability and separation coefficients (non-extreme scores only, as identified by Winsteps software), following COSMIN recommendations. The reliability coefficient measures the percentage of true variation of person scores within the observed variation of the person scores. The person separation coefficient measures the ability of the domain to reliably distinguish between persons of different ability levels. Higher coefficients indicate that a domain contains a sufficient number of items for accurate measurement, with reliability coefficient greater than 0.80 or separation coefficient greater than 2.0. 33 Measurement invariance Differential item functioning (DIF) occurs when young people with the same ability level but different characteristics (e.g. male versus female) respond to the same item in a different way. We examined DIF using a multistep process for three dichotomous subgroups: gender (male n = 199; female n = 98), age (<19 n = 144; 19–22 n = 162), and racial/ethnic identity (White/non-Hispanic n = 128; any other race or ethnic identity n = 164). First, at the model level, we included interaction terms in the RSM between the demographic variable and item difficulty parameters. This extended model was compared with the original RSM using likelihood ratio test; p > 0.05 indicates no DIF in the model. When p < 0.05, we examined the individual item and demographic (gender, age, racial/ethnic identity) interaction terms to identify items with significant DIF, as indicated by adjusted p -values (Benjamini–Hochberg process). Third, for items identified with DIF, we examined item characteristic curves for each subgroup and calculated the weighted area beneath the curve, the absolute differences of item characteristic curves weighted by the sample score distributions. We used a value of the weighted area beneath the curve greater than 0.18 to identify DIF that could produce score differences across subgroups. 34 Fourth, to examine the impact of DIF on domain score, we refitted the RSM including the interaction terms for those items with a value of the weighted area beneath the curve greater than 0.18, and calculated the separated test characteristic curves for each subgroup. If the maximum difference in expected test score was less than 1 point, we concluded that DIF had negligible impact on test score. RESULTS Participants Three-hundred and six young people ( Table 1 ) with developmental disabilities participated, of whom 65.03% identified as male. Across all reported disabilities (participants could report multiple), most reported autism spectrum disorder only (43.1%), followed by intellectual disability (without autism spectrum disorder; 23.9%), other disability (17.6%), and both autism spectrum disorder and intellectual disability (11.4%); 3.9% had no specific disability category reported and identified only generally as a person with a developmental disability. More than half of young people (53.6%) identified as African American/Black, Asian, Hispanic, or any race or ethnicity other than White, non-Hispanic. Many people did not report income (38.89%); however, of those who did, 40.1% reported income less than US$49 999. TABLE 1. Participants characteristics Demographic n = 306 (%) Gender identity Female 98 (32.03) Male 199 (65.03) Other 8 (2.61) Missing/not provided 1 (0.33) All reported disabilities category (primary and secondary) a Autism spectrum disorder 167 (54.58) Intellectual disability 108 (35.29) Speech or language impairment 96 (31.37) Other health impairment 74 (24.18) Specific learning disability 32 (10.46) Other disability 31 (10.13) Orthopedic impairment (e.g. cerebral palsy) 18 (5.88) Multiple disabilities 17 (5.56) Visual impairment 12 (3.92) Emotional disturbance 11 (3.59) Hearing impairment 7 (2.29) Traumatic brain injury 5 (1.63) Deafness 3 (0.98) Missing/not provided b 12 (3.9) Ethnic identity Not Hispanic or Latino(a) 232 (75.82) Hispanic or Latino(a) 55 (17.97) Missing/not provided b 19 (6.21) Racial identity White 167 (54.58) African American or Black 74 (24.18) Asian 13 (4.25) Multiple races 17 (5.56) Other racial identity 9 (2.94) American Indian/Alaskan Native 2 (0.65) Missing/not provided b 24 (7.84) Income (US$) Less than 22 000 27 (8.82) 22 000–30 999 22 (7.19) 31 000–49 999 26 (8.50) 50 000–74 999 26 (8.50) 75 000–99 999 33 (10.78) Above 100 000 53 (17.32) Do not know 60 (19.61) Missing/not provided b 59 (19.28) Open in a new tab a Percentages exceed 100% because participants could report multiple secondary disability categories. Categories are aligned with individualized education plan qualifying categories as defined in the Individuals with Disabilities Education Act. b Participants/guardians could choose not to respond to these questions. Participants were aged 14 to 22 years (mean 18 years 10 months, SD 2 years 5 months). Of the 152 participants who provided IQ scores (88.2% verified), intellectual function ranged from 22 to 126 (mean 69.35, SD 19.41) with a median of 68.00; only two participants, both with verified scores, had an IQ less than 35 (a score historically associated with the most significant needs). Response frequencies and expected score range In all domains, the response category ‘very easy’ was selected most frequently, but response distribution varied by item in each domain ( Table 2 ). Only one item, ‘wash my hands’ (daily activity) had more than 90% of responses in the highest response category (very easy). Respondents chose to skip a response 15.69% or less of the time across all domains and items with the exception of three daily activity items related to gender performance: shave my face (50.33%), put my hair in a ponytail (62.09%), and put on a bra (66.67%). Owing to the high number of ‘skip’ responses, these three daily activity items related to gender performance were not included in further analyses. Following the analytical steps described below, the expected score for daily activities ranged from 27 to 81 (mean 65.7036, SD 10.8757), social/cognitive ranged from 24 to 72 (mean 55.2575, SD 9.3955), and mobility ranged from 21 to 63 (mean 51.6498, SD 8.6056). TABLE 2. PEDI-PRO response scale use and structural validity Domain and response scale Total response a (%) Response for domain items, % (minimum–maximum) Observed Average Measure Infit Outfit Rasch-Andrich Threshold parameter Daily activities Very easy 54.06 17.97–92.16 1.62 1.02 1.06 0.9 A little easy 19.64 4.58–33.33 0.6 0.96 0.86 −0.9 A little hard 15.04 2.29–34.31 −0.24 1.02 1.01 None Skip 11.26 0.33–66.67 N/A N/A N/A N/A Social/cognitive Very easy 47.87 31.7–76.47 1.17 0.98 1.00 0.27 A little easy 28.67 14.71–36.93 0.39 0.91 0.87 −0.27 A little hard 19.51 7.19–29.41 −0.16 1.05 1.12 None Skip 3.95 0.33–15.69 N/A N/A N/A N/A Mobility Very easy 58.56 31.37–89.22 1.68 1.02 1.07 0.12 A little easy 21.54 5.88–35.95 0.52 0.89 0.73 −0.12 A little hard 16.67 3.27–30.72 −0.26 1.07 1.11 None Skip 3.23 0.65–14.38 N/A N/A N/A N/A Open in a new tab a Daily activity domain extreme item response removed from this calculation: very easy responses for the item ‘wash my hands’ (92.16%). Skip responses for the items ‘shave my face’ (50.33%), ‘put my hair in a ponytail’ (62.09%), and ‘put on a bra’ (66.67%). Abbreviations: N/A, not applicable; PEDI-PRO, Pediatric Evaluation of Disability Inventory – Patient Reported Outcome. Structural validity CFA and residual correlations were examined simultaneously. Initial CFA fit indices for the mobility domain were all good, but fit indices for the daily activities and social/cognitive domains did not meet the stated criteria. 23 After removing items with residual correlations greater than 0.2 (see Tables S1 – S3 ), the final CFA fit indices were acceptable ( Table 3 ). Additionally, for the subsequent exploratory factor analysis, the percentage of variance explained in the first factor and the ratio of the first to second eigenvalues met criteria for all domains ( Table 3 ). Some items were maintained to enhance content validity and maximize item difficulty, and the number of remaining item pairs was equal to or less than what would be expected on a unidimensional scale based on our simulated analyses. TABLE 3. Residual correlation coefficient, confirmatory factor analyses, and exploratory factor analysis by PEDI-PRO domain Residual correlation coefficient Final CFA fit indices a EFA Domain Item pairs > 0.2 Items removed Remaining item pairs > 0.2 CFI TLI RMSEA Variance explained b (%) Ratio of first and second eigenvalues c Daily activities 32 6 Two pairs Three items 0.972 0.969 0.035 44 7.2 Social/cognitive 14 8 Two pairs Three items 0.929 0.922 0.053 36 7.8 Mobility 8 7 One pair Two items 0.979 0.977 0.036 47 5.5 Open in a new tab a CFI > 0.95 good, > 0.90 acceptable; TLI > 0.95 good, > 0.90 acceptable; RMSEA < 0.05 good, < 0.08 acceptable. b A value > 20% is good. c A value > 4 is good. Abbreviations: CFA, confirmatory factor analysis; CFI, comparative fit index; EFA, exploratory factor analysis; PEDI-PRO, Pediatric Evaluation of Disability Inventory – Patient Reported Outcome; RMSEA, root mean square error of approximation; TLI, Tucker–Lewis index. Only two items did not meet RSM fit criteria: one daily activity item (‘sign my name in cursive at the doctor’s office’; acceptable infit, poor outfit) and one mobility item (‘ride a bike for 30 minutes’; poor infit and outfit) ( Tables S4 – S6 ). As outfit is less likely to indicate multidimensionality, the daily activity item was retained for content validity. 5 Monotonically increasing threshold parameters suggested the PEDI-PRO response scale functioned as intended ( Table 2 ). Internal consistency Cronbach statistics met acceptability criteria of more than 0.70 for all domains ( Table 4 ). The Rasch reliability coefficient and separation coefficients for persons and items met acceptability criteria (>0.80 and >2 respectively) for all domains. TABLE 4. Internal consistency and precision by PEDI-PRO domain Domain Classical test theory statistics Rasch rating-scale model coefficients a Cronbach’s alpha b Omega coefficient Person reliability coefficient Person separation coefficient Daily activity 0.95 0.92 0.86 2.47 Social/cognitive 0.93 0.89 0.85 2.35 Mobility 0.95 0.91 0.83 2.21 Open in a new tab a Reported coefficients for non-extreme persons only. b Polychoric correlation, available data. Abbreviation: PEDI-PRO, Pediatric Evaluation of Disability Inventory – Patient Reported Outcome. Measurement invariance At the model level, DIF was noted for the daily activities (age, race/ethnicity), social/cognitive (gender), and mobility domains (gender, age, race/ethnicity). Interaction terms were significant for three daily activity, one social/cognitive, and eight mobility items. Their resulting values for weighted area beneath the curve were greater than 0.18 for two daily activity items (both age: ‘put on plastic gloves,’ ‘sign my name in cursive at the doctor’s office’), one social/cognitive item (gender: ‘know how to count money to buy something that costs US$7.79’), and two mobility items (both age: ‘exercise for 15 minutes without taking a break’, ‘move around for 1 hour while I am working’). The impact of the DIF on the total score for each domain was minimal and below the 1.0 score threshold (which is around 10% of the standard deviation of expected scores across each domain): daily activity, 0.45-point decrease; social/cognitive, 0.25-point increase; mobility, 0.65-point decrease. Domain item sets Daily activities Of the 36 items tested, six were removed that did not meet criteria for local independence, and three associated with gender performance could not be included for analyses because of high rates of ‘skip’ response. No items were removed because of poor fit, or because of DIF with an impact greater than 1.0 on the domain score. The domain includes 27 items contributing to the final domain score. In previous studies, young people and clinicians identified activities of daily living related to gender as important. 5 Therefore, the three gender performance items were maintained in the PEDI-PRO software to inform goal setting and intervention planning, but do not contribute information to the daily activity domain score. Social/cognitive Of the 32 items tested, eight were removed as they did not meet criteria for local independence. No items were removed because of poor fit or DIF with a greater than 1.0 impact on domain score. The domain includes 24 items. Mobility Of the 28 items tested, six were removed as they did not meet criteria for local independence, and one was removed because of poor fit. No items removed because of DIF with a greater than 1.0 impact on domain score. The domain includes 21 items. DISCUSSION Findings indicate that the PEDI-PRO, a self-reported measure of functional performance, has acceptable psychometric properties when used by young people with developmental disabilities. There is sufficient evidence for the structural validity of all PEDI-PRO domains across the methods used (i.e. confirmatory/exploratory factor analysis, absolute residual correlations, Rasch RSM item fit, and response scale function). All three domains also demonstrated acceptable internal consistency. The rigorous approach of investigating structural validity and internal consistency using both classical test theory and item response theory, and the investigation of measurement stability, exceeds what is typically used for measures developed for people with developmental disabilities across the lifecourse. 12 , 25 , 35 Also unique to this study is the confirmation of the structural validity and appropriate interpretation of the response scale, which, despite being recognized as a challenging component of self-reporting procedures by young people with developmental disabilities, is typically not investigated. 4 , 12 , 36 This study confirmed that young people with developmental disabilities can use a 3-point Likert response scale in the intended manner. The PEDI-PRO’s cognitively accessible design and innovative measurement framework probably contributed to these promising findings. Other research has been inconclusive about the extent to which self-report measures demonstrate acceptable psychometric properties when used by respondents with developmental disabilities. 25 , 35 Our findings highlight the added value and subsequent quality of PROMs developed in partnership with young people from the ground up, and contribute to the growing body of evidence for inclusive, participatory assessment development methodologies with people with developmental disabilities, including young people and adults. 2 – 5 Technology also provides an opportunity to incorporate innovative measurement approaches used in the PEDI-PRO software, such as administering items by familiar and relatable everyday life situations, while maintaining the ability to generate domain scores for constructs valuable to rehabilitation professionals. The findings of this study suggest such innovations do not compromise the integrity of domain scores. The PEDI-PRO’s comprehensive ‘Learn how to use the PEDI-PRO’ module is another feature that may have contributed to the quality of the measure, and aligns with other research that demonstrates training and practice enhances the appropriate interpretation of self-report measures. 37 This training in the module, complemented with administrative procedures that encourage self-reflection (e.g. think-aloud administration technique), demonstrates the importance of cognitive accessible administration procedures. 9 This study has several limitations. First, it used convenience sampling, and disability categories and IQ was not validated in all participants although all met criteria for developmental disabilities which is the target respondent group for the PEDI-PRO. Additionally, 3.9% did not provide a specific disability category. This reduces confidence that the characteristics of this validation sample are representative of the general population, although most were recruited through organizations that typically serve young people with developmental disabilities such as schools or community-based service providers. Second, two participants had outlier recorded IQ scores less than 35. A specific IQ score was not required for participation in this study, given multiple challenges in assessing and interpreting IQ scores for people with developmental disabilities. 38 Rather, all participants in this sample demonstrated appropriate understanding of the items and response scale through their think-aloud responses. Future research is needed to identify meaningful indicators, other than IQ, of young people who may be most appropriate for the PEDI-PRO. Third, the dichotomous groups created to evaluate measurement invariance did not meet COSMIN criteria for ‘adequate’ sample size. However, our overall sample size met COSMIN standards for ‘very good’ for structural validity and internal reliability, strengthening the confidence in the findings. 20 Future research will be needed to examine DIF across other racial/ethnic identities and establish responsiveness and the minimal difference associated with observable clinical change, 12 , 36 a crucial psychometric property. Responsiveness could be enhanced by the addition of items to the social/cognitive and mobility domains. CONCLUSION The PEDI-PRO addresses a gap in high-quality PROMs that measure priority outcomes for use with young people with developmental disabilities. This study found acceptable structural validity, internal reliability, and measurement invariance. These findings are supplemented with previous evidence for test–retest reliability and content validity, and contribute to a rigorous body of evidence for the PEDI-PRO. Many features of the PEDI-PRO may be adopted for PROMs measuring other constructs valued by young people with developmental disabilities. Supplementary Material Supinfo2 Table S1: PEDI-PRO daily activity domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo2.docx (28.5KB, docx) Supinfo3 Table S2: PEDI-PRO social/cognitive domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo3.docx (27.1KB, docx) Supinfo1 Appendix S1 : PEDI-PRO administration directions. NIHMS2028506-supplement-Supinfo1.pdf (271.7KB, pdf) Supinfo4 Table S3: PEDI-PRO mobility domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo4.docx (26.7KB, docx) Supinfo5 Table S4: Daily activities domain item fit. NIHMS2028506-supplement-Supinfo5.docx (29.8KB, docx) Supinfo6 Table S5: Social/cognitive domain item fit. NIHMS2028506-supplement-Supinfo6.docx (29KB, docx) Supinfo7 Table S6: Mobility domain item fit. NIHMS2028506-supplement-Supinfo7.docx (28.4KB, docx) What this paper adds. All Pediatric Evaluation of Disability Inventory – Patient Reported Outcome (PEDI-PRO) domains have acceptable structural validity, internal reliability, and measurement invariance. Young people with developmental disabilities can use the cognitively accessible response scale appropriately to report their functional performance. The PEDI-PRO can generate daily activity, social/cognitive, and mobility domain scores from items administered across a range of everyday life situations. ACKNOWLEDGEMENTS We thank the young people who shared their perspectives using the PEDI-PRO. The PED-PRO would not be possible without the Inclusive Cool Cats Research team, whose lived experience has shaped the development and evaluation of the PEDI-PRO. We are grateful to all of the families in the Simons Foundation Powering Autism Research for Knowledge (SPARK) and SPARK staff for their support of this project. We also recognize contributions of our research team members. University of Florida: Elizabeth Kennelly-Smith, Savannah Jackson, Shelby Waldron, Amber Nipper, Michelle Duong, Thaila Schug, Sabrina Glover, Samone Cowart, Raine Sims, and Emma Hawkins. From Temple University: Laura Slugg, Amber Davidson. From Colorado State University: Cristina Parsons, Susan Mingles. Thank you to Erik Mugele, Technical Director at AbleLink and the development team lead, for the PEDI-PRO. FUNDING INFORMATION Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under award number R42HD090772. Abbreviations: CFA confirmatory factor analysis COSMIN COnsensus-based Standards for selection of health Measurement INstruments DIF differential item functioning PEDI-PRO Pediatric Evaluation of Disability Inventory – Patient Reported Outcome PROM patient-reported outcome measure RSM rating-scale mode DATA AVAILABILITY STATEMENT The PEDI-PRO manual (available in the PEDI-PRO software) includes a full list of all final items and item maps, and details about residual correlations, and Rasch RSM item difficulty and fit parameters. REFERENCES 1. Havercamp SM, Barnhill LJ, Bonardi A, Chapman RA, Cobranchi C, Fletcher RJ, et al. Straight from the horse’s mouth: Increasing self-report in mental health assessment in individuals with intellectual disability. Journal of Applied Research in Intellectual Disabilities. 2022;35(2):471–9. [ DOI ] [ PubMed ] [ Google Scholar ] 2. Walton K, Krahn GL, Buck A, Andridge R, Lecavalier L, Hollway JA, et al. Putting “ME” into measurement: Adapting self-report health measures for use with individuals with intellectual disability. Research in Developmental Disabilities. 2022. Sep 1;128:104298. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Shogren KA, Bonardi A, Cobranchi C, Krahn G, Murray A, Robinson A, et al. State of the Field: The Need for Self-Report Measures of Health and Quality of Life for People With Intellectual and Developmental Disabilities. Journal of Policy and Practice in Intellectual Disabilities. 2021;18(4):286–95. [ Google Scholar ] 4. Vlot-van Anrooij K, Tobi H, Hilgenkamp TIM, Leusink GL, Naaldenberg J. Self-reported measures in health research for people with intellectual disabilities: an inclusive pilot study on suitability and reliability. BMC Med Res Methodol. 2018. Jul 16;18(1):80. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Kramer JM, Schwartz AE. Development of the Pediatric Disability Inventory-Patient Reported Outcome (PEDI-PRO) measurement conceptual framework and item candidates. Scandinavian journal of occupational therapy. 2018;25(5):335–46. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. King G, Chiarello LA, Ideishi R, D’Arrigo R, Smart E, Ziviani J, et al. The Nature, Value, and Experience of Engagement in Pediatric Rehabilitation: Perspectives of Youth, Caregivers, and Service Providers. Developmental Neurorehabilitation. 2020. Jan 2;23(1):18–30. [ DOI ] [ PubMed ] [ Google Scholar ] 7. Ronen GM, Rosenbaum PL, Streiner DL. Patient perspectives in pediatric neurology: a critical shift in the paradigm of outcome measurement. Dev Med Child Neurol. 2022;64(2):149–55. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Luckasson R, Tassé MJ, Schalock RL. Professional Responsibility in the Field of Intellectual and Developmental Disabilities: Its Definition, Application, and Impacts. Intellectual and Developmental Disabilities. 2022;60(3):183–98. [ DOI ] [ PubMed ] [ Google Scholar ] 9. Kramer JM, Schwartz A. Reducing barriers to patient-reported outcome measures for people with cognitive impairments. Arch Phys Med Rehabil. 2017. Aug;98(8):1705–15. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Coster WJ, Ni P, Slavin MD, Kisala PA, Nandakumar R, Mulcahey MJ, et al. Differential item functioning in the Patient Reported Outcomes Measurement Information System Pediatric Short Forms in a sample of children and adolescents with cerebral palsy. Dev Med Child Neurol. 2016;58(11):1132–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Williams ZJ, Cascio CJ, Woynaroski TG. Measuring subjective quality of life in autistic adults with the PROMIS global–10: Psychometric study and development of an autism-specific scoring method. Autism. 2023. Jan 1;27(1):145–57. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Janssens A, Rogers M, Gumm R, Jenkinson C, Tennant A, Logan A, et al. Measurement properties of multidimensional patient-reported outcome measures in neurodisability: a systematic review of evaluation studies. Dev Med Child Neurol. 2016;58(5):437–51. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Morris C, Janssens A, Shilling V, Allard A, Fellowes A, Tomlinson R, et al. Meaningful health outcomes for paediatric neurodisability: Stakeholder prioritisation and appropriateness of patient reported outcome measures. Health and Quality of Life Outcomes. 2015. Jun 25;13(1):87. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Bendixen RM, Benevides T, Ideishi R, Smythe R, Taylor J, Umeda C, et al. Community Conversations: Stakeholder-Identified Research Priorities to Foster Community Participation for Individuals With Intellectual and Developmental Disability. Frontiers in Rehabilitation Sciences [Internet]. 2022;3. Available from: 10.3389/fresc.2022.873415 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Vargus-Adams JN, Martin LK. Measuring What Matters in Cerebral Palsy: A Breadth of Important Domains and Outcome Measures. Archives of Physical Medicine and Rehabilitation. 2009. Dec 1;90(12):2089–95. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Anaby D, Korner-Bitensky N, Steven E, Tremblay S, Snider L, Avery L, et al. Current rehabilitation practices for children with cerebral palsy: Focus and gaps. Physical & Occupational Therapy In Pediatrics. 2017. Jan 1;37(1):1–15. [ DOI ] [ PubMed ] [ Google Scholar ] 17. Schwartz AE, Kramer JM, PEDI-PRO Youth Team. Inclusive approaches to developing content valid patient-reported outcome measure response scales for youth with intellectual/developmental disabilities. Br J Learn Disabil. 2021. Mar;49(1):100–10. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Kramer JM, Schwartz A. Refining the Pediatric Evaluation of Disability Inventory–Patient-Reported Outcome (PEDI-PRO) item candidates: interpretation of a self-reported outcome measure of functional performance by young people with neurodevelopmental disabilities. Dev Med Child Neurol. 2017;59(10):1083–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Kramer J, Schwartz AE, Greenberg K. Usability and Reliability of the Pediatric Evaluation of Disability Inventory-Patient Reported Outcome (PEDI-PRO) Software Interface for Youth With DD. The American Journal of Occupational Therapy. 2021. Aug 1;75(Supplement_2):7512500004p1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Mokkink LB, Terwee CB, Knol DL, Stratford PW, Alonso J, Patrick DL, et al. The COSMIN checklist for evaluating the methodological quality of studies on measurement properties: A clarification of its content. BMC Medical Research Methodology. 2010. Mar 18;10(1):22. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Haley SM, Coster WJ, Dumas HM, Fragala-Pinkham MA, Kramer J, Ni P, et al. Accuracy and precision of the Pediatric Evaluation of Disability Inventory computer-adaptive tests (PEDI-CAT). Dev Med Child Neurol. 2011;53(12):1100–6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Muthén LK, Muthén BO. Mplus User’s Guide. Eighth Edition. Los Angeles, CA: Muthén & Muthén; 1998. [ Google Scholar ] 23. Reeve BB, Hays RD, Bjorner JB, Cook KF, Crane PK, Teresi JA, et al. Psychometric evaluation and calibration of health-related quality of life item banks: plans for the Patient-Reported Outcomes Measurement Information System (PROMIS). Med Care. 2007. May;45(5 Suppl 1):S22–31. [ DOI ] [ PubMed ] [ Google Scholar ] 24. Salsman JM, Schalet BD, Park CL, George L, Steger MF, Hahn EA, et al. Assessing meaning & purpose in life: development and validation of an item bank and short forms for the NIH PROMIS ® . Qual Life Res. 2020. Aug 1;29(8):2299–310. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Gomez LE, Verdugo MA, Arias B. Validity and reliability of the INICO-FEAPS Scale: An assessment of quality of life for people with intellectual and developmental disabilities. Research in Developmental Disabilities. 2015. Jan 1;36:600–10. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Muthén LK, Muthén BO. Monte Carlo Simulation Studies. In: Mplus User’s Guide (8th edition) Examples [Internet]. 1998. [cited 2024 Aug 22]. Available from: https://www.statmodel.com/usersguide/chapter12.shtml [ Google Scholar ] 27. Wright BD, Linacre JM. Reasonable mean-square fit values. Rasch Measurement Transactions. 1994;(8):370–1. [ Google Scholar ] 28. Linacre JM. Linacre JM (2023). Winsteps ® Rasch measurement computer program (Version 5.6.0). Portland, Oregon: Winsteps.com . [ Google Scholar ] 29. Linacre JM. Optimizing rating scale category effectiveness. J Appl Meas. 2002;3(1):85–106. [ PubMed ] [ Google Scholar ] 30. Gadermann A, Guhn M, Zumbo B. Estimating ordinal reliability for Likert-type and ordinal item response data: A conceptual, empirical, and practical guide. Practical Assessment, Research, and Evaluation [Internet]. 2019. Nov 25;17(1). Available from: https://scholarworks.umass.edu/pare/vol17/iss1/3 [ Google Scholar ] 31. Green SB, Yang Y. Reliability of Summed Item Scores Using Structural Equation Modeling: An Alternative to Coefficient Alpha. Psychometrika. 2009. Mar 1;74(1):155–67. [ Google Scholar ] 32. Watkins MW. The reliability of multidimensional neuropsychological measures: from alpha to omega. Clin Neuropsychol. 2017;31(6–7):1113–26. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Tennant A, Conaghan PG. The Rasch measurement model in rheumatology: what is it and why use it? When should it be applied, and what should one look for in a Rasch paper? Arthritis Rheum. 2007. Dec 15;57(8):1358–62. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Kramer JM, Liljenquist K, Ni P, Coster WJ. Examining differential responses of youth with and without autism on a measure of everyday activity performance. Qual Life Res. 2015. Dec 1;24(12):2993–3000. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Ruddick L, Oliver C. The Development of a Health Status Measure for Self-report by People with Intellectual Disabilities. Journal of Applied Research in Intellectual Disabilities. 2005;18(2):143–50. [ Google Scholar ] 36. Maguire S, Davison J, McLaughlin M, Simms V, Bunting B. Exploring the psychometric properties of self-report instruments used to measure health-related quality of life and subjective wellbeing of adolescents with intellectual disabilities: A Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) systematic review. Journal of Applied Research in Intellectual Disabilities. 2023;36(5):899–915. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Raczka R, Theodore K, Williams J. An initial validation of a new quality of life measure for adults with intellectual disability: The Mini-MANS-LD. J Intellect Disabil. 2020;24(2):177–93. [ DOI ] [ PubMed ] [ Google Scholar ] 38. Hristova M, Manolova H, Staykova S, Markova G. Assessment of Intellectual Functioning among Children with Neurodevelopmental Disorders: Challenges and Implications Beyond the Clinical Practice. Journal of Intellectual Disability- Diagnosis and Treatment. 2017;124–9. [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supinfo2 Table S1: PEDI-PRO daily activity domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo2.docx (28.5KB, docx) Supinfo3 Table S2: PEDI-PRO social/cognitive domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo3.docx (27.1KB, docx) Supinfo1 Appendix S1 : PEDI-PRO administration directions. NIHMS2028506-supplement-Supinfo1.pdf (271.7KB, pdf) Supinfo4 Table S3: PEDI-PRO mobility domain item pairs with residual correlation coefficients. NIHMS2028506-supplement-Supinfo4.docx (26.7KB, docx) Supinfo5 Table S4: Daily activities domain item fit. NIHMS2028506-supplement-Supinfo5.docx (29.8KB, docx) Supinfo6 Table S5: Social/cognitive domain item fit. NIHMS2028506-supplement-Supinfo6.docx (29KB, docx) Supinfo7 Table S6: Mobility domain item fit. NIHMS2028506-supplement-Supinfo7.docx (28.4KB, docx) Data Availability Statement The PEDI-PRO manual (available in the PEDI-PRO software) includes a full list of all final items and item maps, and details about residual correlations, and Rasch RSM item difficulty and fit parameters. ACTIONS View on publisher site PDF (115.7 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Functional difficulties and school limitations of children with epilepsy: findings from the 2009-2010 National Survey of Children with Special Health Care Needs.#214145
• Job Satisfaction, Self-Efficacy, and Quality Performance Among Caregivers of Children with Disability#326020
• Disability, Impairment, and Marginalised Functioning#742693
• [The treatment of adults with intellectual disabilities].#162234
• Functional Abilities in Rett Syndrome#898197