Validity of the Neurology Quality of Life Measurement System (Neuro-QoL) Spanish Social Health Items

SIGNIFICANCE: Social factors, such as the ability to work and socialize with friends, are widely recognized to be important determinants of health. Occupational therapists (OTs) can more effectively influence client health if they have information on these determinants. Social item banks in English and Spanish were created for the National Institute of Neurological Disorders and Stroke (NINDS)-funded Neurology Quality of Life Measurement System (Neuro-QoL) project that can be used to identify the satisfaction and the ability to participate in social roles and activities for individuals with various neurological conditions. This study helps OTs to become familiar with this national outcome measurement effort, enhancing the use of this tool in future research and practice.

INNOVATION: The Neuro-QoL was designed to develop standardized item banks to assess the quality of life (QoL) of individuals with neurologic disorders, making cross-disease, cross-study, and cross-language comparisons of QoL burden or benefits of different interventions possible. Our team adopted the mixed-method and Item Response Theory (IRT)-based approaches to construct both English and Spanish language items that allow for computerized adaptive testing (CAT) or short forms of varying length that provide accurate measurement with low response burden. In this study, we primarily report the validity of the Neuro-QoL Spanish social items. Research questions include the following: What is the Neuro-QoL? Why is it important? What are the measurement properties of Neuro-QoL Spanish social domains? Did persons speaking different languages respond differently in social items?

Effective cross-culture comparisons are warranted if meaningful, culturally appropriate responses are obtained from the instruments developed for various language-speaking populations. This study aimed to provide the validity evidence for the Neuro-QoL social domains that are applicable to assess social function across English- and Spanish-speaking populations.

METHOD: For the study design, we used retrospective analysis of responses to the Neuro-QoL items. We conducted field testing using online panel testing companies. Participants included 549 English-speaking and 253 Spanish-speaking adults from general U.S. populations. Measures included two Neuro-QoL social domains: Ability to Participate in Social Roles (45 items) as well as Activities and Satisfaction With Social Roles and Activities (45 items). We translated English items into Spanish using the Functional Assessment of Chronic Illness Therapy (FACIT) translation methodology. We modeled responses to the Neuro-QoL items using IRT, and we examined the presence and impact of differential item functioning (DIF) by language.

RESULTS: For the Ability domain, two items were removed due to high residual correlation. The final bifactor model fits the data relatively well (comparative fit index [CFI] = .967, Tucker–Lewis index [TLI] = .963, root-mean-square error of approximation [RMSEA] = .072). Two of the remaining 43 items were flagged for DIF on the basis of the R ² = .02 or above criterion, but the impact of language DIF on individual scores and at test level was negligible. For the Satisfaction domain, two items were removed due to high residual correlation and low reliability. The final bifactor model fits the data relatively well (CFI = .977, TLI = .975, RMSEA = .070), and no item was flagged for DIF.

CONCLUSION: This tool shows good validity evidence. English- and Spanish-speaking participants with the same level of social function did not respond differently to the Neuro-QoL social items. These results support the validity of Neuro-QoL social domain to assess social function across diverse language populations.