Academic journal article Psychological Test and Assessment Modeling

Measurement Equivalence of the Patient Reported Outcomes Measurement Information System® (PROMIS®) Applied Cognition - General Concerns, Short Forms in Ethnically Diverse Groups

Academic journal article Psychological Test and Assessment Modeling

Measurement Equivalence of the Patient Reported Outcomes Measurement Information System® (PROMIS®) Applied Cognition - General Concerns, Short Forms in Ethnically Diverse Groups

Article excerpt

Background

Conceptual equivalence of measures implies that questions are understood in the same way by all respondents (Collins, 2003). Differences in race/ethnicity, culture, socioeconomic status, education, and gender can lead to systematic measurement error in interpreting survey responses to standardized questionnaires (Warnecke et al., 1997). Differential item functioning (DIF) analysis in the context of item response theory (IRT) examines whether or not the likelihood of item (category) endorsement is equal across subgroups, conditional on the construct or trait level. For example, DIF is present if different groups of individuals (e.g., males and females) at the same levels of the latent construct exhibit different probabilities of individual item responses (Hulin, 1987).

This paper presents the dimensionality, reliability, information functions, and DIF of the Patient Reported Outcome Measurement Information System® (PROMIS®) Applied Cognition - General Concerns, 8 item short form. This is a measure of self-reported cognitive concerns or complaints, and both terms are used interchangeably to describe the construct assessed. Qualitative methods were used to generate DIF hypotheses for subgroups.

Acknowledgment of the salience of subjective cognitive complaints is relatively new within the field of neurology, and more generally cognitive aging. Early studies of sub- jective cognitive decline focused on memory, e.g., Gurland et al., 1999. Recent findings suggest that subjective complaints are associated with increased risk of dementia (Jessen et al., 2014; Reisberg, Shulman, Torossian, Leng, & Zhu, 2010) and biomarkers of Alzheimer's Disease (Barnes et al., 2006; Sperling et al., 2011) among those presenting with otherwise-normal cognitive function. Subjective cognitive complaints are a key feature of mild cognitive impairment (MCI). However, to date, there is little evidence extant regarding the psychometric performance of such measures, and particularly of their measurement equivalence across subgroups. Moreover, subjective cognitive impairment may be common among people with cancer, especially those undergoing chemotherapy, and this is an important element of health-related quality-of-life for such individuals..

Racial and ethnic differences have been observed in informant-reported cognitive function. For example, examining diagnosis of cognitive impairment no dementia (CIND; based on neuropsychological testing), informant reports of cognitive decline were found to be associated with an increased odds of CIND among Whites, but not African Americans (Potter et al., 2009). Differences have also been observed among Hispanic and nonHispanic White respondents in self-reported cognitive function. For example, 16.9 % of Hispanic or Latino respondents said that they had experienced confusion or memory loss (CML), which was significantly higher than the 12.1 % among Whites (Centers for Disease Control and Prevention, 2013). Differences in self-reported cognition may also occur by gender. Among older adults, reports of subjective memory have been shown to differ between men and women, with women reporting significantly more memory complaints (Gagnon et al., 1994). Further, in a sample of young adults, males and females tended to assess their divergent thinking (i.e., creativity) across traditionally stereotypic lines (Kaufman, 2006); females rated themselves higher on verbal skills, while males rated themselves higher on general analytic thinking. It is also possible, however, that these results reflect DIF, which is to say, for example, when controlling for the overall level of cognitive complaints, females were more likely to endorse higher verbal skills and males to endorse general analytic thinking. DIF analyses are needed to differentiate between true differences and those attributable to DIF.

Previous psychometric investigations of the PROMIS 8-item Applied Cognition - General Concerns short form have been limited to reliability and convergent validity in a community-dwelling sample of adults (Saffer, Lanting, Koehle, Klonsky, & Iverson, 2015). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.