INS NYC 2024 Program

Poster	Poster Session 11 Program Schedule 02/17/2024 10:45 am - 12:00 pm Room: Majestic Complex (Posters 61-120) Poster Session 11: Cultural Neuropsychology \| Education/Training \| Professional Practice Issues Final Abstract #68 Does the Test of Premorbid Functioning (TOPF) Predict Premorbid Intelligence in a Bilingual Veteran Population as it does in a Monolingual Veteran Population? Ida Mohebpour, South Texas Veterans Health Care System, San Antonio, United States Kathleen Bain, South Texas Veterans Health Care System, San Antonio, United States Janice Marceaux, South Texas Veterans Health Care System, San Antonio, United States Julianna Valencia, South Texas Veterans Health Care System, San Antonio, United States Category: Assessment/Psychometrics/Methods (Adult) Keyword 1: bilingualism/multilingualism Keyword 2: premorbid functioning Keyword 3: intelligence Objective: To examine the appropriateness of the Test of Premorbid Functioning (TOPF) for use with bilingual English/Spanish-speaking veterans (BV). This study examined the degree of discrepancy between TOPF predicted and TOPF actual scores, to determine if there was a discrepancy in prediction accuracy between monolingual English-speaking veterans (MV) and BV. The three TOPF FSIQ predictive models (TOPF equated [TOPF-equated], simple demographic without TOPF [TOPF-demographic], and simple demographic with TOPF [TOPF-combined]) were compared to identify whether any one approach was more appropriate than another in predicting actual Wechsler Adult Intelligence Scale–Fourth Edition (WAIS-IV) Full Scale Intelligence Quotient (FSIQ-actual) in MV and BV without cognitive impairment. Participants and Methods: Data from neuropsychological evaluation of 128 clinically referred veterans (M_age= 53.02, SD= 14.54, range= 20-81; M_edu= 14.08, SD= 2.68, range= 6-20; 105 MV, 23 BV) from the patient population of a Veteran Affairs (VA) Medical Center in a large Southwestern city. Inclusion criteria: completion of the TOPF and at least 8 WAIS-IV subtests to calculate a prorated or full FSIQ. Exclusion criteria: PVT failure, neurocognitive disorder, learning disability, substance use disorders, and psychotic disorders. Mean discrepancy between TOPF predicted and TOPF standard scores was computed for the overall sample and within MV and BV. Finally, mean discrepancy between TOPF-predicted FSIQ and FSIQ-actual for each of the three predictive models was computed within the entire sample and well as within each group to determine the most accurate predictive model. Results: The difference between actual TOPF standard score and demographic-predicted TOPF standard score was non-significant in the overall sample (t[129]= -0.36, p= 0.723; d= -0.20 - 0.14), the MV group (t[105]= -0.66, p= 0.509; d= -0.26 - 0.13), and the BV group (t[23]= 0.65, p= 0.524; d= -0.27 - 0.53). ANOVAs with Greenhouse-Geisser corrections indicated WAIS-IV FSIQ and TOPF-predicted FSIQ differed significantly within the overall sample (F[1.94, 246.43]= 4.86, p= .009), where the FSIQ-actual was 2.6 points lower than TOPF-combined. When examined separately, there was no discrepancy between model accuracy within the MV group (F[1.898, 195.52]= 2.28, p= .108), while a significant discrepancy was found among models’ predictive accuracy in the BV group (F[1.90, 43.72]= 9.26, p< .001). Post-hoc analyses found each model to overpredict FSIQ-actual, with TOPF-demographic having the smallest discrepancy (5.54 [95% CI, 0.567 to 10.517] points, p= .023), followed by TOPF-equated (6.00 [95% CI, 1.918 to 10.082] points, p= .002), and TOPF-combined (6.75 [95% CI, 3.287 to 10.231] points, p< .001). Conclusions: Demographic prediction of irregular word reading performance was similar for MV and BV without cognitive impairment. However, there were notable differences in how well each of three IQ predictive models predicted actual FSIQ for MV versus BV. All models overestimated actual FSIQ performance among bilingual, but not monolingual, veterans. Additional research is needed to examine impact of English proficiency and age of acquisition of L2 on TOPF performance among BV, and to determine whether other measures of intellectual abilities or premorbid functioning may yield a more accurate estimate of premorbid intelligence within this population.

Poster

02/17/2024
10:45 am - 12:00 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 11: Cultural Neuropsychology | Education/Training | Professional Practice Issues

Final Abstract #68

Does the Test of Premorbid Functioning (TOPF) Predict Premorbid Intelligence in a Bilingual Veteran Population as it does in a Monolingual Veteran Population?

Ida Mohebpour, South Texas Veterans Health Care System, San Antonio, United States
Kathleen Bain, South Texas Veterans Health Care System, San Antonio, United States
Janice Marceaux, South Texas Veterans Health Care System, San Antonio, United States
Julianna Valencia, South Texas Veterans Health Care System, San Antonio, United States

Category: Assessment/Psychometrics/Methods (Adult)

Keyword 1: bilingualism/multilingualism
Keyword 2: premorbid functioning
Keyword 3: intelligence

Objective:

To examine the appropriateness of the Test of Premorbid Functioning (TOPF) for use with bilingual English/Spanish-speaking veterans (BV). This study examined the degree of discrepancy between TOPF predicted and TOPF actual scores, to determine if there was a discrepancy in prediction accuracy between monolingual English-speaking veterans (MV) and BV. The three TOPF FSIQ predictive models (TOPF equated [TOPF-equated], simple demographic without TOPF [TOPF-demographic], and simple demographic with TOPF [TOPF-combined]) were compared to identify whether any one approach was more appropriate than another in predicting actual Wechsler Adult Intelligence Scale–Fourth Edition (WAIS-IV) Full Scale Intelligence Quotient (FSIQ-actual) in MV and BV without cognitive impairment.

Participants and Methods:

Data from neuropsychological evaluation of 128 clinically referred veterans (M_age= 53.02, SD= 14.54, range= 20-81; M_edu= 14.08, SD= 2.68, range= 6-20; 105 MV, 23 BV) from the patient population of a Veteran Affairs (VA) Medical Center in a large Southwestern city. Inclusion criteria: completion of the TOPF and at least 8 WAIS-IV subtests to calculate a prorated or full FSIQ. Exclusion criteria: PVT failure, neurocognitive disorder, learning disability, substance use disorders, and psychotic disorders. Mean discrepancy between TOPF predicted and TOPF standard scores was computed for the overall sample and within MV and BV. Finally, mean discrepancy between TOPF-predicted FSIQ and FSIQ-actual for each of the three predictive models was computed within the entire sample and well as within each group to determine the most accurate predictive model.

Results:

The difference between actual TOPF standard score and demographic-predicted TOPF standard score was non-significant in the overall sample (t[129]= -0.36, p= 0.723; d= -0.20 - 0.14), the MV group (t[105]= -0.66, p= 0.509; d= -0.26 - 0.13), and the BV group (t[23]= 0.65, p= 0.524; d= -0.27 - 0.53). ANOVAs with Greenhouse-Geisser corrections indicated WAIS-IV FSIQ and TOPF-predicted FSIQ differed significantly within the overall sample (F[1.94, 246.43]= 4.86, p= .009), where the FSIQ-actual was 2.6 points lower than TOPF-combined. When examined separately, there was no discrepancy between model accuracy within the MV group (F[1.898, 195.52]= 2.28, p= .108), while a significant discrepancy was found among models’ predictive accuracy in the BV group (F[1.90, 43.72]= 9.26, p< .001). Post-hoc analyses found each model to overpredict FSIQ-actual, with TOPF-demographic having the smallest discrepancy (5.54 [95% CI, 0.567 to 10.517] points, p= .023), followed by TOPF-equated (6.00 [95% CI, 1.918 to 10.082] points, p= .002), and TOPF-combined (6.75 [95% CI, 3.287 to 10.231] points, p< .001).

Conclusions:

Demographic prediction of irregular word reading performance was similar for MV and BV without cognitive impairment. However, there were notable differences in how well each of three IQ predictive models predicted actual FSIQ for MV versus BV. All models overestimated actual FSIQ performance among bilingual, but not monolingual, veterans. Additional research is needed to examine impact of English proficiency and age of acquisition of L2 on TOPF performance among BV, and to determine whether other measures of intellectual abilities or premorbid functioning may yield a more accurate estimate of premorbid intelligence within this population.