Date of Award


Document Type

Doctoral Research Project

Degree Name

Doctor of Psychology (PsyD)



First Advisor

Anthony LoGalbo

Second Advisor

Julie S. Costopoulos

Third Advisor

Kevin R. Burke

Fourth Advisor

Robert A. Taylor


Introduction: Given the clear impact and prevalence of dementia on the aging population, there is a need for greater research and understanding in this area. As the dementia progresses, medical providers rely heavily on caregiver and family member reports of the patient’s functioning and other aspects of their care. Therefore, it is important to know how accurate caregivers and family members are in judging their loved ones’ cognitive and physical strengths and weaknesses. This study aims to expand upon these previous studies by comparing caregiver/collateral reports of cognitive and functional status on the DSRS to neuropsychological tests and an objective measure of adaptive functioning and exploring whether the informant-rating on the DSRS is able to predict neuropsychological test scores. Furthermore, this study aims to determine whether caregivers are more or less accurate depending on which stage of the disease process their loved one is in. Method: Participants were referred for a brief neuropsychological evaluation (BNE) by their medical provider to help clarify diagnosis and inform treatment and recommendations. The most common referrals were due to concerns about memory and/or other cognitive changes. Patients’ caregivers complete the Dementia Severity Rating Scale (DSRS) during the intake appointment, the DSRS is administered by the licenced social worker. Diagnostic impressions were made by a multidisciplinary team at included a neurologist, a geriatrician or a geriatric nurse practitioner, a neuropsychologist, a pharmacist, a social worker, and clinical psychology doctoral students. The multidisciplinary team utilized medical histories, brain imaging, social histories, and neuropsychological test data to inform diagnosis. 282 participants were administered brief neuropsychological evaluations (BNE) as well as the Dementia Severity Rating Scale (DSRS) at their intake appointment. After removing individuals who did not meet diagnostic criteria for inclusion, a total of 254 participants were included in the current study. Among them, 151 patients were diagnosed with dementia (60.3% female, M = 80.61, SD = 6.54), 64 patients were diagnosed with Mild Cognitive Impairment (MCI; 57.8% female, M = 78.59, SD = 6.95), and 39 were diagnosed with Normal Cognition (NC; 53.8% female, M = 76.64, SD = 7.13. Results: The overall DSRS score was statistically different among diagnostic groups. Specifically, scores from the AD group were significantly higher than both MCI and NC groups, and those from the MCI group were significantly higher than the NC group. Correlation analysis revealed that overall DSRS scores were negatively and significantly correlated with overall TFLS score, suggesting that as DSRS score increases, TFLS score decreases. A linear regression indicated that the DSRS score statistically significantly predicted the TFLS score, accounting for 9.9% of the variance. Further correlations revealed a statistically significant negative correlation between total DSRS and MoCA score, suggesting that as DSRS scores increase (i.e., indicating poorer functioning), MoCA scores decrease (i.e., indicating poorer global cognitive functioning). Correlations among neuropsychological measures of executive functioning with total DSRS score suggested that as DSRS score increases, measures of executive functioning indicate poorer overall functioning, such that a higher DSRS score is correlated with greater time to complete TMT-B, more perseverative errors on M-WCST, fewer correctly completed categories on M-WCST, and fewer correct items on Stroop Color-Word subtest. The degree of agreement between total TFLS and overall DSRS scores revealed that the differences among caregivers who had contact with the patient less than one day per week, one day per week, two days per week, three to four days per week, and 5 or more days per week were not statistically significant. Similarly, there was no significant difference among the degree of agreement between the TFLS and the caregiver’s relationship with the patient. Conclusion: Ultimately, the results of the study showed that collateral reporters using the DSRS were able to accurately distinguish between NC, MCI, and dementia groups. Additionally, the DSRS accurately predicted functional status when compared to an objective measure of adaptive functioning, but only accounted for 9.9% of the variance. It was also reasonably related to neuropsychological measures of global cognitive functioning and executive functioning. However, there was no difference in the accuracy of the DSRS dependent on the collateral reporter’s relationship to the patient, nor how many days per week the collateral reporter spends time with the patient. Therefore, individuals who complete the DSRS for the patient provide information that agrees with the TFLS to the same degree despite their relationship to the patient or how much time they spend with the patient. This finding suggests that information obtained from caregivers via clinical interviews may also be equally accurate regardless of how much time the caregiver spends with the patient. Indeed, it appears that the accuracy of collateral reports received from caregivers and family members is not dependent on the nature of the relationship with the patient or the number of days per week spent together.


Copyright held by author