Non-professional caregiver burden is associated with the severity of patients' cognitive impairment

Background/Objectives To analyse the relationship between caregiver burden and severity of patients' cognitive impairment. Design Data were drawn from the cross-sectional 2015/2016 Adelphi Real World Dementia Disease-Specific Programme. Setting This research was multi-national and studied physicians and their consulting patients with cognitive impairment. Participants 1,201 caregivers completed self-assessment forms. Measurements Validated instruments of caregiver wellbeing and burden (EQ-5D-3L questionnaire, EQ-VAS, Zarit Burden Interview, and Work Productivity and Activity Impairment questionnaire) and number of caregiver hours were analysed by severity of patients' cognitive impairment, categorised according to the Mini-Mental State Examination. Data were analysed using Spearman's rank correlation coefficients and ordinary least squares regression models, to compare outcomes between caregivers of patients with prodromal, mild, moderate, and severe dementia. Results The majority of caregivers were female (69.1%), lived with the patient they cared for (75.8%), and only approximately one third (28.3%) were in part- or full-time employment. There were statistically significant (p<0.001) increases in caregiver time (36.9 versus 108.6 hours per week for prodromal versus severe dementia, respectively) and measures of caregiver burden and health status (EQ-5D-3L, EQ-VAS, and Zarit Burden Interview) and increases in measures of work productivity and activity impairment with increasing severity of patients' disease. Conclusion This study of real-world data confirmed an association between increased caregiver burden and severity of patients' cognitive impairment by analysis of a wide range of validated measures of caregiver burden. These findings suggest that maintaining patients in the earliest stages of their disease for as long as possible may potentially help to protect caregiver wellbeing, although further research is required to confirm this hypothesis.