Background: The aging population and associated high disability rates make the assessment of health-related quality of life (HRQOL) a high priority for elderly people.
Objectives: To compare the Brazilian versions of the Nottingham Health Profile (NHP) and the Short Form Health Survey-36 (SF-36) regarding their measurement properties, such as, ceiling and floor effects, inter-rater/test-retest reliabilities, internal consistency, and the associations between the total scores and those of similar domains.
Methods: The NHP and the SF-36 were randomly administered through interviews to 40 community-dwelling elderly (mean age 70.57; SD=7.42 years). The scores of the five similar domains of the NHP and the SF-36, in percentages, were compared: Energy level/Vitality (EL/V), pain, emotional reactions/mental health (EM/MH), social isolation/social functioning (SI/SF), and physical mobility/physical functioning (PA/PF). Descriptive statistics, Spearman Correlation Coefficient, a-Cronbach, and Intraclass Correlation Coefficients (ICCs) were used for analyses (a=0.05).
Results: Both instruments demonstrated ceiling and floor effects for all similar domains, however the NHP showed, on average, higher scores for individual domains and total scores, as well as, floor effects for the SI domain. The internal consistency (NHP=0.86; SF-36=0.80), the inter-rater reliability (NHP=0.99; SF-36=0.96) and the test-retest reliability (NHP=0.94; SF-36=0.83) were considered adequate. Positive and statistically significant correlations were found between all similar domains (p≤0.01; 0.41≤r(s)≤0.70), except for those related to EL and vitality (r(s)=0.21;p=0.19).
Conclusions: The SF-36 demonstrated to be more adequate regarding the ceiling and floor effects, whereas the NHP presented a higher internal consistency and reliability levels. These findings should be considered for selecting instruments for the assessment of HRQOL of community-dwelling elderly.