Reliability and construct validity of the ACTIVLIM-Hemo and Haemophilia Activities List (HAL) questionnaires in individuals with haemophilia.

Abstract

The objective of this study is to assess the reliability and construct validity of ACTIVLIM-Hemo, a newly developed Rasch-built questionnaire designed to evaluate activity limitations in people with haemophilia (PwH), in comparison with the Haemophilia Activities List (HAL), which was developed using Classical Test Theory. A total of 130 participants with haemophilia A or B were included. They underwent various assessments, including joint health scoring (HJHS), functional tests (TUG and 2MWT) and completed questionnaires such as the BPI, IPAQ, HAL and ACTIVLIM-Hemo. Reliability indices and the minimum detectable change (MDC95) were determined for ACTIVLIM-Hemo and for HAL. Construct validity was evaluated through correlations and multiple linear regression, considering demographic and clinical factors. Both ACTIVLIM-Hemo (Person Separation Index=0.92) and HAL (Cronbach's α=0.98) demonstrated high reliability. The MDC95 for ACTIVLIM-Hemo represented 11.6% of its measurement range, while for HAL, it amounted to 18/100 score points. Activity limitations measured by both instruments were significantly correlated with demographic and clinical factors. Joint health (HJHS), pain severity (BPI) and walking performance (2MWT) emerged as the main predictors of activity limitations, explaining 75% of the variance in ACTIVLIM-Hemo and 60% in HAL. ACTIVLIM-Hemo stands as a reliable and valid instrument for assessing activity limitations in PwH. Both instruments exhibited significant correlations with demographic and clinical factors, but ACTIVLIM-Hemo displayed a more homogeneous construct. Given its linear scale and lower MDC95 and better targeting, ACTIVLIM-Hemo shows promise as a patient-centric instrument for assessing responsiveness to treatment during individual follow-up.