Application of a disease-specific, quality-of-life measure (QoL-AGHDA) in growth hormone-deficient adults and a random population sample in Sweden: validation of the measure by Rasch analysis

Authors


Dr Lena Wirén Research Centre for Endocrinology and Metabolism, Blå Stråket 3, Sahlgrenska University Hospital, Sahlgrenska, S-413 45 Göteborg, Sweden. Fax: + 46 31 41 26 88; E-mail: lena.wiren@medic.gu.se

Abstract

OBJECTIVE

Growth hormone deficiency (GHD) in adults has been associated with impaired health status and quality of life (QoL) in several studies using generic measures, and in a few studies using recently developed disease-specific measures. Theoretically, disease-specific measures may be more sensitive and succinct than generic measures, and hence prove convenient for general use in clinical practice. The present study sought to validate the scaling properties of the disease-specific QoL-AGHDA measure through the implementation of Rasch analysis. The study also sought to compare, by using the QoL-AGHDA, the QoL of a relatively large Swedish cohort of adults with untreated GHD with that of a reference population also from Sweden.

PATIENTS

The QoL of 111 adults with untreated GHD from Stockholm and Göteborg was compared with that of 1448 adult subjects randomly selected from the population of Göteborg.

MEASUREMENTS

The scaling properties of the QoL-AGHDA were assessed by investigating its fit to a dichotomous Rasch model. Rasch-transformed QoL scores from the QoL-AGHDA questionnaire were stratified by age and gender, and 95% confidence intervals were calculated.

RESULTS

Rasch analysis of the QoL-AGHDA indicated the measure to be robust in terms of its unidimensionality and ordering properties, and lack of differential item functioning. The raw scores produced by the QoL-AGHDA are at the ordinal level. Non-overlapping 95% confidence intervals of Rasch-transformed interval scores in most age categories indicated that men and women with GHD had significantly lower QoL than the reference population.

CONCLUSION

The Swedish QoL-AGHDA has good scaling properties, and hence can be considered a robust measure. It is suitable for assessing quality of life in adults with GH deficiency, and for making comparisons with adults who are not growth hormone deficient. Adult GH deficiency is associated with a significant impairment in QoL.

Ancillary