[Quality of life in patients with chronic obstructive lung diseases: psychometric criteria of the German version of the Sickness Impact Profile and initial results of its application].

BACKGROUND

Quality of life has become an important criterion for assessing disease impact and treatment outcome. The Sickness Impact Profile (SIP) is an internationally widely used generic instrument for the measurement of health related quality of life. Furthermore, it has been frequently employed for research in patients suffering from airway diseases. Therefore, we developed a German version of the SIP and investigated its psychometric properties in patients with chronic obstructive lung disease (COPD).

PATIENTS AND METHODS

After translation of the SIP into German, the results of a pilot study suggested several changes. Some items had to be reformulated and the SIP subscales "Emotional and affective behavior", "Alertness" and "Body care and movement" had to be discarded because of insufficient psychometric properties. The reliability, validity and sensitivity of the resulting German SIP version were investigated in a sample of 58 O2-dependent patients with COPD.

RESULTS

Internal consistency (Cronbach Alpha) was 0.93 for the entire instrument, while it was 0.85 for the SIP psycho-social scale and 0.89 for the SIP physical scale. The internal consistency ranged from 0.55 to 0.86 for the subscales of the German SIP version. The nutrition subscale was discarded because it showed an insufficient internal consistency (0.40). Indices of respiratory function (FEV1, FVC, Pa, O2, Sa, O2) and severity of dyspnea as assessed by the Baseline Dyspnea Index (BDI) were associated with the total SIP, psycho-social and physical score reaching from r = -0.19 to r = -0.48. Patients who improved after introduction of liquid oxygen therapy (LOX) showed a better total SIP score (p = 0.023) and less impairment in the SIP physical scale (p = 0.009). In order to evaluate the prognostic validity of the German SIP, a discriminance analysis between those patients who were alive 14 months later and those who had died of their illness was performed. A single discriminant function with an Eigenvalue of 0.59 (canonical correlation r = 0.61; CU2 = 21.213; df = 3; p < 0.001) was found. By means of the FVC, the severity of dyspnea (BDI), the SIP total score and the FEV1 85% of the survivors and 91 % of the fatalities could be predicted correctly.

CONCLUSIONS

The present results demonstrate sufficient psychometric properties of the German SIP version for the use in patients with COPD and hypoxemia. However, the internal consistencies of the subscales "Work" and "Leisure pastimes and recreation" remain to be improved. Furthermore, the results of factor and cluster analyses suggest some changes in the internal structure of the German SIP.