gms | German Medical Science

Objectives: Growing evidence points to the importance of early detection of psychosocial distress in tumor patients to improve quality of life. Psychooncologists see only a fraction of the 25-44% of patients suffering serious distress during illness. This study aims at identifying patients with significant distress by an algorithm-based decision-tree (measuring patients distress-symptoms, illness-adaptation resources, treatment motivation) for referral to treatment (Funding: Commission for Clinical Research, Technical University of Munich, KKF-Project 395/00).

Study design Administration of a 15 items instrument (Screening for Intervention in Psychooncology SIPS), a high score indicates referral to treatment. After a 3x90 min. rater training results showed fair to very good congruence compared to expert rating of a video case. Then 9 trained oncologists applied SIPS in daily routine.

Primary endpoints Concordance of SIPS case definition by an expert rating (blind for oncologists´ ratings) and

1) oncologists´ratings

2) patients´ Hospital Anxiety Depression Score HADS above cut-off

Study population 69 of 91 consecutively recruited inpatients with mostly haematooncologic or digestive system tumors consented.

Mean age 53.9 y

First diagnosis 53.4% tumor recurrence 38.4%

Local tumor 23.3% metastatic 55.1%

Chemotherapy 80.8% curative 43.5% palliative treatment 53.6%

Karnofsky-Index 100-70: 91.7%

Results: A) Significant correlations for expert and oncologists´ ratings in SIPS:

1. 10 items out of 15: 6 items p<.001; 2 items p<.004 resp. p<.009; 2 items p<.019 resp. p<0.025.

2. Sum values of item groups „somatic and psychologic distress symptoms “ p<.001 and „illness-adaptation resources“ p<.001, which were designed to direct referral algorithm.

B) In terms of patients selfratings (HADS scores), oncologists came closer to patients´ anxiety (31.5% p<0.01), whereas the expert came closer to patients´ depression (32,9% p<0.005).

C) Comparison of quality criteria between expert rating and oncologists´ rating resp. patients´ HADS (N= 43) for quality of fit: With regard to expert rating, trained oncologists achieved better case definitions than HADS, concerning sensitivity 78% (HADS 71%), positive predictive value 83% (59%) and accuracy 72% (64%).

Conclusion: In this pilot project, trained oncologists showed to be effective in indicating psychooncological treatment in clinical routine, proving the first validation steps of the Screening-for-Intervention instrument in Psychooncology SIPS.