Article
Cut-offs for patient reported outcomes (PROs) and clinician reported outcomes (ClinROs) in neuro-oncological patients indicating need for support and/or increased distress – results of multicentre observational studies
Cut-Off-Werte für Selbst- und Fremdeinschätzungsfragebögen zur Erfassung des psycho-onkologischen Unterstützungsbedarfs und/oder eines erhöhten Distress bei neuroonkologischen Patienten – Ergebnisse aus multizentrischen Beobachtungsstudien
Search Medline for
Authors
Published: | May 8, 2019 |
---|
Outline
Text
Objective: Neuro-oncological patients need to be assessed adequately. After assessing patients’ psycho-social burden we either statistically determined optimal cut-off scores or re-evaluated existing cut-offs for this patient population.
Methods: Patients with intracranial tumors, participating in our prospective studies, were assessed using the following validated instruments: the Patient Health Questionnaire (PHQ-4, cut-off ≥ 6), the NCCN Distress Thermometer (DT, cut-off ≥ 6), the Hornheide Screening Instrument (HSI), the Supportive Care Needs Survey (SCNS SF-34, cut-off ≥ 3) and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire with its brain cancer module (EORTC QLQ-C30+BN20). Additionally, clinicians completed the Basic Documentation for Psycho-Oncology Short Form (PO-Bado). Optimal cut-off values for the PHQ-4, PO-Bado and EORTC functioning indicating need for support (reference: SCNS-SF34 ≥ 3, HSI, patient wish; as available) or increased distress (reference: DT ≥ 6) were determined applying ROC analyses.
Results: We evaluated 315 patients (male: n=156, 50%), most of which had a glioma (n=193, 61%), and a KPS≥70 (n=280, 89%). A cut-off value of 2.5 for the PHQ-4 was identified to moderately discriminate between patients in or not in distress (AUC 0.77; sensitivity 76.8%, specificity 64.6%) and between patients wishing or not wishing further support (AUC 0.77; sensitivity 82.5%, specificity 59%). Cut-off values for the EORTC functioning scales moderately indicated increased distress (AUC 0.74-0.85, sensitivity 59.6-86.0%, specificity 59.3-80.5%) and need for support (AUC 0.69-0.78, sensitivity 55.2-76.7%), specificity 34.4-84.6%). Thereby emotional functioning had the best discriminatory properties for cut-offs of 62.5 and 70.8 indicating increased distress and need for support, respectively. A PO-Bado total score cut-off of 8.5 (AUC 0.77, sensitivity 71.3%, specificity 67.6%) discriminated moderately between patients in and those not in distress. To distinguish between patients in need of support and those without a need for support a PO-Bado total score cut-off of 9.5 (AUC 0.78, sensitivity 65.1%, specificity 77.7%) was determined.
Conclusion: Cut-off values of existing instruments should be adapted for neuro-oncological patients as we observed that cut-offs in PROs and ClinRO, as recommended in the literature, may be too high in order to detect affected patients with intracranial tumors adequately.