Patient Concern Inventory: Items as a Predictive QOL Screening Tool

Objectives: To explore the Patient Concern Inventory (PCI) items as predictors for a quality of life (QOL) screening instrument in head/neck cancer (HNC) treatment.
Methods: Consecutive HNC patients of the UF Oral Medicine and ENT clinics were invited to complete a questionnaire that included: socio-demographic, PCI, and University of Washington Quality of Life (UW-QOLv4) instruments. The PCI is a clinical questionnaire used in British HNC practices. The UW-QOLv4 physical and social function domains were dichotomized by lower vs. upper 3 quartiles as outcomes. Using recursive partitioning (CHAID), all PCI items (45), sex and age were tested as predictors (Bonferroni adjusted chi-sq <0.05). Significant PCI predictors were produced separately for subjects completing radiation treatment ≤12 months and > 12 months. Sensitivity, specificity, predictive values and false positive/negative rates were produced for utilizing CHAID identified significant predictors in screening for poor QOL scores.
Results: 118 subjects met inclusion criteria. The average age (sd) was 61.8 (9.8) years old; 71% were males. Treatment modalities included radiation (37%) and radiation & surgery (63%). The mean scores for the Physical function (PF) and Social function (SF) domains were 75.7 (13.9) and 77.2 (15.5), respectively. Significant PCI concerns and their cumulative screening utility measures are in Table 1:
Conclusions: These preliminary findings suggest that specific, patient identified PCI concerns coupled with patient characteristic may be found useful as a screening tool to assure identification and support for those with poor QOL while minimizing unnecessary referrals.