FIELD TESTING AND PRELIMINARY ITEM ANALYSIS OF THE PATIENT ASSESSMENT AND MANAGEMENT TOOL PALLIATIVE CARE (PAMT-PC): A 37-ITEM KNOWLEDGE QUESTIONNAIRE AMONG HEALTHCARE PROFESSIONALS IN ANDHRA PRADESH, INDIA

Abstract

Background: Healthcare professionals in India lack validated, context-specific instruments for assessing palliative care knowledge. The only available Indian tool — the Palliative Care Knowledge Questionnaire-Basic (PCKQ-B)— was developed exclusively for Delhi-NCR nursing staff, lacks factor analysis, and omits Communication Skills and Procedural Skills domains mandated by India’s Competency-Based Medical Education (CBME 2019) curriculum. This study reports the development, content validation, pilot testing, field testing, and test-retest reliability assessment of the Patient Assessment and Management Tool- Palliative Care (PAMT-PC), a 37-item, seven-domain palliative care knowledge questionnaire for multidisciplinary healthcare professionals in Andhra Pradesh, India. Methods: A four-phase, COSMIN-guided instrument development study was conducted: (1) systematic item generation from five existing tools and CBME 2019 competencies; (2) expert panel content validity assessment (n = 9 experts, two Delphi rounds); (3) pilot testing (n = 30) with cognitive debriefing; and (4) field testing on n = 230 healthcare professionals (173 nurses, 33 doctors, 24 allied health workers) at Government General Hospital (GGH) Kakinada, Andhra Pradesh; all four phases were completed sequentially between December 2023 and March 2025. Test-retest reliability was assessed in a sub-sample of n = 45 using supervised paper-based administration at a two-week interval. Results: Content validity: S-CVI/Ave = 0.868; mean I-CVI = 0.868 (range 0.778–1.000); mean κ* = 0.860; 15 items revised across two Delphi rounds; all 37 items achieved I-CVI ≥0.778. Pilot testing: median completion time 19.8 min; preliminary KR-20 = 0.551; five items revised. Field test (n = 230): mean score 23.78 ± 4.54/37 (64.3%); KR-20 = 0.720 (acceptable); mean inter-item correlation = 0.063; 22/37 items retained; 12 flagged for review; three items (Q23, Q24, Q32a) showed negative point-biserial correlations and require revision. EFA sampling adequacy: KMO = 0.64; Bartlett’s χ²(666) = 1,768.4, p < 0.001. Test-retest (n = 45): ICC(2,1) = 0.907 (95% CI: 0.837–0.947; excellent); SEM = 1.24 score points; Bland–Altman limits of agreement −3.70 to +3.53. Known-groups validity: prior training significantly predicted higher scores (t(195) = 3.47, p < 0.001, Cohen’s d = 0.626). Conclusion: The PAMT-PC demonstrated near-acceptable content validity, acceptable internal consistency (KR 20 = 0.720), and excellent test-retest stability (ICC = 0.907) in this preliminary validation sample. Three items require immediate revision and 12 require review. Multi-district validation (n ≥ 500) and confirmatory factor analysis are the recommended next steps toward full psychometric validation.