Psychometric properties of the Needs Assessment Tool - Progressive Disease Cancer in UK Primary Care Journal of Pain and Symptom Management

Background: The assessment of patients’ needs for care is a critical step in achieving patient-centred cancer care. Tools can be used to assess needs and inform care planning. The Needs Assessment Tool: Progressive Disease- Cancer (NAT: PD-C) is an Australian oncology clinic tool for assessment by clinicians of patients’ and carers’ palliative care needs. This has not been validated in the UK Primary Care setting. Aim: To test the psychometric properties and acceptability of a UK primary-care adapted NAT:PD-C. Design: Reliability: NAT: PD-C -guided video-recorded consultations were viewed, rated and re-rated by clinicians. Weighted Fleiss’ kappa and PABAK statistics were used. Construct: During a consultation GPs used NAT:PD-C, patient measures (Edmonton Symptom Assessment Scale; Research Utilisation Group Activities of Daily Living; Palliative care Outcome Score; Australian Karnofsky Performance Scale) and carer measures (Carer Strain Index; Carer Support Needs Assessment Tool). Kendall’s Tau-b was used. Setting/Participants: General medical practitioners (GPs), nurses, patients and carers were recruited from primary care practices. Results: Reliability: All patient wellbeing items and 4/5 items in the carer/family ability to care section showed adequate inter-rater reliability. There was moderate test-retest reliability for 5/6 in the patient wellbeing section and 5/5 in the carer/family ability to care section. Construct: There was at least fair agreement for 5/6 of patient wellbeing items; high for daily living (Kendall’s Tau-b =0.57, p<0.001). The NAT:PD-C has adequate carer construct validity (5/8) with strong agreement for 2/8. Over three-quarters of GPs considered the NAT: PD-C to have high acceptability. Conclusion: The NAT PD-C is reliable, valid and acceptable the UK primary care setting. Effectiveness in reducing patient and carer unmet need and issues regarding implementation are yet to be evaluated.