Equity of access and variation in general surgeons’ clinical judgements of patient priority

Abstract

Rationale, aims and objectives: In New Zealand, access to publicly funded elective surgery is rationed by explicit prioritisation of patients, with those scoring above a threshold offered surgery and others referred back to their general practitioner for on-going management. The reliability and validity of the priority scoring systems which underpin this approach to managing the demand for elective surgery have been questioned on many counts, not least their implications for a properly person-centered medicine. The objective of this research was to examine variability in surgeon priority scoring within general surgery.Method: Using an observational cross-sectional study design, 911 patients were assessed by one of 48 general surgeons who rated priority for surgery using a 100mm visual analog scale. Patients completed a questionnaire recording disease severity indicators. A hierarchical Bayesian model was used to estimate between and within surgeon variation in priority scoring, adjusting for patient-reported disease severity.Results: Across 3 study centres and 3conditions, estimated standard deviations ranged from 2.0 (95% CI 0.1-6.5) to 11.6 (7.2-20.3) and from 6.1 (4.7-8.0) to 15.6 (12.8 -19.9) for between and within surgeon variability, respectively. In the majority of comparisons, within-surgeon variation exceeded between-surgeon variation. Using the hierarchical model we estimated the probability that two specialist surgeons scoring patients with identical patient-reported severity indicators would agree to within 10 points. These estimates ranged from 0.31 (95% CI 0.24-0.37) to 0.72 (0.57-0.80).Conclusion: There was, in general, marked variability in surgeons’ scoring of patient priority.  Variation in scoring of patient priority must affect equity of access to surgery and therefore have direct implications for person-centered medicine.. Strategies to reduce this variation are warranted.