Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2018

Utilising non-admitted code sets for an Outpatient cancer patient experience survey (#341)

Peter McGuire 1 , Karen Hay 2 , Eliza Bott 1 , Maree Bransdon 1
  1. Central Integrated Regional Cancer Service, Queensland Health, Brisbane, QLD, Australia
  2. QIMR Berghofer Medical Research Institute, Brisbane, Queensland, Australia


This analysis aimed to determine the prevalence of cancer within the sample population of the Queensland pilot outpatient patient experience survey 2018.  The survey aimed to identify and sample people with cancer who had presented to a hospital outpatient appointment during the month of November 2017. The study aimed to compare the prevalence of cancer for specified Independent Hospital Pricing Authority Tier 2 non-admitted service definitions by sampling data that does not contain patient diagnosis.


The survey sample was drawn from Queensland Health’s non-admitted patient data collection codes for medical oncology, radiation oncology, urology, breast and gynaecology oncology.

The survey analysis included determining if the participant added outpatients related to a cancer diagnosis was analysed using the SURVEYFREQ function on Stata statistical software package (Version 15) with a 95% confidence interval.


Of those participants attending the Tier 2 code clinics 81% (95% CI: 77-84%) reported a cancer-related diagnosis; , however this percentage varied between each clinic type. The distribution of people attending outpatient clinics for a cancer-related diagnosis was Medical Oncology at 99.8% (95% CI: 99.2%-100%) followed by Radiation oncology at 97.2% (95% CI:91.8%-99.1%), Breast/Gynaecology Oncology at 92.5% (95% CI:76.9-97.9) and then Urology at 43.8 % (95% CI:35.6%-50.4%).

If Urology had been excluded from the Tier 2 code clinics captured, the average would have increased to 97.9% (95% CI: 95.2-99.1%) of respondents presenting to outpatient clinics with a cancer-related diagnosis.


Future studies may choose to exclude Urology clinics if the goal is to reach a higher proportion of people with a cancer-related diagnosis than this study achieved.  However, research approaches should consider both the prevalence of cancer and case-mix of cancer types that may be reached through Tier 2 clinic codes.  The study team strongly advocates for data collections that directly support cancer research.