Variation in use of the 2-week referral pathway for suspected cancer: cross-sectional analysis

Abstract

Background: A 2-Week Wait (2WW) referral pathway for earlier diagnosis of suspected cancer was introduced in England in 2000. Nevertheless, a significant proportion of patients with cancer are diagnosed by other routes (detection rate), only a small proportion of 2WW referrals have cancer (conversion rate) and there is considerable between-practice variation. Aim: This study examined use by practices of the 2WW referral in relation to all cancer diagnoses. Design and setting: A cross-sectional analysis of data extracted from the Cancer Waiting Times Database for all 2WW referrals in 2009 and for all patients receiving a first definitive treatment in the same year. Method: The age standardised referral ratio, conversion rate, and detection rate were calculated for all practices in England and the correlation coefficient for each pair of measures. The median detection rate was calculated for each decile of practices ranked by conversion rate and vice versa, performing nonparametric tests for trend in each case. Results: Data for 8049 practices, 865 494 referrals, and 224 984 cancers were analysed. There were significant correlations between referral ratio and conversion rate (inverse) and detection rate (direct). There was also a direct correlation between conversion and detection rates. There was a significant trend in conversion rate for deciles of detection rate, and vice versa, with a marked difference between the lowest and higher deciles. Conclusion: There is a consistent relationship between 2WW referral conversion rate and detection rate that can be interpreted as representing quality of clinical practice. The 2WW referral rate should not be a measure of quality of clinical care.