Fig. 2.

Plots of odds ratios (ORs) for the interactions between age, comorbidity, socioeconomic status and ethnicity on the association between the relative proximity of radiotherapy and surgical services to a patient's residence and the choice of treatment modality. The figure highlights equity differences between patient groups in the odds of receiving radiation therapy (RT) or radical prostatectomy (RP) according to the relative proximity of radiotherapy and surgical services to their residence. Interaction terms derived from a multivariable regression model were used to assess to what extent the observed associations between the relative proximity of the two treatment options and the treatment they actually received varied according to four patient characteristics: age (men older than 70 vs men aged 70 years or younger); comorbidity (men with one or more comorbidity as defined by the Charlson comorbidity index vs men with no comorbidity); socioeconomic status (men from more deprived backgrounds [IMD 3–5] vs men from less deprived backgrounds [IMD 1–2]); and ethnicity (men from ethnic minority groups vs White men). An OR >1 means that the chance of receiving RT is higher than when RT and RP are both offered at the nearest hospital (OR always 1 as it was the reference). The ORs are plotted with 95% CIs (horizontal lines). P values are presented, showing whether the differences observed between the plots within patient groups are statistically significant. For example, the association between age and relative proximity of treatment options on treatment choice varied when comparing the ORs for younger (age ≤ 70 years) and older men (age >70 years; P value <0.001).