Travel time and distance to health care only partially account for the ethnic inequalities in cervical cancer stage at diagnosis and mortality in New Zealand

Objective: To investigate whether travel time or distance to the nearest general practitioner (GP) and/or cancer centre accounts for the ethnic differences in cervical cancer screening, stage at diagnosis and mortality in New Zealand (NZ). Methods: The study involved 1,594 cervical cancer cases registered between 1994 and 2005. Travel time and distance to the GP and cancer centre were estimated using a Geographical Information System. Results: Adjustment for travel time or distance made almost no difference to ethnic differences in screening rates. Adjustment for travel time reduced the excess risk for late-stage diagnosis in M?ori (the odds ratio (OR) reduced from 2.71 (95%CI 1.98-3.72) to 2.59 (1.88- 3.56), a 7% decrease) and 33% in Pacific (the OR reduced from 1.39 (0.76-2.54) to 1.26 (0.68-2.33)) women. Adjustment for travel time reduced the excess risk for mortality by 3% in M?ori (the hazard ratio (HR) reduced from 1.59 (1.21-2.08) to 1.57 (1.19-2.06)) and 13% in Pacific (the HR reduced from 1.92 (1.20-3.08) to 1.80 (1.11-2.91)) women. Similar findings were observed when using travel distance rather than travel time. Conclusions: Travel time and distance are only weakly associated with cervical cancer screening, stage at diagnosis and mortality in NZ. However, travel time may account for a small proportion of the ethnic differences in stage at diagnosis, and to a lesser extent mortality, particularly for Pacific women. Implications: The findings suggest that there may be ethnic variations in access to treatment or treatment quality, which may be related to travel time. © 2012 The Authors.