A population-based analysis of incidence, mortality, and stage-specific survival of cervical cancer patients in Hong Kong: 1997-2006.

ABSTRACT

OBJECTIVES: To determine the trends in incidence and mortality of cervical cancer patients diagnosed during 1997 to 2006, and to describe stage-specific survival using population-based cancer registry data. DESIGN: Retrospective, population-based study. SETTING: Hong Kong. PATIENTS All patients diagnosed with cervical cancer between 1997 and 2006. Patients eligible for survival analysis were followed up till 31 December 2007. MAIN OUTCOME MEASURES: Age-standardised incidence and mortality rates and average annual percent changes in these parameters were calculated using the Poisson regression model. Survival was expressed as relative survival rate using a period approach. Hazard ratios of mortality including 95% confidence intervals for certain variables were estimated using the Cox proportional hazards model. RESULTS: During the 10-year period of the study, overall annual incidence and mortality rates decreased by 4.2% and 6.0%, respectively. Significant rates of reduction were observed in all age-groups except those younger than 45 years. The reduction in incidence of squamous cell carcinoma (3.6% annually) was less than that of adenocarcinoma (5.2%) and other histological types (6.8%). In all, 3807 (86.4%) of the patients were included in survival analysis. The overall 5-year relative survival rate was 71.3% (95% confidence interval, 69.5-73.1%), while the values for stages I, II, III, and IV were 90.9%, 71.0%, 41.7%, and 7.8%, respectively. Age, stage, and histology were independent prognostic factors. Survival of stage IA patients was as good as that of the general population. CONCLUSIONS: As in other industrialised countries, the incidence and mortality rate of cervical cancer were decreasing. Stage-specific population-based cancer survival was available for the first time, and was useful as an indicator of cancer control. Collaboration between public and private sectors to further improve the follow-up data could provide more comprehensive surveillance information.