| Forewords and objectivesLong-term survival rate is an essential outcome measure for patients with cancer. To reflect prognosis of patients and effects of cancer prevention and control,the most up-to-date data should be used to derive updated survival rates.Formerly cohort survival analysis and complete survival analysis are the commonly used methods to calculate survival rates.These two methods request that all or part of the objects in cohorts must be followed up over a certain time.So the results have strong hysteresis and can not reflect recent prognosis of newly diagnosed and treated patients in cases of recent improvement in prognosis.In the 1990s a new method of survival analysis,denoted "period survival analysis",has been proposed by Hermann Brenner and Olaf Gefeller in order to obtain more up-to-date estimates of long-term survival rates.In this approach,patients that diagnosed in recent n years up to interested period are recruited.The updated survival rates are obtained to reflect effects of disease treatments,not have to lag n years.In this approach,only patients that being surviving or diagnosed during the interested period are included to get survival rates.Recent diagnosed cases account for large proportion,so the results of survival rates can timely reflect progresses in medical science and disease diagnosis and treatments.Now the approach of period survival analysis has been used to estimate survival rates,publish the papers and reports in European and American areas.In our domestic cancer registries,the technologies and methods are out of date,and this new method of survival analysis has not been applied.In this study,the objectives are using period method to calculate 5-year survival rates of esophageal and stomach cancer that diagnosed from 1988 to 2004 of population in Linzhou city,analyse variations in 1990～1994,1995～1999 and 2000～2004 three periods,compare the results obtained from cohort survival analysis and evaluate the effects of the secondary prevention,diagnosis and treatments on these cancers in this area.Methods1.All incidence data for esophageal and stomach cancer were drawn from Linzhou Cancer Registry.With the closing date of December 31,2007,all incidence data were linked to death database.Cases in each database were matched.Records with unsure outcomes were followed up to get accurate informations.Dates of migration for settlers were recorded.Those records that were identified as duplicate cases or those belong to death certificate only were excluded.2.The sex-specific tumor cause eliminated abridged life tables were calculated in 1992,1997 and 2002.The probabilities of survival were calculated by formula 1-probability of death.Then the probabilities of survival were split,one year as an age group,to obtained sex-specific and age-specific probabilities of survival in general population.3.Micro program "period" witch compiled by Brenner et al were implemented and the relevant parameters were set up,to get the 5-year observed and relative survival rates and 95%confidence intervals(95%CI) on esophageal and stomach cancer in 1990～1994,1995～1999 and 2000～2004 three periods.The relative survival rate curves were plotted.4.Data were grouped by sex,and cases of stomach cancer were classed to gastric cardiac cancer and non-cardiac cancer by carcinoma sites.The observed and relative survival rates and 95%CI of each group in three periods were calculated,as well as the relative survival rate curves were plotted.5.The approach of cohort survival analysis was used to get survival rates and 95%CI on esophageal and stomach cancer in different periods,and the results were compared with that using period survival analysis to detect the changes in survival rates.Results1.For esophageal cancer,the 5-year observed survival rates(95%CI) among three periods were 14.50%(13.27%～15.74%),18.60%(17.27%～19.93%) and 24.87% (23.36%～26.38%) respectively,the relative survival rates(95%CI) were 28.24% (25.83%～30.65%),35.24%(32.73%～37.75%) and 40.76%(38.29%～43.23%).It showed an increasing trend by periods.During the two former periods,the 5-year observed survival rates in men were lower than that in women,but the 5-year relative survival rates were higher in men than that in women.In 2000～2004 period,the 5-year observed and relative survival rates in women were higher than that in men.2.For stomach cancer,the 5-year observed survival rates(95%CI) among three periods respectively were 12.29%(11.04%～13.54%),15.30%(14.01%～16.59%) and 23.02%(21.49%～24.55%),the 5-year survival rates(95%CI) were 26.66% (23.92%～29.40%),32.01%(29.29%～34.73%) and 40.43%(37.73%～43.14%).It also showed an increasing trend by periods.The 5-year survival rates were higher in men than that in women.During the two former periods,the 5-year survival rates were higher in gastric cardiac cancer than that in non-cardiac cancer.In 2000-2004 period, the 5-year observed and relative survival rates in gastric cardiac cancer were lower than that in non-cardiac cancer.3.Except the relative survival rates on esophageal cancer in 1990-1994 and the relative survival rates on gastric cardiac cancer among 1990～1994 and 2000～2004, other relative survival rates in different periods that calculated using period analysis were higher than that using cohort analysis. Conclusions1.5-year observed and relative survival rates on esophageal and stomach cancer of population in Linzhou city were increasing in 1990～1994,1995～1999,2000～2004 three periods.It indicates a rising status in the secondary prevention and clinical diagnosis and treatments on esophageal and stomach cancer in this area.2.Using approach of period survival analysis to get survival rates can exactly and timely reflect recent conditions in diagnosis and treatments of diseases.Comparing with method of cohort survival analysis,the period method doesn't need to obtain survival rates n years later,so the hysteresis is much more less. |
PDF Full Text Request