PURPOSE: Census tract variables have not been widely available for SEER-wide data due to several technical reasons; thus, prior studies have been conducted on a specific-community basis only or used county-level data. This study is the first to evaluate long-term chronological trends in cancer survival by selected socioeconomic variables of census tract level based on multiple SEER registry data. METHODS: 177,128 breast cancer and 45,615 non-Hodgkin's lymphoma (NHL) cases diagnosed from 1973-2007 and followed through 2009 from 4 SEER registries (Detroit, Hawaii, Utah and Seattle-Puget Sound), were linked to decennial census tract data (1970-2000). Five-year relative survival was calculated using the lifetable method according to census tract poverty levels and by year of diagnosis. The Cox proportional hazard model was used to estimate hazard ratios (HR) for death in 5 years from diagnosis, adjusted for selected covariates and SEER historical stage in the limited models only. RESULTS: Although the 5-year relative survival from both cancers improved similarly across poverty levels as percent per year, absolute increase per year was greater for lower poverty neighborhoods. This trend was most consistently observed for distant stage of cancer. The multivariable HR were significantly higher in the highest poverty group (greater than 20%), 1.41 and 1.33, for breast cancer and NHL respectively, than the lowest (greater than or equal to 5%). Additional adjustment for stage at diagnosis reduced the HR in the highest poverty level in breast cancer patients to 1.30, but had minimal effect on NHL. Socioeconomic disparities in overall survival were more evident in neighborhoods with higher proportions of racial minorities and in middle-aged patients than younger or older patients. CONCLUSIONS: Relative survival for both types of cancer improved over last 35-year period across poverty levels, but absolute differences increased. More studies are needed to develop innovative community-level interventions.
PURPOSE: Census tract variables have not been widely available for SEER-wide data due to several technical reasons; thus, prior studies have been conducted on a specific-community basis only or used county-level data. This study is the first to evaluate long-term chronological trends in cancer survival by selected socioeconomic variables of census tract level based on multiple SEER registry data. METHODS: 177,128 breast cancer and 45,615 non-Hodgkin's lymphoma (NHL) cases diagnosed from 1973-2007 and followed through 2009 from 4 SEER registries (Detroit, Hawaii, Utah and Seattle-Puget Sound), were linked to decennial census tract data (1970-2000). Five-year relative survival was calculated using the lifetable method according to census tract poverty levels and by year of diagnosis. The Cox proportional hazard model was used to estimate hazard ratios (HR) for death in 5 years from diagnosis, adjusted for selected covariates and SEER historical stage in the limited models only. RESULTS: Although the 5-year relative survival from both cancers improved similarly across poverty levels as percent per year, absolute increase per year was greater for lower poverty neighborhoods. This trend was most consistently observed for distant stage of cancer. The multivariable HR were significantly higher in the highest poverty group (greater than 20%), 1.41 and 1.33, for breast cancer and NHL respectively, than the lowest (greater than or equal to 5%). Additional adjustment for stage at diagnosis reduced the HR in the highest poverty level in breast cancerpatients to 1.30, but had minimal effect on NHL. Socioeconomic disparities in overall survival were more evident in neighborhoods with higher proportions of racial minorities and in middle-aged patients than younger or older patients. CONCLUSIONS: Relative survival for both types of cancer improved over last 35-year period across poverty levels, but absolute differences increased. More studies are needed to develop innovative community-level interventions.
Authors: Tisha M Felder; Kathryn L Braun; Lisa Wigfall; Maria Sevoyan; Shraddha Vyas; Samira Khan; Heather M Brandt; Charles Rogers; Sora Tanjasiri; Cheryl A Armstead; James R Hébert Journal: J Cancer Educ Date: 2019-06 Impact factor: 2.037
Authors: Erin E Kent; Nancy Breen; Denise R Lewis; Janet S de Moor; Ashley Wilder Smith; Nita L Seibel Journal: Cancer Causes Control Date: 2015-06-18 Impact factor: 2.506
Authors: S Lamy; C Bettiol; P Grosclaude; G Compaci; G Albertus; C Récher; J C Nogaro; F Despas; G Laurent; C Delpierre Journal: BMC Health Serv Res Date: 2016-08-02 Impact factor: 2.655
Authors: Harry Comber; Marianna De Camargo Cancela; Trutz Haase; Howard Johnson; Linda Sharp; Jonathan Pratschke Journal: PLoS One Date: 2016-12-19 Impact factor: 3.240