Time trends and age-period-cohort analysis of cervical cancer mortality rate in Brazil

Diego Hernan Giunta, Mirian Carvalho de Souza, Maria Beatriz Kneipp Dias, Moyses Szklo, Liz Maria de Almeida

Research output: Contribution to journalArticlepeer-review


Background: Cervical cancer (CC) is a common preventable and curable disease that may lead to death. Our aim was to describe the patterns of time trends in CC mortality rates among women in Brazil from 1980 to 2017, and identify the influence of age, period and birth cohort (APC) stratified by region (North NR, Northeast NER, Southeast SER, South SR, Center-Western region CWR). Methods: We performed a time-series analysis using secondary data bases. Crude (MR) and WHO age-standardized CC mortality rates (aMR) were estimated per 100,000 women. We evaluated time trends using permutation joinpoint regression models (JP) and APC models to estimate the effect of APC on MR. Results: The JP analysis showed a temporal decrease in all regions, except the NR, which had an annual percentage increase of 0.44 (95%CI 0.2 - 0.7). MR in the NR was 2 to 4 times higher than in the other regions. We observed steady increases in MR with age in the NR and NER. A plateau after age 40 was observed in SER, SR, and CWR. The NR and NER MR ratio stabilized around the year 2000. Birth cohort effect showed decreasing MR ratio from 1900 to 1970 for all regions, except the NR, which showed increasing MR rate from older to more recent cohorts. Conclusion: We showed relevant differences in cervical MR by region, which may reflect inequality in access to primary and secondary prevention as well as treatment, particularly in the NR.

Original languageEnglish (US)
Article number100230
JournalJournal of Cancer Policy
StatePublished - Sep 2020


  • Brazil
  • Cause of death
  • Mortality
  • Regression analysis
  • Trends
  • Uterine cervical neoplasms

ASJC Scopus subject areas

  • Oncology
  • Health Policy


Dive into the research topics of 'Time trends and age-period-cohort analysis of cervical cancer mortality rate in Brazil'. Together they form a unique fingerprint.

Cite this