The Norwegian Cancer Registry (CRN) was created in 1953 and contains mandatory information about all new cases of cancer and precancerous lesions. Information from clinical reports, pathological reports and death certificates are the main sources of notification and provide information on location, histological type and disease stage at diagnosis. CRN’s coding and classification system complies with international standards 🇧🇷 The CRN also recorded causes of death for cancer patients (from the Causes of Death Registry administered by the Norwegian Institute of Public Health), available from the 1960s.
The Norwegian Cervical Cancer Screening Program (NCCSP) is an integral part of the national health system. The CRN administers the program and receives mandatory reports from private and public pathology and microbiology laboratories. The program maintains complete records of the results of all Pap smears, histology specimens and HPV tests. Data from individual screenings are recorded and organized into four sub-registries: Cytology Registry, Histology Registry, HPV Test Registry and NIC Registry, the latter containing follow-up and treatment data. The SNOMED coding system, with some local adaptations, is used for classification (cytology and histology).
All residents of Norway are assigned a unique identification number used in all administrative and medical records/databases. This ID number allows for accurate record linking.
This cohort study included all women diagnosed with cervical cancer (International Classification of Diseases (ICD)-10; C53) in Norway during 1953–2013 (n = 21,160). No cases with ICD-10 code 55 were included. For 34 women who had two cervical cancer diagnoses, only the first recorded diagnosis was included.
Diagnosis period (1953–68, 1969–83, 1984–98, and 1999–13), stage distribution (International Federation of Gynecology and Obstetrics: FIGO (1986); stages 1–4), morphology (squamous cell carcinoma, adenocarcinoma, adenosquamous carcinoma, other malignancies and unspecified morphology)  and screening history; Smears obtained from 3.5 years and up to six months before diagnosis (no smear, normal smears only, abnormal smears, and unsatisfactory smears only), by age groups (< 25, 25–29, and ≥ 30 years) were descriptively displayed and analyzed using Contingency Tables. Screening history information was available for cancers diagnosed on or after July 1, 1995 🇧🇷 Chi-square tests were used to assess differences in distributions between age groups. A p-value below 0.05 was considered statistically significant.Crude incidence rates (per 100,000 woman-years) of cervical cancer by age group (< 30 and ≥ 30 years) and age-standardized rates (world standard population), 1955–2014, were calculated separately using supporting CRN data (Fig. 1), taking into account immigration and emigration. Time trends in age-standardized rates were estimated using the National Cancer Institute's Joinpoint Trend Analysis Software (Version 184.108.40.206). We analyzed time trends in incidence rates in women < 30 and ≥ 30 years. For the junction point analysis, we used age-specific cervical cancer figures provided by CRN, population figures from Statistics Norway, over 12 five-year periods (1955–59, …, 2010–14), and age standardization weights for the WHO World Population Standard 🇧🇷 The entire period 1955-2014 was segmented by trend change points, and the annual percentage change (APC) in rates between trend change points was estimated. Subsequently, the mean annual percentage change (AAPC) for the entire study period was calculated as a weighted average of the estimated APC in each segment, using segment lengths as weights. 🇧🇷
Nelson-Aalen cumulative risk function for the risk of dying from cervical cancer, with 95% confidence intervals (CIs), over a 15-year follow-up by age (< 30 and ≥ 30 years) and period of diagnosis ( 1960–73, 1974–86, 1987–99 and 2000–13), was calculated [18, 19]🇧🇷 The start of follow-up was from the date of diagnosis and individuals were followed up until emigration, death, 15 years after diagnosis or end of follow-up on December 31, 2016, whichever occurred first. Follow-up ended 15 years after diagnosis due to the relatively low number of deaths.Risk ratios (HRs) of cervical cancer mortality with 95% CIs, with and without adjustment for stage (FIGO stages 1–4), morphology (squamous cell carcinoma, adenocarcinoma, adenosquamous carcinoma, other malignancies, and unspecified morphology) and time from diagnosis (continuous), were derived from Cox proportional hazards regression models. We also present estimates for specific diagnostic periods (1960–73, 1974–86, 1987–99, and 2000–13).
Data were analyzed using IBM SPSS Statistics 22 (IBM Corporation, Armonk, NY, USA) and Stata/IC 14.0 (StataCorp LP, College Station, TX, USA).