- Email: [email protected]
Effect of organized mammography screening on breast cancer mortality: a population-based cohort study in Norway
European Congress of Epidemiology – “Crises, epidemiological / Revue d’Épidémiologie et de Santé Publique 66S (2018) S233–S276 S6.5
S6.6
A case-control study on cervical cancer and frequency of participation in screening
Effect of organized mammography screening on breast cancer mortality: a population-based cohort study in Norway
L. Fiengo Tanaka ∗ , S. Schülein , D. Schriefer , K. Radde , O. Schoffer , S. Klug Epidemiology, Department of Sport and Health Sciences, Technical University of Munich, Munich, Germany ∗ Corresponding author. E-mail address: [email protected] (L. Fiengo Tanaka) Introduction Since 1971, all women in Germany have been entitled to an annual free Pap smear for cervical cancer screening (CCS), starting at age 20. Despite the existence of an opportunistic program, cervical cancer still occurs at somewhat elevated rates (9.3 cases per 100,000 women, age-standardized rate, European Standard Population) in Germany in comparison to other Western European countries. Few previous studies reported on the association between participation in CCS and cervical cancer. The TeQaZ study is a population-based case-control study investigating the association between participation in CCS, other risk factors and cervical cancer in Germany. Methods Incident cases of cervical cancer (ICD-10 C53), diagnosed between 2012 and 2016 in the German states of Saxony, Rhineland-Palatinate and the neighboring regions of Baden-Württemberg, Hesse, North-Rhine Westphalia, Saarland, Brandenburg, Saxony-Anhalt and Thuringia, were recruited. Cases were matched in a 1:3 ratio with randomly selected population-based controls, recruited via population registries based on age (± 2 years of birth date) and region of residence. Via computer-assisted telephone interviews (CATI, Voxco), cases and controls reported their frequency of CCS participation during the past ten years as well as sociodemographic characteristics, such as income, schooling and other risk factors, namely tobacco smoking, body mass index (BMI, calculated from reported weight, height), use of oral contraceptives, history of chlamydia, condyloma and herpes. Univariable and multivariable conditional logistic regression analyses were performed to estimate the association between CCS participation and cervical cancer using the statistical software SAS. Results A total of 218 cases and 654 controls were included in the analysis. Over 60% of cases and controls were aged 20–39 years. In the ten years prior to the interview, 56.7% of cases and 86.0% of controls attended CCS at least every three years. Fifteen cases (6.9%) and one control (0.2%) had never attended CCS in the past ten years. Among cases, squamous cell carcinoma prevailed (78.9%). In the univariable analyses risk factors for cervical cancer were: CCS participation less frequently than every three years (OR 5.39; 95% CI 3.57–8.14), having ever smoked (OR 1.87; 95% CI 1.34–2.61), having a BMI ≥ 30 (OR 1.69; 95% CI 1.11–2.57), having at least four children (OR 6.96; 95% CI 1.71–28.37) and having had more than one sexual partner (OR 3.11; 95% CI 1.78–5.42). Completing secondary education (OR 0.35; 95% CI 0.23–0.51), engaging sporting activity at least once a week (OR 0.50; 95% CI 0.36–0.70) and a net household income of D 3000 or more (OR 0.29; 95% CI 0.19–0.45) were shown to be protective. In the adjusted conditional logistic regression, participating in CCS less frequently than every three years (OR 4.94; 95% CI 2.74–8.91), having at least four children (OR 6.96; 955 CI 1.71–28.37), having had more than one sexual partner (OR 3.91; 95% CI 1.86–8.23) and having ever had a herpes infection (OR 4.62; 95% CI 1.16–18.40) were statistically significantly associated with developing cervical cancer. Completing secondary education (OR 0.42; 95% CI 0.24–0.73) and a net monthly household income of D 3000 or more (OR 0.43; 95% CI 0.25–2.75) remained protective factors. Conclusion Although infrequent participation (less than every three years) was strongly associated with cervical cancer, more than half of cases had attended CCS in the three years prior to diagnosis, suggesting that it is necessary to further investigate the quality of cytology and adequate treatment and monitoring of precancerous lesions in Germany. Disclosure of interest est.
The authors declare that they have no competing inter-
https://doi.org/10.1016/j.respe.2018.05.041
S249
M.H. Møller a , M.L. Lousdal a , I.S. Kristiansen b , H. Støvring a,∗ Department of Public Health, Aarhus University, Aarhus, Denmark b Department of Health Management and Health Economics, Institute of Health and Society, Oslo University, Oslo, Norway ∗ Corresponding author. E-mail address: [email protected] (H. Støvring)
a
Background Reviews of the randomized controlled trials (RCTs) of mammography screening conducted during the 1970s and 1980s reported a 15–21% reduction in breast cancer mortality. Improved cancer treatment and enhanced awareness may however have reduced the potential net benefit from mammography screening. New RCTs are no longer feasible and observational studies yield widely varying estimates of the reduced breast cancer mortality (0% to 43%), depending on study design. In Norway, mammography screening was introduced gradually during the period 1996–2004. Using this as a natural experiment, we aimed to estimate the effect of organized mammography screening on incidence-based breast cancer mortality by comparing changes in mortality among women eligible for screening to concurrent changes in younger and older ineligible women. The use of un-screened age-groups as control groups allowed us to conduct a balanced incidence-based mortality study in which we aimed to distinguish the screening effect on breast cancer mortality from the effect of the concomitantly established diagnostic centers, which may also have improved treatment. Methods In a national county-wise balanced, open-cohort study in Norway, we used birth cohorts (1896–1982) to construct three age groups in both the historical and screening period: women eligible for screening and younger or older women ineligible for screening. We included women diagnosed with breast cancer who died within the same age-period group during 1987–2010 (n = 4903). We estimated relative incidence-based mortality rate ratios (relative MRR) comparing temporal changes in eligible women to concurrent changes in ineligible women. Additionally, we conducted analyses comparing the change in eligible women to younger, ineligible women with either continued accrual and followup period (eligible women only) or continued follow-up period. Finally, we checked for unadjusted confounding in two negative control analyses expected to produce a null result (relative MRR = 1). First, we compared the mortality change in younger women to the concurrent change in older women. Second, we extended the study period to include the period 1987–2010 for all counties and repeated our analysis as if screening had been introduced at the midpoint of the actual pre-screening period and the actual screening period in each county, respectively. Results All three age groups experienced a reduction in mortality when screening was introduced, but the decrease among eligible women was about the same among ineligible women (relative MRR = 1.05, 95% CI: [0.94–1.18]). Varying the definition of follow-up yields similar results. The negative control analyses all had relative MRRs close to 1 (estimates ranged from 0.91 to 1.04 and with all 95% CIs overlapping 1). Conclusion Mammography screening was not associated with a larger breast cancer mortality reduction than that of women ineligible for mammography screening. Based on the negative control analyses we found limited uncontrolled confounding in the applied observational design. Disclosure of interest est.
The authors declare that they have no competing inter-
https://doi.org/10.1016/j.respe.2018.05.042
S6.6
A case-control study on cervical cancer and frequency of participation in screening
Effect of organized mammography screening on breast cancer mortality: a population-based cohort study in Norway
L. Fiengo Tanaka ∗ , S. Schülein , D. Schriefer , K. Radde , O. Schoffer , S. Klug Epidemiology, Department of Sport and Health Sciences, Technical University of Munich, Munich, Germany ∗ Corresponding author. E-mail address: [email protected] (L. Fiengo Tanaka) Introduction Since 1971, all women in Germany have been entitled to an annual free Pap smear for cervical cancer screening (CCS), starting at age 20. Despite the existence of an opportunistic program, cervical cancer still occurs at somewhat elevated rates (9.3 cases per 100,000 women, age-standardized rate, European Standard Population) in Germany in comparison to other Western European countries. Few previous studies reported on the association between participation in CCS and cervical cancer. The TeQaZ study is a population-based case-control study investigating the association between participation in CCS, other risk factors and cervical cancer in Germany. Methods Incident cases of cervical cancer (ICD-10 C53), diagnosed between 2012 and 2016 in the German states of Saxony, Rhineland-Palatinate and the neighboring regions of Baden-Württemberg, Hesse, North-Rhine Westphalia, Saarland, Brandenburg, Saxony-Anhalt and Thuringia, were recruited. Cases were matched in a 1:3 ratio with randomly selected population-based controls, recruited via population registries based on age (± 2 years of birth date) and region of residence. Via computer-assisted telephone interviews (CATI, Voxco), cases and controls reported their frequency of CCS participation during the past ten years as well as sociodemographic characteristics, such as income, schooling and other risk factors, namely tobacco smoking, body mass index (BMI, calculated from reported weight, height), use of oral contraceptives, history of chlamydia, condyloma and herpes. Univariable and multivariable conditional logistic regression analyses were performed to estimate the association between CCS participation and cervical cancer using the statistical software SAS. Results A total of 218 cases and 654 controls were included in the analysis. Over 60% of cases and controls were aged 20–39 years. In the ten years prior to the interview, 56.7% of cases and 86.0% of controls attended CCS at least every three years. Fifteen cases (6.9%) and one control (0.2%) had never attended CCS in the past ten years. Among cases, squamous cell carcinoma prevailed (78.9%). In the univariable analyses risk factors for cervical cancer were: CCS participation less frequently than every three years (OR 5.39; 95% CI 3.57–8.14), having ever smoked (OR 1.87; 95% CI 1.34–2.61), having a BMI ≥ 30 (OR 1.69; 95% CI 1.11–2.57), having at least four children (OR 6.96; 95% CI 1.71–28.37) and having had more than one sexual partner (OR 3.11; 95% CI 1.78–5.42). Completing secondary education (OR 0.35; 95% CI 0.23–0.51), engaging sporting activity at least once a week (OR 0.50; 95% CI 0.36–0.70) and a net household income of D 3000 or more (OR 0.29; 95% CI 0.19–0.45) were shown to be protective. In the adjusted conditional logistic regression, participating in CCS less frequently than every three years (OR 4.94; 95% CI 2.74–8.91), having at least four children (OR 6.96; 955 CI 1.71–28.37), having had more than one sexual partner (OR 3.91; 95% CI 1.86–8.23) and having ever had a herpes infection (OR 4.62; 95% CI 1.16–18.40) were statistically significantly associated with developing cervical cancer. Completing secondary education (OR 0.42; 95% CI 0.24–0.73) and a net monthly household income of D 3000 or more (OR 0.43; 95% CI 0.25–2.75) remained protective factors. Conclusion Although infrequent participation (less than every three years) was strongly associated with cervical cancer, more than half of cases had attended CCS in the three years prior to diagnosis, suggesting that it is necessary to further investigate the quality of cytology and adequate treatment and monitoring of precancerous lesions in Germany. Disclosure of interest est.
The authors declare that they have no competing inter-
https://doi.org/10.1016/j.respe.2018.05.041
S249
M.H. Møller a , M.L. Lousdal a , I.S. Kristiansen b , H. Støvring a,∗ Department of Public Health, Aarhus University, Aarhus, Denmark b Department of Health Management and Health Economics, Institute of Health and Society, Oslo University, Oslo, Norway ∗ Corresponding author. E-mail address: [email protected] (H. Støvring)
a
Background Reviews of the randomized controlled trials (RCTs) of mammography screening conducted during the 1970s and 1980s reported a 15–21% reduction in breast cancer mortality. Improved cancer treatment and enhanced awareness may however have reduced the potential net benefit from mammography screening. New RCTs are no longer feasible and observational studies yield widely varying estimates of the reduced breast cancer mortality (0% to 43%), depending on study design. In Norway, mammography screening was introduced gradually during the period 1996–2004. Using this as a natural experiment, we aimed to estimate the effect of organized mammography screening on incidence-based breast cancer mortality by comparing changes in mortality among women eligible for screening to concurrent changes in younger and older ineligible women. The use of un-screened age-groups as control groups allowed us to conduct a balanced incidence-based mortality study in which we aimed to distinguish the screening effect on breast cancer mortality from the effect of the concomitantly established diagnostic centers, which may also have improved treatment. Methods In a national county-wise balanced, open-cohort study in Norway, we used birth cohorts (1896–1982) to construct three age groups in both the historical and screening period: women eligible for screening and younger or older women ineligible for screening. We included women diagnosed with breast cancer who died within the same age-period group during 1987–2010 (n = 4903). We estimated relative incidence-based mortality rate ratios (relative MRR) comparing temporal changes in eligible women to concurrent changes in ineligible women. Additionally, we conducted analyses comparing the change in eligible women to younger, ineligible women with either continued accrual and followup period (eligible women only) or continued follow-up period. Finally, we checked for unadjusted confounding in two negative control analyses expected to produce a null result (relative MRR = 1). First, we compared the mortality change in younger women to the concurrent change in older women. Second, we extended the study period to include the period 1987–2010 for all counties and repeated our analysis as if screening had been introduced at the midpoint of the actual pre-screening period and the actual screening period in each county, respectively. Results All three age groups experienced a reduction in mortality when screening was introduced, but the decrease among eligible women was about the same among ineligible women (relative MRR = 1.05, 95% CI: [0.94–1.18]). Varying the definition of follow-up yields similar results. The negative control analyses all had relative MRRs close to 1 (estimates ranged from 0.91 to 1.04 and with all 95% CIs overlapping 1). Conclusion Mammography screening was not associated with a larger breast cancer mortality reduction than that of women ineligible for mammography screening. Based on the negative control analyses we found limited uncontrolled confounding in the applied observational design. Disclosure of interest est.
The authors declare that they have no competing inter-
https://doi.org/10.1016/j.respe.2018.05.042