Cancer risk in individuals with autism spectrum disorder.

BACKGROUND: Whether individuals with autism spectrum disorders (ASDs) have a higher-than-expected risk of cancer remains unknown. PATIENTS AND METHODS: We carried out a population-based cohort study including 2.3 million individuals live-born to mothers from Nordic countries during 1987-2013 in Sweden with follow-up through 2016 (up to age 30 years). Individuals with ASD were ascertained through the Swedish National Patient Register. We estimated the relative risk of cancer in relation to ASD by odds ratios (ORs) and associated 95% confidence intervals (CIs) derived from logistic regression, after detailed adjustment for potential confounders. We also carried out a sibling comparison to address familial confounding and a genetic correlation analysis using the genome-wide association study summary statistics to address confounding due to potential polygenetic pleiotropy between ASD and cancer. RESULTS: We observed an overall increased risk of any cancer among individuals with ASD (OR 1.3, 95% CI 1.2-1.5), compared with individuals without ASD. The association for any cancer was primarily noted for narrowly defined autistic disorder (OR 1.7, 95% CI 1.3-2.1) and ASD with comorbid birth defects (OR 2.1, 95% CI 1.5-2.9) or both birth defects and intellectual disability (ID; OR 4.8, 95% CI 3.4-6.6). An association was also suggested for ASD with comorbid ID (OR 1.4; 95% CI 0.9-2.1), but was not statistically significant. ASD alone (i.e. without comorbid ID or birth defects) was not associated with an increased risk of any cancer (OR 1.0, 95% CI 0.8-1.2). Sibling comparison and genetic correlation analysis showed little evidence for familial confounding or confounding due to polygenetic pleiotropy between ASD and cancer. CONCLUSIONS: ASD per se is not associated with an increased risk for cancer in early life. The increased cancer risk among individuals with ASD is likely mainly attributable to co-occurring ID and/or birth defects in ASD.

A novel causal model for nasopharyngeal carcinoma.

The development of nasopharyngeal carcinoma (NPC) and its unique geographic distribution have long been attributed to a combination of dietary intake of salt-preserved fish, inherited susceptibility, and early-life infection with the Epstein-Barr virus (EBV). New findings from our large, rigorously designed, population-based case-control study of NPC in southern China have enabled substantial revision of this causal model. Here, we briefly summarize these results and provide an updated model of the etiology of NPC. Our new research identifies two EBV genetic variants that may be causally involved in the majority of NPC in southern China, and suggests the rise of modern environmental co-factors accompanying cultural and economic transformation in NPC-endemic regions. These discoveries can be translated directly into clinical and public health advances, including improvement of indoor air quality and oral health, development of an EBV vaccine, enhanced screening strategies, and improved risk prediction. Greater understanding of the roles of environmental, genetic, and viral risk factors can reveal the extent to which these agents act independently or jointly on NPC development. The history of NPC research demonstrates how epidemiology can shed light on the interplay of genes, environment, and infections in carcinogenesis, and how this knowledge can be harnessed for cancer prevention and control.

Inflammatory bowel disease and risk of adenocarcinoma and neuroendocrine tumors in the small bowel.

BACKGROUND: Uncertainty prevails about the magnitude of excess risk of small bowel cancer in patients with inflammatory bowel disease (IBD). PATIENTS AND METHODS: To quantify the risk of small bowel adenocarcinoma and neuroendocrine tumors in patients with ulcerative colitis (UC) and Crohn's disease (CD), we undertook a population-based cohort study of all patients with IBD diagnosed in Norway and Sweden from 1987 to 2016. Patients were followed through linkage to national registers. We calculated standardized incidence ratios (SIRs) with 95% confidence intervals (CIs) of small bowel adenocarcinomas and neuroendocrine tumors for patients with CD and UC. We excluded the first year of follow-up to reduce reverse causality. RESULTS: Among 142 008 patients with a median follow-up of 10.0 years, we identified 66 adenocarcinomas and 57 neuroendocrine tumors in the small bowel. The SIR of small bowel adenocarcinoma was 8.3 (95% CI 5.9-11.3) in CD and 2.0 (95% CI 1.2-3.1) in UC. The incidence rates of adenocarcinomas were highest in CD with stricturing disease and extent limited to the small bowel, at 14.7 (95% CI 8.2-24.2) and 15.8 (95% CI 8.4-27.0) per 100 000 person-years, respectively. The SIR of neuroendocrine tumors was 2.5 (95% CI 1.5-3.9) in CD and 2.0 (95% CI 1.4-2.8) in UC. CONCLUSIONS: Patients with CD experienced an eightfold increased risk of small bowel adenocarcinomas, patients with both UC and CD experienced an about twofold increased risk of neuroendocrine tumors, and patients with UC experienced a twofold increased risk of small bowel adenocarcinoma. The small absolute excess cancer risk suggests that active surveillance to diagnose small intestinal cancer early is unlikely to be cost-effective.

Height as a mediator of sex differences in cancer risk.

BACKGROUND: Globally, age-standardized incidence rates for most cancers at shared sites are substantially and consistently higher in men than in women. Differences in established risk factors are unable to account for much of the sex disparity. We hypothesized that variability in height may be important in explaining sex differences in cancer risk. PATIENTS AND METHODS: We included 49 372 men from the Health Professionals Follow-up Study (1986-2014) and 115 612 women from the Nurses' Health Study (1980-2014). Height was reported at baseline and biennial questionnaires were used to collect information on cancer risk factors. We examined the association between sex and cancer incidence at shared anatomic sites using Cox proportional hazards models and performed mediation analysis to determine the percent of the association that was accounted for by height. RESULTS: Over up to 34 years of follow-up, 21 307 incident cases of cancers at shared sites (7705 men, 13 602 women) were documented. After adjusting for major cancer risk factors, men had a 39% increased risk of shared cancers overall (hazard ratio = 1.39; 95% confidence interval = 1.33-1.45) of which 35% (95% confidence interval = 24-46) was mediated by height. The excess risk of cancer for men was also partially explained by height for several specific cancers (gastrointestinal, melanoma, kidney, brain, hematologic). Mediation by height tended to be stronger among never smokers or those who adhered to a healthy lifestyle, and for cancers with fewer known environmental risk factors. CONCLUSIONS: Differences in height among men and women partially mediated the association between sex and cancer incidence at several shared sites. Hence, mechanisms underlying the relationship between height and cancer may be important determinants of sex disparities in cancer incidence.

Coffee, tea, and caffeine intake and amyotrophic lateral sclerosis mortality in a pooled analysis of eight prospective cohort studies.

BACKGROUND AND PURPOSE: Caffeine is associated with a lower risk of some neurological diseases, but few prospective studies have investigated caffeine intake and risk of amyotrophic lateral sclerosis (ALS) mortality. We therefore determined associations between coffee, tea and caffeine intake, and risk of ALS mortality. METHODS: We conducted pooled analyses of eight international, prospective cohort studies, including 351 565 individuals (120 688 men and 230 877 women). We assessed coffee, tea and caffeine intake using validated food-frequency questionnaires administered at baseline. We used Cox regression to estimate study- and sex-specific risk ratios and 95% confidence intervals (CI) for ALS mortality, which were then pooled using a random-effects model. We conducted analyses using cohort-specific tertiles, absolute common cut-points and continuous measures of all exposures. RESULTS: During follow-up, 545 ALS deaths were documented. We did not observe statistically significant associations between coffee, tea or caffeine intake and risk of ALS mortality. The pooled multivariable risk ratio (MVRR) for ≥3 cups per day vs. >0 to <1 cup per day was 1.04 (95% CI, 0.74-1.47) for coffee and 1.17 (95% CI, 0.77-1.79) for tea. The pooled MVRR comparing the highest with the lowest tertile of caffeine intake (mg/day) was 0.99 (95% CI, 0.80-1.23). No statistically significant results were observed when exposures were modeled as tertiles or continuously. CONCLUSIONS: Our results do not support associations between coffee, tea or total caffeine intake and risk of ALS mortality.

Use of aspirin, other nonsteroidal anti-inflammatory drugs and acetaminophen and risk of endometrial cancer: the Epidemiology of Endometrial Cancer Consortium.

BACKGROUND: Regular use of aspirin has been associated with a reduced risk of cancer at several sites but the data for endometrial cancer are conflicting. Evidence regarding use of other analgesics is limited. PATIENTS AND METHODS: We pooled individual-level data from seven cohort and five case-control studies participating in the Epidemiology of Endometrial Cancer Consortium including 7120 women with endometrial cancer and 16 069 controls. For overall analyses, study-specific odds ratios (ORs) and 95% confidence intervals (CI) were estimated using logistic regression and combined using random-effects meta-analysis; for stratified analyses, we used mixed-effects logistic regression with study as a random effect. RESULTS: At least weekly use of aspirin and non-aspirin nonsteroidal anti-inflammatory drugs (NSAIDs) was associated with an approximately 15% reduced risk of endometrial cancer among both overweight and obese women (OR = 0.86 [95% CI 0.76-0.98] and 0.86 [95% CI 0.76-0.97], respectively, for aspirin; 0.87 [95% CI 0.76-1.00] and 0.84 [0.74-0.96], respectively, for non-aspirin NSAIDs). There was no association among women of normal weight (body mass index < 25 kg/m2, Pheterogeneity = 0.04 for aspirin, Pheterogeneity = 0.003 for NSAIDs). Among overweight and obese women, the inverse association with aspirin was stronger for use 2-6 times/week (OR = 0.81, 95% CI 0.68-0.96) than for daily use (0.91, 0.80-1.03), possibly because a high proportion of daily users use low-dose formulations. There was no clear association with use of acetaminophen. CONCLUSION: Our pooled analysis provides further evidence that use of standard-dose aspirin or other NSAIDs may reduce risk of endometrial cancer among overweight and obese women.

Burden of depressive symptoms and non-alcohol substance abuse; and their association with alcohol use and partner violence: a cross-sectional study in four sub-Saharan Africa countries.

In sub-Saharan Africa, there are limited data on burden of non-alcohol substance abuse (NAS) and depressive symptoms (DS), yet potential risk factors such as alcohol and intimate partner violence (IPV) are common and NAS abuse may be the rise. The aim of this study was to measure the burden of DS and NAS abuse, and determine whether alcohol use and IPV are associated with DS and/or NAS abuse. We conducted a cross-sectional study at five sites in four countries: Nigeria (nurses), South Africa (teachers), Tanzania (teachers) and two sites in Uganda (rural and peri-urban residents). Participants were selected by simple random sampling from a sampling frame at each of the study sites. We used a standardized tool to collect data on demographics, alcohol use and NAS use, IPV and DS and calculated prevalence ratios (PR). We enrolled 1415 respondents and of these 34.6% were male. DS occurred among 383 (32.3%) and NAS use among 52 (4.3%). In the multivariable analysis, being female (PR  =  1.49, p  =  0.008), NAS abuse (PR  =  2.06, p  =  0.02) and IPV (PR  =  2.93, p < 0.001) were significantly associated with DS. Older age [odds ratio (OR) = 0.31, p < 0.001)], female (OR = 0.48, p  =  0.036) were protective of NAS but current smokers (OR = 2.98, p < 0.001) and those reporting IPV (OR  =  2.16, p  =  0.024) were more likely to use NAS. Longitudinal studies should be done to establish temporal relationships with these risk factors to provide basis for interventions.

Different menopausal hormone regimens and risk of breast cancer.

Background: There are considerable knowledge gaps concerning different estrogen and progestin formulations, regimens, and modes of administration of menopausal hormone therapy (HT) and the risk of breast cancer. Our objective was to assess the different treatment options for menopausal HT and the risk of breast cancer. Patients and methods: This Swedish prospective nationwide cohort study included all women who received ≥1 HT prescription during the study period 2005-2012 (290 186 ever-users), group-level matched (1 : 3) to 870 165 never-users; respectively, 6376 (2.2%) and 18 754 (2.2%) developed breast cancer. HT, ascertained from the Swedish Prescribed Drug Register, was subdivided by estrogen and progestogen formulation types, regimens (continuous versus sequential) and modes of administration (oral versus transdermal). The risk of invasive breast cancer was presented as adjusted odds ratios (OR) and 95% confidence intervals. Results: Current use of estrogen-only therapy was associated with a slight excess breast cancer risk [odds ratio (OR) = 1.08 (1.02-1.14)]. The risk for current estrogen plus progestogen therapy was higher [OR = 1.77 (1.69-1.85)] and increased with higher age at initiation [OR = 3.59 (3.30-3.91) in women 70+ years]. In contrast, past use was associated with reduced breast cancer risk. Current continuous estrogen/progestin use was associated with higher risk [OR = 2.18 (1.99-2.40) for progesterone-derived; OR = 2.66 (2.49-2.84) for testosterone-derived] than sequential use [OR = 1.37 (0.97-1.92) for progesterone-derived; OR = 1.12 (0.96-1.30) for testosterone-derived]. The OR for current use was 1.12 (1.04-1.20) for estradiol, 0.76 (0.69-0.84) for estriol, 4.47 (2.67-7.48) for conjugated estrogens, and 1.68 (1.51-1.87) for tibolone. Oral and cutaneous HT showed similar associations. Conclusion: Different HT regimens have profoundly different effects on breast cancer risk. Because of registry limitations some confounders could not be assessed. This knowledge may guide clinical decision-making when HT is considered.

Associations between sun exposure and other lifestyle variables in Swedish women.

PURPOSE: Sun exposure is associated with risk of several chronic diseases including cancer. The study aim is to investigate whether sun behaviors are related to other lifestyle risk factors of cancer. METHODS: We analyzed data collected in 2003-2004 by self-completed questionnaire from 34,402 Swedish women aged 40-61 years, who comprised 70% of a cohort of originally recruited from a population registry in 1991-1992 (n = 49,259). Participants were asked about annual number of sunburns and annual number of weeks of swimming and sunbathing during 1991-2002, solarium use during 1991-1998 and current sunscreen use. RESULTS: Compared to non-drinkers, the prevalence ratio (95% CI) in women who drank >10 g of alcohol per day was 1.64 (1.49, 1.81) for having >1 sunburn per year, 1.39 (1.29, 1.51) for swimming and sunbathing >2.5 weeks per year and 1.55 (1.41, 1.70) for using a solarium >1 time per 2 months, adjusting for demographic and lifestyle variables. Tobacco smokers were less likely to report sunburn and to use sunscreen, and more likely to sunbath and use solaria, compared with non-smokers. Physical activity was associated positively with swimming and sunbathing, and with the separate use of solaria and sunscreens, but not with number of sunburns. The lifestyle variables that explained most of the variation in sun behavior were alcohol and smoking. CONCLUSIONS: Our results suggest that alcohol consumption and tobacco smoking are potential lifestyle confounders which should be adjusted in studies investigating the association that sun and/or solarium exposure may have with risk of several cancer sites.

First-onset mental disorders after cancer diagnosis and cancer-specific mortality: a nationwide cohort study.

BACKGROUND: The diagnosis of cancer is strongly associated with the risk of mental disorders even in patients with no previous history of mental disorders. Accumulating data suggest that mental distress may accelerate tumor progression. We hypothesized therefore that mental disorders after a cancer diagnosis may increase the risk of cancer-specific mortality. PATIENTS AND METHODS: We conducted a nationwide cohort study including 244 261 cancer patients diagnosed in Sweden during 2004-2009 and followed them through 2010. Through the Swedish Patient Register, we obtained clinical diagnoses of all mental disorders and focused on mood-, anxiety-, and substance abuse disorders (ICD10: F10-F16, F18-F19, F32-F33, F40-F41, and F43-45) that are commonly diagnosed among patients with cancer. We further classified the studied mental disorders into first-onset or recurrent mental disorders. We used Cox regression to estimate multivariable hazard ratios (HRs) with 95% confidence intervals (CIs) as a measure of the association between mental disorders after cancer diagnosis and cancer-specific mortality, adjusting for age, sex, calendar period, educational level, cancer stage, and cancer type at diagnosis. RESULTS: After cancer diagnosis, 11 457 patients were diagnosed with mood-, anxiety-, and substance abuse disorders; of which 7236 were first-onset mental disorders. Patients with a first-onset mental disorder were at increased risk of cancer-specific mortality (HR: 1.82, 95% CI: 1.71-1.92) while patients with a recurrent mental disorder had much lower risk elevation (HR: 1.14, 95% CI: 1.05-1.24). The increased cancer-specific mortality by first-onset mental disorders was observed for almost all cancer sites/groups and the association was stronger for localized cancers (HR: 2.00, 95% CI: 1.73-2.31) than for advanced cancers (HR: 1.49, 95% CI: 1.32-1.69). CONCLUSIONS: Patients with a first-onset common mood-, anxiety-, or substance abuse disorder after cancer diagnosis may be at increased risk of cancer-specific death.

Systematic review with meta-analysis: the comparative effectiveness of aspirin vs. screening for colorectal cancer prevention.

BACKGROUND: Both aspirin use and screening with flexible sigmoidoscopy or guaiac faecal occult blood testing (FOBT) may reduce mortality from colorectal cancer, but comparative effectiveness of these interventions is unknown. AIM: To compare aspirin to guaiac FOBT screening with regard to incidence and mortality of colorectal cancer in a network meta-analysis. METHODS: We searched Medline, EMBASE and the COCHRANE central register (CENTRAL) for relevant randomised trials identified until 31 October 2015. Randomised trials in average-risk populations that reported colorectal cancer mortality, colorectal cancer incidence, or both, with a minimum follow-up of 2 years, and more than 100 randomised individuals were included. Three investigators independently extracted data. We calculated relative risks [RR with 95% predictive intervals (PrIs)] for the comparison of the interventions by frequentist network meta-analyses. RESULTS: The effect of aspirin on colorectal cancer mortality was similar to FOBT (RR 1.03; 95% PrI 0.76-1.39) and flexible sigmoidoscopy (RR 1.16; 95% PrI 0.84-1.60). Aspirin was more effective than FOBT (RR 0.36; 95% PrI 0.22-0.59) and flexible sigmoidoscopy (RR 0.37; 95% PrI 0.22-0.62) in preventing death from or cancer in the proximal colon. Aspirin was equally effective as screening in reducing colorectal cancer incidence, while flexible sigmoidoscopy was superior to FOBT (RR 0.84; 95% PrI 0.72-0.97). CONCLUSIONS: Low-dose aspirin seems to be equally effective as flexible sigmoidoscopy or guaiac FOBT screening to reduce colorectal cancer incidence and mortality, and more effective for cancers in the proximal colon. A randomised comparative effectiveness trial of aspirin vs. screening is warranted.

Central adiposity, obesity during early adulthood, and pancreatic cancer mortality in a pooled analysis of cohort studies.

BACKGROUND: Body mass index (BMI), a measure of obesity typically assessed in middle age or later, is known to be positively associated with pancreatic cancer. However, little evidence exists regarding the influence of central adiposity, a high BMI during early adulthood, and weight gain after early adulthood on pancreatic cancer risk. DESIGN: We conducted a pooled analysis of individual-level data from 20 prospective cohort studies in the National Cancer Institute BMI and Mortality Cohort Consortium to examine the association of pancreatic cancer mortality with measures of central adiposity (e.g. waist circumference; n = 647 478; 1947 pancreatic cancer deaths), BMI during early adulthood (ages 18-21 years) and BMI change between early adulthood and cohort enrollment, mostly in middle age or later (n = 1 096 492; 3223 pancreatic cancer deaths). Multivariable hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression models. RESULTS: Higher waist-to-hip ratio (HR = 1.09, 95% CI 1.02-1.17 per 0.1 increment) and waist circumference (HR = 1.07, 95% CI 1.00-1.14 per 10 cm) were associated with increased risk of pancreatic cancer mortality, even when adjusted for BMI at baseline. BMI during early adulthood was associated with increased pancreatic cancer mortality (HR = 1.18, 95% CI 1.11-1.25 per 5 kg/m(2)), with increased risk observed in both overweight and obese individuals (compared with BMI of 21.0 to <23 kg/m(2), HR = 1.36, 95% CI 1.20-1.55 for BMI 25.0 < 27.5 kg/m(2), HR = 1.48, 95% CI 1.20-1.84 for BMI 27.5 to <30 kg/m(2), HR = 1.43, 95% CI 1.11-1.85 for BMI ≥30 kg/m(2)). BMI gain after early adulthood, adjusted for early adult BMI, was less strongly associated with pancreatic cancer mortality (HR = 1.05, 95% CI 1.01-1.10 per 5 kg/m(2)). CONCLUSIONS: Our results support an association between pancreatic cancer mortality and central obesity, independent of BMI, and also suggest that being overweight or obese during early adulthood may be important in influencing pancreatic cancer mortality risk later in life.

No association between adherence to the healthy Nordic food index and cardiovascular disease amongst Swedish women: a cohort study.

BACKGROUND: In several intervention trials, a healthy Nordic diet showed beneficial effects on markers of cardiovascular disease. We investigated the association between a healthy Nordic diet and clinical diagnosis of cardiovascular disease. OBJECTIVE: Our aim was first to examine the association between a healthy Nordic food index (wholegrain bread, oatmeal, apples/pears, root vegetables, cabbages and fish) and the incidence of overall cardiovascular disease (ischaemic heart disease, stroke, arrhythmia, thrombosis and hypertensive disease), and secondly to test for possible effect modification by smoking, body mass index (BMI), alcohol consumption and age. METHODS: We conducted an analysis of data from the prospective Swedish Women's Lifestyle and Health cohort, including 43 310 women who completed a food frequency questionnaire in 1991-1992, and followed up until 31 December 2012 through Swedish registries. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards models. RESULTS: During follow-up, 8383 women developed cardiovascular disease. We found no association between the healthy Nordic food index and overall cardiovascular disease risk or any of the subgroups investigated. There was a statistically significant interaction with smoking status (P = 0.02), with a beneficial effect only amongst former smokers (HR 0.96, 95% CI 0.94-0.99 per 1-point increment). CONCLUSION: The present results do not support an association between a healthy Nordic food index and risk of cardiovascular disease in Swedish women. There was also no effect modification by alcohol intake, BMI or age. Our finding of an interaction with smoking status requires reproduction.

Infertility and incident endometrial cancer risk: a pooled analysis from the epidemiology of endometrial cancer consortium (E2C2).

BACKGROUND: Nulliparity is an endometrial cancer risk factor, but whether or not this association is due to infertility is unclear. Although there are many underlying infertility causes, few studies have assessed risk relations by specific causes. METHODS: We conducted a pooled analysis of 8153 cases and 11 713 controls from 2 cohort and 12 case-control studies. All studies provided self-reported infertility and its causes, except for one study that relied on data from national registries. Logistic regression was used to estimate adjusted odds ratios (OR) and 95% confidence intervals (CI). RESULTS: Nulliparous women had an elevated endometrial cancer risk compared with parous women, even after adjusting for infertility (OR=1.76; 95% CI: 1.59-1.94). Women who reported infertility had an increased risk compared with those without infertility concerns, even after adjusting for nulliparity (OR=1.22; 95% CI: 1.13-1.33). Among women who reported infertility, none of the individual infertility causes were substantially related to endometrial cancer. CONCLUSIONS: Based on mainly self-reported infertility data that used study-specific definitions of infertility, nulliparity and infertility appeared to independently contribute to endometrial cancer risk. Understanding residual endometrial cancer risk related to infertility, its causes and its treatments may benefit from large studies involving detailed data on various infertility parameters.

Socioeconomic disparities in survival from childhood leukemia in the United States and globally: a meta-analysis.

BACKGROUND: Despite advancements in the treatment of childhood leukemia, socioeconomic status (SES) may potentially affect disease prognosis. This study aims to evaluate whether SES is associated with survival from childhood leukemia. METHODS: The US National Cancer Institute Surveillance, Epidemiology and End Results Program (SEER) 1973-2010 data were analyzed; thereafter, results were meta-analyzed along with those from survival (cohort) studies examining the association between SES indices and survival from childhood leukemia (end-of-search date: 31 March 2014). Random-effects models were used to calculate pooled effect estimates (relative risks, RRs); meta-regression was also used. RESULTS: We included 29 studies yielding 28 804 acute lymphoblastic leukemia (ALL), 3208 acute myeloblastic leukemia (AML) and 27 650 'any' leukemia (denoting joint reporting of all subtypes) cases. According to individual-level composite SES indices, children from low SES suffered from nearly twofold higher death rates from ALL (pooled RR: 1.83, 95% confidence interval 1.00-3.34, based on four study arms); likewise, death RRs derived from an array of lower area-level SES indices ranged between 1.17 and 1.33 (based on 11 study arms). Importantly, the survival gap between higher and lower SES seemed wider in the United States, with considerably (by 20%-82%) increased RRs for death from ALL in lower SES. Regarding AML, poorer survival was evident only when area-level SES indices were used. Lastly, remoteness indices were not associated with survival from childhood leukemia. CONCLUSION: Children with lower SES suffering childhood leukemia do not seem to equally enjoy the impressive recent survival gains. Special health policy strategies and increased awareness of health providers might minimize the effects of socioeconomic disparities.

Comparative analysis of breast cancer mortality following mammography screening in Denmark and Norway.

BACKGROUND: Denmark and Norway are the best countries to study effects of mammography screening, because they are the only countries with stepwise introduction of nationwide mammography screening, enabling comparative effectiveness studies of high quality. Although Denmark and Norway are countries with similar populations and health care systems, reported reductions in breast cancer mortality (incidence-based) caused by screening differed vastly; 25% in Denmark versus 10% in Norway. This study explores reasons for this difference. PATIENTS AND METHODS: We compared two published studies from the Danish and Norwegian screening programs (Olsen et al., 2005; Kalager et al., 2010) investigating biennial mammography screening for women age 50-69 years. Four comparison groups of women were constructed ('current' and 'historical screening groups'; 'current' and 'historical nonscreening groups') based on county of residence. We calculated incidence-based breast cancer mortality in the current versus the historical period for screening and nonscreening groups, using mortality rate ratios (MRR) in the two countries, accounting for concomitant changes in breast cancer mortality. RESULTS: In the screening groups, similar reductions in breast cancer mortality were found when periods preceding and following start of screening were compared, in Denmark [25%; MRR 0.75; 95% confidence interval (CI) 0.64% to 0.88%] and in Norway (28%; MRR 0.72; 95% CI 0.63% to 0.81%). However, mortality increased in Denmark in the current nonscreening group compared with the historical nonscreening group; for women >59 years, breast cancer mortality increased by 14% (MRR 1.14, 95% CI 1.07-1.22), whereas in Norway a 19% reduction was seen (MRR 0.81, 95% CI 0.72-0.92). This increase accounts for the different relative effect of screening in Denmark and Norway; 25% breast cancer mortality reduction in Denmark, 10% in Norway. CONCLUSIONS: The seemingly larger effect of screening in Denmark may not be solely attributable to screening itself, but to increased breast cancer mortality in women older than 59 years not invited to screening.