Methods for pooling results of epidemiologic studies: the Pooling Project of Prospective Studies of Diet and Cancer

Identifying modifiable risk factors to prevent aggressive colorectal cancer

It remains unclear if pre-diagnostic factors influence the developmental pathways of colorectal cancer (CRC) that could enhance tumor aggressiveness. This study used prospective data from 205,489 cancer-free US health professionals to investigate the associations of 31 known or putative risk factors with the risk of aggressive CRC. Tumor aggressiveness was characterized by three endpoints: aggressive CRC (cancer that causes death within 5 years of diagnosis), fatal CRC, and tumor stage at diagnosis. The data augmentation method was used to assess the difference in the associations between risk factors and endpoints. We documented 3201 CRC cases, of which 899 were aggressive. The protective associations of undergoing lower endoscopy (hazard ratios [HR] 0.43, 95% confidence interval (CI) 0.37, 0.49 for aggressive versus HR 0.61, 95% CI 0.56, 0.67 for non-aggressive) and regular use of aspirin (HR 0.70, 95% CI 0.61, 0.81 versus HR 0.84, 95% CI 0.77, 0.92) were stronger for aggressive than non-aggressive CRC (pHeterogeneity <0.05). Lower intake of whole grains or cereal fiber and greater dietary inflammatory potential were associated with a higher risk of aggressive but not non-aggressive CRC. The remaining risk factors showed comparable associations with aggressive CRC and non-aggressive CRC. Aggressive cases were more likely to have KRAS-mutated tumors but less likely to have distal or MSI-high tumors (p < .007). Similar results were observed for fatal CRC and advanced tumor stages at diagnosis. These findings provide initial evidence for the role of pre-diagnostic risk factors in the pathogenesis of aggressive CRC and suggest research priorities for preventive interventions.

Harmonizing the CBCL and SDQ ADHD scores by using linear equating, kernel equating, item response theory and machine learning methods

Introduction

A problem that applied researchers and practitioners often face is the fact that different institutions within research consortia use different scales to evaluate the same construct which makes comparison of the results and pooling challenging. In order to meaningfully pool and compare the scores, the scales should be harmonized. The aim of this paper is to use different test equating methods to harmonize the ADHD scores from Child Behavior Checklist (CBCL) and Strengths and Difficulties Questionnaire (SDQ) and to see which method leads to the result.

Methods

Sample consists of 1551 parent reports of children aged 10-11.5 years from Raine study on both CBCL and SDQ (common persons design). We used linear equating, kernel equating, Item Response Theory (IRT), and the following machine learning methods: regression (linear and ordinal), random forest (regression and classification) and Support Vector Machine (regression and classification). Efficacy of the methods is operationalized in terms of the root-mean-square error (RMSE) of differences between predicted and observed scores in cross-validation.

Results and discussion

Results showed that with single group design, it is the best to use the methods that use item level information and that treat the outcome as interval measurement level (regression approach).

Fruits and vegetables intake and bladder cancer risk: a pooled analysis from 11 case-control studies in the BLadder cancer Epidemiology and Nutritional Determinants (BLEND) consortium

PURPOSE

High consumption of fruits and vegetables decrease the risk of bladder cancer (BC). The evidence of specific fruits and vegetables and the BC risk is still limited.

METHODS

Fruit and vegetable consumptions in relation to BC risk was examined by pooling individual participant data from case-control studies. Unconditional logistic regression was used to estimate study-specific odds ratio's (ORs) with 95% confidence intervals (CIs) and combined using a random-effects model for intakes of total fruits, total vegetables, and subgroups of fruits and vegetables.

RESULTS

A total of 11 case-control studies were included, comprising 5637 BC cases and 10,504 controls. Overall, participants with the highest intakes versus the lowest intakes of fruits in total (OR 0.79; 95% CI 0.68-0.91), citrus fruits (OR 0.81; 95% CI 0.65-0.98), pome fruits (OR 0.76; 95% CI 0.65-0.87), and tropical fruits (OR 0.84; 95% CI 0.73-0.94) reduced the BC risk. Greater consumption of vegetables in total, and specifically shoot vegetables, was associated with decreased BC risk (OR 0.82; 95% CI 0.68-0.96 and OR 0.87; 95% CI 0.78-0.96, respectively). Substantial heterogeneity was observed for the associations between citrus fruits and total vegetables and BC risk.

CONCLUSION

This comprehensive study provides compelling evidence that the consumption of fruits overall, citrus fruits, pome fruits and tropical fruits reduce the BC risk. Besides, evidence was found for an inverse association between total vegetables and shoot vegetables intake.

Perspective: Beyond the Mediterranean Diet-Exploring Latin American, Asian, and African Heritage Diets as Cultural Models of Healthy Eating

The Mediterranean diet is a well-studied cultural model of healthy eating, yet research on healthy models from other cultures and cuisines has been limited. This perspective article summarizes the components of traditional Latin American, Asian, and African heritage diets, their association with diet quality and markers of health, and implications for nutrition programs and policy. Though these diets differ in specific foods and flavors, we present a common thread that emphasizes healthful plant foods and that is consistent with high dietary quality and low rates of major causes of disability and deaths. In this perspective, we propose that nutrition interventions that incorporate these cultural models of healthy eating show promise, though further research is needed to determine health outcomes and best practices for implementation.

Dietary Fiber Intake and Risk of Advanced and Aggressive Forms of Prostate Cancer: A Pooled Analysis of 15 Prospective Cohort Studies

BACKGROUND

Evidence of an association between dietary fiber intake and risk of advanced and aggressive forms of prostate cancer (PC) and PC mortality is limited.

OBJECTIVE

The aim of this study was to examine associations between intakes of dietary fiber overall and by food source and risk of advanced and aggressive forms of PC.

DESIGN

The study design was a pooled analysis of the primary data from 15 cohorts in 3 continents. Baseline dietary fiber intake was assessed using a validated food frequency questionnaire or diet history in each study.

PARTICIPANTS/SETTING

There were 842 149 men followed for up to 9 to 22 years between 1985 and 2009 across studies.

MAIN OUTCOME MEASURES

The primary outcome measures were advanced (stage T4, N1, or M1 or PC mortality), advanced restricted (excluded men with missing stage and those with localized PC who died of PC), and high-grade PC (Gleason score ≥8 or poorly differentiated/undifferentiated) and PC mortality.

STATISTICAL ANALYSIS PERFORMED

Study-specific multivariable hazard ratios (MVHR) were calculated using Cox proportional hazards regression and pooled using random effects models.

RESULTS

Intake of dietary fiber overall, from fruits, and from vegetables was not associated with risk of advanced (n = 4863), advanced restricted (n = 2978), or high-grade PC (n = 9673) or PC mortality (n = 3097). Dietary fiber intake from grains was inversely associated with advanced PC (comparing the highest vs lowest quintile, MVHR 0.84; 95% CI 0.76-0.93), advanced restricted PC (MVHR 0.85; 95% CI 0.74-0.97), and PC mortality (MVHR 0.78; 95% CI 0.68-0.89); statistically significant trends were noted for each of these associations (P ≤ .03), and a null association was observed for high-grade PC for the same comparison (MVHR 1.00; 95% CI 0.93-1.07). The comparable results were 1.06 (95% CI 1.01-1.10; P value, test for trend = .002) for localized PC (n = 35,199) and 1.05 (95% CI 0.99-1.11; P value, test for trend = .04) for low/intermediate grade PC (n = 34 366).

CONCLUSIONS

Weak nonsignificant associations were observed between total dietary fiber intake and risk of advanced forms of PC, high-grade PC, and PC mortality. High dietary fiber intake from grains was associated with a modestly lower risk of advanced forms of PC and PC mortality.

Partial substitution of red meat or processed meat with plant-based foods and the risk of colorectal cancer

OBJECTIVES

Shifting from animal-based to plant-based diets could reduce colorectal cancer (CRC) incidence. Currently, the impacts of these dietary shifts on CRC risk are ill-defined. Therefore, we examined partial substitutions of red or processed meat with whole grains, vegetables, fruits or a combination of these in relation to CRC risk in Finnish adults.

METHODS

We pooled five Finnish cohorts, resulting in 43 788 participants aged ≥ 25 years (79% men). Diet was assessed by validated food frequency questionnaires at study enrolment. We modelled partial substitutions of red (100 g/week) or processed meat (50 g/week) with corresponding amounts of plant-based foods. Cohort-specific hazard ratios (HR) for CRC were calculated using Cox proportional hazards models and pooled together using random-effects models. Adjustments included age, sex, energy intake and other relevant confounders.

RESULTS

During the median follow-up of 28.8 years, 1124 CRCs were diagnosed. We observed small risk reductions when red meat was substituted with vegetables (HR 0.97, 95% CI 0.95 - 0.99), fruits (0.97, 0.94 - 0.99), or whole grains, vegetables and fruits combined (0.97, 0.95 - 0.99). For processed meat, these substitutions yielded 1% risk reductions. Substituting red or processed meat with whole grains was associated with a decreased CRC risk only in participants with < median whole grain intake (0.92, 0.86 - 0.98; 0.96, 0.93 - 0.99, respectively; pinteraction=0.001).

CONCLUSIONS

Even small, easily implemented substitutions of red or processed meat with whole grains, vegetables or fruits could lower CRC risk in a population with high meat consumption. These findings broaden our insight into dietary modifications that could foster CRC primary prevention.

Reproductive factors, hormonal interventions, and gastric cancer risk in the Stomach cancer Pooling (StoP) Project

BACKGROUND

Gastric cancer incidence is higher in men, and a protective hormone-related effect in women is postulated. We aimed to investigate and quantify the relationship in the Stomach cancer Pooling (StoP) Project consortium.

METHODS

A total of 2,084 cases and 7,102 controls from 11 studies in seven countries were included. Summary odds ratios (ORs) and 95% confidence intervals (CIs) assessing associations of key reproductive factors and menopausal hormone therapy (MHT) with gastric cancer were estimated by pooling study-specific ORs using random-effects meta-analysis.

RESULTS

A duration of fertility of ≥ 40 years (vs. < 20), was associated with a 25% lower risk of gastric cancer (OR = 0.75; 95% CI: 0.58-0.96). Compared with never use, ever, 5-9 years and ≥ 10 years use of MHT in postmenopausal women, showed ORs of 0.73 (95% CI: 0.58-0.92), 0.53 (95% CI: 0.34-0.84) and 0.71 (95% CI: 0.50-1.00), respectively. The associations were generally similar for anatomical and histologic subtypes.

CONCLUSION

Our results support the hypothesis that reproductive factors and MHT use may lower the risk of gastric cancer in women, regardless of anatomical or histologic subtypes. Given the variation in hormones over the lifespan, studies should address their effects in premenopausal and postmenopausal women. Furthermore, mechanistic studies may inform potential biological processes.

Effect of multivitamin-mineral supplementation versus placebo on cognitive function: results from the clinic subcohort of the COcoa Supplement and Multivitamin Outcomes Study (COSMOS) randomized clinical trial and meta-analysis of 3 cognitive studies within COSMOS

BACKGROUND

Longer effects of multivitamin-mineral (MVM) supplementation on late-life cognitive function remain untested using in-person, detailed neuropsychological assessments. Furthermore, insufficient evidence exists for healthcare providers to recommend daily MVM supplements to prevent cognitive decline.

OBJECTIVES

This study aimed to test MVM effects on cognitive change using in-person, detailed neuropsychological assessments and conduct a meta-analysis within COSMOS (COcoa Supplement and Multivitamin Outcomes Study) cognitive substudies for a robust evaluation of MVM effects on cognition.

METHODS

COSMOS is a 2 × 2 factorial trial of cocoa extract (500 mg flavanols/d) and/or a daily MVM supplement for cardiovascular disease and cancer prevention among 21,442 United States adults aged ≥60 y. There were 573 participants in the clinic subcohort of COSMOS (that is, COSMOS-Clinic) who completed all cognitive tests administered at baseline. For the meta-analysis, we included nonoverlapping participants across 3 COSMOS cognitive substudies: COSMOS-Clinic (n = 573); COSMOS-Mind (n = 2158); COSMOS-Web (n = 2472).

RESULTS

In COSMOS-Clinic, we observed a modest benefit of MVM compared with placebo on global cognition over 2 y {mean difference [95% confidence interval (CI)] = 0.06 SD units (SU) (-0.003, 0.13)}, with a significantly more favorable change in episodic memory [mean difference (95% CI) = 0.12 SU (0.002, 0.23)] but not in executive function or attention [mean difference (95% CI) = 0.04 SU (-0.04, 0.11)]. The meta-analysis of COSMOS substudies showed clear evidence of MVM benefits on global cognition [mean difference (95% CI) = 0.07 SU (0.03, 0.11); P = 0.0009] and episodic memory [mean difference (95% CI) = 0.06 SU (0.03, 0.10); P = 0.0007]; the magnitude of effect on global cognition was equivalent to reducing cognitive aging by 2 y.

CONCLUSIONS

In COSMOS-Clinic, daily MVM supplementation leads to a significantly more favorable 2-y change in episodic memory. The meta-analysis within COSMOS cognitive substudies indicates that daily MVM significantly benefits both global cognition and episodic memory. These findings within the COSMOS trial support the benefits of a daily MVM in preventing cognitive decline among older adults. This trial was registered at COSMOS-clinicaltrials.gov as NCT02422745, at COSMOS-Mind-clinicaltrials.gov as NCT03035201, and at COSMOS-Web-clinicaltrials.gov as NCT04582617.

Systematic Review of Injuries and Chronic Musculoskeletal Pain Among High-speed Boat Operators

INTRODUCTION

High-speed boat operators constitute a population at risk of work-related injuries and disabilities. This review aimed to summarize the available knowledge on workplace-related injuries and chronic musculoskeletal pain among high-speed boat operators.

MATERIALS AND METHODS

In this systematic review, we searched Medline, Embase, Scopus, and the Cochrane Library Database for studies, published from 1980 to 2022, on occupational health and hazards onboard high-speed boats. Studies and reports were eligible for inclusion if they evaluated, compared, used, or described harms associated with impact exposure onboard high-speed boats. Studies focusing on recreational injuries and operators of non-planing boats were excluded. The primary outcome of interest was the incidence of acute injuries. The secondary outcome measures comprised the presence of chronic musculoskeletal disorders, pain medication use, and days off work.

RESULTS

Of the 163 search results, 5 (2 prospective longitudinal and 3 cross-sectional cohort studies) were included in this systematic review. A total of 804 cases with 3,312 injuries sustained during 3,467 person-years onboard high-speed boats were included in the synthesis of the results. The pooled incidence rate was 1.0 per person-year. The most common injuries were related to the lower back (26%), followed by neck (16%) and head (12%) injuries. The pooled prevalence of chronic pain was 74% (95% CI: 73-75%) and 60% (95% CI: 59-62%) of the cohort consumed analgesics.

CONCLUSIONS

Despite very limited data, this review found evidence that high-speed boat operators have a higher rate of injuries and a higher prevalence of chronic pain than other naval service operators and the general workforce. Given the low certainty of these findings, further prospective research is required to verify the injury incidence and chronic pain prevalence among high-speed boat operators.

Red meat intake and risk of type 2 diabetes in a prospective cohort study of United States females and males

BACKGROUND

Studies with methodological advancements are warranted to confirm the relation of red meat consumption to the incidence of type 2 diabetes (T2D).

OBJECTIVE

We aimed to assess the relationships of intakes of total, processed, and unprocessed red meat to risk of T2D and to estimate the effects of substituting different protein sources for red meats on T2D risk.

METHODS

Our study included 216,695 participants (81% females) from the Nurses' Health Study (NHS), NHS II, and Health Professionals Follow-up Study (HPFS). Red meat intakes were assessed with semiquantitative food frequency questionnaires (FFQs) every 2 to 4 y since the study baselines. We used multivariable-adjusted proportional hazards models to estimate the associations between red meats and T2D.

RESULTS

Over 5,483,981 person-years of follow-up, we documented 22,761 T2D cases. Intakes of total, processed, and unprocessed red meat were positively and approximately linearly associated with higher risks of T2D. Comparing the highest to the lowest quintiles, hazard ratios (HR) were 1.62 (95% confidence interval [CI]: 1.53, 1.71) for total red meat, 1.51 (95% CI: 1.44, 1.58) for processed red meat, and 1.40 (95% CI: 1.33, 1.47) for unprocessed red meat. The percentage lower risk of T2D associated with substituting 1 serving/d of nuts and legumes for total red meat was 30% (HR = 0.70, 95% CI: 0.66, 0.74), for processed red meat was 41% (HR = 0.59, 95% CI: 0.55, 0.64), and for unprocessed red meat was 29% (HR = 0.71, 95% CI: 0.67, 0.75); Substituting 1 serving/d of dairy for total, processed, or unprocessed red meat was also associated with significantly lower risk of T2D. The observed associations became stronger after we calibrated dietary intakes to intakes assessed by weighed diet records.

CONCLUSIONS

Our study supports current dietary recommendations for limiting consumption of red meat intake and emphasizes the importance of different alternative sources of protein for T2D prevention.

Exposure to Parental Violence and Self-Reported Sexual Violence among Unpartnered Adolescent Girls in Nigeria: Evidence from a National Survey

Objectives

Despite the high prevalence of sexual violence among young and adolescent women in Nigeria, there is a paucity of studies on the extent of sexual violence among adolescent girls, especially unpartnered girls, and the role of parental violence. This study assesses the prevalence of self-reported sexual violence and the influence of exposure to parental violence among unpartnered adolescent girls (aged 15-19) in Nigeria.

Methods

The women's data (n = 5,145) from the 2013 and 2018 Nigeria Demographic and Health Surveys were pooled and analyzed with descriptive analysis and multinomial logistic regressions.

Results

Of all adolescent girls, 5.65% had ever experienced sexual violence, 94.09% said they never did, and 0.26% did not respond. Adolescent Girls exposed to parental violence were more likely to have ever experienced sexual violence than the unexposed girls (aRRR= 1.90; 95% CI: 1.29-2.79).

Conclusion

Interventions to prevent sexual violence among adolescent girls should sensitize parents on the potential negative implications of parental violence for their daughters' wellbeing.

Prospective and Mendelian randomization analyses on the association of circulating fatty acid binding protein 4 (FABP-4) and risk of colorectal cancer

BACKGROUND

Fatty acid binding protein 4 (FABP-4) is a lipid-binding adipokine upregulated in obesity, which may facilitate fatty acid supply for tumor growth and promote insulin resistance and inflammation and may thus play a role in colorectal cancer (CRC) development. We aimed to investigate the association between circulating FABP-4 and CRC and to assess potential causality using a Mendelian randomization (MR) approach.

METHODS

The association between pre-diagnostic plasma measurements of FABP-4 and CRC risk was investigated in a nested case-control study in 1324 CRC cases and the same number of matched controls within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. A two-sample Mendelian randomization study was conducted based on three genetic variants (1 cis, 2 trans) associated with circulating FABP-4 identified in a published genome-wide association study (discovery n = 20,436) and data from 58,131 CRC cases and 67,347 controls in the Genetics and Epidemiology of Colorectal Cancer Consortium, Colorectal Cancer Transdisciplinary Study, and Colon Cancer Family Registry.

RESULTS

In conditional logistic regression models adjusted for potential confounders including body size, the estimated relative risk, RR (95% confidence interval, CI) per one standard deviation, SD (8.9 ng/mL) higher FABP-4 concentration was 1.01 (0.92, 1.12) overall, 0.95 (0.80, 1.13) in men and 1.09 (0.95, 1.25) in women. Genetically determined higher FABP-4 was not associated with colorectal cancer risk (RR per FABP-4 SD was 1.10 (0.95, 1.27) overall, 1.03 (0.84, 1.26) in men and 1.21 (0.98, 1.48) in women). However, in a cis-MR approach, a statistically significant association was observed in women (RR 1.56, 1.09, 2.23) but not overall (RR 1.23, 0.97, 1.57) or in men (0.99, 0.71, 1.37).

CONCLUSIONS

Taken together, these analyses provide no support for a causal role of circulating FABP-4 in the development of CRC, although the cis-MR provides some evidence for a positive association in women, which may deserve to be investigated further.

Folate Intake and Ovarian Cancer Risk among Women with Endometriosis: A Case-Control Study from the Ovarian Cancer Association Consortium

BACKGROUND

Although folate intake has not been associated with an increased risk of ovarian cancer overall, studies of other cancer types have suggested that high folate intake may promote carcinogenesis in precancerous lesions. Women with endometriosis (a potential precancerous lesion) have an increased risk of developing ovarian cancer; however, whether high folate intake increases risk in this group is unknown.

METHODS

We conducted a pooled analysis of six case-control studies from the Ovarian Cancer Association Consortium to investigate the association between folate intake and risk of ovarian cancer among women with and without self-reported endometriosis. We included 570 cases/558 controls with and 5,171/7,559 without endometriosis. We used logistic regression to estimate odds ratios (OR) and 95% confidence intervals for the association between folate intake (dietary, supplemental, and total) and ovarian cancer risk. Finally, we used Mendelian randomization (MR) to evaluate our results using genetic markers as a proxy for folate status.

RESULTS

Higher dietary folate intake was associated with an increased risk of ovarian cancer for women with endometriosis [OR, 1.37 (1.01-1.86)] but not for women without endometriosis. There was no association between supplemental folate intake and ovarian cancer risk for women with or without endometriosis. A similar pattern was seen using MR.

CONCLUSIONS

High dietary folate intake may be associated with an increased risk of ovarian cancer among women with endometriosis.

IMPACT

Women with endometriosis with high folate diets may be at increased risk of ovarian cancer. Further research is needed on the potential cancer-promoting effects of folate in this group.

Leisure-time physical activity and gastric cancer risk: A pooled study within the Stomach cancer Pooling (StoP) Project

BACKGROUND

Although physical activity (PA) has been recognized as a favourable factor in the prevention of various diseases, including certain forms of cancer, the relationship between PA and gastric cancer (GC) is not yet fully understood. This study aims to provide data from a pooled analysis of case-control studies within the Stomach cancer Pooling (StoP) Project to estimate the association between leisure-time PA and the occurrence of GC.

METHODS

Six case-control studies from StoP project collected data on leisure-time PA, for a total of 2,343 cases and 8,614 controls. Subjects were classified into three leisure-time PA categories, either none/low, intermediate or high, based on study-specific tertiles. We used a two-stage approach. Firstly, we applied multivariable logistic regression models to obtain study-specific odds ratios (ORs) and corresponding 95% confidence intervals (CIs) then, we used a random-effect models to obtain pooled effect estimates. We performed stratified analyses according to demographic, lifestyle and clinical covariates.

RESULTS

The meta-analysis showed ORs of GC with no significant differences between intermediate vs low and high vs low PA level (OR 1.05 [95%CI 0.76-1.45]; OR 1.23 [95%CI 0.78-1.94], respectively). GC risk estimates did not strongly differ across strata of selected covariates except for age ≤ 55 years old (high vs low level: OR 0.72 [95%CI 0.55-0.94]) and for control population-based studies (high vs low level: OR 0.79 [95%CI 0.68-0.93]).

CONCLUSIONS

No association was found between leisure time PA and GC, apart from a slight suggestion of decreased risk below age 55 and in control population-based studies. These results may reflect specific characteristics of GC at a younger age, or the presence of a cohort effect mediating and interacting with socioeconomic determinants of GC The different distribution of PA levels among hospitalized controls could have led to an underestimated effect of PA on GC risk.

Hypertensive conditions of pregnancy, preterm birth, and premenopausal breast cancer risk: a premenopausal breast cancer collaborative group analysis

PURPOSE

Women with preeclampsia are more likely to deliver preterm. Reports of inverse associations between preeclampsia and breast cancer risk, and positive associations between preterm birth and breast cancer risk are difficult to reconcile. We investigated the co-occurrence of preeclampsia/gestational hypertension with preterm birth and breast cancer risk using data from the Premenopausal Breast Cancer Collaborative Group.

METHODS

Across 6 cohorts, 3096 premenopausal breast cancers were diagnosed among 184,866 parous women. We estimated multivariable hazard ratios (HR) and 95% confidence intervals (CI) for premenopausal breast cancer risk using Cox proportional hazards regression.

RESULTS

Overall, preterm birth was not associated (HR 1.02, 95% CI 0.92, 1.14), and preeclampsia was inversely associated (HR 0.86, 95% CI 0.76, 0.99), with premenopausal breast cancer risk. In stratified analyses using data from 3 cohorts, preterm birth associations with breast cancer risk were modified by hypertensive conditions in first pregnancies (P-interaction = 0.09). Preterm birth was positively associated with premenopausal breast cancer in strata of women with preeclampsia or gestational hypertension (HR 1.52, 95% CI: 1.06, 2.18), but not among women with normotensive pregnancy (HR = 1.09, 95% CI: 0.93, 1.28). When stratified by preterm birth, the inverse association with preeclampsia was more apparent, but not statistically different (P-interaction = 0.2), among women who did not deliver preterm (HR = 0.82, 95% CI 0.68, 1.00) than those who did (HR = 1.07, 95% CI 0.73, 1.56).

CONCLUSION

Findings support an overall inverse association of preeclampsia history with premenopausal breast cancer risk. Estimates for preterm birth and breast cancer may vary according to other conditions of pregnancy.

Different carbohydrate exposures and weight gain-results from a pooled analysis of three population-based studies

BACKGROUND

The role of carbohydrate quantity and quality in weight gain remains unsolved, and research on carbohydrate subcategories is scarce. We examined total carbohydrates, dietary fiber, total sugar, and sucrose intake in relation to the risk of weight gain in Finnish adults.

METHODS

Our data comprised 8327 adults aged 25-70 years in three population-based prospective cohorts. Diet was assessed by a validated food frequency questionnaire and nutrient intakes were calculated utilizing the Finnish Food Composition Database. Anthropometric measurements were collected according to standard protocols. Two-staged pooling was applied to derive relative risks across cohorts for weight gain of at least 5% by exposure variable intake quintiles in a 7-year follow-up. Linear trends were examined based on a Wald test.

RESULTS

No association was observed between intakes of total carbohydrate, dietary fiber, total sugar or sucrose and the risk of weight gain of at least 5%. Yet, total sugar intake had a borderline protective association with the risk of weight gain in participants with obesity (RR 0.63; 95% CI 0.40-1.00 for highest vs. lowest quintile) and sucrose intake in participants with ≥10% decrease in carbohydrate intake during the follow-up (RR 0.78; 95% CI 0.61-1.00) after adjustments for sex, age, baseline weight, education, smoking, physical activity, and energy intake. Further adjustment for fruit consumption strengthened the associations.

CONCLUSIONS

Our findings do not support an association between carbohydrate intake and weight gain. However, the results suggested that concurrent changes in carbohydrate intake might be an important determinant of weight change and should be further examined in future studies.

Association of protein intake with recurrence and survival following primary treatment of ovarian cancer

BACKGROUND

Malnutrition is common during treatment for ovarian cancer and one in three report multiple symptoms affecting food intake after primary treatment. Little is known about diet post-treatment in relation to ovarian cancer survival, however, general recommendations for cancer survivors are to maintain a higher level of protein intake to support recovery and minimize nutritional deficits.

OBJECTIVE

To investigate whether intake of protein and protein food sources following primary treatment for ovarian cancer is associated with recurrence and survival.

DESIGN

Intake levels of protein and protein food groups were calculated from dietary data collected about 12 months post-diagnosis using a validated food frequency questionnaire in an Australian cohort of females with invasive epithelial ovarian cancer. Disease recurrence and survival status were abstracted from medical records (median 4.9 years follow-up). Cox proportional hazards regression was used to calculate adjusted hazard ratios (HR) and 95% confidence intervals (CI) for protein intake and progression-free and overall survival.

RESULTS

Among 591 females who were progression-free at 12-months follow-up, 329 (56%) subsequently experienced cancer recurrence and 231 (39%) died. A higher level of protein intake was associated with better progression-free survival (>1-1.5 vs. ≤1 grams per kilogram body weight (g/kg): HR_adjusted=0.69, 95% CI 0.48, 1.00; >1.5 vs. ≤1g/kg: HR_adjusted=0.61, 95% CI 0.41, 0.90; >20% vs. ≤20% total energy intake from protein: HR_adjusted=0.77, 95% CI 0.61, 0.96). There was no evidence for better progression-free survival with any particular protein food sources. There was a suggestion of better overall survival among those with higher total intakes of animal-based protein foods, particularly dairy products (HR=0.71, 95% CI 0.51, 0.99 for highest vs. lowest tertiles of total dairy intake).

CONCLUSIONS

After primary treatment for ovarian cancer, a higher level of protein intake may benefit progression-free survival. Ovarian cancer survivors should avoid dietary practices that limit intake of protein-rich foods.

Partial substitution of red or processed meat with plant-based foods and the risk of type 2 diabetes

High consumption of red and processed meat has been associated with increased type 2 diabetes (T2D) risk. These kinds of diets are also environmentally unsustainable. We examined a modeled association between a partial substitution of red meat or processed meat with plant-based foods (legumes, vegetables, fruit, cereals, or a combination of these) and T2D risk among Finnish adults. We used pooled data from five Finnish cohorts (n = 41,662, 22% women, aged ≥ 25 years, 10.9 years median follow-up with 1750 incident T2D cases). Diet was assessed by a validated food frequency questionnaire. In the substitution models, 100 g/week of red meat or 50 g/week of processed meat were substituted with similar amounts of plant-based substitutes. Cohort-specific hazard ratios (HRs) were estimated by Cox proportional hazards multivariable model and pooled using a two-staged random-effects model. We observed small, but statistically significant, reductions in T2D risk in men when red or processed meat were partially substituted with fruits (red meat: HR 0.98, 95% CI 0.97-1.00, P = 0.049, processed meat: 0.99, 0.98-1.00, P = 0.005), cereals (red meat: 0.97, 0.95-0.99, P = 0.005, processed meat: 0.99, 0.98-1.00, P = 0.004) or combination of plant-based foods (only processed meat: 0.99, 0.98-1.00, P = 0.004) but not with legumes or vegetables. The findings of women were similar but not statistically significant. Our findings suggest that even small, easily implemented, shifts towards more sustainable diets may reduce T2D risk particularly in men.

Physical activity in association with mortality among Black women diagnosed with breast cancer in the Southern Community Cohort Study

PURPOSE

Physical activity (PA) is associated with many health benefits. While PA has been associated with reduced mortality after breast cancer diagnosis in many studies, few studies have examined the role of PA in breast cancer survival among underserved and minority populations, including Black women. We investigated PA in association with mortality among Black predominantly low-income breast cancer survivors in the Southern Community Cohort Study (SCCS).

METHODS

Study participants were women diagnosed with incident breast cancer (n = 949) in the SCCS, which is a prospective cohort study of predominantly low-income adults aged 40-79 years recruited from 12 Southeastern states between 2002 and 2009. Participants completed a detailed baseline questionnaire, with annual follow-up for mortality via registry linkages. Cox regression models were used to estimate adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for associations of pre-diagnosis PA (measured via a validated questionnaire) with all-cause and breast cancer-specific mortality.

RESULTS

Breast cancer survivors had a mean age of 61.1 years and most (79.3%) had a household income of < $25,000. In adjusted models, higher levels of total PA (MET-hours/day) were inversely associated with all-cause mortality with HRs (95% CIs): 0.79 (0.59-1.06), 0.66 (0.49-0.90), and 0.60 (0.43-0.84), for Q2, Q3, and Q4 (reference: Q1), respectively, ptrend ≤ 0.01. A similar inverse association was found for breast cancer-specific mortality.

CONCLUSION

Higher levels of pre-diagnosis PA were associated with improved survival among low-income Black breast cancer survivors. Resources to reduce barriers to PA participation and increase support for education and intervention efforts to promote PA among Black women are needed.

Dietary patterns and risk of colorectal cancer: a comparative analysis

BACKGROUND

Although several dietary patterns have been associated with incident colorectal cancer (CRC), it is unclear which diet is optimal.

METHODS

Participants included 48 409 men and 169 772 women from three USA-based prospective cohort studies. We compared the associations of 18 dietary patterns with CRC risk, including two reference scores. The reference scores were derived based on the dietary recommendations for cancer prevention and CRC-specific dietary risk factors mentioned in the 2018 World Cancer Research Fund/American Institute of Cancer Research (WCRF/AICR) Third Expert Report. Multivariable Cox proportional hazards models were fitted to estimate hazard ratios (HRs) and 95% CIs.

RESULTS

Most dietary patterns showed moderate correlations with the WCRF dietary score (absolute values of Spearman correlation coefficients: 0.45-0.63), except the Plant-based diet index, low-carbohydrate diets, the Empirical dietary index for hyperinsulinemia (EDIH) and Empirical dietary inflammation pattern (EDIP). HR for the 10th-90th percentile difference in the score was 0.86 (95% CI: 0.78-0.94) for the Dietary Approaches to Stop Hypertension score (DASH), 1.15 (1.06-1.26) for Western dietary pattern, 1.20 (1.10-1.31) for EDIH and 1.23 (1.13-1.34) for EDIP. These associations between patterns and CRC risk persisted after adjusting for the two reference scores.

CONCLUSIONS

Although further research is needed to improve the WCRF/AICR dietary recommendations, our comprehensive assessment of dietary patterns revealed that the DASH, Western dietary pattern, EDIH and EDIP may be the most relevant diets for preventing CRC.

Clinical and Genomic Characterization of Interval Colorectal Cancer in 3 Prospective Cohorts

BACKGROUND & AIMS

Interval colorectal cancers (CRCs), cancers diagnosed after a screening/surveillance examination in which no cancer is detected, and before the date of next recommended examination, reflect an unprecedented challenge in CRC detection and prevention. To better understand this poorly characterized CRC variant, we examined the clinical and mutational characteristics of interval CRCs in comparison with screen detected CRCs.

METHODS

We included 1175 CRCs documented in the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial and 3661 CRCs in the Nurses' Health Study (NHS) and Health Professionals Follow-up Study (HPFS). Multivariable Cox models were performed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of death risk. Whole exome sequencing was conducted in 147 PLCO cases and 796 NHS/HPFS cases.

RESULTS

A total of 619 deaths (312 CRC-specific) and 2404 deaths (1904 CRC-specific) were confirmed during follow-up of PLCO and NHS/HPFS, respectively. Compared with screen detected CRCs, interval CRCs had a multivariate-adjusted HR (95% CI) of 1.47 (1.21-1.78) for CRC-specific mortality and 1.27 (1.09-1.47) for overall mortality (meta-analysis combining all 3 cohorts). However, we did not observe significant differences in mutational features between interval and screen detected CRCs (false discovery rate adjusted P > .05).

CONCLUSION

Interval CRCs had a significantly increased risk of death compared with screen detected CRCs that were not explained by established clinical prognostic factors, including stage at diagnosis. The survival disadvantage of interval CRCs did not appear to be explained by differences in the genomic landscape of tumors characterized by whole exome sequencing.

Disease classification for whole-blood DNA methylation: Meta-analysis, missing values imputation, and XAI

BACKGROUND

DNA methylation has a significant effect on gene expression and can be associated with various diseases. Meta-analysis of available DNA methylation datasets requires development of a specific workflow for joint data processing.

RESULTS

We propose a comprehensive approach of combined DNA methylation datasets to classify controls and patients. The solution includes data harmonization, construction of machine learning classification models, dimensionality reduction of models, imputation of missing values, and explanation of model predictions by explainable artificial intelligence (XAI) algorithms. We show that harmonization can improve classification accuracy by up to 20% when preprocessing methods of the training and test datasets are different. The best accuracy results were obtained with tree ensembles, reaching above 95% for Parkinson's disease. Dimensionality reduction can substantially decrease the number of features, without detriment to the classification accuracy. The best imputation methods achieve almost the same classification accuracy for data with missing values as for the original data. XAI approaches have allowed us to explain model predictions from both populational and individual perspectives.

CONCLUSIONS

We propose a methodologically valid and comprehensive approach to the classification of healthy individuals and patients with various diseases based on whole-blood DNA methylation data using Parkinson's disease and schizophrenia as examples. The proposed algorithm works better for the former pathology, characterized by a complex set of symptoms. It allows to solve data harmonization problems for meta-analysis of many different datasets, impute missing values, and build classification models of small dimensionality.

Anthropometric traits and risk of multiple myeloma: a pooled prospective analysis

BACKGROUND

Obesity is a risk factor for multiple myeloma (MM), yet results of prior studies have been mixed regarding the importance of early and/or later adult obesity; other measures of body composition have been less well studied.

METHODS

We evaluated associations of early adult (ages 18-21) and usual adult body mass index (BMI), waist circumference, and predicted fat mass with MM by pooling data from six U.S. prospective cohort studies comprising 544,016 individuals and 2756 incident diagnoses over 20-37 years of follow-up. We used Cox proportional hazards models to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for associations, adjusted for age and other risk factors.

RESULTS

Each 5 kg/m2 increase in usual adult BMI was associated with a 10% increased risk of MM (HR: 1.10; 95% CI: 1.05-1.15). Positive associations were also noted for early adult BMI (HR per 5 kg/m2: 1.14; 95% CI: 1.04-1.25), height (HR per 10 cm: 1.28; 95% CI: 1.20-1.37), waist circumference (HR per 15 cm: 1.09; 95% CI: 1.00-1.19), and predicted fat mass (HR per 5 kg: 1.06; 95% CI: 1.01-1.11).

CONCLUSIONS

These findings highlight the importance of avoidance of overweight/obesity and excess adiposity throughout adulthood as a potential MM risk-reduction strategy.

Diet Quality Is Associated with Glucose Regulation in a Cohort of Young Adults

Young-onset type 2 diabetes and prediabetes is a growing epidemic. Poor diet is a known risk factor for T2D in older adults, but the contribution of diet to risk factors for T2D is not well-described in youth. Our objective was to examine the relationship of diet quality with prediabetes, glucose regulation, and adiposity in young adults. A cohort of young adults (n = 155, age 17-22) was examined between 2014-2018, and 89 underwent a follow-up visit from 2020-2022. At each visit, participants completed diet and body composition assessments and an oral glucose tolerance test. Adherence to four dietary patterns was assessed: Dietary Approaches to Stop Hypertension (DASH), Healthy Eating Index (HEI), Mediterranean diet, and Diet Inflammatory Index (DII). Regression analyses were used to determine adjusted associations of diet with risk for prediabetes and adiposity. Each one-point increase in DASH or HEI scores between visits reduced the risk for prediabetes at follow-up by 64% (OR, 95% CI: 0.36, 0.17-0.68) and 9% (OR, 95% CI: 0.91, 0.85-0.96), respectively. The DASH diet was inversely associated with adiposity, while DII was positively associated with adiposity. In summary, positive changes in HEI and DASH scores were associated with reduced risk for prediabetes in young adults.

Spline Analysis of Biomarker Data Pooled from Multiple Matched/Nested Case-Control Studies

Pooling biomarker data across multiple studies enables researchers to obtain precise estimates of the association between biomarker measurements and disease risks due to increased sample sizes. However, biomarker measurements often vary significantly across different assays and laboratories; therefore, calibration of the local laboratory measurements to a reference laboratory is necessary before pooling data. We propose two methods for estimating the dose-response curves that allow for a nonlinear association between the continuous biomarker measurements and log relative risk in pooling projects of matched/nested case-control studies. Our methods are based on full calibration and internalized calibration methods. The full calibration method uses calibrated biomarker measurements for all subjects, even for people with reference laboratory measurements, while the internalized calibration method uses the reference laboratory measurements when available and otherwise uses the calibrated biomarker measurements. We conducted simulation studies to compare these methods, as well as a naive method, where data are pooled without calibration. Our simulation and theoretical results suggest that, in estimating the dose-response curves for biomarker-disease relationships, the internalized and full calibration methods perform substantially better than the naive method, and the full calibration approach is the preferred method for calibrating biomarker measurements. We apply our methods in a pooling project of nested case-control studies to estimate the association of circulating Vitamin D levels with risk of colorectal cancer.

"True" Helicobacter pylori infection and non-cardia gastric cancer: A pooled analysis within the Stomach Cancer Pooling (StoP) Project

BACKGROUND

Helicobacter pylori is the most important risk factor for non-cardia gastric cancer (NCGC); however, the magnitude of the association varies across epidemiological studies. This study aimed to quantify the association between H. pylori infection and NCGC, using different criteria to define infection status.

METHODS

A pooled analysis of individual-level H. pylori serology data from eight international studies (1325 NCGC and 3121 controls) from the Stomach Cancer Pooling (StoP) Consortium was performed. Cases and controls with a negative H. pylori infection status were reclassified as positive considering the presence of anti-Cag A antibodies, gastric atrophy, or advanced stage at diagnosis, as available and applicable. A two-stage approach was used to pool study-specific adjusted odds ratios (OR), and 95% confidence intervals (95% CI). A meta-analysis of published prospective studies assessing H. pylori seropositivity in NCGCs was conducted.

RESULTS

The OR for the association between serology-defined H. pylori and NCGC was 1.45 (95% CI: 0.87-2.42), which increased to 4.79 (95% CI: 2.39-9.60) following the reclassification of negative H. pylori infection. The results were consistent across strata of sociodemographic characteristics, clinical features and lifestyle factors, though significant differences were observed according to geographic region-a stronger association in Asian studies. The pooled risk estimates from the literature were 3.01 (95% CI: 2.22-4.07) for ELISA or EIA and 9.22 (95% CI: 3.12-27.21) for immunoblot or multiplex serology.

CONCLUSION

The NCGC risk estimate from StoP based on the reclassification of H. pylori seronegative individuals is consistent with the risk estimates obtained from the literature. Our classification algorithm may be useful for future studies.

Obesity is Associated With Increased Risk of Crohn's disease, but not Ulcerative Colitis: A Pooled Analysis of Five Prospective Cohort Studies

BACKGROUND AND AIMS

It is unclear whether obesity is associated with the development of inflammatory bowel disease despite compelling data from basic science studies. We therefore examined the association between obesity and risk of Crohn's disease (CD) and ulcerative colitis (UC).

METHODS

We conducted pooled analyses of 5 prospective cohorts with validated anthropometric measurements for body mass index (BMI) and waist-hip ratio and other lifestyle factors. Diagnoses of CD and UC were confirmed through medical records or ascertained using validated definitions. We used Cox proportional hazards modeling to calculate pooled multivariable-adjusted HRs (aHRs) and 95% confidence intervals (CIs).

RESULTS

Among 601,009 participants (age range, 18-98 years) with 10,110,018 person-years of follow-up, we confirmed 563 incident cases of CD and 1047 incident cases of UC. Obesity (baseline BMI ≥30 kg/m2) was associated with an increased risk of CD (pooled aHR, 1.34; 95% CI, 1.05-1.71, I2 = 0%) compared with normal BMI (18.5 to <25 kg/m2). Each 5 kg/m2 increment in baseline BMI was associated with a 16% increase in risk of CD (pooled aHR, 1.16; 95% CI, 1.05-1.22; I2 = 0%). Similarly, with each 5 kg/m2 increment in early adulthood BMI (age, 18-20 years), there was a 22% increase in risk of CD (pooled aHR, 1.22; 95% CI, 1.05-1.40; I2 = 13.6%). An increase in waist-hip ratio was associated with an increased risk of CD that did not reach statistical significance (pooled aHR across quartiles, 1.08; 95% CI, 0.97-1.19; I2 = 0%). No associations were observed between measures of obesity and risk of UC.

CONCLUSIONS

In an adult population, obesity as measured by BMI was associated with an increased risk of older-onset CD but not UC.

Oral Contraceptive and Menopausal Hormone Therapy Use and Risk of Pituitary Adenoma: Cohort and Case-Control Analyses

CONTEXT

No prospective epidemiologic studies have examined associations between use of oral contraceptives (OCs) or menopausal hormone therapy (MHT) and risk of pituitary adenoma in women.

OBJECTIVE

Our aim was to determine the association between use of OC and MHT and risk of pituitary adenoma in two separate datasets.

METHODS

We evaluated the association of OC/MHT with risk of pituitary adenoma in the Nurses' Health Study and Nurses' Health Study II by computing multivariable-adjusted hazard ratios (MVHR) of pituitary adenoma by OC/MHT use using Cox proportional hazards models. Simultaneously, we carried out a matched case-control study using an institutional data repository to compute multivariable-adjusted odds ratios (MVOR) of pituitary adenoma by OC/MHT use.

RESULTS

In the cohort analysis, during 6 668 019 person-years, 331 participants reported a diagnosis of pituitary adenoma. Compared to never-users, neither past (MVHR = 1.05; 95% CI, 0.80-1.36) nor current OC use (MVHR = 0.72; 95% CI, 0.40-1.32) was associated with risk. For MHT, compared to never-users, both past (MVHR = 2.00; 95% CI, 1.50-2.68) and current use (MVHR = 1.80; 95% CI, 1.27-2.55) were associated with pituitary adenoma risk, as was longer duration (MVHR = 2.06; 95% CI, 1.42-2.99 comparing more than 5 years of use to never, P trend = .002). Results were similar in lagged analyses, when stratified by body mass index, and among those with recent health care use. In the case-control analysis, we included 5469 cases. Risk of pituitary adenoma was increased with ever use of MHT (MVOR = 1.57; 95% CI, 1.35-1.83) and OC (MVOR = 1.27; 95% CI, 1.14-1.42) compared to never.

CONCLUSION

Compared to never use, current and past MHT use and longer duration of MHT use were positively associated with higher risk of pituitary adenoma in 2 independent data sets. OC use was not associated with risk in the prospective cohort analysis and was associated with only mildly increased risk in the case-control analysis.

Pesticide Residue Intake From Fruit and Vegetable Consumption and Risk of Glioma

We aimed to determine whether intake of pesticide residues from fruits and vegetables was associated with glioma. Within 3 prospective cohorts from 1998-2016-the Nurses' Health Study (NHS), Nurses' Health Study II (NHSII), and Health Professionals Follow-up Study-we computed multivariable-adjusted hazard ratios (MVHRs) and 95% confidence intervals (CI) for glioma by quintiles of intake of low- and high-pesticide-residue fruits and vegetables using Cox proportional hazards regression. Fruits and vegetables were categorized as high or low residue using a validated method based on pesticide surveillance data. We confirmed 275 glioma cases across 2,745,862 person-years. A significant association was observed between intake of high-residue fruits and vegetables and glioma in NHS (MVHR = 2.99, 95% CI: 1.38, 6.44 comparing highest with lowest quintile, P for trend = 0.02). This was not identified in NHSII (MVHR = 0.52, 95% CI: 0.19, 1.45, P for trend = 0.20) or Health Professionals Follow-up Study (MVHR = 1.01, 95% CI: 0.42, 2.45, P for trend = 0.39). No significant associations were observed by intake of low-residue fruits and vegetables; overall intake was not significantly associated with glioma in any cohort. We found no evidence for an inverse relationship of fruit and vegetable intake with glioma. Although limited in power, this study suggests a possible association between fruit-and-vegetable pesticide residue intake and risk of glioma that merits further study.

Salt intake and gastric cancer: a pooled analysis within the Stomach cancer Pooling (StoP) Project

PURPOSE

Previous studies show that consuming foods preserved by salting increases the risk of gastric cancer, while results on the association between total salt or added salt and gastric cancer are less consistent and vary with the exposure considered. This study aimed to quantify the association between dietary salt exposure and gastric cancer, using an individual participant data meta-analysis of studies participating in the Stomach cancer Pooling (StoP) Project.

METHODS

Data from 25 studies (10,283 cases and 24,643 controls) from the StoP Project with information on salt taste preference (tasteless, normal, salty), use of table salt (never, sometimes, always), total sodium intake (tertiles of grams/day), and high-salt and salt-preserved foods intake (tertiles of grams/day) were used. A two-stage approach based on random-effects models was used to pool study-specific adjusted (sex, age, and gastric cancer risk factors) odds ratios (aORs), and the corresponding 95% confidence intervals (95% CI).

RESULTS

Gastric cancer risk was higher for salty taste preference (aOR 1.59, 95% CI 1.25-2.03), always using table salt (aOR 1.33, 95% CI 1.16-1.54), and for the highest tertile of high-salt and salt-preserved foods intake (aOR 1.24, 95% CI 1.01-1.51) vs. the lowest tertile. No significant association was observed for the highest vs. the lowest tertile of total sodium intake (aOR 1.08, 95% CI 0.82-1.43). The results obtained were consistent across anatomic sites, strata of Helicobacter pylori infection, and sociodemographic, lifestyle and study characteristics.

CONCLUSION

Salty taste preference, always using table salt, and a greater high-salt and salt-preserved foods intake increased the risk of gastric cancer, though the association was less robust with total sodium intake.

Harmonized Phenotypes for Anxiety, Depression, and Attention-Deficit Hyperactivity Disorder (ADHD)

In multi-cohort consortia, the problem often arises that a phenotype is measured using different questionnaires. This study aimed to harmonize scores based on the Child Behaviour Check List (CBCL) and the Strength and Difficulties Questionnaire (SDQ) for anxiety/depression and ADHD. To link the scales, we used parent reports on 1330 children aged 10–11.5 years from the Raine study on both SDQ and CBCL. Harmonization was done based on Item Response Theory. We started from existing CBCL and SDQ scales related to anxiety/depression and ADHD (theoretical approach). Next, we conducted a data-driven approach using factor analysis to validate the theoretical approach. Both approaches yielded similar scales, validating the combination of existing scales. In addition, we studied the impact of harmonized (IRT-based) scores on the statistical power of the results in meta-analytic gene-finding studies. The results showed that the IRT-based harmonized scores increased the statistical power of the results compared to sum scores, even with an equal sample size. These findings can help future researchers to harmonize data from different samples and/or different questionnaires that measure anxiety, depression, and ADHD, in order to obtain the larger sample sizes, to compare research results across subpopulations or to increase generalizability, the validity or statistical power of research results. We recommend using our item parameters to estimate harmonized scores that represent commensurate phenotypes across cohorts, and we explained in detail how other researchers can use our results to harmonize data in their studies.

Environmental triggers of Parkinson's disease - Implications of the Braak and dual-hit hypotheses

Idiopathic Parkinson's disease (PD) may take decades to develop, during which many risk or protective factors may come into play to initiate the pathogenesis or modify its progression to clinical PD. The lack of understanding of this prodromal phase of PD and the factors involved has been a major hurdle in the study of PD etiology and preventive strategies. Although still controversial, the Braak and dual-hit hypotheses that PD may start peripherally in the olfactory structures and/or the gut provides a theoretical platform to identify the triggers and modifiers of PD prodromal development and progression. This is particularly true for the search of environmental causes of PD as the olfactory structures and gut are the major human mucosal interfaces with the environment. In this review, we lay out our personal views about how the Braak and dual-hit hypotheses may help us search for the environmental triggers and modifiers for PD, summarize available experimental and epidemiological evidence, and discuss research gaps and strategies.

Alcohol intake and risk of pituitary adenoma

BACKGROUND

The association between alcohol intake and incidence of pituitary adenoma has not been reported previously. We examined this association in three large, prospective cohort studies.

METHODS

Using data from the Nurses' Health Study, Nurses' Health Study II, and Health Professionals Follow-Up Study, we computed multivariable-adjusted hazard ratios (MVHR) and 95% confidence intervals (CI) for pituitary adenoma by levels of alcohol intake using Cox proportional hazards regression.

RESULTS

We identified 292 incident cases of pituitary adenoma (225 among women, 67 among men) among 235,973 participants with 6,548,732 person-years of follow-up. Compared with intake of ≤ 0.5 g/day, cumulative average alcohol intake in all categories was associated with reduced risk of pituitary adenoma (MVHR = 0.60, 95% CI 0.43-0.83 for 0.5-≤ 2 g/day, MVHR = 0.57, 95% CI 0.41-0.79 for > 2.0-≤ 8.0, MVHR = 0.70, 95% CI 0.47-1.04 for > 8.0-≤ 15.0, and MVHR = 0.51, 95% CI 0.32-0.83 for > 15.0 g/day). Significant inverse findings were present in women and were similar but non-significant in men. For specific alcoholic beverages, inverse associations were statistically significant for total wine (MVHR = 0.58, 95% CI 0.43-0.79 comparing 0.5-≤ 2 to ≤ 0.5 g/day), red wine (MVHR = 0.65, 95% CI 0.46-0.92 comparing 0.5-≤ 2 to ≤ 0.5 g/day), and white wine (MVHR = 0.72, 95% CI 0.53-0.97 comparing 0.5-≤ 2 to ≤ 0.5 g/day). Results were consistent using baseline intake, recent intake, and with an 8-year lag.

CONCLUSION

In three prospective cohorts, compared to almost no consumption, alcohol consumption was associated with reduced risk of pituitary adenoma. Sensitivity analyses suggest that these results are unlikely to be the result of reverse causation or diagnostic bias.

Urban environment and cognitive and motor function in children from four European birth cohorts

BACKGROUND

The urban environment may influence neurodevelopment from conception onwards, but there is no evaluation of the impact of multiple groups of exposures simultaneously. We investigated the association between early-life urban environment and cognitive and motor function in children.

METHODS

We used data from 5403 mother-child pairs from four population-based birth-cohorts (UK, France, Spain, and Greece). We estimated thirteen urban home exposures during pregnancy and childhood, including: built environment, natural spaces, and air pollution. Verbal, non-verbal, gross motor, and fine motor functions were assessed using validated tests at five years old. We ran adjusted multi-exposure models using the Deletion-Substitution-Addition algorithm.

RESULTS

Higher greenness exposure within 300 m during pregnancy was associated with higher verbal abilities (1.5 points (95% confidence interval 0.4, 2.7) per 0.20 unit increase in greenness). Higher connectivity density within 100 m and land use diversity during pregnancy were related to lower verbal abilities. Childhood exposure to PM2.5 mediated 74% of the association between greenness during childhood and verbal abilities. Higher exposure to PM2.5 during pregnancy was related to lower fine motor function (-1.2 points (-2.1, -0.4) per 3.2 μg/m3 increase in PM2.5). No associations were found with non-verbal abilities and gross motor function.

DISCUSSION

This study suggests that built environment, greenness, and air pollution may impact child cognitive and motor function at five years old. This study adds evidence that well-designed urban planning may benefit children's cognitive and motor development.

A repeated measures approach to pooled and calibrated biomarker data

Participant-level meta-analysis across multiple studies increases the sample size for pooled analyses, thereby improving precision in effect estimates and enabling subgroup analyses. For analyses involving biomarker measurements as an exposure of interest, investigators must first calibrate the data to address measurement variability arising from usage of different laboratories and/or assays. In practice, the calibration process involves reassaying a random subset of biospecimens from each study at a central laboratory and fitting models that relate the study-specific "local" and central laboratory measurements. Previous work in this area treats the calibration process from the perspective of measurement error techniques and imputes the estimated central laboratory value among individuals with only a local laboratory measurement. In this work, we propose a repeated measures method to calibrate biomarker measurements pooled from multiple studies with study-specific calibration subsets. We account for correlation between measurements made on the same person and between measurements made at the same laboratory. We demonstrate that the repeated measures approach provides valid inference, and compare it to existing calibration approaches grounded in measurement error techniques in an example describing the association between circulating vitamin D and stroke.

Association between Body Mass Index and Chronic Kidney Disease in Asian Populations: A Participant-level Meta-Analysis

Obesity and chronic kidney disease (CKD) are major public health problems worldwide. However, the association between body mass index (BMI) and CKD is inconclusive in Asians. In this meta-analysis, eight population-based studies, from China, India, Russia (Asian), Singapore and South Korea, provided individual-level data (n=50037). CKD was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m². BMI was analyzed both as a continuous variable and in three categories: <25kg/m², normal; 25-29.9kg/m², overweight; and ≥30kg/m², obese. The association between BMI and CKD was evaluated in each study using multivariable logistic regression models and individual estimates were pooled using random-effect meta-analysis to obtain the pooled odds ratio (OR) and 95% confidence interval (CI). Associations were also evaluated in subgroups of age, gender, smoking, diabetes, and hypertension status. Of 50037 adults, 4258 (8.5%) had CKD. 13328 (26.6%) individuals were overweight while 4440 (8.9%) were obese. The prevalence of any CKD ranged from 3.5% to 29.1% across studies. In pooled analysis, both overweight and obesity were associated with increased odds of CKD, with pooled OR (95% CI) of 1.15 (1.03-1.29) and 1.23 (1.06-1.42), respectively. In subgroup analyses, significant associations between BMI and CKD were observed in adult males, non-smokers, and those with diabetes and arterial hypertension (all p<0.05). When evaluated as a continuous variable, BMI was not significantly associated with CKD. If confirmed in longitudinal studies, these results may have clinical implications in risk stratification and preventive measures, given that obesity and CKD are two major chronic diseases with substantial public health burden worldwide.

Identifying the profile of Helicobacter pylori negative gastric cancers: a case only analysis within the Stomach cancer Pooling (StoP) Project

BACKGROUND

The prevalence of Helicobacter pylori negative gastric cancer (HpNGC) can be as low as 1%, when infection is assessed using more sensitive tests or considering the presence of gastric atrophy. HpNGC may share a high-risk profile contributing to the occurrence of cancer in the absence of infection. We estimated the proportion of HpNGC, using different criteria to define infection status, and compared HpNGC and positive cases regarding gastric cancer risk factors.

METHODS

Cases from 12 studies from the Stomach cancer Pooling (StoP) Project providing data on H. pylori infection status determined by serological test were included. HpNGC was reclassified as positive (eight studies) when cases presented CagA markers (four studies), gastric atrophy (six studies), or advanced stage at diagnosis (three studies), and were compared with positive cases. A two-stage approach (random-effects models) was used to pool study-specific prevalence and adjusted odds ratios (ORs).

RESULTS

Among non-cardia cases, the pooled prevalence of HpNGC was 22.4% (n=166/853) and decreased to 7.0% (n=55) when considering CagA status; estimates for all criteria were 21.8% (n=276/1325) and 6.6% (n=97), respectively. HpNGC had a family history of gastric cancer more often (OR=2.18, 95% confidence interval [CI]:1.03-4.61) and were current smokers (OR=2.16, 95%CI:0.52-9.02).

CONCLUSION

This study found a low prevalence of HpNGC, who are more likely to have a family history of gastric cancer in first-degree relatives.

IMPACT

Our results support that H. pylori infection is present in most non-cardia gastric cancers, and suggest that HpNGC may have distinct patterns of exposure to other risk factors.

Experiences from the harmonization of Finnish national population-based health survey data

Aims: There are several advantages to pooling survey data from individual studies over time or across different countries. Our aim is to share our experiences on harmonizing data from 13 Finnish health examination surveys covering the years 1972–2017 and to describe the challenges related to harmonizing different variable types using two questionnaire variables – blood pressure measurement and total cholesterol assessment – as examples. Methods: Data from Finnish national population-based health surveys were harmonized as part of the research project ‘Projections of the Burden of Disease and Disability in Finland – Health Policy Prospects’, including variables from questionnaires, objective health measurements and results from the laboratory analysis of biological samples. The process presented in the Maelstrom Research guidelines for data harmonization was followed with minor adjustments. Results: The harmonization of data from objective measurements and biomarkers was reasonably straightforward, but questionnaire items proved more challenging. Some questions and response options had changed during the covered time period. This concerned, for example, questionnaire items on the availability and use of medication and diet. Conclusions: The long time period – 45 years – made harmonization more complicated. The survey questions or response options had changed for some topics due to changes in society. However, common core variables for topics that were especially relevant for the project, such as lifestyle factors and certain diseases or conditions, could be harmonized with sufficient comparability. For future surveys, the use of standardized survey methods and the proper documentation of data collection are recommended to facilitate harmonization.

Exploring the interactions between Helicobacter pylori (Hp) infection and other risk factors of gastric cancer: A pooled analysis in the Stomach cancer Pooling (StoP) Project

Helicobacter pylori (Hp) is crucial in gastric carcinogenesis, but infection alone is not a sufficient cause, and the interaction between Hp infection and other risk factors has not been adequately studied. We conducted a pooled analysis of seven case-control studies from the Stomach cancer Pooling (StoP) Project, comprising 1377 cases and 2470 controls, to explore the interaction among Hp infection and tobacco smoking, alcohol drinking, socioeconomic status (SES) and dietary salt intake on the risk of gastric cancer. We estimated summary odds ratios (ORs) and the corresponding 95% confidence intervals (CIs) by multivariate unconditional logistic regression. The analysis showed no consistent interaction between Hp infection and cigarette smoking, while interaction was more than multiplicative for alcohol drinking (OR = 1.38, 95% CI: 1.07-1.77, P-interaction 0.02) and high intake of salt (OR = 2.62, 95% CI: 1.88-3.65, P-interaction = 0.04). The interaction with SES followed the multiplicative model (P = 0.49), resulting in a weakening among infected individuals of the protective effect of high SES among observed Hp-negative individuals. The interactions found were more pronounced in subjects with history of peptic ulcer. The interactions with Hp infection were stronger for cigarette smoking and dietary salt in the case of noncardia cancer, and for alcohol and SES in the case of cardia cancer. No differences were found when stratifying for histologic type. This large-scale study aimed to quantify the interaction between Hp infection and other modifiable risk factors of gastric cancer revealed that the benefit of combined Hp eradication and lifestyle modification on gastric cancer prevention may be larger than commonly appreciated.

Alcohol intake and risk of glioma: results from three prospective cohort studies

PURPOSE

The association between alcohol intake and glioma remains unclear. We evaluated the association between alcohol intake and incidence of glioma in three large, prospective cohort studies with repeated alcohol assessments.

METHODS

We harnessed data from three studies with repeat alcohol assessment to compute hazard ratios (HR) and 95% confidence intervals (CI) for glioma by overall alcohol intake and intake from specific beverages using Cox proportional hazards regression, adjusted for age, cohort, body mass index, smoking status, and caloric intake. Analyses were conducted separately for glioma overall and for glioblastoma (GBM).

RESULTS

We confirmed 554 incident glioma cases (362 GBM) among 237,505 participants with 6,216,378 person-years of follow up. Cumulative average alcohol intake was associated with reduced risk of glioma (HR = 0.75, 95%CI:0.56-0.99 comparing > 8-15 to ≤ 0.5 g/d; HR = 0.71, 95%CI:0.53-0.96 comparing > 15 g/d to ≤ 0.5 g/d). When stratified by sex, for the same comparisons, the HRs for men were 0.57 (95%CI:0.36-0.89) and 0.79 (0.53-1.16), and for women 0.90 (95%CI:0.62-1.30) and 0.62, 95%CI:0.39-0.97. Results were consistent when examining cumulative average, baseline, and recent intake, and with a 4 year lag.

CONCLUSION

These results provide evidence against a positive association between alcohol intake and glioma risk. Alcohol intake was associated with reduced risk of glioma in both men and women.

International strategy in cancer epidemiology: Japan's involvement in global projects and future role

In recent years, collaboration among researchers in the field of cancer epidemiology has been accelerating in various forms. Here, we review recent trends in international collaborative research activities in the cancer epidemiology field in Japan. These include not only support for other countries with less developed cancer statistics infrastructures, but also large-scale compilations and international comparisons through collaborative studies, as well as integration with analytical epidemiology and clinical research. Formation of international cohort consortia and estimates of cancer and risk factors in each country have contributed to raising the skill levels of cancer epidemiologists as well as to expanding research networks and activities among cancer epidemiologists. Molecular and genome epidemiological studies on cancer have progressed over decades and these continue to increase in size and dimension. Application of evidence from this area in prevention is still underway and needs further effort. Japanese epidemiologists have great potential to assume international leadership roles by taking advantage of the uniqueness, originality and characteristics of Japanese cohorts.

Dairy foods, calcium, and risk of breast cancer overall and for subtypes defined by estrogen receptor status: a pooled analysis of 21 cohort studies

BACKGROUND

Epidemiologic studies examining the relations between dairy product and calcium intakes and breast cancer have been inconclusive, especially for tumor subtypes.

OBJECTIVE

To evaluate the associations between intakes of specific dairy products and calcium and risk of breast cancer overall and for subtypes defined by estrogen receptor (ER) status.

METHOD

We pooled the individual-level data of over 1 million women who were followed for a maximum of 8-20 years across studies. Associations were evaluated for dairy product and calcium intakes and risk of incident invasive breast cancer overall (n = 37,861 cases) and by subtypes defined by ER status. Study-specific multivariable hazard ratios (HRs) were estimated and then combined using random-effects models.

RESULTS

Overall, no clear association was observed between the consumption of specific dairy foods, dietary (from foods only) calcium, and total (from foods and supplements) calcium, and risk of overall breast cancer. Although each dairy product showed a null or very weak inverse association with risk of overall breast cancer (P, test for trend >0.05 for all), differences by ER status were suggested for yogurt and cottage/ricotta cheese with associations observed for ER-negative tumors only (pooled HR = 0.90, 95% CI: 0.83, 0.98 comparing ≥60 g/d with <1 g/d of yogurt and 0.85, 95% CI: 0.76, 0.95 comparing ≥25 g/d with <1 g/d of cottage/ricotta cheese). Dietary calcium intake was only weakly associated with breast cancer risk (pooled HR = 0.98, 95% CI: 0.97, 0.99 per 350 mg/d).

CONCLUSION

Our study shows that adult dairy or calcium consumption is unlikely to associate with a higher risk of breast cancer and that higher yogurt and cottage/ricotta cheese intakes were inversely associated with the risk of ER-negative breast cancer, a less hormonally dependent subtype with poor prognosis. Future studies on fermented dairy products, earlier life exposures, ER-negative breast cancer, and different racial/ethnic populations may further elucidate the relation.

Statistical methods for analysis of combined biomarker data from multiple nested case-control studies

By combining data across multiple studies, researchers increase sample size, statistical power, and precision for pooled analyses of biomarker-disease associations. However, researchers must adjust for between-study variability in biomarker measurements. Previous research often treats the biomarker measurements from a reference laboratory as a gold standard, even though those measurements are certainly not equal to their true values. This paper addresses measurement error and bias arising from both the reference and study-specific laboratories. We develop two calibration methods, the exact calibration method and approximate calibration method, for pooling biomarker data drawn from nested or matched case-control studies, where the calibration subset is obtained by randomly selecting controls from each contributing study. Simulation studies are conducted to evaluate the empirical performance of the proposed methods. We apply the proposed methods to a pooling project of nested case-control studies to evaluate the association between circulating 25-hydroxyvitamin D (25(OH)D) and colorectal cancer risk.

Family History and Gastric Cancer Risk: A Pooled Investigation in the Stomach Cancer Pooling (STOP) Project Consortium

Simple Summary

Research is still required to establish the relationship between family history (FH) and gastric cancer (GC) in relation to different histological types and anatomical sites. The present work aimed to examine the influence of first-degree FH on the risk of GC, also according to the GC location and histological type, including 5946 cases and 12,776 controls from 17 studies of 11 countries in three continents participating in the Stomach Cancer Pooling (StoP) Project consortium. This analysis confirms the effect of FH on the risk of GC, reporting an approximately doubled risk, and provides further quantification of the risk of GC according to the subsite and histotype.

Abstract

Although there is a clear relationship between family history (FH) and the risk of gastric cancer (GC), quantification is still needed in relation to different histological types and anatomical sites, and in strata of covariates. The objective was to analyze the risk of GC according to first-degree FH in a uniquely large epidemiological consortium of GC. This investigation includes 5946 cases and 12,776 controls from 17 studies of the Stomach Cancer Pooling (StoP) Project consortium. Summary odds ratios (OR) and the corresponding 95% confidence intervals (CIs) were calculated by pooling study-specific ORs using fixed-effect model meta-analysis techniques. Stratified analyses were carried out by sex, age, tumor location and histological type, smoking habit, socioeconomic status, alcohol intake and fruit consumption. The pooled OR for GC was 1.84 (95% CI: 1.64–2.04; I2 = 6.1%, P heterogeneity = 0.383) in subjects with vs. those without first-degree relatives with GC. No significant differences were observed among subgroups of sex, age, geographic area or study period. Associations tended to be stronger for non-cardia (OR = 1.82; 95% CI: 1.59–2.05 for subjects with FH) than for cardia GC (OR = 1.38; 95% CI: 0.98–1.77), and for the intestinal (OR = 1.92; 95% CI: 1.62–2.23) than for the diffuse histotype (OR = 1.62; 95% CI: 1.28–1.96). This analysis confirms the effect of FH on the risk of GC, reporting an approximately doubled risk, and provides further quantification of the risk of GC according to the subsite and histotype. Considering these findings, accounting for the presence of FH to carry out correct prevention and diagnosis measures is of the utmost importance.

Birth Weight and Adult-Onset Hearing Loss

OBJECTIVES

Among low-birth-weight infants, exposure to stress or undernutrition in utero may adversely affect cochlear development. As cochlear reserve declines, the risk of hearing loss may increase with age. While low birth weight is associated with a higher risk of neonatal hearing loss, our objective was to examine whether birth weight was associated with adult-onset, self-reported hearing loss in the Nurses' Health Studies (NHS) I and II (n = 113,130).

DESIGN

We used Cox proportional hazards regression to prospectively examine whether birth weight, as well as gestational age at birth, is associated with adult-onset hearing loss. Participants reported their birth weight in 1992 in NHS I and 1991 in NHS II. Mothers of NHS II participants reported gestational age at birth in a substudy (n = 28,590). The primary outcome was adult-onset, self-reported moderate or greater hearing loss, based on questionnaires administered in 2012/2016 in NHS I and 2009/2013 in NHS II.

RESULTS

Our results suggested a higher risk of hearing loss among those with birth weight <5.5 lbs compared with birth weight 7 to <8.5 lbs (pooled multivariable-adjusted hazard ratio 1.14, 95% confidence interval = 1.04-1.23; p trend = 0.01). Additionally, participants with gestational age at birth ≥42 weeks had a higher risk of hearing loss, compared with gestational age 38 to <42 weeks (multivariable-adjusted hazard ratio 1.33, 95% confidence interval = 1.06-1.65).

CONCLUSIONS

Birth weight <5.5 lbs was independently associated with higher risk of self-reported, adult-onset hearing loss. In addition, gestational age at birth ≥42 weeks was also associated with higher risk.