Cohort Profile: The UK Women's Cohort Study (UKWCS)

Nutrient intake and risk of multimorbidity: a prospective cohort study of 25,389 women

BACKGROUND

Multimorbidity is becoming an increasingly serious public health challenge in the aging population. The impact of nutrients on multimorbidity remains to be determined and was explored using data from a UK cohort study.

METHOD

Our research analysis is mainly based on the data collected by the United Kingdom Women's Cohort Study (UKWCS), which recruited 35,372 women aged 35-69 years at baseline (1995 to 1998), aiming to explore potential associations between diet and chronic diseases. Daily intakes of energy and nutrients were estimated using a validated 217-item food frequency questionnaire at recruitment. Multimorbidity was assessed using the Charlson comorbidity index (CCI) through electronic linkages to Hospital Episode Statistics up to March 2019. Cox's proportional hazards models were used to estimate associations between daily intakes of nutrients and risk of multimorbidity. Those associations were also analyzed in multinomial logistic regression as a sensitivity analysis. In addition, a stratified analysis was conducted with age 60 as the cutoff point.

RESULTS

Among the 25,389 participants, 7,799 subjects (30.7%) were confirmed with multimorbidity over a median follow-up of 22 years. Compared with the lowest quintile, the highest quintile of daily intakes of energy and protein were associated with 8% and 12% increased risk of multimorbidity respectively (HR 1.08 (95% CI 1.01, 1.16), p-linearity = 0.022 for energy; 1.12 (1.04, 1.21), p-linearity = 0.003 for protein). Higher quintiles of daily intakes of vitamin C and iron had a slightly lowered risk of multimorbidity, compared to the lowest quintile. A significantly higher risk of multimorbidity was found to be linearly associated with higher intake quintiles of vitamin B12 and vitamin D (p-linearity = 0.001 and 0.002, respectively) in Cox models, which became insignificant in multinomial logistic regression. There was some evidence of effect modification by age in intakes of iron and vitamin B1 associated with the risk of multimorbidity (p-interaction = 0.006 and 0.025, respectively).

CONCLUSIONS

Our findings highlight a link between nutrient intake and multimorbidity risk. However, there is uncertainty in our results, and more research is needed before definite conclusions can be reached.

Mediterranean diet associated with lower frailty risk: A large cohort study of 21,643 women admitted to hospitals

BACKGROUND

Mediterranean diet is traditionally considered as a healthy dietary pattern, while its association with frailty has not been confirmed. This study investigated associations between Mediterranean diet and risk of frailty among women admitted to hospitals in England from an older-aged women's cohort study.

METHODS

A modified Mediterranean diet was evaluated from a validated 217-item food frequency questionnaire. Incident frailty was determined using a hospital frailty risk score based on linkage to Hospital Episode Statistics up to March 2019. Cox proportional hazard models were conducted to estimate hazard ratios (HR) and 95% confidence intervals (CI). Further subgroup analyses stratified by age and body mass index (BMI), and sensitivity analyses were additionally explored.

RESULTS

Over a mean follow-up of 13 years, there were 14,838 (68.6%) cases of frailty out of 21,643 individuals included in this study. Compared with low adherence to Mediterranean diet, moderate adherence was associated with 5% (HR = 0.95, 95%CI: 0.91, 0.99) lower risk of frailty, with high adherence associated with even lower risk (HR = 0.89, 95%CI: 0.85, 0.94). The magnitude of above associations remained consistent in subgroups stratified by age and BMI, except the association between moderate adherence and risk of frailty was attenuated in the ≥60-year (HR = 0.99, 95%CI: 0.93, 1.06) and the BMI > 24.9 kg/m2 (HR = 0.97, 95%CI: 0.91, 1.03) subgroups.

CONCLUSIONS

Adherence to Mediterranean diet was associated with lower risk of frailty. The better the adherence, the greater the magnitude of the protective association. Older and overweight women may potentially benefit from greater adherence to the Mediterranean diet regarding frailty prevention.

Foods, Nutrients, and Risk of In-Hospital Frailty in Women: Findings from a Large Prospective Cohort Study

Frailty is increasingly prevalent worldwide because of aging populations. Diet may play a role as a modifiable risk factor. This study aimed to investigate associations between dietary factors and risk of frailty in the UK Women's Cohort admitted to hospitals in England. Consumption of foods and nutrients was estimated using a validated 217-item food frequency questionnaire at baseline. Incident frailty was assessed via a hospital frailty risk score based on linkage with hospital episode statistics. Out of 25,186 participants admitted to hospitals, 6919 (27%) were identified with frailty and 10,562 (42%) with pre-frailty over a mean follow-up of 12.7 years. After adjustment for confounding, we observed a 12% increase in risk of frailty with each additional 10 g/MJ intake of total meat (HR = 1.12, 95%CI: 1.07, 1.17), with the highest risk observed for processed meats (HR = 1.45, 95%CI: 1.21, 1.73). Similar associations were observed with pre-frailty. Vegetable intake was associated with slightly lower risk of frailty (HR = 0.98, 95%CI: 0.97, 1.00). There was no evidence of association between most nutrient intakes and in-hospital frailty risk. Overall, our findings suggest that reducing consumption of meat, especially processed meat, in adults may be beneficial regarding the development of frailty.

Major dietary patterns in the United Kingdom Women's Cohort Study showed no evidence of prospective association with pancreatic cancer risk

Diet is a modifiable risk factor for pancreatic cancer. We hypothesized that specific dietary patterns would increase/decrease pancreatic cancer risk. We evaluated the association of dietary patterns with pancreatic cancer risk in the UK Women's Cohort Study. Dietary patterns were assessed at enrollment using: (1) self-reported practice of vegan/vegetarian dietary habits, (2) diet quality indices (World Health Organization Healthy Diet Indicator and Mediterranean Diet Score), and (3) principal component analysis-derived dietary patterns. The association of dietary patterns with pancreatic cancer incidence was quantified using Cox regression survival analysis. Over a median follow-up of 19 years of 35,365 respondents, there were 136 incident cases of pancreatic cancer. No association between dietary habits/quality and pancreatic cancer incidence was evident after adjustments (hazard ratio (95% confidence interval): self-reported omnivores vs vegan/vegetarian dietary habit: 1.13 (0.73-1.76); per-unit increase in World Health Organization Healthy Diet Indicator scores: 0.99 (0.91-1.09); per-unit increase in Mediterranean Diet Score: 0.92 (0.83-1.02). Similarly, no association of principal component analysis-derived dietary patterns with pancreatic cancer risk was evident ("prudent:" 1.02 [0.94-1.10]; ``meat-based:'' 1.00 [0.92-1.09]; ``fast-food, sugar-sweetened beverages, and carbohydrate-rich snacks:'' 0.96 [0.86-1.07]; ``cereal and dairy-rich:'' 1.04 [0.94-1.16], and ``low-diversity and lowfat:'' 1.00 [0.89-1.13]). In this prospective cohort of women, several major dietary patterns were of poor quality. There was no evidence of a prospective association between any of the dietary patterns explored and pancreatic cancer incidence.

Association between adherence to the Mediterranean Diet and the Eatwell Guide and changes in weight and waist circumference in post-menopausal women in the UK Women's Cohort Study

OBJECTIVE

This study investigated the associations between adherence to the Mediterranean Diet and the Eatwell Guide (EWG) and changes in weight and waist circumference in post-menopausal women.

STUDY DESIGN

Post-hoc analysis of post-menopausal women from the UK Women's Cohort Study.

MAIN OUTCOME MEASURES

Changes in weight, waist circumference and the risk of abdominal and general obesity.

RESULTS

4162 post-menopausal women were selected. Higher adherence to both the EWG and the Mediterranean Diet was associated with smaller increases in waist circumference over 4 years (EWG: β -0.47, CI -0.75, -0.20 per 1 tertile increase in score), (Mediterranean Diet: β -0.29, CI -0.58, -0.01 per 1 tertile increase in score); and lower risk of abdominal obesity (EWG: OR 0.55, CI 0.43, 0.70 third versus the first tertile), (Mediterranean Diet: OR 0.60, CI 0.46, 0.76 third versus the first tertile), but was not associated with weight changes (EWG: β 0.14, CI -0.07, 0.36 per 1 tertile increase in score), (Mediterranean Diet: β 0.03, CI -0.19, 0.25 per 1 tertile increase in score) or risk of becoming overweight or obese (EWG: OR 1.09, CI 0.77, 1.52 third versus the first tertile), (Mediterranean Diet: OR 0.91, CI 0.65, 1.27 third versus the first tertile).

CONCLUSIONS

The results suggest that adherence to either the Mediterranean Diet or the EWG can help to prevent abdominal obesity in post-menopausal women.

Foods, nutrients and hip fracture risk: A prospective study of middle-aged women

BACKGROUND AND AIMS

Hip fracture affects 1.6 million people globally each year, and increases morbidity and mortality. There is potential for risk reduction through diet modification, but prospective evidence for associations between intake of several foods and nutrients and hip fracture risk is limited. This study aimed to investigate associations between food and nutrient intakes and hip fracture risk in the UK Women's Cohort Study, and to determine the role of body mass index (BMI) as a potential effect modifier.

METHODS

Dietary, lifestyle, anthropometric, and socio-economic information of UK women, ages 35-69 years, were collected in a survey at recruitment (1995-1998), and included a validated 217-item food frequency questionnaire. Hip fracture cases were identified by linking participant data at recruitment with their Hospital Episode Statistics (HES) up to March 2019. Cox regression models were used to estimate associations between standard portions of food and nutrient intakes and hip fracture risk over a median follow-up time of 22.3 years.

RESULTS

Among 26,318 women linked to HES data (556,331 person-years), 822 hip fracture cases were identified. After adjustment for confounders, every additional cup of tea or coffee per day was associated with a 4% lower risk of hip fracture (HR (95% CI): 0.96 (0.92, 1.00)). A 25 g/day increment of dietary protein intake was also associated with a 14% lower risk of hip fracture (0.86 (0.73, 1.00)). In subgroup analyses, BMI modified linear associations between dietary intakes of protein, calcium, total dairy, milk, and tea and hip fracture risk (p_interaction = 0.02, 0.002, 0.003, 0.001, and 0.003, respectively); these foods and nutrients were associated with a reduced risk of hip fracture in underweight but not healthy or overweight participants. In particular, risk of hip fracture in underweight participants (28 cases, 545 participants) was 45% lower for every 25 g/day protein consumed (0.55 (0.38, 0.78)).

CONCLUSIONS

This is the first prospective cohort study internationally of multiple food and nutrient intakes in relation to hip fracture risk by BMI using linkage to hospital records. Results suggest that the potential roles of some foods and nutrients in hip fracture prevention, particularly protein, tea and coffee in underweight women, merit confirmation.

PROTOCOL REGISTRATION

Clinicaltrials.gov NCT05081466.

Risk of hip fracture in meat-eaters, pescatarians, and vegetarians: results from the UK Women's Cohort Study

BACKGROUND

The risk of hip fracture in women on plant-based diets is unclear. We aimed to investigate the risk of hip fracture in occasional meat-eaters, pescatarians, and vegetarians compared to regular meat-eaters in the UK Women's Cohort Study and to determine if potential associations between each diet group and hip fracture risk are modified by body mass index (BMI).

METHODS

UK women, ages 35-69 years, were classified as regular meat-eaters (≥ 5 servings/week), occasional meat-eaters (< 5 servings/week), pescatarian (ate fish but not meat), or vegetarian (ate neither meat nor fish) based on a validated 217-item food frequency questionnaire completed in 1995-1998. Incident hip fractures were identified via linkage to Hospital Episode Statistics up to March 2019. Cox regression models were used to estimate the associations between each diet group and hip fracture risk over a median follow-up time of 22.3 years.

RESULTS

Amongst 26,318 women, 822 hip fracture cases were observed (556,331 person-years). After adjustment for confounders, vegetarians (HR (95% CI) 1.33 (1.03, 1.71)) but not occasional meat-eaters (1.00 (0.85, 1.18)) or pescatarians (0.97 (0.75, 1.26)) had a greater risk of hip fracture than regular meat-eaters. There was no clear evidence of effect modification by BMI in any diet group (p-interaction = 0.3).

CONCLUSIONS

Vegetarian women were at a higher risk of hip fracture compared to regular meat-eaters. Further research is needed to confirm this in men and non-European populations and to identify factors responsible for the observed risk difference. Further research exploring the role of BMI and nutrients abundant in animal-sourced foods is recommended.

TRIAL REGISTRATION

ClinicalTrials.gov , NCT05081466.

A comparison of time to event analysis methods, using weight status and breast cancer as a case study

Survival analysis with cohort study data has been traditionally performed using Cox proportional hazards models. Random survival forests (RSFs), a machine learning method, now present an alternative method. Using the UK Women's Cohort Study (n = 34,493) we evaluate two methods: a Cox model and an RSF, to investigate the association between Body Mass Index and time to breast cancer incidence. Robustness of the models were assessed by cross validation and bootstraping. Histograms of bootstrap coefficients are reported. C-Indices and Integrated Brier Scores are reported for all models. In post-menopausal women, the Cox model Hazard Ratios (HR) for Overweight (OW) and Obese (O) were 1.25 (1.04, 1.51) and 1.28 (0.98, 1.68) respectively and the RSF Odds Ratios (OR) with partial dependence on menopause for OW and O were 1.34 (1.31, 1.70) and 1.45 (1.42, 1.48). HR are non-significant results. Only the RSF appears confident about the effect of weight status on time to event. Bootstrapping demonstrated Cox model coefficients can vary significantly, weakening interpretation potential. An RSF was used to produce partial dependence plots (PDPs) showing OW and O weight status increase the probability of breast cancer incidence in post-menopausal women. All models have relatively low C-Index and high Integrated Brier Score. The RSF overfits the data. In our study, RSF can identify complex non-proportional hazard type patterns in the data, and allow more complicated relationships to be investigated using PDPs, but it overfits limiting extrapolation of results to new instances. Moreover, it is less easily interpreted than Cox models. The value of survival analysis remains paramount and therefore machine learning techniques like RSF should be considered as another method for analysis.

Dietary Patterns Derived from UK Supermarket Transaction Data with Nutrient and Socioeconomic Profiles

Poor diet is a leading cause of death in the United Kingdom (UK) and around the world. Methods to collect quality dietary information at scale for population research are time consuming, expensive and biased. Novel data sources offer potential to overcome these challenges and better understand population dietary patterns. In this research we will use 12 months of supermarket sales transaction data, from 2016, for primary shoppers residing in the Yorkshire and Humber region of the UK (n = 299,260), to identify dietary patterns and profile these according to their nutrient composition and the sociodemographic characteristics of the consumer purchasing with these patterns. Results identified seven dietary purchase patterns that we named: Fruity; Meat alternatives; Carnivores; Hydrators; Afternoon tea; Beer and wine lovers; and Sweet tooth. On average the daily energy intake of loyalty card holders -who may buy as an individual or for a household- is less than the adult reference intake, but this varies according to dietary purchase pattern. In general loyalty card holders meet the recommended salt intake, do not purchase enough carbohydrates, and purchase too much fat and protein, but not enough fibre. The dietary purchase pattern containing the highest amount of fibre (as an indicator of healthiness) is bought by the least deprived customers and the pattern with lowest fibre by the most deprived. In conclusion, supermarket sales data offer significant potential for understanding population dietary patterns.

Development and Relative Validity of a Semiquantitative Food Frequency Questionnaire to Estimate Dietary Intake among a Multi-Ethnic Population in the Malaysian Cohort Project

Measuring dietary intakes in a multi-ethnic and multicultural setting, such as Malaysia, remains a challenge due to its diversity. This study aims to develop and evaluate the relative validity of an interviewer-administered food frequency questionnaire (FFQ) in assessing the habitual dietary exposure of The Malaysian Cohort (TMC) participants. We developed a nutrient database (with 203 items) based on various food consumption tables, and 803 participants were involved in this study. The output of the FFQ was then validated against three-day 24-h dietary recalls (n = 64). We assessed the relative validity and its agreement using various methods, such as Spearman’s correlation, weighed Kappa, intraclass correlation coefficient (ICC), and Bland–Altman analysis. Spearman’s correlation coefficient ranged from 0.24 (vitamin C) to 0.46 (carbohydrate), and almost all nutrients had correlation coefficients above 0.3, except for vitamin C and sodium. Intraclass correlation coefficients ranged from −0.01 (calcium) to 0.59 (carbohydrates), and weighted Kappa exceeded 0.4 for 50% of nutrients. In short, TMC’s FFQ appears to have good relative validity for the assessment of nutrient intake among its participants, as compared to the three-day 24-h dietary recalls. However, estimates for iron, vitamin A, and vitamin C should be interpreted with caution.

Traditional and Novel Adiposity Indicators and Pancreatic Cancer Risk: Findings from the UK Women's Cohort Study

(1) Background: We studied the association of both conventional (BMI, waist and hip circumference and waist-hip ratio) and novel (UK clothing sizes) obesity indices with pancreatic cancer risk in the UK women's cohort study (UKWCS). (2) Methods: The UKWCS recruited 35,792 women from England, Wales and Scotland from 1995 to 1998. Cancer diagnosis and death information were obtained from the National Health Service (NHS) Central Register. Cox's proportional hazards regression was used to evaluate the association between baseline obesity indicators and pancreatic cancer risk. (3) Results: This analysis included 35,364 participants with a median follow-up of 19.3 years. During the 654,566 person-years follow up, there were 136 incident pancreatic cancer cases. After adjustments for age, smoking, education and physical activity, each centimetre increase in hip circumference (HR: 1.03, 95% CI: 1.01-1.05, p = 0.009) and each size increase in skirt size (HR: 1.12, 95% CI: 1.02-1.23, p = 0.041) at baseline increased pancreatic cancer risk. Baseline BMI became a significant predictor of pancreatic cancer risk (HR: 1.04, 95% CI: 1.00-1.08, p = 0.050) when latent pancreatic cancer cases were removed. Only baseline hip circumference was associated with pancreatic cancer risk (HR: 1.03, 95% CI: 1.00-1.05, p = 0.017) when participants with diabetes at baseline were excluded to control for reverse causality. (4) Conclusion: Hip circumference and skirt size were significant predictors of pancreatic cancer risk in the primary analysis. Thus, hip circumference is useful to assess body shape relationships. Additionally, standard skirt sizes offer an economical and objective alternative to conventional obesity indices for evaluating pancreatic cancer risk in women.

Association Between Reproductive Life Span and Incident Nonfatal Cardiovascular Disease: A Pooled Analysis of Individual Patient Data From 12 Studies

Importance

Early menarche and early menopause are associated with increased risk of cardiovascular disease (CVD) in midlife, but little is known about the association between reproductive life span and the risk of CVD.

Objective

To investigate the association between the length of reproductive life span and risk of incident CVD events, while also considering the timing of menarche and menopause.

Design, Setting, and Participants

Individual-level data were pooled from 12 studies participating in the International Collaboration for a Life Course Approach to Reproductive Health and Chronic Disease Events consortium. Women provided complete information on the timing of menarche and menopause, nonfatal CVD events, and covariates. Cox proportional hazards models were used to estimate hazard ratios and 95% CIs, adjusted for covariates. The association between reproductive life span and CVD was adjusted for age at menarche and age at menopause separately. Analysis began March 2018 and ended December 2019.

Exposures

Reproductive life span was calculated by subtracting age at menarche from age at menopause and categorized as younger than 30, 30 to 32, 33 to 35, 36 to 38 (reference group), 39 to 41, 42 to 44, and 45 years or older.

Main Outcomes and Measures

First nonfatal CVD event, including coronary heart disease and stroke events.

Results

A total of 307 855 women were included. Overall, the mean (SD) ages at menarche, menopause, and reproductive life span were 13.0 (1.5) years, 50.2 (4.4) years, and 37.2 (4.6) years, respectively. Pooled analyses showed that women with a very short reproductive life span (<30 years) were at 1.71 (95% CI, 1.58-1.84) times higher risk of incident CVD events than women with a reproductive life span of 36 to 38 years after adjustment for covariates. This association remained unchanged when adjusted for age at menarche but was attenuated to 1.26 (95% CI, 1.09-1.46) when adjusted for age at menopause. There was a significant interaction between reproductive life span and age at menarche associated with CVD risk (P < .001). Women who had both short reproductive life span (<33 years) and early menarche (age ≤11 years) had the highest risk of CVD (hazard ratio, 2.06; 95% CI, 1.76-2.41) compared with those with a reproductive life span of 36 to 38 years and menarche at age 13 years.

Conclusions and Relevance

Short reproductive life span was associated with an increased risk of nonfatal CVD events in midlife, and the risk was significantly higher for women with early age at menarche.

Foods, nutrient intakes and Mediterranean dietary pattern in midlife are not associated with reaction times: a longitudinal analysis of the UK Women’s Cohort Study

Associations between dietary factors and general cognition in the elderly have been documented; however, little is known about reaction time ability in relation to midlife diet. The present study aimed to investigate associations between reaction time and midlife dietary factors, specifically foods, nutrients and Mediterranean diet (MeDi) pattern. The UK Women’s Cohort Study collected dietary information from middle-aged women (52 (sd 9·4) years old) using a validated 217-item FFQ in 1995–1998. In 2010–2011, a sub-group of 664 participants completed online reaction time ability tests including simple reaction time (SRT) and choice reaction time; 503 participants were eligible for analysis. Participants were grouped into fast and slow groups by their median reaction time. The intake of particular foods, nutrients, adherence to the MeDi and cooking methods (roasting/baking, frying and barbecuing/grilling) were explored in relation to reaction times. We did not find any significant associations between reaction times and investigated foods, nutrients or adherence to the MeDi in adjusted models. However, consumers of roasted/baked fish and fried vegetables were associated with slower SRT (adjusted OR 1·46, 95 % CI 1·00, 2·13, P = 0·049; and adjusted OR 1·64, 95 % CI 1·12, 2·39, P = 0·010, respectively) compared with non-consumers of that particularly cooked food. Overall, our findings show no significant associations between midlife diet and reaction time ability 10–15 years later.

Age at natural menopause and risk of incident cardiovascular disease: a pooled analysis of individual patient data

Background

Early menopause is linked to an increased risk of cardiovascular disease mortality; however, the association between early menopause and incidence and timing of cardiovascular disease is unclear. We aimed to assess the associations between age at natural menopause and incidence and timing of cardiovascular disease.

Methods

We harmonised and pooled individual-level data from 15 observational studies done across five countries and regions (Australia, Scandinavia, the USA, Japan, and the UK) between 1946 and 2013. Women who had reported their menopause status, age at natural menopause (if postmenopausal), and cardiovascular disease status (including coronary heart disease and stroke) were included. We excluded women who had hysterectomy or oophorectomy and women who did not report their age at menopause. The primary endpoint of this study was the occurrence of first non-fatal cardiovascular disease, defined as a composite outcome of incident coronary heart disease (including heart attack and angina) or stroke (including ischaemic stroke or haemorrhagic stroke). We used Cox proportional hazards models to estimate multivariate hazard ratios (HRs) and 95% CIs for the associations between age at menopause and incident cardiovascular disease event. We also adjusted the model to account for smoking status, menopausal hormone therapy status, body-mass index, and education levels. Age at natural menopause was categorised as premenopausal or perimenopausal, younger than 40 years (premature menopause), 40–44 years (early menopause), 45–49 years (relatively early), 50–51 years (reference category), 52–54 years (relatively late), and 55 years or older (late menopause).

Findings

Overall, 301 438 women were included in our analysis. Of these 301 438 women, 12 962 (4·3%) had a first non-fatal cardiovascular disease event after menopause, of whom 9369 (3·1%) had coronary heart disease and 4338 (1·4%) had strokes. Compared with women who had menopause at age 50–51 years, the risk of cardiovascular disease was higher in women who had premature menopause (age <40 years; HR 1·55, 95% CI 1·38–1·73; p<0·0001), early menopause (age 40–44 years; 1·30, 1·22–1·39; p<0·0001), and relatively early menopause (age 45–49 years; 1·12, 1·07–1·18; p<0·0001), with a significantly reduced risk of cardiovascular disease following menopause after age 51 years (p<0·0001 for trend). The associations persisted in never smokers, and were strongest before age 60 years for women with premature menopause (HR 1·88, 1·62–2·20; p<0·0001) and early menopause (1·40, 1·27–1·54; p<0·0001), but were attenuated at age 60–69 years, with no significant association observed at age 70 years and older.

Interpretation

Compared with women who had menopause at age 50–51 years, women with premature and early menopause had a substantially increased risk of a non-fatal cardiovascular disease event before the age of 60 years, but not after age 70 years. Women with earlier menopause need close monitoring in clinical practice, and age at menopause might also be considered as an important factor in risk stratification of cardiovascular disease for women.

Funding

Australian National Health and Medical Research Council.

Diet and risk of breast, endometrial and ovarian cancer: UK Women's Cohort Study

This study aimed to investigate the association between diet and the risk of breast, endometrial and ovarian cancer in the UK Women's Cohort Study. A total of 35 372 women aged 35-69 years were enrolled between 1995 and 1998 and completed a validated 217-item FFQ. The individual foods were collapsed into sixty-four main food groups and compared using Cox proportional models, adjusting for potential confounders. Hazard ratio (HR) estimates are presented per portion increase in food items. After approximately 18 years of follow-up, there were 1822, 294 and 285 cases of breast, endometrial and ovarian cancer, respectively. A high consumption of processed meat and total meat was associated with an increased risk of breast and endometrial cancer. High intake of tomatoes (HR 0·87, 99 % CI 0·75, 1·00) and dried fruits (HR 0·60, 99 % CI 0·37, 0·97) was associated with a reduced risk of breast and endometrial cancer, respectively. Mushroom intake was associated with a higher risk of ovarian cancer (HR 1·57, 99 % CI 1·09, 2·26). Subgroup analysis by pre- or postmenopausal cancer further demonstrated an association between processed meat intake and both postmenopausal breast cancer and endometrial cancer. Intake of dried fruits was associated with a reduced risk of postmenopausal endometrial cancer (HR 0·55, 99 % CI 0·31, 0·98). Our findings suggest that while some foods may trigger the risk of these cancers, some foods may also be protective; supporting the call for further randomised controlled trials of dietary interventions to reduce the risk of cancer among pre- and postmenopausal women.

Fruit and Vegetable Consumption and Their Polyphenol Content Are Inversely Associated with Sleep Duration: Prospective Associations from the UK Women's Cohort Study

This study aims to investigate the prospective associations between fruit and vegetable (FV) intakes and their polyphenol content with subsequent sleep duration in UK women. In this study, 13,958 women with ~4 years of follow-up in the UK Women's Cohort Study were included in the analyses. FV intakes were assessed at baseline using a food frequency questionnaire (FFQ), and average hours of sleep per day were self-reported in follow-up. Polyphenol intake was calculated by matching FV items from the FFQ with the Phenol-Explorer database. Linear regression models, adjusting for confounders, were used for the analyses. Consuming an additional portion of apples, kiwi, oranges, pineapple, and 100% pure juice were associated with shorter sleep. Similarly, an additional portion of cabbage, celery, aubergine, olives, and peppers were inversely associated with sleep duration. An additional gram of total polyphenols was associated with shorter sleep by 18 min (99% CI -31 to -4, p < 0.001). FV consumption and total polyphenol content were inversely associated with sleep duration; however, effect sizes were small, and polyphenol classes from FV intakes were not associated with sleep duration. Future intervention studies considering the time of FV consumption in relation to sleep are needed to clarify the underlying mechanisms.

Associations of clothing size, adiposity and weight change with risk of postmenopausal breast cancer in the UK Women's Cohort Study (UKWCS)

OBJECTIVES

Breast cancer is associated with overweight and obesity after menopause. However, clothing size as a proxy of adiposity in predicting postmenopausal breast cancer is not widely studied. We aimed to explore the relationships between postmenopausal breast cancer risk with adipose indicators (including clothing sizes) and weight change over adulthood.

DESIGN

Prospective cohort study.

SETTING

England, Wales and Scotland.

PARTICIPANTS

17 781 postmenopausal women from the UK Women's Cohort Study.

PRIMARY OUTCOME MEASURE

Incident cases of malignant breast cancers (International Classification of Diseases (ICD) 9 code 174 and ICD 10 code C50).

RESULTS

From 282 277 person-years follow-up, there were 946 incident breast cancer cases with an incidence rate of 3.35 per 1000 women. Body mass index (HR: 1.04; 95% CI: 1.02 to 1.07), blouse size (HR: 1.10; 1.03 to 1.18), waist circumference (HR: 1.07; 1.01 to 1.14) and skirt size (HR: 1.14;1.06 to 1.22) had positive associations with postmenopausal breast cancer after adjustment for potential confounders. Increased weight over adulthood (HR: 1.02; 1.01 to 1.03) was also associated with increased risk for postmenopausal breast cancer.

CONCLUSIONS

Blouse and skirt sizes can be used as adipose indicators in predicting postmenopausal breast cancer. Maintaining healthy body weight over adulthood is an effective measure in the prevention of postmenopausal breast cancer.

Can big data solve a big problem? Reporting the obesity data landscape in line with the Foresight obesity system map

Background

Obesity research at a population level is multifaceted and complex. This has been characterised in the UK by the Foresight obesity systems map, identifying over 100 variables, across seven domain areas which are thought to influence energy balance, and subsequent obesity. Availability of data to consider the whole obesity system is traditionally lacking. However, in an era of big data, new possibilities are emerging. Understanding what data are available can be the first challenge, followed by an inconsistency in data reporting to enable adequate use in the obesity context. In this study we map data sources against the Foresight obesity system map domains and nodes and develop a framework to report big data for obesity research. Opportunities and challenges associated with this new data approach to whole systems obesity research are discussed.

Methods

Expert opinion from the ESRC Strategic Network for Obesity was harnessed in order to develop a data source reporting framework for obesity research. The framework was then tested on a range of data sources. In order to assess availability of data sources relevant to obesity research, a data mapping exercise against the Foresight obesity systems map domains and nodes was carried out.

Results

A reporting framework was developed to recommend the reporting of key information in line with these headings: Background; Elements; Exemplars; Content; Ownership; Aggregation; Sharing; Temporality (BEE-COAST). The new BEE-COAST framework was successfully applied to eight exemplar data sources from the UK. 80% coverage of the Foresight obesity systems map is possible using a wide range of big data sources. The remaining 20% were primarily biological measurements often captured by more traditional laboratory based research.

Conclusions

Big data offer great potential across many domains of obesity research and need to be leveraged in conjunction with traditional data for societal benefit and health promotion.

The Mediterranean diet and risk of colorectal cancer in the UK Women's Cohort Study

Background

Evidence from epidemiological studies investigating associations between adherence to the Mediterranean diet and colorectal cancer is inconsistent. The aim of this study is to assess in the UK Women's Cohort Study whether adherence to the Mediterranean dietary pattern is associated with reduced incidence of cancers of the colon and rectum.

Method

A total of 35 372 women were followed for a median of 17.4 years. A 10-component score indicating adherence to the Mediterranean diet was generated for each cohort participant, using a 217-item food frequency questionnaire. The Mediterranean diet score ranged from 0 for minimal adherence to 10 for maximal adherence. Cox proportional hazards regression was used to provide adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for colon and rectal cancer risk.

Results

A total of 465 incident colorectal cancer cases were documented. In the multivariable adjusted model, the test for trend was positive (HR = 0.88, 95% CI: 0.78 to 0.99; Ptrend = 0.03) for a 2-point increment in the Mediterranean diet score. For rectal cancer, a 2-point increment in the Mediterranean diet score resulted in an HR (95% CI) of 0.69 (0.56 to 0.86), whereas a 62% linear reduced risk (HR 0.38; 95% CI: 0.20 to 0.74; Ptrend < 0.001) was observed for women within the highest vs the lowest category of the MD score. Estimates for an association with colon cancer were weak (Ptrend = 0.41).

Conclusions

Findings suggest that women adhering to a Mediterranean dietary pattern may have a lower risk of colorectal cancer, especially rectal cancer.

Common dietary patterns and risk of cancers of the colon and rectum: Analysis from the United Kingdom Women's Cohort Study (UKWCS)

Few prospective cohort studies in the UK have specifically focused on the associations between commonly consumed dietary patterns and colorectal cancer (CRC). The aim of our study was to assess whether red meat, poultry, fish and vegetarian dietary patterns are associated with differences in the incidence of cancers of colon and rectum in the UKWCS. Four common dietary patterns were defined based on a hierarchy of consumption of red meat, poultry and fish for each cohort participant, using a 217‐item food frequency questionnaire. Cox proportional hazards regression was used to provide adjusted hazard ratios (HR) and 95% confidence intervals (CI) for CRC. A total of 32,147 women recruited and surveyed between 1995 and 1998 were followed up for a mean of 17.2 years (426,798 person‐years). A total of 462 incident CRC cases were documented; 335 colon cancers (172 proximal and 119 distal) and 152 in the rectum. In multivariable‐adjusted models, there was no evidence of a reduction in risk of overall CRC (HR = 0.86, 95% CI: 0.66–1.12), colon cancer (HR = 0.77, 95% CI: 0.56–1.05) or rectal cancer (HR = 1.04, 95% CI: 0.66–1.63) when comparing grouped red meat free diets with diets containing red meat. Exploratory analysis suggested a reduced risk of distal colon cancer in grouped red meat free diets (HR = 0.56, 95% CI: 0.34–0.95), though numbers with this outcome were small. These results indicate that a protective association of red meat free diets specifically on distal colon cancer merits confirmation in a larger study.

What's new?

Eating red meat may affect one's risk of colon cancer. Here, the authors looked for a relationship between colorectal cancer risk and red meat, poultry, fish, and vegetarian diets. They used data from the UK Womens Cohort study, with dietary information collected by survey. No statistically significant association was found for total CRC risk, but when they looked at specific cancer sub‐sites, a correlation emerged. Although numbers were small, a reduction in distal CRC risk was observed among those who did not eat red meat. A larger study would be desirable to confirm these findings.

Does adherence to the World Cancer Research Fund/American Institute of Cancer Research cancer prevention guidelines reduce risk of colorectal cancer in the UK Women's Cohort Study?

Evidence on adherence to diet-related cancer prevention guidelines and associations with colorectal cancer (CRC) risk is limited and conflicting. The aim of this cohort analysis is to evaluate associations between adherence to the World Cancer Research Fund/American Institute of Cancer Research (WCRF/AICR) 2007 recommendations and incident CRC. The UK Women's Cohort Study comprises over 35 372 women who filled in a FFQ at baseline in 1995. They were followed up for CRC incidence for a median of 17·4 years, an individual score linking adherence to eight of the WCRF/AICR recommendations was constructed. Cox proportional hazards regression provided hazard ratios (HR) and 95 % CI for the estimation of CRC risk, adjusting for confounders. Following exclusions, 444 CRC cases were identified. In the multivariate-adjusted model, women within the second and third (highest) categories of the WRCF/AICR score had HR of 0·79 (95 % CI 0·62, 1·00) and 0·73 (95 % CI 0·48, 1·10), respectively, for CRC compared with those in the lowest, reference category. The overall linear trend across the categories was not significant (P=0·17). No significant associations were observed between the WCRF/AICR score and proximal colon, distal colon and rectal cancers separately. Of the individual score components, a BMI within the normal weight range was borderline significantly protective only for rectal cancer in the fully adjusted model. In view of the likely different causes of CRC subtypes, further research is needed to identify the optimal dietary patterns associated with reducing colon and rectal cancer risk, respectively.

Associations between Nut Consumption and Health Vary between Omnivores, Vegetarians, and Vegans

Regular nut consumption is associated with reduced risk factors for chronic disease; however, most population-based studies lack consideration of effect modification by dietary pattern. The UK Women’s Cohort Study (UKWCS) provides an ideal opportunity to examine relationships between nut consumption and chronic disease risk factors in a large sample with diverse dietary patterns. Nut and nutrient intake from 34,831 women was estimated using a food frequency questionnaire among self-identified omnivores, vegetarians and vegans. In this cross-sectional analysis, higher nut consumption was associated with lower body weight (difference between highest and lowest consumption categories from adjusted model: 6.1 kg; 95% CI: 4.7, 7.6) body mass index (BMI, 2.4 units difference; 95% CI: 1.9, 2.9), and waist circumference (2.6 cm difference; 95% CI: 1.4, 3.8) (all p for linear trend < 0.001). Higher nut consumption was also associated with reduced prevalence of high cholesterol and high blood pressure; having a history of heart attack, diabetes and gallstones; and markers of diet quality (all adjusted p for linear trend ≤ 0.011). Higher nut consumption appeared overall to be associated with greater benefits amongst omnivores compared to vegetarians and vegans. Findings support existing literature around beneficial effects of nut consumption and suggest that benefits may be larger among omnivores. Nut promotion strategies may have the highest population impact by specifically targeting this group.

Early menarche, nulliparity and the risk for premature and early natural menopause

STUDY QUESTION

Are parity and the timing of menarche associated with premature and early natural menopause?

SUMMARY ANSWER

Early menarche (≤11 years) is a risk factor for both premature menopause (final menstrual period, FMP <40 years) and early menopause (FMP 40-44 years), a risk that is amplified for nulliparous women.

WHAT IS KNOWN ALREADY

Women with either premature or early menopause face an increased risk of chronic conditions in later life and of early death. Findings from some studies suggest that early menarche and nulliparity are associated with early menopause, however overall the evidence is mixed. Much of the evidence for a direct relationship is hampered by a lack of comparability across studies, failure to adjust for confounding factors and inadequate statistical power.

STUDY DESIGN, SIZE, DURATION

This pooled study comprises 51 450 postmenopausal women from nine observational studies in the UK, Scandinavia, Australia and Japan that contribute to the International collaboration for a Life course Approach to reproductive health and Chronic disease Events (InterLACE).

PARTICIPANTS/MATERIALS, SETTING, METHODS

Age at menarche (categorized as ≤11, 12, 13, 14 and 15 or more years) and parity (categorized as no children, one child and two or more children) were exposures of interest. Age at FMP was confirmed by at least 12 months of cessation of menses where this was not the result of an intervention (such as surgical menopause due to bilateral oophorectomy or hysterectomy) and categorized as premature menopause (FMP before age 40), early menopause (FMP 40-44 years), 45-49 years, 50-51 years, 52-53 years and 54 or more years. We used multivariate multinomial logistic regression models to estimate relative risk ratio (RRR) and 95% CI for associations between menarche, parity and age at FMP adjusting for within-study correlation.

MAIN RESULTS AND THE ROLE OF CHANCE

The median age at FMP was 50 years (interquartile range 48-53 years), with 2% of the women experiencing premature menopause and 7.6% early menopause. Women with early menarche (≤11 years, compared with 12-13 years) were at higher risk of premature menopause (RRR 1.80, 95% CI 1.53-2.12) and early menopause (1.31, 1.19-1.44). Nulliparity was associated with increased risk of premature menopause (2.26, 1.84-2.77) and early menopause (1.32, 1.09-1.59). Women having early menarche and nulliparity were at over 5-fold increased risk of premature menopause (5.64, 4.04-7.87) and 2-fold increased risk of early menopause (2.16, 1.48-3.15) compared with women who had menarche at ≥12 years and two or more children.

LIMITATIONS, REASONS FOR CAUTION

Most of the studies (except the birth cohorts) relied on retrospectively reported age at menarche, which may have led to some degree of recall bias.

WIDER IMPLICATIONS OF THE FINDINGS

Our findings support early monitoring of women with early menarche, especially those who have no children, for preventive health interventions aimed at mitigating the risk of adverse health outcomes associated with early menopause.

STUDY FUNDING/COMPETING INTEREST(S)

InterLACE project is funded by the Australian National Health and Medical Research Council project grant (APP1027196). G.D.M. is supported by Australian Research Council Future Fellowship (FT120100812). There are no competing interests.

The InterLACE study: Design, data harmonization and characteristics across 20 studies on women's health

OBJECTIVES

The International Collaboration for a Life Course Approach to Reproductive Health and Chronic Disease Events (InterLACE) project is a global research collaboration that aims to advance understanding of women's reproductive health in relation to chronic disease risk by pooling individual participant data from several cohort and cross-sectional studies. The aim of this paper is to describe the characteristics of contributing studies and to present the distribution of demographic and reproductive factors and chronic disease outcomes in InterLACE.

STUDY DESIGN

InterLACE is an individual-level pooled study of 20 observational studies (12 of which are longitudinal) from ten countries. Variables were harmonized across studies to create a new and systematic synthesis of life-course data.

MAIN OUTCOME MEASURES

Harmonized data were derived in three domains: 1) socio-demographic and lifestyle factors, 2) female reproductive characteristics, and 3) chronic disease outcomes (cardiovascular disease (CVD) and diabetes).

RESULTS

InterLACE pooled data from 229,054 mid-aged women. Overall, 76% of the women were Caucasian and 22% Japanese; other ethnicities (of 300 or more participants) included Hispanic/Latin American (0.2%), Chinese (0.2%), Middle Eastern (0.3%), African/black (0.5%), and Other (1.0%). The median age at baseline was 47 years (Inter-quartile range (IQR): 41-53), and that at the last follow-up was 56 years (IQR: 48-64). Regarding reproductive characteristics, half of the women (49.8%) had their first menstruation (menarche) at 12-13 years of age. The distribution of menopausal status and the prevalence of chronic disease varied considerably among studies. At baseline, most women (57%) were pre- or peri-menopausal, 20% reported a natural menopause (range 0.8-55.6%) and the remainder had surgery or were taking hormones. By the end of follow-up, the prevalence rates of CVD and diabetes were 7.2% (range 0.9-24.6%) and 5.1% (range 1.3-13.2%), respectively.

CONCLUSIONS

The scale and heterogeneity of InterLACE data provide an opportunity to strengthen evidence concerning the relationships between reproductive health through life and subsequent risks of chronic disease, including cross-cultural comparisons.

Can a dietary quality score derived from a short-form FFQ assess dietary quality in UK adult population surveys?

Objective

To devise a measure of diet quality from a short-form FFQ (SFFFQ) for population surveys. To validate the SFFFQ against an extensive FFQ and a 24 h diet recall.

Design

Population-based cross-sectional survey.

Setting

East Leeds and Bolton in Northern England.

Subjects

Adults (n1999) were randomly selected from lists of those registered with a general practitioner in the study areas, contacted by mail and asked to complete the SFFFQ. Responders were sent a longer FFQ to complete and asked if they would take part in a telephone-based 24 h diet recall.

Results

Results from 826 people completing the SFFFQ, 705 completing the FFQ and forty-seven completing the diet recall were included in the analyses. The dietary quality score (DQS), based on fruit, vegetable, oily fish, non-milk extrinsic sugar and fat intakes, showed significant agreement between the SFFFQ and the FFQ (κ=0·38,P<0·001). The DQS for the SFFFQ and the diet recall did not show significant agreement (κ=0·04,P=0·312). A number of single items on the SFFFQ predicted a ‘healthy’ DQS when calculated from the FFQ. The odds of having a healthy diet were increased by 27 % (95 % CI 9, 49 %,P<0·001) for an increase in fruit of 1 portion/d and decreased by 67 % (95 % CI 47, 79 %,P<0·001) for an increase in crisps of 1 portion/d.

Conclusions

The SFFFQ has been shown to be an effective method of assessing diet quality. It provides an important method for determining variations in diet quality within and across different populations.