Scores Obtained from a Simple Cognitive Test of Visuospatial Episodic Memory Performed Decades before Death Are Associated with the Ultimate Presence of Alzheimer Disease Pathology.

BACKGROUND: Community- or population-based longitudinal studies of cognitive ability with a brain donation end point offer an opportunity to examine relationships between pathology and cognitive state prior to death. Discriminating the earliest signs of dementing disorders, such as Alzheimer disease (AD), is necessary to undertake early interventions and treatments. METHODS: The neuropathological profile of brains donated from The University of Manchester Longitudinal Study of Cognition in Normal Healthy Old Age, including CERAD (Consortium to Establish a Registry for Alzheimer's Disease) and Braak stage, was assessed by immunohistochemistry. Cognitive test scores collected 20 years prior to death were correlated with the extent of AD pathology present at death. RESULTS: Baseline scores from the Memory Circle test had the ability to distinguish between individuals who developed substantial AD pathology and those with no, or low, AD pathology. Predicted test scores at the age of 65 years also discriminated between these pathology groups. The addition of APOE genotype further improved the discriminatory ability of the model. CONCLUSIONS: The results raise the possibility of identifying individuals at future risk of the neuropathological changes associated with AD over 20 years before death using a simple cognitive test. This work may facilitate early interventions, therapeutics and treatments for AD by identifying at-risk and minimally affected (in pathological terms) individuals.

How serious is bias in effect estimation in randomised trials with survival data given risk heterogeneity and informative censoring?

It is often assumed that randomisation will prevent bias in estimation of treatment effects from clinical trials, but this is not true of the semiparametric Proportional Hazards model for survival data when there is underlying risk heterogeneity. Here, a new formula is proposed for estimation of this bias, improving on a previous formula through ease of use and clarity regarding the role of the mid-study cumulative hazard rate, shown to be an important factor for the bias magnitude. Informative censoring (IC) is recognised as a source of bias. Here, work on selection effects among survivors due to risk heterogeneity is extended to include IC. A new formula shows that bias in causal effect estimation under IC has two sources: one consequent on heterogeneity and one from the additional impact of IC. The formula provides new insights not previously shown: there may less bias under IC than when there is no IC and even, in principle, zero bias. When tested against simulated data, the new formulae are shown to be very accurate for prediction of bias in Proportional Hazards and accelerated failure time analyses which ignore heterogeneity. These data are also used to investigate the performance of accelerated failure time models which explicitly model risk heterogeneity ('frailty models') and G estimation. These methods remove bias when there is simple censoring but not with informative censoring when they may lead to overestimation of treatment effects. The new formulae may be used to help researchers judge the possible extent of bias in past studies. Copyright © 2017 John Wiley & Sons, Ltd.

Changes in selected metabolic parameters in patients over 65 receiving hydrochlorothiazide plus amiloride, atenolol or placebo in the MRC elderly trial.

BACKGROUND: Treatment of hypertension reduces incidence of stroke, myocardial infarction and heart failure perhaps partly by controlling different metabolic parameters. There is limited information regarding the changes in potassium, sodium, weight, cholesterol and glucose levels in patients using anti-hypertensives. This study aimed to determine changes in potassium, sodium, glucose, cholesterol, weight, urea and urate levels in patients using anti-hypertensives. Furthermore, to describe these changes and differences between the atenolol, hydrochlorothiazide plus amiloride and placebo arms of the Medical Research Council (MRC) elderly randomised controlled trial. METHODS: Patients were randomly allocated to one of the three treatment arms. Measurements were taken at baseline, end of year one and end of year two in 4396 subjects. Linear Mixed Models (LMM) were used to determine the longitudinal profiles of sodium, potassium, weight, cholesterol, glucose, urea and urate. Estimates of changes within groups and difference between groups were obtained. RESULTS: Patients randomised to receive hydrochlorothiazide + amiloride experienced a significantly greater mean reduction in potassium, sodium and weight compared to placebo at end of year one - mean differences in change -0.18 mmol/L, (95 % CI: -0.21, -0.15); -1.45 mmol/L, (95 % CI: -1.62, -1.29) and -0.46 kgs (95 % CI: -0.73, -0.20) respectively, and greater increases in cholesterol, urea and urate - mean differences in change 0.16 mmol/L, (95 % CI: 0.10,0.22); 0.77 mmol/L, (95 % CI: 0.68, 0.87) and 53.10 µmol/L, (95 % CI: 49.35, 56.85) respectively. Changes were in the same direction but smaller in the atenololarm except for potassium and weight (increases). No group differences in glucose were found. CONCLUSION: Results were in line with expectation except for lack of change in glucose in the hydrochlorothiazide + amiloride arms.

Trends in incidence of occupational asthma, contact dermatitis, noise-induced hearing loss, carpal tunnel syndrome and upper limb musculoskeletal disorders in European countries from 2000 to 2012.

OBJECTIVES: The European Union (EU) strategy for health and safety at work underlines the need to reduce the incidence of occupational diseases (OD), but European statistics to evaluate this common goal are scarce. We aim to estimate and compare changes in incidence over time for occupational asthma, contact dermatitis, noise-induced hearing loss (NIHL), carpal tunnel syndrome (CTS) and upper limb musculoskeletal disorders across 10 European countries. METHODS: OD surveillance systems that potentially reflected nationally representative trends in incidence within Belgium, the Czech Republic, Finland, France, Italy, the Netherlands, Norway, Spain, Switzerland and the UK provided data. Case counts were analysed using a negative binomial regression model with year as the main covariate. Many systems collected data from networks of 'centres', requiring the use of a multilevel negative binomial model. Some models made allowance for changes in compensation or reporting rules. RESULTS: Reports of contact dermatitis and asthma, conditions with shorter time between exposure to causal substances and OD, were consistently declining with only a few exceptions. For OD with physical causal exposures there was more variation between countries. Reported NIHL was increasing in Belgium, Spain, Switzerland and the Netherlands and decreasing elsewhere. Trends in CTS and upper limb musculoskeletal disorders varied widely within and between countries. CONCLUSIONS: This is the first direct comparison of trends in OD within Europe and is consistent with a positive impact of European initiatives addressing exposures relevant to asthma and contact dermatitis. Taking a more flexible approach allowed comparisons of surveillance data between and within countries without harmonisation of data collection methods.

Occupation exposures and sperm morphology: a case-referent analysis of a multi-centre study.

OBJECTIVE: We examined occupational exposures and sperm morphology to establish whether exposures implicated differed from those affecting motile sperm concentration. METHODS: Computer aided sperm morphometric assessment was undertaken on morphology slides obtained as part of a multi-centre study in 1999-2002 of occupational factors in male infertility. Men attending 14 fertility clinics across the UK were recruited and gave a semen sample. Before results of the semen analysis were known, the men completed detailed questionnaires about their employment and lifestyle. Occupational exposures were assessed by occupational hygienists. Data were analysed using an unmatched case-referent design, allowing for clustering and for confounders. Three case definitions were used: poor morphology (normal morphology <4%), low motile sperm count (MSC) (<4.8×10(6)) and either condition. RESULTS: Morphology results were available for 1861/2011 men employed at the time of recruitment. Of these 1861, 296 (15.9%) had poor morphology; of the 2011with sperm count, 453 (22.5%) had low MSC; 654/1981 (33.0%) had either condition. Poor morphology, adjusted for confounding, was related to self-reported lifetime exposure to lead (OR=1.33; 95% CI 1.00 to 1.75). Low MSC was also related to self-reported lead and to hygienist-assessed glycol ether exposure. Self-reported use of paint stripper (OR=1.47; 95% CI 1.07 to 2.03) and lead, but not glycol ether, were significantly related to the combined case definition. CONCLUSIONS: While this study did not identify any occupational exposure uniquely related to sperm morphology, the capacity of the study to detect risk was increased by including morphology with sperm concentration and motility.

Pesticide exposure and screen-positive neuropsychiatric disease in British sheep farmers.

BACKGROUND: Epidemiological evidence linking low dose pesticide exposure and chronic ill-health in UK sheep farmers is limited. Our aim was to examine whether neuropsychiatric disorders were associated with low dose chronic and/or more acute pesticide exposure in sheep farmers. METHODS: A cohort of British farmers working in the 1970s was sent a screening questionnaire which asked about their health and work history. The prevalence of screen-positive depression, dementia, Parkinsonism and neuropathy was determined using a priori algorithms. Self-reported pesticide exposure was assessed by whether the participant had ever handled the pesticide concentrate (for low dose chronic exposure) or sought advice for pesticide poisoning (acute exposure) and participants categorised into those with only acute or chronic exposure, those with both acute and chronic exposure and those with neither acute nor chronic exposure. Associations between acute and chronic pesticide exposure, and screen-positive ill-health were determined after adjustment for demographic, lifestyle, occupation and somatic severity scores and other variables. RESULTS: In those participants, who had never sought advice for pesticide poisoning, handling the pesticide concentrate for treating sheep was associated with elevated ORs for screen-positive neuropathy (ORadi 1.57 95%CI 0.97-2.54) and Parkinsonism (ORadj 1.56 95%CI 0.95-2.56) but not depression or dementia. In those participants who had handled the pesticide concentrate, seeking advice for pesticide poisoning was associated with screen-positive depression (Odds ratio, ORadj=9.97 95%CI 4.76-20.8 ), dementia (OR=6.94 95%CI 3.44-14.0), Parkinsonism (ORadj=4.77 95% 2.39-9.52), and neuropathy (ORadj=4.77 95%CI 2.39-9.52). Adjustment for somatic severity score modified little the associations with pesticide handling in those not acutely exposed but reduced the ORs for seeking advice for pesticide poisoning in those exposed chronically. Furthermore, stratification of results based upon somatic severity score indicated that the highest ORs for handling the pesticide concentrate associated with neuropathy and Parkinsonism were found in those participants whose somatic score was minimal. CONCLUSIONS: Results are consistent with low-dose exposure to pesticides being associated with screen-positive neuropathy and Parkinsonism but the stronger associations between seeking advice for pesticide poisoning and screen-positive ill-health suggest that acute pesticide exposure remains an important determinant of ill-health. Further work is required to better delineate to what extent low dose exposures may contribute to ill-health in populations without acute exposures. Somatising tendency does not appear to play an important role in this population.

Applying optimal model selection in principal stratification for causal inference.

Noncompliance to treatment allocation is a key source of complication for causal inference. Efficacy estimation is likely to be compounded by the presence of noncompliance in both treatment arms of clinical trials where the intention-to-treat estimate provides a biased estimator for the true causal estimate even under homogeneous treatment effects assumption. Principal stratification method has been developed to address such posttreatment complications. The present work extends a principal stratification method that adjusts for noncompliance in two-treatment arms trials by developing model selection for covariates predicting compliance to treatment in each arm. We apply the method to analyse data from the Esprit study, which was conducted to ascertain whether unopposed oestrogen (hormone replacement therapy) reduced the risk of further cardiac events in postmenopausal women who survive a first myocardial infarction. We adjust for noncompliance in both treatment arms under a Bayesian framework to produce causal risk ratio estimates for each principal stratum. For mild values of a sensitivity parameter and using separate predictors of compliance in each arm, principal stratification results suggested that compliance with hormone replacement therapy only would reduce the risk for death and myocardial reinfarction by about 47% and 25%, respectively, whereas compliance with either treatment would reduce the risk for death by 13% and reinfarction by 60% among the most compliant. However, the results were sensitive to the user-defined sensitivity parameter.

Time trends in the incidence of work-related mental ill-health and musculoskeletal disorders in the UK.

OBJECTIVES: To determine UK trends (from 1996 to 2009) in incidence of work-related mental ill-health and musculoskeletal disorders, for all industry as well as for health and social care employees. Second, to investigate whether there may have been a recent shift from a physical to psychological perspective in how patients present their illness by comparing reporting trends for back pain and 'other work stress'. METHODS: Multilevel models were used to investigate changes in incidence of work-related illness, as diagnosed by specialist physicians. The dependent variable comprised case reports to The Health and Occupation Research network. Comparisons were made between medical specialties, industry (health and social care vs all other employees), gender and diagnosis. RESULTS: Trends for Occupational Physicians' (OP) reporting mental ill-health (average annual increase +3.7% (95% CI +2.2% to +5.2%)) differed significantly (p<0.001) from psychiatrists' reporting over the same time period (-5.9% (95% CI -7.6% to -4.2%)). For OPs' reporting, the rate of increase was greater for females and for health and social care employees. A fall in incidence of musculoskeletal disorders for OPs of -5.8% (95% CI -7.3% to -4.3%) and rheumatologists' reporting -6.6% (95% CI -8.3% to -4.8%) was found, with little variation by gender or industry. Within health and social care, an increase in incidence of 'other work stress' was accompanied by a similar decrease in 'spine/back pain'. CONCLUSIONS: The evidence presented is consistent with a shift in the presentation of ill-health from a physical to psychological perspective, although changes in hazards, prevention measures and physician awareness should also be considered as explanations.

Has the short-term effect of black smoke exposure on pneumonia mortality been underestimated because hospitalisation is ignored: findings from a case-crossover study.

BACKGROUND: Short-term associations have been demonstrated between air pollution and respiratory mortality including pneumonia. Studies typically estimate exposure based only on place of residence, yet many are in hospital prior to death. This study investigates lag length and tests the hypothesis that the effect of 'black smoke' is greater when restricted to pneumonia deaths in the community - Community Deaths from Pneumonia. METHODS: A time-stratified case-crossover design using conditional logistic regression estimated the daily percentage increase in risk of pneumonia mortality in relation to 'black smoke' in the preceding 30 days. Cases were pneumonia deaths in Edinburgh 1981-1996. Multiple 'control' periods, were defined using the same weekdays for the same month as the case death. Lag structure was investigated by a stratified lag model with five 6-day periods and by distributed lag models. Hospital admissions data, defined a community death as someone who had not been in hospital in any of the 30 days before death. RESULTS: Of 14,346 subjects who died from pneumonia, 7,536 were community deaths. Larger estimated increases in risks were seen in the community for all lag periods. Both stratified and distributed lag methods suggested positive effect estimates for 18 days after exposure and negative thereafter; the average percent increase per day across the 18 days was 0.70% (95% C.I. 0.29-1.14) for community subjects and 0.30% (95% C.I. 0.03-0.59) for all subjects. CONCLUSIONS: Studies which fail to account for hospitalisation may underestimate exposure effects as stronger pollution effects on mortality were evident in community based subjects.

An Epidemiological Study of Lung Cancer and Selected Other Cancers among Namibian Uranium Workers.

The Rössing Uranium Limited (RUL) open-cast uranium mine in Namibia has operated since 1976. Studies of underground uranium miners from Europe and North America have shown increased cancer risks (principally lung cancer). We explored the association between radiation doses and selected cancers in RUL mineworkers. Employees with at least one-year of continuous employment between 1976 and 2010 were included. Incident cancer cases [lung, extra-thoracic airways (ETA), leukemia, brain and kidney] occurring before the end of 2015 were identified from the Namibian and South African National Cancer Registries, and RUL's occupational health provider. Using a case-cohort design, data on exposure and confounding factors were collected for all cancer cases among the study cohort and a stratified random sample (sub-cohort) of the cohort, including cases. Radiation doses were estimated based on annual dose records held by RUL. In total, 76 cancer cases (32 lung, 18 ETA, 8 leukemia, 9 brain, 9 kidney) and a sub-cohort of 1,121 sampled from 7,901 RUL employees were included. A weighted Cox model, adjusted for available known confounders, produced a rate ratio (95% CI) for lung cancer of 1.42 (0.42, 4.77) and 1.22 (0.26, 5.68), respectively, for medium and higher cumulative lung dose categories compared to the lower category, and 1.04 (0.95, 1.13) for a dose increase of 10 mSv. This study faced considerable challenges with respect to case ascertainment, exposure estimates, and ensuring accuracy of key variables. Persuasive consistent evidence for elevated cancer risk was not found for radiation or other exposures studied at the Rössing uranium mine.

Increased risk of miscarriage and ectopic pregnancy among women with irritable bowel syndrome.

BACKGROUND & AIMS: Irritable bowel syndrome (IBS) is the most commonly diagnosed gastrointestinal condition and is most prevalent in women of reproductive age. We investigated the effects of IBS on risk for adverse outcomes from pregnancy. METHODS: We conducted a cohort study by using the United Kingdom General Practice Research Database. The study cohort consisted of 100,000 women selected by stratified random sampling from all women with a diagnosis of pregnancy from January 1, 1990, to December 31, 2008. Those with a recorded diagnosis of IBS before pregnancy were identified (n = 26,543). Outcome measures were spontaneous miscarriage, ectopic pregnancy, preeclampsia, and stillbirth. Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between IBS and pregnancy outcomes were estimated by using logistic regression adjusted for several potential confounders. RESULTS: Of women diagnosed with IBS before pregnancy, 6578 (7%) had a spontaneous miscarriage, 741 (0.74%) had an ectopic pregnancy, 425 (0.43%) developed preeclampsia, and 217 (0.22%) had a stillbirth. Maternal IBS was associated with a moderately increased risk of miscarriage (OR, 1.21; 95% CI, 1.13-1.30) and ectopic pregnancy (OR, 1.28%; 95% CI, 1.06-1.55). There did not appear to be an association between IBS and preeclampsia (OR, 1.09; 95% CI, 0.85-1.39) or stillbirth (OR, 1.00; 95% CI, 0.69-1.44). CONCLUSIONS: IBS, a common disorder in women of reproductive age, appears to increase the risk of miscarriage and ectopic pregnancy. These findings indicate the importance of prenatal care for women with IBS.

Performance of bias-correction methods for exposure measurement error using repeated measurements with and without missing data.

It is known that measurement error leads to bias in assessing exposure effects, which can however, be corrected if independent replicates are available. For expensive replicates, two-stage (2S) studies that produce data 'missing by design', may be preferred over a single-stage (1S) study, because in the second stage, measurement of replicates is restricted to a sample of first-stage subjects. Motivated by an occupational study on the acute effect of carbon black exposure on respiratory morbidity, we compare the performance of several bias-correction methods for both designs in a simulation study: an instrumental variable method (EVROS IV) based on grouping strategies, which had been recommended especially when measurement error is large, the regression calibration and the simulation extrapolation methods. For the 2S design, either the problem of 'missing' data was ignored or the 'missing' data were imputed using multiple imputations. Both in 1S and 2S designs, in the case of small or moderate measurement error, regression calibration was shown to be the preferred approach in terms of root mean square error. For 2S designs, regression calibration as implemented by Stata software is not recommended in contrast to our implementation of this method; the 'problematic' implementation of regression calibration although substantially improved with use of multiple imputations. The EVROS IV method, under a good/fairly good grouping, outperforms the regression calibration approach in both design scenarios when exposure mismeasurement is severe. Both in 1S and 2S designs with moderate or large measurement error, simulation extrapolation severely failed to correct for bias.

Evaluating interventions aimed at reducing occupational exposure to latex and rubber glove allergens.

OBJECTIVE: Concerns over occupational exposures to blood-borne viruses resulted in increased protective glove use; consequentially latex allergy became a hazard for some occupational groups. Interventions aimed at managing this problem included substitution measures (eg, non-powdered/non-latex gloves), but such changes may not occur simultaneously across occupational sectors. This study evaluated whether the incidence of occupational dermatoses fell after interventions aiming to reduce exposure to 'latex and rubber glove allergens' ('latex') were introduced, and whether these interventions were more effective for healthcare workers (HCWs), compared with non-HCWs. METHODS: Incidence rate ratios (IRRs) comparing cases reported to EPIDERM (a UK-wide surveillance scheme) during post versus pre-intervention periods were calculated, both where 'latex' was cited and for cases associated with other exposures ('controls'). RESULTS: Among HCWs, cases of contact urticaria and allergic contact dermatitis (ACD) where 'latex' was cited showed significant downward trends post-intervention, with IRRs of 0.72, 95% CI; 0.52 to 1.00 and 0.47, 95% CI; 0.35 to 0.64 respectively. For HCWs, this fall in 'latex' associated ACD was significantly greater (p=0.02) than for other exposures ('controls') IRR=0.85, 95% CI; 0.57 to 1.28, and greater than that among non-HCWs (IRR 0.75, 95% CI; 0.61 to 0.93). Increases over time were seen for irritant contact dermatitis (ICD) reporting for HCWs, both for cases associated with 'latex' (IRR 1.47, 95% CI: 1.02 to 2.13) and for other exposures ('controls') IRR 1.36, 95% CI 1.06 to 1.76, but not for non-HCWs. CONCLUSIONS: A reduction in overall ACD, particularly in HCWs, coincided with interventions aimed at managing workplace contact dermatoses associated with 'latex' exposure. A coincidental rise in ICD reporting is also important, both for hand care and for infection control strategies.

Work-related asthma in France: recent trends for the period 2001-2009.

OBJECTIVE: Knowledge on the time-course (trends) of work-related asthma (WRA) remains sparse. The aim of this study was to describe WRA trends in terms of industrial activities and the main causal agents in France over the period 2001-2009. METHOD: Data were collected from the French national network of occupational health surveillance and prevention (Réseau National de Vigilance et de Prévention des Pathologies Professionnelles (RNV3P)). Several statistical models (non-parametric test, zero-inflated negative binomial, logistic regression and time-series models) were used and compared with assess trends. RESULTS: Over the study period, 2914 WRA cases were included in the network. A significant decrease was observed overall and for some agents such as isocyanates (p = 0.007), aldehydes (p = 0.01) and latex (p = 0.01). Conversely, a significant increase was observed for cases related to exposure to quaternary ammonium compounds (p = 0.003). The health and social sector demonstrated both a growing number of cases related to the use of quaternary ammonium compounds and a decrease of cases related to aldehyde and latex exposure. CONCLUSIONS: WRA declined in France over the study period. The only significant increase concerned WRA related to exposure to quaternary ammonium compounds. Zero-inflated negative binomial and logistic regression models appear to describe adequately these data.

Time-sampled versus continuous-time reporting for measuring incidence.

BACKGROUND: Accuracy of incidence estimates may be affected by biases that depend on frequency of approach to reporters and reporting window length. A time-sampling strategy enables infrequent approaches with short windows but has never been evaluated. METHODS: A randomized crossover trial compared incidence estimates of work-related diseases using time-sampled versus continuous-time reporting. Physicians were randomly allocated either to report every month (12/12) in 2004 and for 1 randomly chosen month (1/12) in 2005, or to the reverse sequence. Numbers of new cases of work-related disease reported per reporter per month for 1/12 and 12/12 reporting periods were compared. RESULTS: Response rates were high (87%). Withdrawal from the study was higher under 12/12 reporting. The rate ratio for 1/12 versus 12/12 reporting was 1.26 (95% confidence interval = 1.11-1.42). Rates declined gradually in the 12/12 groups over the year, consistent with reporting fatigue. CONCLUSIONS: Increased frequency of data collection may reduce incidence estimates.

Performance of statistical methods for analysing survival data in the presence of non-random compliance.

Noncompliance often complicates estimation of treatment efficacy from randomized trials. Under random noncompliance, per protocol analyses or even simple regression adjustments for noncompliance, could be adequate for causal inference, but special methods are needed when noncompliance is related to risk. For survival data, Robins and Tsiatis introduced the semi-parametric structural Causal Accelerated Life Model (CALM) which allows time-dependent departures from randomized treatment in either arm and relates each observed event time to a potential event time that would have been observed if the control treatment had been given throughout the trial. Alternatively, Loeys and Goetghebeur developed a structural Proportional Hazards (C-Prophet) model for when there is all-or-nothing noncompliance in the treatment arm only. Whitebiet al. proposed a 'complier average causal effect' method for Proportional Hazards estimation which allows time-dependent departures from randomized treatment in the active arm. A time-invariant version of this estimator (CHARM) consists of a simple adjustment to the Intention-to-Treat hazard ratio estimate. We used simulation studies mimicking a randomized controlled trial of active treatment versus control with censored time-to-event data, and under both random and non-random time-dependent noncompliance, to evaluate performance of these methods in terms of 95 per cent confidence interval coverage, bias and root mean square errors (RMSE). All methods performed well in terms of bias, even the C-Prophet used after treating time-varying compliance as all-or-nothing. Coverage of the latter method, as implemented in Stata, was too low. The CALM method performed best in terms of bias and coverage but had the largest RMSE.

Undiagnosed coeliac disease in a father does not influence birthweight and preterm birth.

There is conflicting evidence regarding the effect of coeliac disease (CD) in the father on birthweight and preterm birth. We investigated the association between paternal CD and birthweight and preterm birth. Medical records of all singleton live-born children in Denmark between 1 January 1979 and 31 December 2004 were linked to information about parents' diseases. Fathers who were diagnosed with CD were then identified. Fathers with CD were considered treated if they were diagnosed before pregnancy and untreated if they were diagnosed after the date of conception. The outcome measures were: birthweight, small-for-gestational age (birthweight<10th centile for gestational age) and preterm birth (<37 weeks). We compared the offspring of men without CD (n = 1 472 352) and offspring of those with CD [untreated (n = 138) and treated (n = 473)]. There was no significant association between untreated CD in the father and birthweight (adjusted mean difference = -3 g; [95% CI -46, 40]) or preterm birth (adjusted odds ratio (OR) = 0.86, [95% CI 0.53, 1.37]) (compared with no CD). There was some evidence for an association between treated paternal CD and birthweight (adjusted mean difference = -81 g; [95% CI -161, -3]), but not preterm birth (adjusted OR = 1.76, [95% CI 0.95, 3.26]). Untreated paternal CD was not associated with an increased risk of reduced birthweight, or of preterm birth. There was some evidence that diagnosis and presumed treatment of paternal CD with a gluten-free diet is associated with reduced birthweight.

Agreement in diagnosing occupational asthma by occupational and respiratory physicians who report to surveillance schemes for work-related ill-health.

OBJECTIVES: To assess diagnostic agreement for occupational asthma, and to identify case and rater characteristics associated with this diagnosis. METHODS: Summaries of possible occupational asthma cases were sent to 104 occupational and respiratory physicians. Raters assigned likelihood scores (0-100%) of occupational asthma based on case histories (phase 1), and on histories plus investigative procedures (phase 2). Interclass correlation coefficients were calculated as statistical measures of reliability for occupational asthma scores. Comparisons between mean scores were assessed for statistical significance using tests based on multilevel models. RRs were calculated to summarise effects of raters' demographics, and of supplying investigative procedures information. RESULTS: Occupational asthma scores showed limited agreement within each group of (occupational or respiratory) physicians, but scores were not systematically different. The difference between mean overall scores was 2.1% (52.1% occupational physicians; 50.0% respiratory physicians) in phase 1 (95% CI -2.6 to 6.8, p=0.37). In phase 2, mean overall scores were 46.1% (occupational physicians) and 41.5% (respiratory physicians); the difference in mean overall scores was 4.6% (95% CI -3.5 to 12.5, p=0.27). Raters with General Medical Council registration > or =1986 were more likely to give a positive occupational asthma diagnosis. In phase 2, male raters were more likely to label cases as occupational asthma than female raters (RR 4.5, 95% CI 3.3 to 6.0). CONCLUSIONS: The RR of a positive occupational asthma diagnosis was unaffected by clinical specialty. Further work on why physicians consider cases to be occupational asthma will assist better diagnosis and prevention of this disease.

Phytoestrogen intake and other dietary risk factors for low motile sperm count and poor sperm morphology.

BACKGROUND: Few potentially modifiable risk factors of male infertility have been identified, and while different diets and food groups have been associated with male infertility, evidence linking dietary factors including phytoestrogens and semen quality is limited and contradictory. OBJECTIVES: To study the associations between phytoestrogen intake and other dietary factors and semen quality. MATERIALS AND METHODS: A case-referent study was undertaken of the male partners, of couples attempting conception with unprotected intercourse for 12 months or more without success, recruited from 14 UK assisted reproduction clinics. A total of 1907 participants completed occupational, lifestyle and dietary questionnaires before semen quality (concentration, motility and morphology) were assessed. Food intake was estimated by a 65-item food frequency questionnaire (FFQ) covering the 12 months prior to recruitment. Analyses of dietary risk factors for low motile sperm concentration (MSC: <4.8 × 106 /mL) and poor sperm morphology (PM: <4% normal morphology) used unconditional logistic regression, accounting for clustering of subjects within the clinics, first without, and then with, adjustment for confounders associated with that outcome. RESULTS: High consumption of daidzein (≥13.74 µg/d), a phytoestrogen found in soy products, was a protective factor for MSC with an odds ratio (95%CI) of 0.58 (0.42-0.82) after adjustment for clustering and potential confounding. Dietary risk factors for PM after similar adjustment showed that drinking whole milk (OR 0.67, 95%CI 0.47-0.96) and eating red meat were protective with an OR 0.67 (0.46-0.99) for eating red meat >3 times/wk. DISCUSSION: In this case-referent study of men attending an infertility clinic for fertility diagnosis, we have identified that low MSC is inversely associated with daidzein intake. In contrast, daidzein intake was not associated with PM but eating red milk and drinking whole milk were protective. CONCLUSIONS: Dietary factors associated with semen quality were identified, suggesting that male fertility might be improved by dietary changes.

Intention to treat, per protocol, as treated and instrumental variable estimators given non-compliance and effect heterogeneity.

We consider the behaviour of three approaches to efficacy estimation--using so-called 'as treated' (AT), 'per protocol' (PP) and 'instrumental variable' (IV) analyses--and of the Intention to Treat estimator, in a two-arm randomized treatment trial with a Normally distributed outcome when there is treatment effect heterogeneity and non-random compliance with assigned treatment. Formulae are derived for the bias of estimators when used either to estimate average treatment effect (ACE) or to estimate complier average treatment effect (CACE) under several models for the relationship between compliance and potential outcomes. These enable the expected values of AT, PP and IV estimators to be ranked in relation to ACE, and show that AT and PP estimators are generally biased for both ACE and CACE even under homogeneity. However, we show that the difference between any pair of (AT, PP, IV) estimates can be used to estimate the correlation between the latent variable determining compliance behaviour and one potential outcome. In the absence of measures that predict compliance, bounds for ACE can only be set given strong assumptions. Regarding the Intention to Treat estimator, while this is 'biased towards the null' if viewed as a measure of CACE, we show that it is not always so in relation to ACE. Finally we discuss the behaviour of the estimators under weak and strong null hypotheses.