Finite-Sample Bias of the Linear Excess Relative Risk in Cohort Studies of Computed Tomography-Related Radiation Exposure and Cancer.

The linear excess relative risk (ERR) is the most commonly reported measure of association in radiation epidemiological studies, when individual dose estimates are available. While the asymptotic properties of the ERR estimator are well understood, there is evidence of small sample bias in case-control studies of treatment-related radiation exposure and second cancer risk. Cohort studies of cancer risk after exposure to low doses of radiation from diagnostic procedures, e.g., computed tomography (CT) examinations, typically have small numbers of cases and risks are small. Therefore, understanding the properties of the estimated ERR is essential for interpretation and analysis of such studies. We present results of a simulation study that evaluates the finite-sample bias of the ERR estimated by time-to-event analyses and its confidence interval using simulated data, resembling a retrospective cohort study of radiation-related leukemia risk after CT examinations in childhood and adolescence. Furthermore, we evaluate how the Firth-corrected estimator reduces the finite-sample bias of the classical estimator. We show that the ERR is overestimated by about 30% for a cohort of about 150,000 individuals, with 42 leukemia cases observed on average. The bias is reduced for higher baseline incidence rates and for higher values of the true ERR. As the number of cases increases, the ERR is approximately unbiased. The Firth correction reduces the bias for all cohort sizes to generally around or under 5%. Epidemiological studies showing an association between radiation exposure from pediatric CT and cancer risk, unless very large, may overestimate the magnitude of the relationship, while there is no evidence of an increased chance for false-positive results. Conducting large studies, perhaps by pooling individual studies to increase the number of cases, should be a priority. If this is not possible, Firth correction should be applied to reduce small-sample bias.

Role of circulating fibrocytes in the diagnosis of acute appendicitis.

BACKGROUND: Improved diagnostic biomarkers are required for acute appendicitis. The circulating fibrocyte percentage (CFP) is increased in inflammatory states, but has not been studied in acute appendicitis. This study aimed to determine CFP in acute appendicitis and compare diagnostic accuracy with standard serological biomarkers. METHODS: A prospective cohort study was carried out between June 2015 and February 2016 at University Hospital Limerick. The CFP was determined by dual-staining peripheral venous samples for CD45 and collagen I using fluorescence-activated cell sorting, and correlated with histopathological diagnoses. The accuracy of CFP in determining histological acute appendicitis was characterized and compared with the white cell count, C-reactive protein concentration, neutrophil count, lymphocyte count and neutrophil : lymphocyte ratio. RESULTS: Of 95 adults recruited, 15 were healthy individuals and 80 had suspected appendicitis at presentation. Forty-six of these 80 patients had an appendicectomy, of whom 34 had histologically confirmed appendicitis. The CFP was statistically higher in patients with pathologically proven acute appendicitis than in healthy controls (median 6·1 (i.q.r. 1·6-11·6) versus 2·3 (0·9-3·4) per cent respectively; P = 0·008). The diagnostic accuracy of CFP, as determined using the area under the receiver operating characteristic (ROC) curve, was similar to that of standard biomarkers. In multinomial regression analysis, only raised CFP was retained as an independent prognostic determinant of acute appendicitis (odds ratio 1·57, 95 per cent c.i. 1·05 to 2·33; P = 0·027). CONCLUSION: The CFP is increased in histologically confirmed acute appendicitis and is as accurate as standard serological biomarkers in terms of diagnosis.

Strengthening the Reporting of Observational Studies in Epidemiology - nutritional epidemiology (STROBE-nut): An extension of the STROBE statement.

Concerns have been raised about the quality of reporting in nutritional epidemiology. Research reporting guidelines such as the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement can improve quality of reporting in observational studies. Herein, we propose recommendations for reporting nutritional epidemiology and dietary assessment research by extending the STROBE statement into Strengthening the Reporting of Observational Studies in Epidemiology - Nutritional Epidemiology (STROBE-nut). Recommendations for the reporting of nutritional epidemiology and dietary assessment research were developed following a systematic and consultative process, co-ordinated by a multidisciplinary group of 21 experts. Consensus on reporting guidelines was reached through a three-round Delphi consultation process with 53 external experts. In total, 24 recommendations for nutritional epidemiology were added to the STROBE checklist. When used appropriately, reporting guidelines for nutritional epidemiology can contribute to improve reporting of observational studies with a focus on diet and health.

A meta-analysis of Hodgkin lymphoma reveals 19p13.3 TCF3 as a novel susceptibility locus.

Recent genome-wide association studies (GWAS) of Hodgkin lymphoma (HL) have identified associations with genetic variation at both HLA and non-HLA loci; however, much of heritable HL susceptibility remains unexplained. Here we perform a meta-analysis of three HL GWAS totaling 1,816 cases and 7,877 controls followed by replication in an independent set of 1,281 cases and 3,218 controls to find novel risk loci. We identify a novel variant at 19p13.3 associated with HL (rs1860661; odds ratio (OR)=0.81, 95% confidence interval (95% CI) = 0.76-0.86, P(combined) = 3.5 × 10(-10)), located in intron 2 of TCF3 (also known as E2A), a regulator of B- and T-cell lineage commitment known to be involved in HL pathogenesis. This meta-analysis also notes associations between previously published loci at 2p16, 5q31, 6p31, 8q24 and 10p14 and HL subtypes. We conclude that our data suggest a link between the 19p13.3 locus, including TCF3, and HL risk.

Estimating the asbestos-related lung cancer burden from mesothelioma mortality.

BACKGROUND: Quantifying the asbestos-related lung cancer burden is difficult in the presence of this disease's multiple causes. We explore two methods to estimate this burden using mesothelioma deaths as a proxy for asbestos exposure. METHODS: From the follow-up of 55 asbestos cohorts, we estimated ratios of (i) absolute number of asbestos-related lung cancers to mesothelioma deaths; (ii) excess lung cancer relative risk (%) to mesothelioma mortality per 1000 non-asbestos-related deaths. RESULTS: Ratios varied by asbestos type; there were a mean 0.7 (95% confidence interval 0.5, 1.0) asbestos-related lung cancers per mesothelioma death in crocidolite cohorts (n=6 estimates), 6.1 (3.6, 10.5) in chrysotile (n=16), 4.0 (2.8, 5.9) in amosite (n=4) and 1.9 (1.4, 2.6) in mixed asbestos fibre cohorts (n=31). In a population with 2 mesothelioma deaths per 1000 deaths at ages 40-84 years (e.g., US men), the estimated lung cancer population attributable fraction due to mixed asbestos was estimated to be 4.0%. CONCLUSION: All types of asbestos fibres kill at least twice as many people through lung cancer than through mesothelioma, except for crocidolite. For chrysotile, widely consumed today, asbestos-related lung cancers cannot be robustly estimated from few mesothelioma deaths and the latter cannot be used to infer no excess risk of lung or other cancers.

Larval nutritional stress affects vector immune traits in adult yellow fever mosquito Aedes aegypti (Stegomyia aegypti).

We report key physiological traits that link larval nutritional experience to adult immune status in the yellow fever mosquito Aedes aegypti L. (Stegomyia aegypti) (Diptera: Culicidae). Many lines of defence make up the innate immune system of mosquitoes. Among defences, the epithelium-lined midgut is the first barrier, circulating haemocytes are cellular components of innate immunity and, when triggered, the Toll and Imd pathways signal production of antimicrobial peptides (AMP) as part of humoral defences. We quantified three lines of defence in Ae. aegypti in response to larval nutritional stress, and our data show that important female immune functions are modified by the larval rearing environment. Adult midgut basal lamina thickness was not affected by larval nutrient stress as has been observed in another Aedes sp. However, nutrient stresses experienced by larvae lead to a reduced number of haemocytes in females. Transcripts of Spaetzle (upstream regulator of Toll pathway that leads to induction of AMPs) and some immune-related genes were less abundant in stressed larvae but showed increased expression in females derived from stressed larvae. Results indicate a potential for compensation by the humoral branch for a reduced cellular branch of innate immunity in adults in response to larval nutrient stress.

Morphological predictors of BRCA1 germline mutations in young women with breast cancer.

BACKGROUND: Knowing a young woman with newly diagnosed breast cancer has a germline BRCA1 mutation informs her clinical management and that of her relatives. We sought an optimal strategy for identifying carriers using family history, breast cancer morphology and hormone receptor status data. METHODS: We studied a population-based sample of 452 Australian women with invasive breast cancer diagnosed before age 40 years for whom we conducted extensive germline mutation testing (29 carried a BRCA1 mutation) and a systematic pathology review, and collected three-generational family history and tumour ER and PR status. Predictors of mutation status were identified using multiple logistic regression. Areas under receiver operator characteristic (ROC) curves were estimated using five-fold stratified cross-validation. RESULTS: The probability of being a BRCA1 mutation carrier increased with number of selected histology features even after adjusting for family history and ER and PR status (P<0.0001). From the most parsimonious multivariate model, the odds ratio for being a carrier were: 9.7 (95% confidence interval: 2.6-47.0) for trabecular growth pattern (P=0.001); 7.8 (2.7-25.7) for mitotic index over 50 mitoses per 10 high-powered field (P=0.0003); and 2.7 (1.3-5.9) for each first-degree relative with breast cancer diagnosed before age 60 years (P=0.01).The area under the ROC curve was 0.87 (0.83-0.90). CONCLUSION: Pathology review, with attention to a few specific morphological features of invasive breast cancers, can identify almost all BRCA1 germline mutation carriers among women with early-onset breast cancer without taking into account family history.

Oral contraceptives, reproductive history and risk of colorectal cancer in the European Prospective Investigation into Cancer and Nutrition.

BACKGROUND: Oral contraceptive use and reproductive factors may initiate long-term changes to the hormonal milieu and thereby, possibly influence colorectal cancer risk. METHODS: We examined the association of hormonal and reproductive factors with risk of colorectal cancer among 337,802 women in the European Prospective Investigation into Cancer and Nutrition, of whom 1878 developed colorectal cancer. RESULTS: After stratification for center and age, and adjustment for body mass index, smoking, diabetes mellitus, physical activity and alcohol consumption, ever use of oral contraceptives was marginally inversely associated with colorectal cancer risk (hazard ratio (HR), 0.92; 95% confidence interval (CI), 0.83-1.02), although this association was stronger among post-menopausal women (HR, 0.84; 95% CI: 0.74-0.95). Duration of oral contraceptive use and reproductive factors, including age at menarche, age at menopause, type of menopause, ever having an abortion, parity, age at first full-term pregnancy and breastfeeding, were not associated with colorectal cancer risk. CONCLUSION: Our findings provide limited support for a potential inverse association between oral contraceptives and colorectal cancer risk.

Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials.

BACKGROUND: Early administration of intravenous recombinant tissue plasminogen activator (rt-PA) after ischaemic stroke improves outcome. Previous analysis of combined data from individual patients suggested potential benefit beyond 3 h from stroke onset. We re-examined the effect of time to treatment with intravenous rt-PA (alteplase) on therapeutic benefit and clinical risk by adding recent trial data to the analysis. METHODS: We added data from ECASS III (821 patients) and EPITHET (100 patients) to a pool of common data elements from six other trials of alteplase for acute stroke (2775 patients). We used multivariate logistic regression to assess the relation of stroke onset to start of treatment (OTT) with treatment on favourable 3-month outcome (defined as modified Rankin score 0-1), mortality, and occurrence and outcome of clinically relevant parenchymal haemorrhage. The presence of an arterial occlusion was inferred from the patient's symptoms and absence of haemorrhage or other causes of ischaemic stroke. Vascular imaging was not a requirement in the trials. All patients with confirmed OTT within 360 min were included in the analysis. FINDINGS: Treatment was started within 360 min of stroke onset in 3670 patients randomly allocated to alteplase (n=1850) or to placebo (n=1820). Odds of a favourable 3-month outcome increased as OTT decreased (p=0.0269) and no benefit of alteplase treatment was seen after around 270 min. Adjusted odds of a favourable 3-month outcome were 2.55 (95% CI 1.44-4.52) for 0-90 min, 1.64 (1.12-2.40) for 91-180 min, 1.34 (1.06-1.68) for 181-270 min, and 1.22 (0.92-1.61) for 271-360 min in favour of the alteplase group. Large parenchymal haemorrhage was seen in 96 (5.2%) of 1850 patients assigned to alteplase and 18 (1.0%) of 1820 controls, with no clear relation to OTT (p=0.4140). Adjusted odds of mortality increased with OTT (p=0.0444) and were 0.78 (0.41-1.48) for 0-90 min, 1.13 (0.70-1.82) for 91-180 min, 1.22 (0.87-1.71) for 181-270 min, and 1.49 (1.00-2.21) for 271-360 min. INTERPRETATION: Patients with ischaemic stroke selected by clinical symptoms and CT benefit from intravenous alteplase when treated up to 4.5 h. To increase benefit to a maximum, every effort should be taken to shorten delay in initiation of treatment. Beyond 4.5 h, risk might outweigh benefit. FUNDING: None.

Hyporesponsiveness to re-challenge dose following pneumococcal polysaccharide vaccine at 12 months of age, a randomized controlled trial.

BACKGROUND: To evaluate the immunological impact of the 23-valent pneumococcal polysaccharide vaccine (23vPPS) at 12 months, for children who have received zero to three infant doses of seven-valent pneumococcal conjugate vaccine (PCV), on responses to a subsequent exposure to a small dose of 23vPPS (mPPS). METHODS: Five hundred and fifty-two Fijian infants were stratified by ethnicity and randomized into eight groups to receive zero, one, two, or three PCV doses at 14 weeks, six and 14 weeks, or six, ten, and 14 weeks. Within each group, half received 23vPPS at 12 months and all received mPPS at 17 months. Sera were taken prior and one month post-mPPS. FINDINGS: By 17 months, geometric mean antibody concentrations (GMC) to all 23 serotypes in 23vPPS were significantly higher in children who had received 23vPPS at 12 months compared to those who had not. Post-mPPS, children who had not received the 12 month 23vPPS had a significantly higher GMC for all PCV serotypes compared with those who had (each p<0.02). For the non-PCV serotypes, children who had not received the 12 month 23vPPS had significantly higher GMC for six of 16 non-PCV serotypes (7F, 9N, 12F, 19A, 22F, 33F) than those who did (each p<0.02). After adjusting for the pre-mPPS level, exposure to 23vPPS was associated with a lower response to mPPS for all serotypes (each p<0.001). INTERPRETATION: Despite higher antibody concentrations at 17 months in children who had received 23vPPS at 12 months, the response to a re-challenge was poor for all 23 serotypes compared to children who had not received the 12 month 23vPPS.

Prospects for epigenetic research within cohort studies of psychological disorder: a pilot investigation of a peripheral cell marker of epigenetic risk for depression.

Cohort studies have considerable prima facie value for investigating epigenetic processes in psychological disorder; however, the future prospects for such studies will depend on valid peripheral markers. The purpose of this pilot study was to investigate association between buccal cell methylation and risk for depression. Epigenotyping was limited to promoter methylation of the serotonin transporter gene (5HTT). A transcription limiting VNTR in the 5HTT promoter (5HTTLPR) was also genotyped. A nested sample of 25 depressed and 125 non-depressed adolescents was drawn from an established longitudinal study of adolescent health. There was no association between depressive symptoms and either buccal cell 5HTT methylation or 5HTTLPR. However, depressive symptoms were more common among those with elevated buccal cell 5HTT methylation who carried 5HTTLPR short-allele (OR 4.9, CI 1.9-13, p=0.001). Both complete and partial (as little as 10%) methylation of a 5HTT reporter gene in an expressing cell line reduced 5HTT activity. Replication is needed.

Sibship analysis of associations between SNP haplotypes and a continuous trait with application to mammographic density.

Haplotype-based association studies have been proposed as a powerful comprehensive approach to identify causal genetic variation underlying complex diseases. Data comparisons within families offer the additional advantage of dealing naturally with complex sources of noise, confounding and population stratification. Two problems encountered when investigating associations between haplotypes and a continuous trait using data from sibships are (i) the need to define within-sibship comparisons for sibships of size greater than two and (ii) the difficulty of resolving the joint distribution of haplotype pairs within sibships in the absence of parental genotypes. We therefore propose first a method of orthogonal transformation of both outcomes and exposures that allow the decomposition of between- and within-sibship regression effects when sibship size is greater than two. We conducted a simulation study, which confirmed analysis using all members of a sibship is statistically more powerful than methods based on cross-sectional analysis or using subsets of sib-pairs. Second, we propose a simple permutation approach to avoid errors of inference due to the within-sibship correlation of any errors in haplotype assignment. These methods were applied to investigate the association between mammographic density (MD), a continuously distributed and heritable risk factor for breast cancer, and single nucleotide polymorphisms (SNPs) and haplotypes from the VDR gene using data from a study of 430 twins and sisters. We found evidence of association between MD and a 4-SNP VDR haplotype. In conclusion, our proposed method retains the benefits of the between- and within-pair analysis for pairs of siblings and can be implemented in standard software.

Fruits and vegetables consumption and the risk of histological subtypes of lung cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC).

OBJECTIVE: To examine the association between fruit and vegetable consumption and risk of different histological subtypes of lung cancer among participants of the European Prospective Investigation into Cancer and Nutrition study. METHODS: Multivariable Cox proportional hazard models were used to analyze the data. A calibration study in a subsample was used to reduce dietary measurement errors. RESULTS: During a mean follow-up of 8.7 years, 1,830 incident cases of lung cancer (574 adenocarcinoma, 286 small cell, 137 large cell, 363 squamous cell, 470 other histologies) were identified. In line with our previous conclusions, we found that after calibration a 100 g/day increase in fruit and vegetables consumption was associated with a reduced lung cancer risk (HR 0.94; 95% CI 0.89-0.99). This was also seen among current smokers (HR 0.93; 95% CI 0.90-0.97). Risks of squamous cell carcinomas in current smokers were reduced for an increase of 100 g/day of fruit and vegetables combined (HR 0.85; 95% CI 0.76-0.94), while no clear effects were seen for the other histological subtypes. CONCLUSION: We observed inverse associations between the consumption of vegetables and fruits and risk of lung cancer without a clear effect on specific histological subtypes of lung cancer. In current smokers, consumption of vegetables and fruits may reduce lung cancer risk, in particular the risk of squamous cell carcinomas.

Immunogenicity following one, two, or three doses of the 7-valent pneumococcal conjugate vaccine.

The aim was to identify an appropriate infant pneumococcal vaccination strategy for resource poor countries. Fijian infants received zero, one, two, or three doses of 7-valent pneumococcal conjugate vaccine (PCV) in early infancy. Following three PCV doses, geometric mean concentration (GMC) to all seven serotypes were > or = 1.0 microg/mL, and >85% of children achieved antibody levels > or = 0.35 microg/mL at 18 weeks. Following two doses, GMC were lower for 6B, 14, and 23F, but higher for 19F compared with three doses. Following a single dose, significant responses were seen for all serotypes post-primary series compared with the unvaccinated. By 12 months, differences between two and three doses persisted for serotype 14 only. Although GMC following three doses are higher than after two doses, the differences were small. A single dose may offer some protection for most serotypes.

Hospitalized community-acquired pneumonia in the elderly: an Australian case-cohort study.

This study describes the epidemiology of community-acquired pneumonia (CAP) in elderly Australians for the first time. Using a case-cohort design, cases with CAP were in-patients aged > or = 65 years with ICD-10-AM codes J10-J18 admitted over 2 years to two tertiary hospitals. The cohort sample was randomly selected from all hospital discharges, frequency-matched to cases by month. Logistic regression was used to estimate risk ratios for factors predicting CAP or associated mortality. A total of 4772 in-patients were studied. There were 1952 cases with CAP that represented 4% of all elderly admissions: mean length of stay was 9.0 days and 30-day mortality was 18%. Excluding chest radiograph, 520/1864 (28%) cases had no investigations performed. The strongest predictors of CAP were previous pneumonia, history of other respiratory disease, and aspiration. Intensive-care-unit admission, renal disease and increasing age were the strongest predictors of mortality, while influenza vaccination conferred protection. Hospitalization with CAP in the elderly is common, frequently fatal and a considerable burden to the Australian community. Investigation is ad hoc and management empirical. Influenza vaccination is associated with reduced mortality. Patient characteristics can predict risk of CAP and subsequent mortality.

A meta-analysis of ambulatory versus inpatient laparoscopic cholecystectomy.

BACKGROUND: Laparoscopic cholecystectomy is increasingly used on an ambulatory basis. This study aimed to examine its effectiveness for carefully selected patients. METHODS: A systematic review of Cochrane, Embase, and Medline using the keywords "ambulatory," "laparoscopic," and "cholecystectomy" was performed. Postoperative complications leading to admissions and readmissions were compared between day care and inpatient laparoscopic cholecystectomy groups. Postoperative quality of life, patient satisfaction, and cost effectiveness also were analyzed. RESULTS: The search process identified seven clinical trials suitable for meta-analysis. These trials, consisting of 598 patients, compared day care and inpatient procedures. The unplanned admission rate in the ambulatory group was comparable with the prolonged hospitalization of inpatients (odds ratio [OR], 1.979; 95% confidence interval [CI], 0.846-4.628). There was no significant difference between the readmission rates of the two groups (OR, 0.964; 95% CI, 0.318-2.922). The quality-of-life indicators were similar for the ambulatory and overnight-stay patients (p = 0.195). The cost effectiveness was better for the day care procedures because of the shorter mean hospital stay. CONCLUSION: Ambulatory laparoscopic cholecystectomy can be performed safely for selected patients, with reduced cost and a high level of patient satisfaction.

Does smoking among friends explain apparent genetic effects on current smoking in adolescence and young adulthood?

We used data from a prospective cohort study of twins to investigate the influence of unmeasured genetic and measured and unmeasured environmental factors on the smoking behaviour of adolescents and young adults. Twins were surveyed in 1988 (aged 11-18 years), 1991, 1996 and 2004 with data from 1409, 1121, 732 and 758 pairs analysed from each survey wave, respectively. Questionnaires assessed the smoking behaviour of twins and the perceived smoking behaviour of friends and parents. Using a novel logistic regression analysis, we simultaneously modelled individual risk and excess concordance for current smoking as a function of zygosity, survey wave, parental smoking and peer smoking. Being concordant for having peers who smoked was a predictor of concordance for current smoking (P<0.001). After adjusting for peer smoking, monozygotic (MZ) pairs were no more alike than dizygotic pairs for current smoking at waves 2, 3 and 4. Genetic explanations are not needed to explain the greater concordance for current smoking among adult MZ pairs. However, if they are invoked, the role of genes may be due to indirect effects acting through the social environment. Smoking prevention efforts may benefit more by targeting social factors than attempting to identify genetic factors associated with smoking.

ICD-10 codes are a valid tool for identification of pneumonia in hospitalized patients aged > or = 65 years.

This study examines the validity of using ICD-10 codes to identify hospitalized pneumonia cases. Using a case-cohort design, subjects were randomly selected from monthly cohorts of patients aged > or = 65 years discharged from April 2000 to March 2002 from two large tertiary Australian hospitals. Cases had ICD-10-AM codes J10-J18 (pneumonia); the cohort sample was randomly selected from all discharges, frequency matched to cases by month. Codes were validated against three comparators: medical record notation of pneumonia, chest radiograph (CXR) report and both. Notation of pneumonia was determined for 5098/5101 eligible patients, and CXR reports reviewed for 3349/3464 (97%) patients with a CXR. Coding performed best against notation of pneumonia: kappa 0.95, sensitivity 97.8% (95% CI 97.1-98.3), specificity 96.9% (95% CI 96.2-97.5), positive predictive value (PPV) 96.2% (95% CI 95.4-97.0) and negative predictive value (NPV) 98.2% (95% CI 97.6-98.6). When medical record notation of pneumonia is used as the standard, ICD-10 codes are a valid method for retrospective ascertainment of hospitalized pneumonia cases and appear superior to use of complexes of symptoms and signs, or radiology reports.

COMT Val(158)Met and 5HTTLPR functional loci interact to predict persistence of anxiety across adolescence: results from the Victorian Adolescent Health Cohort Study.

We investigated whether a composite genetic factor, based on the combined actions of catechol-O-methyltransferase (COMT) (Val(158)Met) and serotonin transporter (5HTTLPR) (Long-Short) functional loci, has a greater capacity to predict persistence of anxiety across adolescence than either locus in isolation. Analyses were performed on DNA collected from 962 young Australians participating in an eight-wave longitudinal study of mental health and well-being (Victorian Adolescent Health Cohort Study). When the effects of each locus were examined separately, small dose-response reductions in the odds of reporting persisting generalized (free-floating) anxiety across adolescence were observed for the COMT Met(158) [odds ratio (OR) = 0.85, 95% confidence interval (CI) = 0.76-0.95, P = 0.004] and 5HTTLPR Short alleles (OR = 0.88, CI = 0.79-0.99, P = 0.033). There was no evidence for a dose-response interaction effect between loci. However, there was a double-recessive interaction effect in which the odds of reporting persisting generalized anxiety were more than twofold reduced (OR = 0.45, CI = 0.29-0.70, P < 0.001) among carriers homozygous for both the COMT Met(158) and the 5HTTLPR Short alleles (Met(158)Met + Short-Short) compared with the remaining cohort. The double-recessive effect remained after multivariate adjustment for a range of psychosocial predictors of anxiety. Exploratory stratified analyses suggested that genetic protection may be more pronounced under conditions of high stress (insecure attachments and sexual abuse), although strata differences did not reach statistical significance. By describing the interaction between genetic loci, it may be possible to describe composite genetic factors that have a more substantial impact on psychosocial development than individual loci alone, and in doing so, enhance understanding of the contribution of constitutional processes in mental health outcomes.