Win-loss parameters for right-censored event data, with application to recurrent events.

The win ratio has become a popular method for comparing multiple event data between two groups in clinical cohort studies. The win ratio compares the event data in prioritized order, where the first prioritized event is death and a typical example for the second prioritized event is hospitalization. Literature is sparse on inference for win and loss parameters, including the win ratio, for censored event data. Inference for two prioritized censored event times has been developed for independent right-censoring. Many clinical studies include recurrent event data such as hospitalizations. In this article, we suggest inference for win-loss parameters for death and a recurrent event outcome under independent right-censoring. The small sample properties of the proposed method are studied in a simulation study showing that the variance formula is accurate even for small samples. The method is applied on a data set from a randomized clinical trial.

Comparing performance between clinics of an embryo evaluation algorithm based on time-lapse images and machine learning.

PURPOSE: This article aims to assess how differences in maternal age distributions between IVF clinics affect the performance of an artificial intelligence model for embryo viability prediction and proposes a method to account for such differences. METHODS: Using retrospectively collected data from 4805 fresh and frozen single blastocyst transfers of embryos incubated for 5 to 6 days, the discriminative performance was assessed based on fetal heartbeat outcomes. The data was collected from 4 clinics, and the discrimination was measured in terms of the area under ROC curves (AUC) for each clinic. To account for the different age distributions between clinics, a method for age-standardizing the AUCs was developed in which the clinic-specific AUCs were standardized using weights for each embryo according to the relative frequency of the maternal age in the relevant clinic compared to the age distribution in a common reference population. RESULTS: There was substantial variation in the clinic-specific AUCs with estimates ranging from 0.58 to 0.69 before standardization. The age-standardization of the AUCs reduced the between-clinic variance by 16%. Most notably, three of the clinics had quite similar AUCs after standardization, while the last clinic had a markedly lower AUC both with and without standardization. CONCLUSION: The method of using age-standardization of the AUCs that is proposed in this article mitigates some of the variability between clinics. This enables a comparison of clinic-specific AUCs where the difference in age distributions is accounted for.

Regression models for censored time-to-event data using infinitesimal jack-knife pseudo-observations, with applications to left-truncation.

Jack-knife pseudo-observations have in recent decades gained popularity in regression analysis for various aspects of time-to-event data. A limitation of the jack-knife pseudo-observations is that their computation is time consuming, as the base estimate needs to be recalculated when leaving out each observation. We show that jack-knife pseudo-observations can be closely approximated using the idea of the infinitesimal jack-knife residuals. The infinitesimal jack-knife pseudo-observations are much faster to compute than jack-knife pseudo-observations. A key assumption of the unbiasedness of the jack-knife pseudo-observation approach is on the influence function of the base estimate. We reiterate why the condition on the influence function is needed for unbiased inference and show that the condition is not satisfied for the Kaplan-Meier base estimate in a left-truncated cohort. We present a modification of the infinitesimal jack-knife pseudo-observations that provide unbiased estimates in a left-truncated cohort. The computational speed and medium and large sample properties of the jack-knife pseudo-observations and infinitesimal jack-knife pseudo-observation are compared and we present an application of the modified infinitesimal jack-knife pseudo-observations in a left-truncated cohort of Danish patients with diabetes.

Replacing Red Meat with Other Nonmeat Food Sources of Protein is Associated with a Reduced Risk of Type 2 Diabetes in a Danish Cohort of Middle-Aged Adults.

BACKGROUND: Few cohort studies have modelled replacements of red meat with other sources of protein on subsequent risk of type 2 diabetes using dietary changes. OBJECTIVES: To determine whether replacing red meat with other food sources of protein is associated with a lower risk of type 2 diabetes. METHODS: We used data from the Danish Diet, Cancer, and Health cohort (n = 39,437) of middle-aged (55-72 years old) men and women who underwent 2 dietary assessments roughly 5 years apart to investigate dietary changes. The pseudo-observation method was used to model the average exposure effect of decreasing the intake of red meat while increasing the intake of either poultry, fish, eggs, milk, yogurt, cheese, whole grains, or refined grains on the subsequent 10-year risk of developing type 2 diabetes, compared with no changes in the intakes of these foods. RESULTS: Replacing 1 serving/day (100 g/day) of red meat with 1 serving/day of eggs [risk difference (RD), -2.7%; 95% CI: -4.0 to -1.1%; serving size: 50 g/day], milk (RD, -1.2%; 95% CI: -2.1 to -0.4%; 200 g/day), yogurt (RD, -1.5%; 95% CI: -2.4 to -0.7%; 70 g/day), whole grains (RD, -1.7%; 95% CI: -2.5 to -0.9%; 30 g/day), or refined grains (RD, -1.2%; 95% CI: -2.0 to -0.3%; 30 g/day) was associated with a reduced risk of type 2 diabetes. Analyses of replacements with poultry or cheese, but not fish, also suggested a lower risk, but with wide CIs. After further adjustment for potential mediators (BMI, waist circumference, and history of hypertension or hypercholesterolemia), only the replacement with eggs was associated with a reduced risk (RD, -1.7%; 95% CI: -3.0 to -0.5%; 50 g/day). CONCLUSIONS: Replacing red meat with eggs in middle-aged adults may reduce the risk of type 2 diabetes. In models not adjusted for potential mediators, replacing red meat with milk, yogurt, whole grains, or refined grains was also associated with a reduced risk of type 2 diabetes.

Neonatal jaundice and autism spectrum disorder: a systematic review and meta-analysis.

BACKGROUND: Two meta-analyses concluded that jaundice was associated with an increased risk of autism. We hypothesize that these findings were due to methodological limitations of the studies included. Neonatal jaundice affects many infants and risks of later morbidity may prompt physicians towards more aggressive treatment. METHODS: To conduct a systematic literature review and a meta-analysis of the association between neonatal jaundice and autism with particular attention given to low risk of bias studies. Pubmed, Scopus, Embase, Cochrane, and Google Scholar were searched for publications until February 2019. Data was extracted by use of pre-piloted structured sheets. Low risk of bias studies were identified through predefined criteria. RESULTS: A total of 32 studies met the inclusion criteria. The meta-analysis of six low risk of bias studies showed no association between neonatal jaundice and autism; cohort studies risk ratio 1.09, 95% CI, 0.99-1.20, case-control studies odds ratio 1.29 95% CI 0.95, 1.76. Funnel plot of all studies suggested a high risk of publication bias. CONCLUSIONS: We found a high risk of publication bias, selection bias, and potential confounding in all studies. Based on the low risk of bias studies there was no convincing evidence to support an association between neonatal jaundice and autism. IMPACT: Meta-analysis of data from six low risk of bias studies indicated no association between neonatal jaundice and autism spectrum disorder. Previous studies show inconsistent results, which may be explained by unadjusted confounding and selection bias. Funnel plot suggested high risk of publication bias when including all studies. There is no evidence to suggest jaundice should be treated more aggressively to prevent autism.

Changes in intake of dairy product subgroups and risk of type 2 diabetes: modelling specified food substitutions in the Danish Diet, Cancer and Health cohort.

PURPOSE: We investigated the association between an increased intake of one dairy product subgroup at the expense of another within a 5-year period and the subsequent 10-year risk of type 2 diabetes. METHODS: The cohort included 39,393 adults with two measurements of diet assessed using food frequency questionnaires (FFQ) administered in 1993-1997 and 1999-2003. Dairy products were milk (skimmed, semi-skimmed, whole fat), buttermilk, low-fat yogurt, whole-fat yogurt, cheese and butter. Type 2 diabetes cases were ascertained from the Danish National Diabetes Register. The pseudo-observation method was used to calculate risk differences (RD) with 95% confidence intervals (CI). The data were analysed in age strata to fulfil the assumption of independent entry. RESULTS: Among participants aged 56-59 years at completion of the follow-up FFQ, increased intake of whole-fat yogurt in place of skimmed, semi-skimmed or whole-fat milk was associated with a reduced risk (RD% [95% CI]: - 0.8% [- 1.3, - 0.2]; - 0.6% [- 1,1, - 0.1]; - 0.7 [- 1.2, - 0.1]; per 50 g/d, respectively). Among participants aged 60-64 and 65-72, substitution of skimmed milk for semi-skimmed milk was associated with an increased risk of type 2 diabetes (0.5% [0.2, 0.7]; 0.4% [0.1, 0.7]; per 50 g/d, respectively). Similar patterns of associations were found after adjustment for potential mediators. CONCLUSION: Our results suggest that substitution of whole-fat yogurt for milk among those aged 56-59 decreases risk of type 2 diabetes and substitution of skimmed milk for semi-skimmed milk may increase the risk among those aged 60-64 and 65-72.

Repeat pulmonary vein isolation in patients with atrial fibrillation: low ablation index is associated with increased risk of recurrent arrhythmia.

OBJECTIVES: We aimed to investigate the predictors of recurrent arrhythmia after repeated pulmonary vein isolation (PVI) performed in the era of contact force without additional substrate ablation. One of the predictors studied, ablation index (AI), incorporates power, contact force, and time in a weighted formula and is reported to predict lesion size in animals. Design. Consecutive patients (n = 108) undergoing repeat PVI without additional substrate modification using a contact force sensing catheter were included retrospectively at a tertiary center. All ablation points were analyzed offline. A new variable, normalized AI (AI corrected for the location of the lesion-anterior vs. posterior) was calculated. The patients were systematically followed with clinical visit and 12-lead ECG as well as review of the regional electronic patient files at 3 and 12 months after the procedure with 5-day Holter at 12 months. Results. Electrical reconnection to at least one pulmonary vein (PV) was seen in 97% of the patients. The recurrence rate was 35%. There was no recurrence in patients with nAI above 1.15 (n = 26). Patients with electrical reconnection of up to two PVs had a higher risk of recurrence compared with patients having electrical reconnection of three or four PVs (p = .003), and this risk was especially high in patients with persistent atrial fibrillation (69 [39-91]%). Conclusions. The risk of recurrence is higher in patients with ablations performed with low levels of AI and in patients with reconnection to up to two PVs. Our data may indicate the need for higher target levels of AI during repeat PVI than normally used during de-novo PVI.

Depression and Uptake of Oral Anticoagulation Therapy in Patients With Atrial Fibrillation: A Danish Nationwide Cohort Study.

BACKGROUND: Oral anticoagulation therapy (OAT) in patients with atrial fibrillation (AF) is a highly important preventive intervention, perhaps especially in those with comorbid depression, who have a worse prognosis. However, OAT may pose particular challenges in depressed patients. OBJECTIVES: To assess whether AF patients with depression have lower OAT uptake. METHODS: This nationwide register-based 2005-2016 cohort study of all Danes with AF and OAT indication (CHA2DS2VASc stroke risk score ≥2) assessed OAT initiation within 90 days in those with incident AF (N=147,162) and OAT prevalence in those with prevalent AF (N=192,656). The associations of depression with both outcomes were estimated in regression analyses with successive adjustment for socioeconomic characteristics and somatic and psychiatric comorbidity. RESULTS: Comorbid depression was significantly associated with lower frequency of OAT initiation in incident AF patients {adjusted proportion differences (aPDs): -6.6% [95% confidence interval (CI), -7.4 to -5.9]} and lower prevalence of OAT [aPD: -4.2% (95% CI, -4.7 to -3.8)] in prevalent AF patients. Yet, the OAT uptake increased substantially during the period, particularly in depressed patients [aPD for OAT prevalence in 2016: -0.8% (95% CI, -1.6 to -0.0)]. CONCLUSIONS: Comorbid depression was associated with a significantly lower OAT uptake in patients with AF, which questions whether depressed patients receive sufficient support to manage this consequential cardiac condition. However, a substantial increase in the overall OAT uptake and a decrease of the depression-associated deficit in OAT were seen over the period during which OAT was developed through the introduction of new oral anticoagulation therapy.

Cumulative risk regression in case-cohort studies using pseudo-observations.

Case-cohort studies are useful when information on certain risk factors is difficult or costly to ascertain. Particularly, a case-cohort study may be well suited in situations where several case series are of interest, e.g. in studies with competing risks, because the same sub-cohort may serve as a comparison group for all case series. Previous analyses of this kind of sampled cohort data most often involved estimation of rate ratios based on a Cox regression model. However, with competing risks this method will not provide parameters that directly describe the association between covariates and cumulative risks. In this paper, we study regression analysis of cause-specific cumulative risks in case-cohort studies using pseudo-observations. We focus mainly on the situation with competing risks. However, as a by-product, we also develop a method by which absolute mortality risks may be analyzed directly from case-cohort survival data. We adjust for the case-cohort sampling by inverse sampling probabilities applied to a generalized estimation equation. The large-sample properties of the proposed estimator are developed and small-sample properties are evaluated in a simulation study. We apply the methodology to study the effect of a specific diet component and a specific gene on the absolute risk of atrial fibrillation.

Maternal Smoking During Pregnancy and Timing of Puberty in Sons and Daughters: A Population-Based Cohort Study.

Because early puberty has been linked to diseases later in life, identification of modifiable causes of early puberty is of interest. We explored the possible associations between maternal smoking during pregnancy and pubertal development in sons and daughters. Between 2012 and 2017, 15,819 children from the Danish National Birth Cohort, born during 2000-2003, provided half-yearly information on puberty from the age of 11 years. We estimated adjusted age differences (in months) at attaining various pubertal milestones, including Tanner stages, per 10 daily cigarettes smoked in the first trimester of gestation. In sons, exposure to smoking in utero was associated with earlier genital development (Tanner 2, -1.3 months, 95% confidence interval (CI): -2.5, 0.0; Tanner 5, -3.7 months, 95% CI: -5.3, -2.0), pubic hair development (Tanner 2, -1.8 months, 95% CI: -2.9, -0.6; Tanner 5, -2.9 months, 95% CI: -4.2, -1.7), and voice break (-2.4 months, 95% CI: -3.6, -1.3). In daughters, maternal smoking was associated with earlier breast development (Tanner 2, -3.4 months, 95% CI: -5.3, -1.5; Tanner 5, -4.7 months, 95% CI: -6.5, -2.9), pubic hair development stages 3-5 (Tanner 5, -2.5 months, 95% CI: -4.1, -1.0), and menarche (-3.1 months, 95% CI: -4.0, -2.3). Fetal exposure to tobacco smoke might advance timing of puberty in boys and girls.

Acetaminophen (Paracetamol) Exposure During Pregnancy and Pubertal Development in Boys and Girls From a Nationwide Puberty Cohort.

This study explored the association between exposure to acetaminophen during pregnancy and pubertal development using data from 15,822 boys and girls in the longitudinal Puberty Cohort, nested within the Danish National Birth Cohort. Use of acetaminophen was reported 3 times during pregnancy and 6 months postpartum. In total, 54% of mothers indicated use at least once during pregnancy. Between 2012 and 2017, sons and daughters provided information on a wide range of pubertal milestones-including Tanner stages, axillary hair growth, and age at menarche or voice break and first ejaculation-every 6 months from 11 years of age until full sexual maturation. Data were analyzed using a regression model for interval-censored data, providing adjusted mean monthly differences in age at attaining the pubertal milestones according to intrauterine cumulative (weeks) and trimester-specific acetaminophen exposure. Our results suggested a tendency towards slightly earlier attainment of almost all studied markers of female pubertal development with increasing number of weeks of exposure (i.e., about 1.5-3 months earlier age at pubic hair, axillary hair, and acne development comparing unexposed with those prenatally exposed for more than 12 weeks). Male pubertal development had no strong association with acetaminophen exposure.

Birth seasonality and risk of autism spectrum disorder.

Season of birth has been hypothesized to be a risk factor for autism spectrum disorder (ASD). However, the evidence has been mixed and limited due to methodological challenges. We examine ASD birth trends for 5,464,628 births across 5 countries. ASD birth prevalence data were obtained from the International Collaboration for Autism Registry Epidemiology database, including children born in Denmark, Finland, Norway, Sweden, and Western Australia. Empirical mode decomposition and cosinor modeling were used to assess seasonality. We show seasonal variation in ASD births for the countries of Finland and Sweden. There was a modest increase in risk for children born in the fall and a modest decrease in risk for children born in the spring. Solar radiation levels around conception and the postnatal period were inversely correlated with seasonal trends in ASD risk. In the first multinational study of birth seasonality of ASD, there was evidence supporting the presence of seasonal trends in Finland and Sweden. The observations that risk was highest for fall births (i.e., conceived in the winter) and lowest for spring births (i.e., conceived in the summer), and sunlight levels during critical neurodevelopmental periods explained much of the seasonal trends, are consistent with the hypothesis that a seasonally fluctuating risk factor may influence risk of ASD.

Apgar score and risk of autism.

Low Apgar score has been associated with higher risk for several neurological and psychiatric disorders, including cerebral palsy and intellectual disability. Studies of the association between Apgar score and autism spectrum disorder (ASD) have been inconsistent. We aimed to investigate (1) the association between low Apgar score at 5 min and risk for ASD, and (2) the modifying effects of gestational age and sex on this association in the largest multinational database of ASD. We included prospective data from 5.5 million individuals and over 33,000 cases of ASD from Norway, Sweden, Denmark and Western Australia who were born between 1984 and 2007. We calculated crude and adjusted risk ratios (RR) with 95% confidence intervals (95% CIs) for the associations between low Apgar score and ASD. All analyses for ASD were repeated for autistic disorder (AD). We used interaction terms and stratified analysis to investigate the effects of sex, gestational age, and birth weight on the association. In fully adjusted models, low Apgar scores (1-3) (RR, 1.42; 95% CI, 1.16-1.74), and intermediate Apgar scores (4-6) (RR, 1.50; 95% CI, 1.36-1.65) were associated with a higher RR of ASD than optimal Apgar score (7-10). The point estimates for low (RR, 1.88; 95% CI, 1.41-2.51) and intermediate Apgar score (RR, 1.54; 95% CI, 1.32-1.81) were larger for AD than for ASD. This study suggests that low Apgar score is associated with higher risk of ASD, and in particular AD. We did not observe any major modifying effects of gestational age and sex, although there seems to be substantial confounding by gestational age and birth weight on the observed association.

Hypertensive Disorders of Pregnancy and Genital Anomalies in Boys: A Danish Nationwide Cohort Study.

BACKGROUND: Although congenital abnormalities in the male reproductive tract are common, their causes remain poorly understood. We studied associations between hypertensive disorders of pregnancy (pregestational hypertension, gestational hypertension, and preeclampsia) and the genital anomalies, cryptorchidism (undescended testes), and hypospadias (ventrally displaced urethral meatus). METHODS: We established a population of 1,073,026 Danish boys born alive between 1 January 1978 and 31 December 2012. By means of Cox regression analyses, we estimated hazard ratios with 95% confidence intervals for cryptorchidism and hypospadias according to type and severity of hypertensive disorder. Further, we used restricted cubic spline analyses to investigate the association between gestational age at onset of severe and moderate preeclampsia and the two genital anomalies. RESULTS: We found associations between pregestational hypertension and cryptorchidism (HR: 1.3; 95% CI = 1.1, 1.6) and hypospadias (HR: 1.7; 95% CI = 1.3, 2.3), whereas gestational hypertension was only associated with cryptorchidism (HR: 1.2; 95% CI = 1.1, 1.4). Boys of mothers with preeclampsia had the highest occurrence of cryptorchidism and hypospadias, increasing with preeclampsia severity. Women with HELLP syndrome faced the highest risk of having a child with both cryptorchidism (HR: 2.1; 95% CI = 1.4, 3.2) and hypospadias (HR: 3.9; 95% CI = 2.5, 6.1). Further, the occurrence increased with early onset of preeclampsia diagnosis. CONCLUSIONS: These findings support the hypotheses that preeclampsia and genital anomalies share common etiologic factors and that placental dysfunction and androgen deficiency in early pregnancy are important in the etiology of male genital anomalies.

EARLY REGULATION IN CHILDREN WHO ARE LATER DIAGNOSED WITH AUTISM SPECTRUM DISORDER. A LONGITUDINAL STUDY WITHIN THE DANISH NATIONAL BIRTH COHORT.

Studies have shown that children later diagnosed with autism spectrum disorders (ASD) in their first years of life might show symptoms in main developmental areas and that these signs might be sensed by the parents. The present study investigated in a large birth cohort if children later diagnosed with ASD had deviations at 6 and 18 months in areas such as the ability to self-regulate emotions, feeding, and sleeping. The study was based on prospective information collected from 76,322 mothers who participated in the Danish National Birth Cohort. When the children reached an average age of 11 years, 973 children with ASD and a control group of 300 children with intellectual disability (IDnoASD) were identified via Danish health registries. Associations were found between short periods of breast-feeding and the children later diagnosed with ASD and IDnoASD as well as associations at 18 months to deviations in regulation of emotions and activity. The similarities in these associations emphasize how difficult it is to distinguish between diagnoses early in life.

Causal inference in survival analysis using pseudo-observations.

Causal inference for non-censored response variables, such as binary or quantitative outcomes, is often based on either (1) direct standardization ('G-formula') or (2) inverse probability of treatment assignment weights ('propensity score'). To do causal inference in survival analysis, one needs to address right-censoring, and often, special techniques are required for that purpose. We will show how censoring can be dealt with 'once and for all' by means of so-called pseudo-observations when doing causal inference in survival analysis. The pseudo-observations can be used as a replacement of the outcomes without censoring when applying 'standard' causal inference methods, such as (1) or (2) earlier. We study this idea for estimating the average causal effect of a binary treatment on the survival probability, the restricted mean lifetime, and the cumulative incidence in a competing risks situation. The methods will be illustrated in a small simulation study and via a study of patients with acute myeloid leukemia who received either myeloablative or non-myeloablative conditioning before allogeneic hematopoetic cell transplantation. We will estimate the average causal effect of the conditioning regime on outcomes such as the 3-year overall survival probability and the 3-year risk of chronic graft-versus-host disease. Copyright © 2017 John Wiley & Sons, Ltd.

Estimating a population cumulative incidence under calendar time trends.

BACKGROUND: The risk of a disease or psychiatric disorder is frequently measured by the age-specific cumulative incidence. Cumulative incidence estimates are often derived in cohort studies with individuals recruited over calendar time and with the end of follow-up governed by a specific date. It is common practice to apply the Kaplan-Meier or Aalen-Johansen estimator to the total sample and report either the estimated cumulative incidence curve or just a single point on the curve as a description of the disease risk. METHODS: We argue that, whenever the disease or disorder of interest is influenced by calendar time trends, the total sample Kaplan-Meier and Aalen-Johansen estimators do not provide useful estimates of the general risk in the target population. We present some alternatives to this type of analysis. RESULTS: We show how a proportional hazards model may be used to extrapolate disease risk estimates if proportionality is a reasonable assumption. If not reasonable, we instead advocate that a more useful description of the disease risk lies in the age-specific cumulative incidence curves across strata given by time of entry or perhaps just the end of follow-up estimates across all strata. Finally, we argue that a weighted average of these end of follow-up estimates may be a useful summary measure of the disease risk within the study period. CONCLUSIONS: Time trends in a disease risk will render total sample estimators less useful in observational studies with staggered entry and administrative censoring. An analysis based on proportional hazards or a stratified analysis may be better alternatives.

Risk of epilepsy and autism in full and half siblings-A population-based cohort study.

OBJECTIVE: Epilepsy and autism spectrum disorder (ASD) often occur together in the same individual. However, it remains unknown whether siblings of children with ASD have an increased risk of epilepsy and vice versa. This study determines the risk of ASD and epilepsy among younger siblings of children with ASD and epilepsy. DESIGN: The study included all children born in Denmark between January 1, 1980 and 31 December 2006 who participated in follow-up until December 31, 2012 (1,663,302 children). We used Cox regression to calculate the adjusted hazard ratio (aHR) and the Kaplan-Meier method to calculate the cumulative incidence. RESULTS: The overall aHR of epilepsy in younger siblings increased by 70% (aHR 1.70, 95% confidence interval [CI] 1.34-2.16%) if the older sibling had ASD compared with siblings where the older sibling did not have ASD. The cumulative incidence of epilepsy at 20 years of age was 2.54% (95% CI 1.97-3.26%) if the older sibling had ASD, whereas the cumulative incidence of epilepsy at 20 years of age was 1.63% (95% CI 1.60-1.66%) if the older sibling did not have ASD. The overall aHR of ASD in younger siblings increased by 54% if the older sibling had epilepsy (aHR 1.54, 95% CI 1.32-1.80) compared with siblings where the older sibling did not have epilepsy. The cumulative incidence of ASD at 20 years of age was 2.06% (95% CI 1.84-2.32%) if the older sibling had epilepsy, whereas the cumulative incidence of ASD at 20 years of age was 1.27% (95% CI 1.25-1.29%) if the older sibling did not have epilepsy. SIGNIFICANCE: The cross-disorder sibling risk of epilepsy and ASD was increased for the two disorders, which suggests that genes or environmental factors shared by family members may play a causal role in the co-occurrence of ASD and epilepsy.

Instrumental variable method for time-to-event data using a pseudo-observation approach.

Observational studies are often in peril of unmeasured confounding. Instrumental variable analysis is a method for controlling for unmeasured confounding. As yet, theory on instrumental variable analysis of censored time-to-event data is scarce. We propose a pseudo-observation approach to instrumental variable analysis of the survival function, the restricted mean, and the cumulative incidence function in competing risks with right-censored data using generalized method of moments estimation. For the purpose of illustrating our proposed method, we study antidepressant exposure in pregnancy and risk of autism spectrum disorder in offspring, and the performance of the method is assessed through simulation studies.

Early development in children that are later diagnosed with disorders of attention and activity: a longitudinal study in the Danish National Birth Cohort.

Not much is known about the early development in children that are later diagnosed with disorders of attention and activity (ADHD). Using prospective information collected from mothers in the Danish National Birth Cohort (DNBC), we investigated if developmental deviations in the first years of life are associated with later ADHD. In the DNBC 76,286 mothers were interviewed about their child's development and behaviour at age 6 and 18 months. At the end of follow-up, when the children were 8-14 years of age, 2034 were registered in Danish health registers with a clinical diagnosis of ADHD. The Hazard Ratio of ADHD was estimated using Cox regression model. At 6 months of age deviations in development showed associations with the child later being diagnosed with ADHD such as duration of breastfeeding, motor functioning, and incessant crying. At 18 months, many observations clearly associated with ADHD as for example the child not being able to fetch things on request [HR 3.0 (95 % CI 2.4; 3.7)], or the child being significantly more active than average [HR 2.0 (95 % CI 1.8; 2.2)]. An association to ADHD was shown, especially at 18 months, if the mother found it difficult to handle the child [HR 2.9 (95 % CI 2.4-3.5)]. However, it goes for all observations that the positive predictive values were low. Many children with ADHD showed signs of developmental deviations during the first years of their life. In general, however, ADHD cannot be identified solely on basis of the questions in DNBC.