Handling missing data in matched case-control studies using multiple imputation

Prediction of suicidal thoughts and behaviors based on the diurnal cortisol pattern and THC dosage in continued cannabis users, a 5 year population-based matched cohort study

It appears that the THC dosage is the link between dysregulation of the hypothalamic pituitary adrenal (HPA) axis and suicidal thoughts and behaviors (STB). We proposed a new model to understand the underlying pathophysiological mechanism of STB based on the interaction of cortisol and THC dosage. From September 1, 2019, to January 1, 2024, we conducted a population-based, matched-pair, nested case-control study resulting from a three-wave complete longitudinal, multicenter cohort study on a sample of congress 60 clients. A total of 368 male continued cannabis users (CCu) were allocated to four categories, including low, moderate and high THC dosages and relapse, using optimal matching. Several HPA axis measures were analyzed in the saliva using liquid chromatography with tandem mass spectrometry (LC-MS-MS), and carboxylic acids levels in the urine were assessed via gas chromatography/mass spectrometry (GC-MS). We used structural equation modeling (SEM) to examine the relationship between the variables of interest and the model fit test, and used the Akaike information criterion (AIC) to compare the model fit and select the best-fitting model. Population attributable fractions (PAFs) and cumulative risk score were also calculated for the best-fitting pattern. The analysis showed that the likelihood of STB in individuals with a cortisol awakening response (CAR) and a blunted diurnal cortisol slope (DCS) and higher area under the curve (AUC) who reported heavy cannabis use was more than three times higher than the control group (OR 3.2, 95 % CI 2.4-4.1). These findings indicate the importance of the specific cortisol secretion pattern in the increased clinical expression of STB and may be an important factor for guiding preventive efforts in this area.

The contribution of childhood maltreatment to the incidence of heavy cannabis use in Iran (IRNS-CCI): A multicenter, matched-pairs, nested, case-control study

BACKGROUND

Previous studies have shown the role of the interaction between the endocannabinoid system (ECS) and life's adversities in the formation of addiction, including alcohol abuse.

OBJECTIVE

Our objective was to identify childhood maltreatment (CM) patterns with the strongest impact on the probability of heavy cannabis use (THCCOOH concentrations ≥150 ng/mL) in Iran.

PARTICIPANTS AND SETTING

Using survivor sampling, 350 adult participants were selected, and they were then allocated to three categories based on an optimal algorithm: 1) Sexual abuse, 2) Physical abuse, and 3) Physical neglect.

METHODS

From 1 September 2019 to 1 May 2023, we implemented a multicenter, matched-pairs, nested, case-control study based on the wave 3-wave 6 data of a longitudinal, multicenter, cohort study. The cases and controls (n = 350 men) were defined according to the severity of CM. The THC potency was evaluated with the delta-9 carboxy tetrahydrocannabinol (THC-COOH) levels in urine using gas chromatography/mass spectrometry (GC/MS). We calculated the population attributable fractions (PAFs) to identify the patterns of maltreatment associated with the highest odds of high-potency cannabis use.

RESULTS

Accumulating CM, including sexual abuse, physical abuse, and physical neglect, carried more than three times the risk of heavy cannabis use (OR 3.4 95 % CI 2.9-4.1), and the combination of the three indicators of maltreatment and a high BMI (25-29.9) carried more than four times the risk of heavy cannabis use (OR 4.7 95 % CI 2.7-4.1) compared to the controls. We estimated that in the case of zero CM for each of the three indicators, over 20 % of new cases of heavy cannabis use can be prevented.

CONCLUSIONS

The findings show the significance of CM as a predicator of heavy cannabis use in adulthood and in the abstinence phase.

Exogenous female sex steroid hormones and new-onset asthma in women: a matched case-control study

BACKGROUND

Evidence on the role of exogenous female sex steroid hormones in asthma development in women remains conflicting. We sought to quantify the potential causal role of hormonal contraceptives and menopausal hormone therapy (MHT) in the development of asthma in women.

METHODS

We conducted a matched case-control study based on the West Sweden Asthma Study, nested in a representative cohort of 15,003 women aged 16-75 years, with 8-year follow-up (2008-2016). Data were analyzed using Frequentist and Bayesian conditional logistic regression models.

RESULTS

We included 114 cases and 717 controls. In Frequentist analysis, the odds ratio (OR) for new-onset asthma with ever use of hormonal contraceptives was 2.13 (95% confidence interval [CI] 1.03-4.38). Subgroup analyses showed that the OR increased consistently with older baseline age. The OR for new-onset asthma with ever MHT use among menopausal women was 1.17 (95% CI 0.49-2.82). In Bayesian analysis, the ORs for ever use of hormonal contraceptives and MHT were, respectively, 1.11 (95% posterior interval [PI] 0.79-1.55) and 1.18 (95% PI 0.92-1.52). The respective probability of each OR being larger than 1 was 72.3% and 90.6%.

CONCLUSIONS

Although use of hormonal contraceptives was associated with an increased risk of asthma, this may be explained by selection of women by baseline asthma status, given the upward trend in the effect estimate with older age. This indicates that use of hormonal contraceptives may in fact decrease asthma risk in women. Use of MHT may increase asthma risk in menopausal women.

Proportion of suicides in Denmark attributable to bereavement by the suicide of a first-degree relative or partner: Nested case-control study

OBJECTIVE

To provide the first estimates of the risk of suicide after bereavement by the suicide of any first-degree relative and the proportion of suicides in Denmark attributable to suicide bereavement.

METHODS

We conducted a nationwide nested case-control study defining cases as all Danish-born individuals who died by suicide in Denmark between 01 January 1980 and 31 December 2016 (n = 32,248), age-matched to four living controls. Using three exposure categories (bereavement by the suicide of a relative [parent, offspring, sibling, and spouse/cohabitee]; non-suicide bereavement; no bereavement) and conditional logistic regression adjusted for pre-specified covariates we estimated the odds of exposure to suicide bereavement in cases versus controls. We tested whether associations differed for men and women, estimated the population attributable fraction (PAF) of suicides in our population at risk that could be attributed to a first-degree relative's suicide loss, and estimated the attributable fraction among the exposed (AFe).

RESULTS

Suicide bereavement was associated with an increased odds of suicide when compared with no bereavement (OR_adj2  = 2.90, 95% CI: 2.46-3.40) or non-suicide bereavement (OR_adj2  = 1.48, 95% CI: 1.25-1.74). There was no evidence to support any interaction with sex. PAF (0.69%; 95% CI: 0.62%-0.77%) and AFe (60.12%; 95% CI: 53.19%-66.03%) estimates suggested that in Denmark 0.69% of suicides, and 60% of suicides among suicide-bereaved relatives, could be prevented if it was possible to address all factors increasing suicide risk in suicide-bereaved relatives.

CONCLUSION

Suicide bereavement in relatives and partners contributes to at least one in 145 suicides in Denmark.

Modifiable (Sleeping Pattern and Stress) and Non-Modifiable Risk Factors Associated with Breast Cancer: A Matched Case-Control Study in Delhi, India

BACKGROUND

An utmost increase of breast cancer burden during the last several decades was reported in Asian countries. Findings from literature confirm that risk factors of breast cancers can be modifiable and non-modifiable in nature.

OBJECTIVE

The present study is designed to identify specific modifiable and non-modifiable risk factors associated with breast cancer.

METHODS

A matched case-control study was conducted considering 187 cases as women diagnosed with breast cancer and 187 hospital-controls as women without having breast cancer visiting the hospital. Other than standard risk factors, stress is measured using Perceived Stress Scale (PSS) and stress is measured using Pittsburgh Sleep Quality Index (PSQI). Several modifiable and non-modifiable risk factors were assessed using conditional logistic regression to find out significant association with breast cancer.

RESULTS

Regular multi-vitamin uptake (OR = 3.38; 95%CI = 1.69 - 6.77; p-value = 0.001), poor sleep (OR = 11.29; 95%CI = 4.36 - 29.25; p-value < 0.001), irregular sleep (OR = 34.11; 95%CI = 10.03 - 115.92; p-value < 0.001) and severe stress (OR = 6.74; 95%CI = 3.06 - 14.81; p-value < 0.001) were found to be the highest odds ratio among all modifiable risk factor of breast cancer. Also, age at first childbirth less than 30 years (OR = 0.44; 95%CI = 0.25 - 0.78; p-value = 0.005) was found protective against breast cancer.

CONCLUSION

In our study, stress, sleeping pattern, and regular multi-vitamin uptake were found to be significant modifiable risk factors of breast cancer. None of the non-modifiable risk factors were found to be significantly associated with the risk of breast cancer.

Risk factors of hepatocellular carcinoma in type 2 diabetes patients: A two-centre study in a developing country

Type 2 diabetes mellitus (T2DM) is increasingly known as a risk factor of hepatocellular carcinoma (HCC). In this study, we determined the risk factors associated with HCC in T2DM patients. This was a matched case-control study conducted at two hepatobiliary referral centres in a developing country. Patients' sociodemographic, clinical, and biochemical characteristics between 1 January 2012 and 30 June 2018 were extracted from the electronic medical records and analysed using multivariate logistic regression analysis. A total of 212 case-control pairs were included. Significant risk factors included Chinese and Malay ethnicities that interacted with viral hepatitis (adjusted odds ratio [AOR] = 11.77, 95% confidence interval [CI]: 1.39-99.79) and (AOR = 37.94, 95% CI: 3.92-367.61) respectively, weight loss (AOR = 5.28, 95% CI: 2.29-12.19), abdominal pain/ discomfort (AOR = 6.73, 95% CI: 3.34-13.34), alcohol (AOR = 4.08, 95% CI: 1.81-9.22), fatty liver (AOR = 3.29, 95% CI: 1.40-7.76), low platelet (AOR = 4.03, 95% CI:1.90-8.55), raised alanine transaminase (AOR = 2.11, 95% CI: 1.16-3.86). and alkaline phosphatase (ALP) levels (AOR = 2.17, 95% CI: 1.17-4.00). Statins reduced the risk of HCC by 63% (AOR = 0.37, 95% CI: 0.21-0.65). The identification of these factors aids the risk stratification for HCC among T2DM patients for early detection and decision-making in patient management in the primary care setting.

Associations of pre-pregnancy impaired fasting glucose and body mass index among pregnant women without pre-existing diabetes with offspring being large for gestational age and preterm birth: a cohort study in China

INTRODUCTION

Associations of pre-pregnancy impaired fasting glucose (IFG) and body mass index (BMI) with large for gestational age (LGA) and preterm birth (PTB) have been poorly understood. We aimed to investigate the associations of maternal BMI, separately and together with pre-pregnancy IFG, with LGA and PTB in Chinese population. We also aimed to quantify these associations by maternal age.

RESEARCH DESIGN AND METHODS

This was a retrospective cohort study of women from the National Free Preconception Health Examination Project with singleton birth from 121 counties/districts in 21 cities of Guangdong Province, China, from January 1, 2013 to December 31, 2017. Women were included if they did not have pre-existing chronic diseases (diabetes, hypertension, etc). Participants were divided into eight groups according to their BMI (underweight (BMI <18.5 kg/m²), normal weight (18.5-23.9 kg/m²), overweight (24.0-27.9 kg/m²), and obesity (≥28.0 kg/m²)) and pre-pregnancy fasting glucose status (normoglycemia (fasting glucose concentration <6.1 mmol/L) and IFG (6.1-7.0 mmol/L)). Adjusted incidence risk ratios (aIRRs) and 95% CIs of LGA, severe LGA, PTB and early PTB were estimated.

RESULTS

We included 634 030 women. The incidences of LGA, severe LGA, PTB and early PTB for the study population were 7.1%, 2.5%, 5.1% and 1.1%, respectively. Compared with normal weight mothers with normoglycemia, overweight and obese mothers irrespective of IFG had a higher risk of LGA (eg, obesity with IFG aIRR 1.85 (1.60-2.14)) and severe LGA (eg, obesity with IFG 2.19 (1.73-2.79)). The associations of BMI and pre-pregnancy fasting glucose status with LGA were similar found among women of all age groups. Underweight with normoglycemia had 6.0% higher risk of PTB (1.06 (1.03-1.09)) and 8.0% higher risk of early PTB (1.08 (1.02-1.17)), underweight with IFG had 14.0% higher risk of PTB (1.14 (1.02-1.27)), and obese mothers with IFG had 45.0% higher risk of PTB (1.45 (1.18-1.78)). The associations of BMI and pre-pregnancy fasting glucose status with PTB differed by maternal age.

CONCLUSION

Overweight and obesity regardless of IFG were associated with an increased risk of LGA, and these associations were similarly observed among mothers of all age groups. Underweight regardless of IFG, and obesity with IFG were associated with an increased risk of PTB, but the associations differed by maternal age. Findings from this study may have implications for risk assessment and counselling before pregnancy.

A low-cost laboratory-based method for predicting newly diagnosed biopsy-proven diabetic nephropathy in people with type 2 diabetes

AIMS

To identify significant prognostic factors for newly diagnosed biopsy-proven diabetic nephropathy using routine laboratory measures, from which to derive a low-cost explanatory model, and to use this model to examine associations between the potential low-cost test panels and the risk of diabetic nephropathy in people with type 2 diabetes with normal kidney function.

METHOD

A population-based case-control study was undertaken to test the association between diabetic nephropathy and 47 laboratory variables using a 'hypothesis-free' strategy and five routinely recorded factors in diabetes care (BMI, systolic and diastolic blood pressure, HbA_1c , fasting glucose). Factors that were significant after Bonferroni correction were included in different test panels and used to develop diabetic nephropathy (outcome) explanatory models. Models were derived using risk-set sampling among 950 biopsy-proven diabetic nephropathy cases newly diagnosed in the period between 2012 and 2018 and among 4750 age- and gender-matched controls.

RESULTS

A total of 15 Bonferroni-corrected significant laboratory predictors in the three test panels (blood cell, serum electrolytes and blood coagulation) were identified through multivariable analysis and used to develop the three explanatory models. The optimism-adjusted C-statistics and calibration slope were 0.725 (95% CI 0.723-0.728) and 0.978 (95% CI 0.912-0.999) for the blood cell model, 0.688 (95% CI 0.686-0.690) and 0.923 (95% CI 0.706-0.977) for the serum electrolytes model, 0.648 (95% CI 0.639-0.658) and 0.914 (95% CI 0.641-1.115) for the blood coagulation model, respectively.

CONCLUSIONS

A total of 15 predictors were significantly associated with newly diagnosed biopsy-proven diabetic nephropathy in type 2 diabetes. The blood cell model appeared to be the low-cost model with the best predictive ability.

GDF-15 is associated with sudden cardiac death due to incident myocardial infarction

AIMS

Preventing sudden cardiac death (SCD) due to acute myocardial infarction (MI) in previously healthy patients is challenging. Proteomic analysis may lead to an understanding of biological mechanisms and provide predictive biomarkers.

METHODS

In this prospective, nested case-control study from northern Sweden, 87 candidate cardiovascular protein biomarkers were studied in 244 individuals who later died within 24 h from an incident MI and 244 referents without MI and individually matched for age, sex and date of health examination and alive at the date of event in the index person. Association analysis was conducted using conditional logistic regression. Bonferroni correction was applied to avoid false positive findings.

RESULTS

Ten proteins were associated with future SCD due to acute MI in the non-adjusted analysis. The strongest association were found for growth differentiation factor 15 (GDF-15) with an odds ratio (OR) of 1.79 (95% confidence interval [CI] 1.41, 2.25) per standard deviation increase in protein, and urokinase-type plasminogen activator receptor with an OR of 1.66 (95% CI 1.34, 2.06). In models adjusted for lipid levels, body mass index, education, smoking, hypertension and C-reactive protein, only association with GDF-15 remained (OR 1.47 (95% 1.11, 1.95)).

CONCLUSION

Elevated levels of GDF-15 are associated with increased risk of SCD within 24 h of incident MI. Further research may enable the use of GDF-15 together with other clinical and biological markers to guide primary preventive interventions for individuals at high risk for SCD.

Association between anthropometric indices of obesity and risk of cardiovascular disease in Japanese men

Objectives

We aimed to compare the association of body mass index (BMI), waist circumference (WC), and waist‐to‐height ratio (WHtR) with risk of cardiovascular disease (CVD) among middle‐aged working Japanese men.

Methods

A nested case‐control study was performed among middle‐aged male employees who underwent periodic health checkup. A total of 241 CVD cases were identified and matched individually on age, gender, and worksite with 1205 controls. Data on BMI, WC, WHtR, smoking, hypertension, diabetes, and dyslipidemia collected at 4 years before the event/index date were retrieved. Associations between BMI, WC, WHtR, and CVD risk were assessed by using conditional logistic regression models.

Results

The strength of the association of BMI, WC, and WHtR with CVD risk was similar. The smoking‐adjusted odds ratio (95% confidence interval) for CVD was 1.60 (1.38‐1.85), 1.53 (1.33‐1.78), and 1.56 (1.35‐1.81) for a 1 SD unit increase in BMI, WC, and WHtR respectively. After further adjustment for hypertension, diabetes, and dyslipidemia, these associations were attenuated but remained statistically significant.

Conclusions

Measures of general (BMI) and abdominal (WC and WHtR) obesity were similarly associated with CVD in middle‐aged Japanese men.

Patient-reported outcomes in hip resurfacing versus conventional total hip arthroplasty: a register-based matched cohort study of 726 patients

Background and purpose - The theoretical mechanical advantages of metal-on-metal hip resurfacing (MoM-HR) compared with conventional total hip arthroplasty (THA) have been questioned. Studies including measures of patient-reported function, physical activity, or health-related quality of life have been sparse. We compared patient-reported outcomes in MoM-HR patients with a matched group of patients with conventional THA at 7 years post-surgery. Patients and methods - Patients and patient data were retrieved from the Swedish Hip Arthroplasty Register. The case group, consisting of 363 patients with MoM-HR, was matched 1:1 with a control group, consisting of patients with a conventional THA. Patients were sent a postal patient-reported outcome measures (PROM) questionnaire including the Hip Disability and Osteoarthritis Outcome Score (HOOS), EQ-5D, and VAS pain. We used multivariable linear regression analyses to investigate the influence of prosthesis type. Results - 569 patients (78%) returned the questionnaire with complete responses (299 MoM-HRs and 270 conventional THAs). MoM-HR was associated with better scores in HOOS function of daily living (4 percentage units) and HOOS function in sport and recreation (8 percentage units) subscales. Type of prosthesis did not influence HOOS quality of life, HOOS pain, HOOS symptoms, EQ-5D index, hip pain, or satisfaction as measured with visual analog scales. Interpretation - At mean 7 years post-surgery, patients with hip resurfacing had somewhat better self-reported hip function than patients with conventional THA. The largest difference between groups was seen in the presumed most demanding subscale, i.e., function in sport and recreation.

A hybrid of whale optimization and late acceptance hill climbing based imputation to enhance classification performance in electronic health records

Electronic health records (EHR) are a major source of information in biomedical informatics. Yet, missing values are prominent characteristics of EHR. Prediction on dataset with missing values results in inaccurate inferences. Nearest neighbour imputation based on lazy learning approach is a proven technique for missing data imputation and is recognized as one among the top ten data mining algorithms due to its simplicity and understandability. But its performance is deteriorated due to the curse of dimensionality as unimportant features are likely to dominate. We address this problem by proposing a novel approach for feature weighting based on a hybrid of metaheuristic whale optimization algorithm (WOA) and local search late acceptance hill climbing algorithm (LAHCA) on nearest neighbour imputation method. Our proposed approach Metaheuristic and Local Search based Feature Weighted Nearest Neighbour Imputation (kNN+LAHCAWOA) also learns different k values for different test points. Our approach is tested on benchmark EHR datasets with three proven classifiers Support Vector Machines(SVM), Random forest(RF) and Deep neural networks(DNN). The results prove that kNN+LAHCAWOA is an effective imputation strategy and aids in improving the classification performance when compared with its competitor methods.

Association of Medication Intensity and Stages of Airflow Limitation With the Risk of Hospitalization or Death in Patients With Heart Failure and Chronic Obstructive Pulmonary Disease

IMPORTANCE

In heart failure (HF), chronic obstructive pulmonary disease (COPD) increases the risk of poor outcomes, but the effect of COPD severity is unknown. This information is important for early intervention tailored to the highest-risk groups.

OBJECTIVES

To determine the associations between COPD medication intensity or stage of airflow limitation and the risk of hospitalization or death in patients with HF.

DESIGN, SETTING, AND PARTICIPANTS

This UK population-based, nested case-control study with risk-set sampling used the Clinical Practice Research Datalink linked to Hospital Episode Statistics between January 1, 2002, to January 1, 2014. Participants included patients aged 40 years and older with a new diagnosis of HF in their family practice clinical record. Data analysis was conducted from 2017 to 2018.

EXPOSURES

In patients with HF, those with COPD were compared with those without it. International COPD (Global Initiative for Chronic Obstructive Lung Disease [GOLD]) guidelines were used to stratify patients with COPD by 7 medication intensity levels and 4 airflow limitation severity stages using automatically recorded prescriptions and routinely requested forced expiratory volume in 1 second (FEV1) data.

MAIN OUTCOMES AND MEASURES

First all-cause admission or all-cause death.

RESULTS

There were 50 114 patients with new HF (median age, 79 years [interquartile range, 71-85 years]; 46% women) during the study period. In patients with HF, COPD (18 478 [13.8%]) was significantly associated with increased mortality (adjusted odds ratio [AOR], 1.31; 95% CI, 1.26-1.36) and hospitalization (AOR, 1.33; 95% CI, 1.26-1.39). The 3 most severe medication intensity levels showed significantly increasing mortality associations from full inhaler therapy (AOR, 1.17; 95% CI, 1.06-1.29) to oral corticosteroids (AOR, 1.69; 95% CI, 1.57-1.81) to oxygen therapy (AOR, 2.82; 95% CI, 2.42-3.28). The respective estimates for hospitalization were AORs of 1.17 (95% CI, 1.03-1.33), 1.75 (95% CI, 1.59-1.92), and 2.84 (95% CI, 1.22-3.63). Availability of spirometry data was limited but showed that increasing airflow limitation was associated with increased risk of mortality, with the following AORs: FEV1 80% or more, 1.63 (95% CI, 1.42-1.87); FEV1 50% to 79%, 1.69 (95% CI, 1.56-1.83); FEV1 30% to 49%, 2.21 (95% CI, 2.01-2.42); FEV1 less than 30%, 2.93 (95% CI, 2.49-3.43). The strength of associations between FEV1 and hospitalization risk were similar among stages ranging from FEV1 80% or more (AOR, 1.48; 95% CI, 1.31-1.68) to FEV1 less than 30% (AOR, 1.73; 95% CI, 1.40-2.12).

CONCLUSIONS AND RELEVANCE

In the UK HF community setting, increasing COPD severity was associated with increasing risk of mortality and hospitalization. Prescribed COPD medication intensity and airflow limitation provide the basis for targeting high-risk groups.

Effectiveness of aggression replacement training in reducing criminal recidivism among convicted adult offenders

BACKGROUND

Aggression replacement training (ART) is a widely used cognitive behavioural intervention for reducing aggression-related recidivism among criminal offenders. Its effectiveness in reducing offending, however, remains uncertain.

AIM

To examine the effect of ART on adult offenders' criminal recidivism rates.

METHOD

We compared 1,124 convicted adult offenders who began ART in the Swedish Prison and Probation Services 2003-2009 with 3,372 offenders in the system at the same time who did not participate in ART. Linkage with nationwide, longitudinal registries allowed extensive propensity score-matched controlling for baseline differences (e.g. sociodemographics, criminal history, psychiatric morbidity, and substance misuse) between groups.

RESULTS

Intent-to-treat analyses suggested similar 1-year general reconviction rates (according to the National Crime Register) between the two groups (ART participants 50% [n = 465]: comparison participants 51% [n = 1,492]; hazard ratio [HR] = 0.97, 95% CI [0.88, 1.07]) and similar one-year violent recidivism (ART participants 19% [n = 174]: comparison participants 18% [n = 547]; HR = 1.02, 95% CI [0.89, 1.17]). For ART completers, findings suggested a marginal decrease in reconvictions for any recidivism, but not for violent recidivism specifically. Sensitivity analyses yielded similar results.

CONCLUSIONS AND IMPLICATIONS

Our findings add to the emerging literature suggesting no effect of ART on reoffending among adult offenders. Originally designed for adolescents, it may be that the programme should include components for more adult-specific needs. Further, although group differences in reoffending did not emerge, research with a wider range of outcomes may be worthwhile before abandoning this approach with offenders.

Proteomic Biomarkers for Incident Aortic Stenosis Requiring Valvular Replacement

Background:

Aortic valve stenosis (AS) is the most common indication for cardiac valve surgery; untreated AS is linked to high mortality. The etiological background of AS is unknown. Previous human studies were typically based on case-control studies. Biomarkers identified in prospective studies could lead to novel mechanistic insights.

Methods:

Within a large population survey with blood samples obtained at baseline, 334 patients were identified who later underwent surgery for AS (median age [interquartile range], 59.9 [10.4] years at survey and 68.3 [12.7] at surgery; 48% female). For each case, 2 matched referents were allocated. Plasma was analyzed with the multiplex proximity extension assay for screening of 92 cardiovascular candidate proteins. Conditional logistic regression models were used to assess associations between each protein and AS, with correction for multiple testing. A separate set of 106 additional cases with 212 matched referents was used in a validation study.

Results:

Six proteins (growth differentiation factor 15, galectin-4, von Willebrand factor, interleukin 17 receptor A, transferrin receptor protein 1, and proprotein convertase subtilisin/kexin type 9) were associated with case status in the discovery cohort; odds ratios ranged from 1.25 to 1.37 per SD increase in the protein signal. Adjusting the multivariable models for classical cardiovascular risk factors at baseline yielded similar results. Subanalyses of case-referent triplets (n=133) who showed no visible coronary artery disease at the time of surgery in the index person supported associations between AS and growth differentiation factor 15 (odds ratio, 1.40; 95% confidence interval, 1.10-1.78) and galectin-4 (odds ratio, 1.27; 95% confidence interval, 1.02-1.59), but these associations were attenuated after excluding individuals who donated blood samples within 5 years before surgery. In triplets (n=201), which included index individuals with concurrent coronary artery disease at the time of surgery, all 6 proteins were robustly associated with case status in all sensitivity analyses. In the validation study, the association of all but 1 (interleukin 17 receptor A) of these proteins were replicated in patients with AS with concurrent coronary artery disease but not in patients with AS without coronary artery disease.

Conclusions:

We provide evidence that 5 proteins were altered years before AS surgery and that the associations seem to be driven by concurrent atherosclerotic disease.

Multiple imputation of missing data in nested case-control and case-cohort studies

The nested case-control and case-cohort designs are two main approaches for carrying out a substudy within a prospective cohort. This article adapts multiple imputation (MI) methods for handling missing covariates in full-cohort studies for nested case-control and case-cohort studies. We consider data missing by design and data missing by chance. MI analyses that make use of full-cohort data and MI analyses based on substudy data only are described, alongside an intermediate approach in which the imputation uses full-cohort data but the analysis uses only the substudy. We describe adaptations to two imputation methods: the approximate method (MI-approx) of White and Royston (2009) and the “substantive model compatible” (MI-SMC) method of Bartlett et al. (2015). We also apply the “MI matched set” approach of Seaman and Keogh (2015) to nested case-control studies, which does not require any full-cohort information. The methods are investigated using simulation studies and all perform well when their assumptions hold. Substantial gains in efficiency can be made by imputing data missing by design using the full-cohort approach or by imputing data missing by chance in analyses using the substudy only. The intermediate approach brings greater gains in efficiency relative to the substudy approach and is more robust to imputation model misspecification than the full-cohort approach. The methods are illustrated using the ARIC Study cohort. Supplementary Materials provide R and Stata code.

Tryptophan metabolic profile in term and preterm breast milk: implications for health

Breast milk is the only source of the essential amino acid tryptophan (TRP) in breast-fed infants. Low levels of TRP could have implications for infant neurodevelopment. The objectives of the present study were to compare the relationship of TRP and its neuroactive pathway metabolites kynurenine (Kyn) and kynurenic acid (KynA) in preterm and term expressed breast milk (EBM) in the first 14 d following birth, and the relationship of TRP metabolism to maternal stress and immune status. A total of twenty-four mothers were recruited from Cork University Maternity Hospital: twelve term (>38 weeks) and twelve preterm (<35 weeks). EBM samples were collected on days 7 and 14. Free TRP, Kyn and KynA were measured using HPLC, total TRP using MS, cytokines using the Meso Scale Discovery (MSD) assay system, and cortisol using a cortisol ELISA kit. Although total TRP was higher in preterm EBM in comparison with term EBM (P < 0·05), free TRP levels were lower (P < 0·05). Kyn, KynA and the Kyn:TRP ratio increased significantly in term EBM from day 7 to day 14 (P < 0·05), but not in preterm EBM. TNF-α, IL-6 and IL-8 were higher in day 7 preterm and term EBM in comparison with day 14. There were no significant differences between term and preterm EBM cortisol levels. Increased availability of total TRP, lower levels of free TRP and alterations in the temporal dynamics of TRP metabolism in preterm compared with term EBM, coupled with higher EBM inflammatory markers on day 7, may have implications for the neurological development of exclusively breast-fed preterm infants.

Lung Cancer Risk and Residential Exposure to Air Pollution: A Korean Population-Based Case-Control Study

PURPOSE

To investigate the association between long-term exposure to ambient air pollution and lung cancer incidence in Koreans.

MATERIALS AND METHODS

This was a population-based case-control study covering 908 lung cancer patients and 908 controls selected from a random sample of people within each Korean province and matched according to age, sex, and smoking status. We developed land-use regression models to estimate annual residential exposure to particulate matter (PM₁₀) and nitrogen dioxide (NO₂) over a 20-year exposure period. Logistic regression was used to estimate odds ratios (ORs) and their corresponding 95% confidence intervals (CI).

RESULTS

Increases in lung cancer incidence (expressed as adjusted OR) were 1.09 (95% CI: 0.96-1.23) with a ten-unit increase in PM₁₀ (μg/m³) and 1.10 (95% CI: 1.00-1.22) with a ten-unit increase in NO₂ (ppb). Tendencies for stronger associations between air pollution and lung cancer incidence were noted among never smokers, among those with low fruit consumption, and among those with a higher education level. Air pollution was more strongly associated with squamous cell and small cell carcinomas than with adenocarcinoma of the lung.

CONCLUSION

This study provides evidence that PM10 and NO₂ contribute to lung cancer incidence in Korea.

Association Between Type 2 Diabetes and All-Cause Hospitalization and Mortality in the UK General Heart Failure Population: Stratification by Diabetic Glycemic Control and Medication Intensification

OBJECTIVES

This study sought to investigate in the general heart failure (HF) population, whether the associations between type 2 diabetes (T2D) and risk of hospitalization and death, are modified by changing glycemic or drug treatment intensity.

BACKGROUND

In the general HF population, T2D confers a higher risk of poor outcomes, but whether this risk is modified by the diabetes status is unknown.

METHODS

A nested case-control study in an incident HF database cohort (2002 to 2014) compared patients with T2D with those without for risk of all-cause first hospitalization and death. T2D was stratified by categories of glycosylated hemoglobin (HbA_1c) or drug treatments measured 6 months before hospitalization and 1 year before death and compared with the HF group without T2D.

RESULTS

In HF, T2D was associated with risk of first hospitalization (adjusted odds ratio [aOR]: 1.29; 95% confidence interval [CI]: 1.24 to 1.34) and mortality (aOR: 1.24; 95% CI: 1.29 to 1.40). Stratification of T2D by HbA_1c levels, compared with the reference HF group without T2D, showed U-shaped associations with both outcomes. Highest risk categories were HbA_1c >9.5% (hospitalization, aOR: 1.75; 95% CI: 1.52 to 2.02; mortality, aOR: 1.30; 95% CI: 1.24 to 1.47) and <5.5% (hospitalization, aOR: 1.42; 95% CI: 1.12 to 1.80; mortality, aOR: 1.29; 95% CI: 1.10 to 1.51, respectively). T2D group with change in HbA_1c of >1% decrease was associated with hospitalization (aOR: 1.33; 95% CI: 1.18 to 1.49) and mortality (aOR: 1.36; 95% CI: 1.24 to 1.48). T2D drug group associations with hospitalization were no medication (aOR: 1.12; 95% CI: 1.04 to 1.19), oral antihyperglycemic only (aOR: 1.34; 95% CI: 1.27 to 1.41), oral antihyperglycemic+insulin (aOR: 1.36; 95% CI: 1.21 to 1.52), and insulin only (aOR: 1.61; 95% CI: 1.43 to 1.81); and with mortality for the same drug groups were 1.31 (95% CI: 1.23 to 1.39), 1.16 (95% CI: 1.11 to 1.22), 1.19 (95% CI: 1.06 to 1.34), and 1.43 (95% CI: 1.31 to 1.57), respectively. The T2D group with reduced drug treatments were associated with hospitalization (aOR: 2.13; 95% CI: 1.68 to 2.69) and mortality (aOR: 2.09; 95% CI: 1.81 to 2.41).

CONCLUSIONS

In the general HF population, T2D stratified by glycemic control and drug treatments showed differential risk associations. Routine measures of dynamic diabetes status provide important prognostic indication of poor outcomes in HF.

The use of multiple imputation method for the validation of 24-h food recalls by part-time observation of dietary intake in school

External validation of food recall over 24 h in schoolchildren is often restricted to eating events in schools and is based on direct observation as the reference method. The aim of this study was to estimate the dietary intake out of school, and consequently the bias in such research design based on only part-time validated food recall, using multiple imputation (MI) conditioned on the information on child age, sex, BMI, family income, parental education and the school attended. The previous-day, web-based questionnaire WebCAAFE, structured as six meals/snacks and thirty-two foods/beverage, was answered by a sample of 7-11-year-old Brazilian schoolchildren (n 602) from five public schools. Food/beverage intake recalled by children was compared with the records provided by trained observers during school meals. Sensitivity analysis was performed with artificial data emulating those recalled by children on WebCAAFE in order to evaluate the impact of both differential and non-differential bias. Estimated bias was within ±30 % interval for 84·4 % of the thirty-two foods/beverages evaluated in WebCAAFE, and half of the latter reached statistical significance (P<0·05). Rarely (<3 %) consumed dietary items were often under-reported (fish/seafood, vegetable soup, cheese bread, French fries), whereas some of those most frequently reported (meat, bread/biscuits, fruits) showed large overestimation. Compared with the analysis restricted to fully validated data, MI reduced differential bias in sensitivity analysis but the bias still remained large in most cases. MI provided a suitable statistical framework for part-time validation design of dietary intake over six daily eating events.