Marginal Structural Models for Partial Exposure Regimes

Intensive care unit (ICU) patients are ell known to be highly susceptible for nosocomial (i.e. hospital-acquired) infections due to their poor health and many invasive therapeutic treatments. The effects of acquiring such infections in ICU on mortality are however ill understood. Our goal is to quantify these effects using data from the National Surveillance Study of Nosocomial Infections in Intensive Care Units (Belgium). This is a challenging problem because of the presence of time-dependent confounders (such as exposure to mechanical ventilation)which lie on the causal path from infection to mortality. Standard statistical analyses may be severely misleading in such settings and have shown contradicting results. While inverse probability weighting for marginal structural models can be used to accommodate time-dependent confounders, inference for the effect of ?ICU acquired infections on mortality under such models is further complicated (a) by the fact that marginal structural models infer the effect of acquiring infection on a given, fixed day ?in ICU?, which is not well defined when ICU discharge comes prior to that day; (b) by informative censoring of the survival time due to hospital discharge; and (c) by the instability of the inverse weighting estimation procedure. We accommodate these problems by developing inference under a new class of marginal structural models which describe the hazard of death for patients if, possibly contrary to fact, they stayed in the ICU for at least a given number of days s and acquired infection or not on that day. Using these models we estimate that, if patients stayed in the ICU for at least s days, the effect of acquiring infection on day s would be to multiply the subsequent hazard of death by 2.74 (95 per cent conservative CI 1.48; 5.09).

Maternal tobacco smoking and decreased leukocytes, including dendritic cells, in neonates

Maternal smoking in pregnancy is associated with respiratory diseases in the offspring, possibly due to prenatal influences on the developing immune system. We investigated whether maternal smoking in pregnancy was associated with cord blood leukocyte numbers, including precursor dendritic cells, adjusting for concomitant factors. In a prospective healthy birth cohort study, total leukocyte counts were reduced in neonates of smoking mothers [10.7 (8.4-13.0), n=14] compared with nonexposed infants [14.7 (13.7-15.7), n=74, p=0.002] [geometric mean cells x 10(3)/microL (95% confidence interval)]. All leukocyte subsets were decreased, most prominently segmented neutrophils [4.3 (2.8-5.7) versus 6.2 (5.5-6.8), p=0.021], lymphocytes [3.8 (2.9-4.8) versus 5.0 (4.5-5.6), p=0.036], and myeloid precursor dendritic cells [12.7 cells/microL (9.1-17.8) versus 18.3 (15.8-21.2), p=0.055]. These differences persisted after adjustment for possible confounders. Predictors of myeloid precursor dendritic cell numbers in multivariable models were maternal smoking (-5.1 cells/microL, p=0.042), age (-0.5 cells/microL/y, p=0.035), and, marginally, asthma (+8.1 cells/microL, p=0.075). The decrease of all leukocytes in neonates of smoking mothers could be clinically significant and suggests a decreased cell production, increased peripheral recruitment, or retention in bone marrow. Given the importance of dendritic cells in early immune responses, their decrease might reflect an impact of maternal smoking on the developing fetal immune system.

Causal Inference in Observational Studies with Outcome-Dependent Sampling

In this paper, we consider estimation of the causal effect of a treatment on an outcome from observational data collected in two phases. In the first phase, a simple random sample of individuals are drawn from a population. On these individuals, information is obtained on treatment, outcome, and a few low-dimensional confounders. These individuals are then stratified according to these factors. In the second phase, a random sub-sample of individuals are drawn from each stratum, with known, stratum-specific selection probabilities. On these individuals, a rich set of confounding factors are collected. In this setting, we introduce four estimators: (1) simple inverse weighted, (2) locally efficient, (3) doubly robust and (4)enriched inverse weighted. We evaluate the finite-sample performance of these estimators in a simulation study. We also use our methodology to estimate the causal effect of trauma care on in-hospital mortality using data from the National Study of Cost and Outcomes of Trauma.

Cardiovascular disease and periodontitis: an update on the associations and risk

BACKGROUND: Associations between periodontitis and cardiovascular diseases have been recognized. MATERIAL AND METHODS: New literature since the last European Workshop on Periodontology has been reviewed. RESULTS: The lack of reliable epidemiological data on disease prevalence makes an assessment of the associations and risks between periodontitis and cardiovascular diseases difficult. Two recent meta-analysis reports have identified associations between periodontitis and cardiovascular diseases (odds ratios: 1.1-2.2). Different surrogate markers for both disease entities, including serum biomarkers, have been investigated. Brachial artery flow-mediated dilatation, and carotid intima media thickness have in some studies been linked to periodontitis. Studies are needed to confirm early results of improvements of such surrogate markers following periodontal therapy. While intensive periodontal therapy may enhance inflammatory responses and impair vascular functions, studies are needed to assess the outcome of periodontal therapies in subjects with confirmed cardiovascular conditions. Tooth eradication may also reduce the systemic inflammatory burden of individuals with severe periodontitis. The role of confounders remain unclear. CONCLUSIONS: Periodontitis may contribute to cardiovascular disease and stroke in susceptible subjects. Properly powered longitudinal case-control and intervention trials are needed to identify how periodontitis and periodontal interventions may have an impact on cardiovascular diseases.

[Periodontitis and cardiovascular disease]

The prevalence of periodontitis and cardiovascular disease (CVD) is high. A mixed infectious biofilm etiology of periodontitis is known but not fully established in CVD. Cofactors; smoking habits, stress, ethnicity, genetics, socioeconomics and age contribute to both diseases. The objectives of this report are to summarize factors in regards to CVD and periodontitis that are clinically relevant. The hypothesis behind a relationship between the two conditions can be founded in (I) shared infections etiology, (II) shared inflammatory response, (III) epidemiological and case-control studies, and (IV) periodontal studies demonstrating improvements of CVD markers. Streptococcus species in the S. mitis group, and S. anginosus group have been identified in periodontitis and are known as pathogens in endocarditis possibly transported from the oral cavity to the heart through bacteremia during dental therapies, and tooth brushing. Other periodontal bacteria such as Porphyromonas gingivalis, Fusobacterium nucleatum and Parvimonas micra are beta-lactamase producing and may contribute to antibiotic resistance (extended spectrum beta-lactamases). Other bacteria in CVD and periodontitis include Staphylococcus aureus, and Pseudomonas aeruginosa. Chlamydia pneumoniae and P. gingivalis lipopolyysaccharide capsels share homology and induce heat-shock protein activity and a cascade of proinflammatory cytokines. Associations between periodontitis and CVD have been presented in many studies when controlling for confounders. Other studies have demonstrated that periodontal therapies increase brachial artery flow rate and reduce serum inflammatory cytokine levels. Thus, physicians caring for subjects at CVD risk should consult with dentists/periodontists. Dentists must improve their medical knowledge and also learn to consult with physicians when treating patients at CVD risk.

Childhood leukaemia and socioeconomic status: what is the evidence?

The objectives of this systematic review are to summarise the current literature on socioeconomic status (SES) and the risk of childhood leukaemia, to highlight methodological problems and formulate recommendations for future research. Starting from the systematic review of Poole et al. (Socioeconomic status and childhood leukaemia: a review. Int. J. Epidemiol. 2006;35(2):370-384.), an electronic literature search was performed covering August 2002-April 2008. It showed that (1) the results are heterogeneous, with no clear evidence to support a relation between SES and childhood leukaemia; (2) a number of factors, most importantly selection bias, might explain inconsistencies between studies; (3) there is some support for an association between SES at birth (rather than later in childhood) and childhood leukaemia and (4) if there are any associations, these are weak, limited to the most extreme SES groups (the 10-20% most or least deprived). This makes it unlikely that they would act as strong confounders in research addressing associations between other exposures and childhood leukaemia. Future research should minimise case and control selection bias, distinguish between different SES measures and leukaemia subtypes and consider timing of exposures and cancer outcomes.

Residence near power lines and mortality from neurodegenerative diseases: longitudinal study of the Swiss population

The relation between residential magnetic field exposure from power lines and mortality from neurodegenerative conditions was analyzed among 4.7 million persons of the Swiss National Cohort (linking mortality and census data), covering the period 2000-2005. Cox proportional hazard models were used to analyze the relation of living in the proximity of 220-380 kV power lines and the risk of death from neurodegenerative diseases, with adjustment for a range of potential confounders. Overall, the adjusted hazard ratio for Alzheimer's disease in persons living within 50 m of a 220-380 kV power line was 1.24 (95% confidence interval (CI): 0.80, 1.92) compared with persons who lived at a distance of 600 m or more. There was a dose-response relation with respect to years of residence in the immediate vicinity of power lines and Alzheimer's disease: Persons living at least 5 years within 50 m had an adjusted hazard ratio of 1.51 (95% CI: 0.91, 2.51), increasing to 1.78 (95% CI: 1.07, 2.96) with at least 10 years and to 2.00 (95% CI: 1.21, 3.33) with at least 15 years. The pattern was similar for senile dementia. There was little evidence for an increased risk of amyotrophic lateral sclerosis, Parkinson's disease, or multiple sclerosis.

The association between beliefs about low back pain and work presenteeism

OBJECTIVE: To examine the influence of beliefs about low back pain (LBP) on reduced productivity at work ("presenteeism") caused by LBP. METHODS: Two thousand five hundred seven individuals completed the Back Beliefs Questionnaire, the Fear Avoidance Beliefs questionnaire (FABQ), and questions about LBP-related work-absence, reduced work-productivity, pain, comorbidity, and demographics. RESULTS: Six hundred seventy (25%) individuals were of working age, employed and reported current LBP. Univariate models showed beliefs were more "negative" in individuals with work-absence and reduced productivity (P = 0.0001). In multivariable analysis, controlling for confounders, "FABQwork" was a unique predictor of both absenteeism and presenteeism (each, P = 0.0001), though with small effect sizes. CONCLUSIONS: Negative beliefs about LBP are associated with both work absence and reduced work-productivity. Further investigations should examine their potential as a target for educational interventions when considering initiatives to reduce the socioeconomic costs of LBP.

Comparison of drug retention rates and causes of drug discontinuation between anti-tumor necrosis factor agents in rheumatoid arthritis

OBJECTIVE: Tumor necrosis factor (TNF) inhibitors have revolutionized the treatment of severe rheumatoid arthritis (RA), yet drug discontinuation is common. The aim of this study was to compare treatment retention rates and specific causes of anti-TNF discontinuation in a population-based RA cohort. METHODS: All patients treated with etanercept, infliximab, or adalimumab within the Swiss Clinical Quality Management RA cohort between 1997 and 2006 were included in the study. Causes of treatment discontinuation were broadly categorized as adverse events (AEs) or nontoxic causes, and further subdivided into specific categories. Specific causes of treatment interruption were analyzed using a Cox proportional hazards model and adjusted for potential confounders. RESULTS: A total of 2,364 anti-TNF treatment courses met the inclusion criteria. Treatment discontinuation was reported 803 times: 309 with etanercept, 249 with infliximab, and 245 with adalimumab. Drug inefficacy represented the largest single cause of treatment discontinuation (55.8% of cases). The median time of receiving anti-TNF therapy was 37 months, but discontinuation rates differed between the 3 anti-TNF agents (P < 0.001), with shorter retention rates for infliximab (hazard ratio [HR] 1.24, 99% confidence interval [99% CI] 1.01-1.51). The specific causes of treatment discontinuation revealed an increased risk of AEs with infliximab (HR 1.4, 99% CI 1.003-1.96), mostly due to an increased risk of infusion or allergic reactions (HR 2.11, 99% CI 1.23-3.62). Other discontinuation causes were equally distributed between the anti-TNF agents. CONCLUSION: In this population, infliximab was associated with higher overall discontinuation rates compared with etanercept and adalimumab, which is mainly due to an increased risk of infusion or allergic reactions.

Mortality from HIV and TB coinfections is higher in Eastern Europe than in Western Europe and Argentina

BACKGROUND AND OBJECTIVES: Tuberculosis (TB) is a leading cause of death in HIV-infected patients worldwide. We aimed to study clinical characteristics and outcome of 1075 consecutive patients diagnosed with HIV/TB from 2004 to 2006 in Europe and Argentina. METHODS: One-year mortality was assessed in patients stratified according to region of residence, and factors associated with death were evaluated in multivariable Cox models. RESULTS: At TB diagnosis, patients in Eastern Europe had less advanced immunodeficiency, whereas a greater proportion had a history of intravenous drug use, coinfection with hepatitis C, disseminated TB, and infection with drug-resistant TB (P < 0.0001). In Eastern Europe, fewer patients initiated TB treatment containing at least rifamycin, isoniazid, and pyrazinamide or combination antiretroviral therapy (P < 0.0001). Mortality at 1 year was 27% in Eastern Europe, compared with 7, 9 and 11% in Central/Northern Europe, Southern Europe, and Argentina, respectively (P < 0.0001). In a multivariable model, the adjusted relative hazard of death was significantly lower in each of the other regions compared with Eastern Europe: 0.34 (95% confidence interval 0.17-0.65), 0.28 (0.14-0.57), 0.34 (0.15-0.77) in Argentina, Southern Europe and Central/Northern Europe, respectively. Factors significantly associated with increased mortality were CD4 cell count less than 200 cells/microl [2.31 (1.56-3.45)], prior AIDS [1.74 (1.22-2.47)], disseminated TB [2.00 (1.38-2.85)], initiation of TB treatment not including rifamycin, isoniazid and pyrazinamide [1.68 (1.20-2.36)], and rifamycin resistance [2.10 (1.29-3.41)]. Adjusting for these known confounders did not explain the increased mortality seen in Eastern Europe. CONCLUSION: The poor outcome of patients with HIV/TB in Eastern Europe deserves further study and urgent public health attention.

Alcohol-selling outlets and mortality in Switzerland--the Swiss National Cohort

AIM To examine the association of alcohol-related mortality and other causes of death with neighbourhood density of alcohol-selling outlets for on-site consumption. DESIGN, SETTING AND PARTICIPANTS Longitudinal study of the adult Swiss population (n = 4 376 873) based on census records linked to mortality data from 2001 to 2008. MEASUREMENTS Sex-specific hazard ratios (HR) for death and 95% confidence intervals (95%CI) were calculated using Cox models adjusting for age, educational level, occupational attainment, marital status and other potential confounders. The density of alcohol-selling outlets within 1000 m of the residence was calculated using geocodes of outlets and residences. FINDINGS Compared with >17 outlets within 1000 m the HR for alcohol-related mortality in men was 0.95 (95%CI: 0.89-1.02) for 8-17 outlets, 0.84 (95%CI: 0.77-0.90) for 3-7 outlets, 0.76 (95%CI: 0.68-0.83) for 1-2 outlets and 0.60 (95%CI: 0.51-0.72) for 0 outlets. The gradient in women was somewhat steeper, with a HR comparing 0 with >17 outlets of 0.39 (95%CI: 0.26-0.60). Mortality from mental and behavioural causes and lung cancer were also associated with density of alcohol-selling outlets: HRs comparing 0 outlets with >17 outlets were 0.64 (95%CI: 0.52-0.79) and 0.79 (95%CI: 0.72-0.88), respectively, in men and 0.46 (95%CI: 0.27-0.78) and 0.63 (95%CI: 0.52-0.77), respectively, in women. There were weak associations in the same direction with all-cause mortality in men but not in women. CONCLUSIONS In Switzerland, alcohol-related mortality is associated with the density of outlets around the place of residence. Community-level interventions to reduce alcohol outlet density may usefully complement existing interventions.

Catch-up alveolarization in ex-preterm children: evidence from (3)He magnetic resonance

RATIONALE Histologic data from fatal cases suggest that extreme prematurity results in persisting alveolar damage. However, there is new evidence that human alveolarization might continue throughout childhood and could contribute to alveolar repair. OBJECTIVES To examine whether alveolar damage in extreme-preterm survivors persists into late childhood, we compared alveolar dimensions between schoolchildren born term and preterm, using hyperpolarized helium-3 magnetic resonance. METHODS We recruited schoolchildren aged 10-14 years stratified by gestational age at birth (weeks) to four groups: (1) term-born (37-42 wk; n = 61); (2) mild preterm (32-36 wk; n = 21); (3) extreme preterm (<32 wk, not oxygen dependent at 4 wk; n = 19); and (4) extreme preterm with chronic lung disease (<32 wk and oxygen dependent beyond 4 wk; n = 18). We measured lung function using spirometry and plethysmography. Apparent diffusion coefficient, a surrogate for average alveolar dimensions, was measured by helium-3 magnetic resonance. MEASUREMENTS AND MAIN RESULTS The two extreme preterm groups had a lower FEV1 (P = 0.017) compared with term-born and mild preterm children. Apparent diffusion coefficient was 0.092 cm(2)/second (95% confidence interval, 0.089-0.095) in the term group. Corresponding values were 0.096 (0.091-0.101), 0.090 (0085-0.095), and 0.089 (0.083-0.094) in the mild preterm and two extreme preterm groups, respectively, implying comparable alveolar dimensions across all groups. Results did not change after controlling for anthropometric variables and potential confounders. CONCLUSIONS Alveolar size at school age was similar in survivors of extreme prematurity and term-born children. Because extreme preterm birth is associated with deranged alveolar structure in infancy, the most likely explanation for our finding is catch-up alveolarization.

The association between admission systolic blood pressure and mortality in significant traumatic brain injury: A multi-centre cohort study

INTRODUCTION Low systolic blood pressure (SBP) is an important secondary insult following traumatic brain injury (TBI), but its exact relationship with outcome is not well characterised. Although a SBP of <90mmHg represents the threshold for hypotension in consensus TBI treatment guidelines, recent studies suggest redefining hypotension at higher levels. This study therefore aimed to fully characterise the association between admission SBP and mortality to further inform resuscitation endpoints. METHODS We conducted a multicentre cohort study using data from the largest European trauma registry. Consecutive adult patients with AIS head scores >2 admitted directly to specialist neuroscience centres between 2005 and July 2012 were studied. Multilevel logistic regression models were developed to examine the association between admission SBP and 30 day inpatient mortality. Models were adjusted for confounders including age, severity of injury, and to account for differential quality of hospital care. RESULTS 5057 patients were included in complete case analyses. Admission SBP demonstrated a smooth u-shaped association with outcome in a bivariate analysis, with increasing mortality at both lower and higher values, and no evidence of any threshold effect. Adjusting for confounding slightly attenuated the association between mortality and SBP at levels <120mmHg, and abolished the relationship for higher SBP values. Case-mix adjusted odds of death were 1.5 times greater at <120mmHg, doubled at <100mmHg, tripled at <90mmHg, and six times greater at SBP<70mmHg, p<0.01. CONCLUSIONS These findings indicate that TBI studies should model SBP as a continuous variable and may suggest that current TBI treatment guidelines, using a cut-off for hypotension at SBP<90mmHg, should be reconsidered.

Tenofovir or zidovudine in second-line antiretroviral therapy after stavudine failure in southern Africa

BACKGROUND There is debate over using tenofovir or zidovudine alongside lamivudine in second-line antiretroviral therapy (ART) following stavudine failure. We analyzed outcomes in cohorts from South Africa, Zambia and Zimbabwe METHODS: Patients aged ≥16 years who switched from a first-line regimen including stavudine to a ritonavir-boosted lopinavir-based second-line regimen with lamivudine or emtricitabine and zidovudine or tenofovir in seven ART programs in southern Africa were included. We estimated the causal effect of receiving tenofovir or zidovudine on mortality and virological failure using Cox proportional hazards marginal structural models. Its parameters were estimated using inverse probability of treatment weights. Baseline characteristics were age, sex, calendar year and country. CD4 cell count, creatinine and hemoglobin levels were included as time-dependent confounders. RESULTS 1,256 patients on second-line ART, including 958 on tenofovir, were analyzed. Patients on tenofovir were more likely to have switched to second-line ART in recent years, spent more time on first-line ART (33 vs. 24 months) and had lower CD4 cell counts (172 vs. 341 cells/μl) at initiation of second-line ART. The adjusted hazard ratio comparing tenofovir with zidovudine was 1.00 (95% confidence interval 0.59-1.68) for virologic failure and 1.40 (0.57-3.41) for death. CONCLUSIONS We did not find any difference in treatment outcomes between patients on tenofovir or zidovudine; however, the precision of our estimates was limited. There is an urgent need for randomized trials to inform second-line ART strategies in resource-limited settings.

Chronic NSAID use and long-term decline of renal function in a prospective rheumatoid arthritis cohort study

OBJECTIVES Non-steroidal anti-inflammatory drugs (NSAIDs) may cause kidney damage. This study assessed the impact of prolonged NSAID exposure on renal function in a large rheumatoid arthritis (RA) patient cohort. METHODS Renal function was prospectively followed between 1996 and 2007 in 4101 RA patients with multilevel mixed models for longitudinal data over a mean period of 3.2 years. Among the 2739 'NSAID users' were 1290 patients treated with cyclooxygenase type 2 selective NSAIDs, while 1362 subjects were 'NSAID naive'. Primary outcome was the estimated glomerular filtration rate according to the Cockroft-Gault formula (eGFRCG), and secondary the Modification of Diet in Renal Disease and Chronic Kidney Disease Epidemiology Collaboration formula equations and serum creatinine concentrations. In sensitivity analyses, NSAID dosing effects were compared for patients with NSAID registration in ≤/>50%, ≤/>80% or ≤/>90% of assessments. FINDINGS In patients with baseline eGFRCG >30 mL/min, eGFRCG evolved without significant differences over time between 'NSAID users' (mean change in eGFRCG -0.87 mL/min/year, 95% CI -1.15 to -0.59) and 'NSAID naive' (-0.67 mL/min/year, 95% CI -1.26 to -0.09, p=0.63). In a multivariate Cox regression analysis adjusted for significant confounders age, sex, body mass index, arterial hypertension, heart disease and for other insignificant factors, NSAIDs were an independent predictor for accelerated renal function decline only in patients with advanced baseline renal impairment (eGFRCG <30 mL/min). Analyses with secondary outcomes and sensitivity analyses confirmed these results. CONCLUSIONS NSAIDs had no negative impact on renal function estimates but in patients with advanced renal impairment.