Search

The ability to critically read a research paper is a skill that all postgraduate students and academicians require because the findings of a study must be interpreted in the context of its strengths and limitations. This article summarizes a recent study on continuation vs. discontinuation of antidepressants during pregnancy; preterm birth and low-birth weight were the outcomes of interest. The strengths and limitations of the study are considered, as are the best and worst case scenarios related to antidepressant use during pregnancy. It is hoped that this exercise will increase the reader's awareness of statistical and methodological issues that emerge when a study is critically examined.

Small relative effect sizes are common in observational studies of exposure in environmental and public health. However, such effects can still have considerable policy importance when the baseline rate of the health outcome is high, and many persons are exposed. Assessing the certainty of the evidence based on these effect sizes is challenging because they can be prone to residual confounding due to the non-randomized nature of the evidence. When applying GRADE, a precise relative risk >2.0 increases the certainty in an existing effect because residual confounding is unlikely to explain the association. GRADE also suggests rating up when opposing plausible residual confounding exists for other effect sizes. In this concept paper, we propose using the E-value, defined as the smallest effect size of a confounder that still can reduce an observed RR to the null value, and a reference confounder to assess the likelihood of residual confounding. We propose a 4-step approach. 1. Assess the association of interest for relevant exposure levels. 2. Calculate the E-value for this observed association. 3. Choose a reference confounder with sufficient strength and information and assess its effect on the observed association using the E-value. 4. Assess how likely it is that residual confounding will still bias the observed RR. We present three case studies and discuss the feasibility of the approach.

Objectives To describe lung function in correlation with information on exposure to dust and fibers in soft tissue paper-producing factories in Germany. Methods Ambient monitoring was performed for inhalable, respirable dust and fibers in nine soft tissue paper-producing factories. In a study group of 1,047 workers (189 control subjects, 240 workers with moderate exposure, and 618 workers with high exposure), spirometry (FVC, FEV 1 ) was performed. Information on occupational history, duration of exposure, workshop within the company, former occupational exposures, and smoking habits were collected. By employing multiple linear regression modeling, the potentially confounding effects of age, sex, body mass index, smoking habits, and factory were incorporated into the analysis of FVC, FEV 1 , and FEV 1 in percent of FVC (FEV 1 %FVC). By employing a logistic regression model, odds ratios were calculated for FVC Results The mean concentrations for inhalable, respirable, and fibrous dusts were 12.4 mg/m 3 , 0.28 mg/m 3 , and 420,000 fibers per cubic meter. With relation to cumulative dust and fiber exposure, a decrease of FVC from 105.4% predicted to 96.9% predicted (dust) and 97.1% predicted (fibers) in the subgroup with highest cumulative exposure was observed. For FEV 1 , a decrease from 107.3% predicted to 103.0% predicted (dust) and 102.8% predicted (fibers) was found. The parameter estimates show dose-response relationships that are more pronounced for FVC compared to FEV 1 . FEV 1 %FVC did not change significantly with increasing cumulative exposure, indicating a restrictive pattern of the findings. Conclusions Due to high ambient dust concentrations and the observed adverse effects on lung function, a reduction of dust exposure and secondary preventive measures is advised.

A recent study of the relationship between tamsulosin and dementia found that, in propensity score-matched analyses that adjusted for measured risk factors for dementia, the use of tamsulosin to treat benign prostatic hyperplasia (BPH) was associated with a small but significant increase in the risk of incident dementia relative to untreated BPH, BPH treated with the 5α-reductase inhibitors dutasteride and finasteride, and BPH treated with the α adrenoceptor blockers doxazosin, terazosin, and alfuzosin. The choice of control groups addressed confounding by indication, confounding by disease severity, and confounding by pharmacologic drug class. The authors of the study provided a wealth of detail in their main paper and in supplementary materials, allowing an almost forensic examination of the findings. This article discusses the study from the perspective of whether the hypothesis relating tamsulosin to dementia was set a priori or emerged after an exploratory exercise; whether tamsulosin crosses the blood-brain barrier for dementia as the event of interest to be attributable to treatment; whether the action of tamsulosin is plausibly related to dementia as a possible outcome; whether confounding was adequately addressed in the analyses; whether the duration of follow-up was sufficient for the event of interest to be attributable to tamsulosin; whether the absolute increase in the rate of incident dementia was clinically significant; and what the curiosities in the dose-dependence and subgroup analysis findings could imply. These matters provide considerable food for thought. Thus, this article is intended to serve as an example of how to read a paper critically and how to think about what the findings might suggest.

Background: Despite peer review, publications in scientific journals are not always well written, sometimes contain errors, and often exhibit deliberate or unintended biases. It is necessary to learn how to identify such limitations. It is also necessary to learn how to read between and beyond the lines of papers no matter how well written they are and no matter how highly ranked the journal is. Materials and Methods: This paper critically examines an important article in a leading journal with a view to help the reader learn how to place the findings of a study in perspective, understand its limitations, and glean information beyond that actually presented and discussed in the text. Results: Several issues are examined; these relate to case-control research designs, confounding, propensity matching, absolute risk, confidence intervals, interpretation of findings, real-world relevance, ecological validity, and definition of a cause-effect relationship. Conclusions: The issues examined in this paper reflect common themes in research, and a reader aware of these themes will more easily identify them in his future readings.

Linear regression is the equation which provides of straight line that best describes the association between two continuous variables, x and y. However, it is often the case that the dependent variable y is influenced by more than one variable and not just a single x variable. Multivariate analysis is a statistical modeling technique wherein multiple x variables are analysed simultaneously for their effect on y, resulting in an additive model (via an equation) that explains the observation/s and corrects for confounding association/s using one dependent and several independent variables, assigning a gradient to each of these independent variables, and with all product terms of gradient and magnitude of the independent variables adding up to estimate 'y'. This paper outlines these various techniques and applications.

Objective: To investigate time differences in recording observations and an early warning score using traditional paper charts and a novel e-Obs system in clinical practice. Methods: Researchers observed the process of recording observations and early warning scores across 3 wards in 2 university teaching hospitals immediately before and after introduction of the e-Obs system. The process of recording observations included both measurement and documentation of vital signs. Interruptions were timed and subtracted from the measured process duration. Multilevel modeling was used to compensate for potential confounding factors. Results: In all, 577 nurse events were observed (281 paper, 296 e-Obs). The geometric mean time to take a complete set of vital signs was 215 s (95% confidence interval [CI], 177 s–262 s) on paper, and 150 s (95% CI, 130 s–172 s) electronically. The treatment effect ratio was 0.70 (95% CI, 0.57–0.85, P Discussion: Introduction of an e-Obs system was associated with a statistically significant reduction in overall time to measure and document vital signs electronically compared to paper documentation. The reductions in time varied among wards and were of clinical significance on only 1 of 3 wards studied. Conclusion: Our results suggest that introduction of an e-Obs system could lower nursing workload as well as increase documentation quality.

OBJECTIVES: Studies in traditional paper mills have indicated an excess cancer risk, and mutagenic compounds have been identified in the industry. No studies have reported on risk of cancer in paper recycling. Therefore the cancer incidence in Danish paper recycling mills was investigated. METHODS: 5377 employees in five paper recycling plants were included in a historical cohort study. The workers had been employed in paper recycling in 1965-90, and the cohort was followed up until 31 December 1993. The expected number of cancer cases was calculated from national rates. RESULTS: There was significantly more pharyngeal cancer among male workers (seven observed (standardised incidence ratio (SIR) 3.33, 95% confidence interval (95% CI) 1.34 to 6.87)). There was slightly more lung cancer among male workers in production (39 observed, SIR 1.21, 95% CI 0.86 to 1.65). Risk of Hodgkin9s disease was doubled in male production worker (four observed, SIR 1.90, 95% CI 0.51 to 4.85). CONCLUSIONS: The increased risk of pharyngeal cancer found in this study is interesting but may be influenced by confounders such as smoking and alcohol intake. This study also indicates an excess risk of Hodgkin9s disease, which is in accordance with some studies in the traditional paper mills. As this is the first report on risk of cancer in paper recycling, further studies are needed.

Genetic factors may influence the natural course of diabetic peripheral neuropathy and explain some of its variability. The aim of this review was to examine the association between apolipoprotein E (apoE) gene polymorphisms and diabetic peripheral neuropathy. Four relevant studies were identified. The two earlier works provided evidence that the ɛ4 allele is a risk factor for this complication, while the two more recent studies were negative. Important differences in the methodology used and in the populations included are obvious, rendering difficult the comparison between studies. In conclusion, the association between APOE gene polymorphisms and diabetic peripheral neuropathy is still unclear. Available evidence is rather limited and results have so far been contradictory. Future studies should employ more robust methodology, adjusting for potential confounders and for the prevalence of neuropathy in the general population with diabetes.

Objectives A cohort study of female pulp and paper workers in Norway has shown a significantly increased risk of ovarian cancer. Other than the involvement of hormonal and reproductive factors, little is known of the etiology of ovarian cancer. Asbestos and talc are two agents hypothesized to influence the development of the disease. The present study aimed to investigate the association between ovarian cancer and occupational exposure to asbestos, talc, and total dust among Norwegian pulp and paper workers Methods Forty-six cases of ovarian cancer, with four controls each, were included in the study. Occupational exposure was assessed by combining work histories from personnel files, questionnaire information about production processes, and exposure assessments from the mills. To obtain information about possible confounders, cases and controls were invited to participate in a personal interview. Results The odds ratio for asbestos exposure was 2.02, 95% confidence interval 0.72-5.66. For talc exposure, the odds ratio was 1.10, and for ever exposure to total dust, it was below 1.00. The risk estimates did not essentially differ after adjustment for possible confounding variables. Conclusions The results do not confirm an association between exposure to asbestos, talc, and total dust and ovarian cancer among Norwegian pulp and paper workers. However, the odds ratio for asbestos exposure was doubled, and control for established nonoccupational risk factors did not change the estimate. Therefore, the possibility that exposure to substances in the work environment contributes to the elevated risk cannot be rejected.

We have read with interest the article by Sang-Hee Yoon et al. [1] entitled ‘Bilateral salpingectomy can reduce the risk of ovarian cancer in the general population: a meta-analysis’ published recently in the Journal. We have found some methodological issues in the paper that deserve a commentary. First, the authors did not use adjusted effects for the meta-analysis. In consequence, the effect could be affected by confounders which have not been considered [2]. Second, the effects expressed as odds ratio (OR) and hazard ratio (HR) could have been analysed together in the analysis. The paper of Falconer et al. [3] showed its results as HR which could have been transformed, using the methodology of Chinn [4], into OR.

Summary Allergic rhinitis (AR) is a common condition with quality of life and economic implications for those affected. Numerous studies have attempted to evaluate treatments for rhinitis, seeking clinically meaningful efficacy and safety results to enable evidence-based treatment decisions. Traditional studies of medications for AR are hampered by many confounding environmental factors as well as suboptimal medication compliance. They are also an unsuitable setting for determination of precise pharmacodynamic properties of medications, including onset and duration of action. Allergen challenge chambers (ACCs) were developed to provide predetermined, controlled allergen levels and to limit variables inherent in traditional studies. An ACC hosts a number of allergen-sensitive subjects who may receive either medication or placebo in a closed environment regulated for temperature, humidity and other variables. Subjects' allergic responses are monitored using subjective and objective assessments throughout the study, and the resultant information contributes significantly to the clinical profile of a medication. This consensus paper provides an in-depth review of the role of ACCs as a means to evaluate treatments in AR, and concludes that ACC trials fulfil an important supportive role in the assessment of anti-allergic medication.

BACKGROUND: Most clinicians read only the abstract of papers in scientific journals. Therefore, it is very important that abstracts contain as much information as possible, to summarize the data succinctly. Our objectives were to evaluate the quality of information in abstracts reporting human fetal outcomes following drug exposure during pregnancy. METHODS: We developed quality criteria based on previous work, modifying them for use with pregnancy outcomes. Quality scores were calculated as present/absent for all of the equally weighted criteria, then expressed as percentages (present/[present + absent]). We examined a random sample of 100 abstracts obtained through searches of MEDLINE, EMBASE, and the Web of Science databases from 1990 to 2005. Average quality scores were compared across designs (cohort, case-control, meta-analysis, and mixed design) Using Kruskal-Wallis ANOVA and structured/unstructured formats using Student's t test. RESULTS: the overall average quality was 59.2% ± 14% (median, 61.5%; range, 15.4–83.3%). Quality was not significantly different across designs (P = .16) or between structured and unstructured abstracts (P = .44). Quality scores increased over time (Rho = 0.23, P = .02). Most frequently absent were baseline risk (94%), drug dose (91%), nonsignificant P values (72%), confounders (69%), significant P values (57%), and risk difference (48%). CONCLUSIONS: Abstracts provide insufficient information, particularly baseline risk values, for readers to make evidence-based decisions regarding drug use during pregnancy. Efforts need to be made to improve the quality of abstracts and include critical information such as baseline risk. Birth Defects Research (Part A), 2006. © 2006 Wiley-Liss, Inc.

Objectives: To provide a participant’s account of the development of a paper commissioned by the tobacco industry examining the reliability of self reported smoking status; to redress the distorted report of this Japanese spousal smoking study which evaluated the reliability and validity of self reported smoking status, and estimated confounding by diet and lifestyle factors. Design: Repeated interviews on smoking status and its verification by environmental and biological markers for environmental tobacco smoke (ETS) exposure. Setting: Urban wives in Osaka City and Sizuoka City, Japan Participants: Semi-random sampling of 200 wives in each city. From the Osaka subjects, 100 non-smoking wives were selected for the validity study. Main outcome measures: Kappa coefficient for reliability of self reported smoking status. Correlation coefficients between environmental nicotine concentration, cotinine in saliva and urine, and self reported smoking status. Results: The κ coefficient for the repeated interview was high suggesting sufficient reliability of the response. The proportion of self reported current smokers misclassified as non-smokers was equivalent to the misclassified self reported non-smokers. Ambient concentration of nicotine and personal exposure to nicotine correlated with each other and also with salivary cotinine and self reported ETS exposure but not with urinary cotinine/creatinine ratio (CCR). There was no major difference in diet and lifestyle related to husband’s smoking status. Conclusion: Self reported smoking status by Japanese wives shows high reliability. It also shows high validity when verified by both nicotine exposure and salivary cotinine, but not by CCR. A previous report questioning the credibility of self reported smoking status, based on questionable CCR, could thus be of dubious validity. In addition, possible dietary and lifestyle confounding factors associated with smoking husbands were not demonstrable, a finding not reported previously. Using all the data from this project changes the conclusion of the previous published report. In addition to the distortion of scientific findings by a tobacco industry affiliated researcher, anti-smoking campaigners made attempts to intimidate and suppress scientific activities. These distortions of science should be counteracted.

The objective of this review was to explore the relationship between lactational performance and health. We focused on lactational incidence rates (LIR) of the production diseases that are of economic importance to the dairy industry. Based on a review of 11 epidemiological and 14 genetic studies we found little evidence that high yielding cows have increased risk of dystocia, retained placenta, metritis and left-displaced abomasum. Results for periparturient paresis were inconsistent. Whilst we found no phenotypical relationship between milk yield and the risk of ketosis and lameness, selection for higher milk yields will probably increase LIR for these diseases. Mastitis was the only disease where there was a clear relationship between milk yield and risk of infection. Continued selection for higher milk yields will worsen this situation. However, our overall conclusion is that reviewing existing literature, even with a structured literature selection, is inadequate to the task of elucidating the relationship between lactational performance and risk of production diseases. There are substantial problems with confounding effects and unaccounted for biological correlations. In the second part of the review we argue towards a common basis for addressing production diseases. We propose abnormal body mobilisation and immune competence as common currencies for metabolic and immune status and argue for the development of indicators of metabolic imbalance and the early development of diseases. Furthermore, we suggest the use of indicators of ‘imbalance’ to guide feeding according to the needs of individual cows with their specific genotype and management history. We believe that this approach has the potential to provide new diagnostic and decision support tools to improve animal health and reproduction, whilst simultaneously maintaining optimal production and efficiency. Further research is needed to identify and validate new indicators and individual feeding strategies.

This paper summarizes the themes and recommendations that emerged from the papers presented at the "International Symposium on Lifestyle Factors and Human Lung Cancer". In terms of our current knowledge, the following are considered important: (1) recognition and confirmation of lung cancer as a multifactorial disease; (2) the strong association known to exist between cigarette smoking and lung cancer does not explain the observation that the lung cancer death rate is lower than the smoking rate in certain groups of individuals, e.g. Chinese farmers; (3) the high incidence of adenocarcinoma in non-smoking women, e.g. Chinese housewives, suggests that factors other than cigarette smoking are involved for the development of lung cancer; (4) indoor air pollution is a significant risk factor for lung cancer in China; (5) there is incomplete information on the chemical nature and exposure assessment of indoor air pollutants; and (6) the need to better define the role of diet and nutrients as confounding factors for lung cancer. Based on the above, several recommendations were offered for continuing research to further our understanding of the etiology and pathogenesis of lung cancer.

Sirs: Thank you very much for your letters to the editor.1,2 We hope we can help clarify and address some of the concerns raised. Our study actually did address many of these concerns using multivariate logistic regression and risk adjustment. This method simultaneously analyzes the effect of multiple independent variables on a dependent variable in relation to one another. Logistic regression aims to find which of the independent variables (ie, operative time, age, BMI, sex, smoking status) has a true relationship to the observed outcome (postoperative complication) and to adjust the reported effect of the variable of interest by weighing the influence of the other variables. Essentially, by using this model, we were able to adjust for confounders of operative time. Thus, we did in fact control for the confounders and addressed the patient characteristics and various procedures statistically. There are some flaws in our data, including some aberrant operative times, which were subject to errors based on what was reported by nurses, time outs, and surgical reports. …

Aims: Data from the Northern Hemisphere support an excess of winter-spring births of individuals who later develop schizophrenia when compared with the general population. The data from the Southern Hemisphere have been less consistent. This paper will present a systematic review and mete-analysis of relevant data from the Southern Hemisphere. Methods: To identify relevant studies we searched electronic databases, reviewed citations from target publications and wrote letters to published authors in the field. The counts for observed and expected births were assessed in four planned comparisons. In the absence of significant heterogeneity, the data were combined using Mantel-Haenzel odds ratio in a fixed effect model. Results: Twelve studies were identified. Published and unpublished data from eight of these were able to be included in the analyses. For the two seasonal comparisons (n=20 017), small but non-significant excesses were found in the first comparison (winter versus other seasons; OR=1.04, 0.99-1.08) and for the second comparison (winter and spring versus other seasons; OR=1.03, 0.99-1.07). For the two quarterly comparisons (n= 14 799), there was a small but non-significant excess found in the third comparison (third quarter versus other quarters; OR=1.03, 0.98-1.09), and a small but non-significant deficit in the fourth comparison (third and fourth quarter versus other quarters OR=0.99, 0.95-1.04). Conclusions: Assuming that season of birth acts as a proxy marker for fluctuating non-genetic risk-modifying factors for schizophrenia, this review suggests that in the Southern Hemisphere these factors may be weaker, less prevalent, less regular, and/or may be modified by other confounding or modifying variables. (C) 1999 Published by Elsevier Science B.V. All rights reserved.

Objective To identify variables that must be coded when synthesizing primary studies that use quasi-experimental designs. Study Design and Setting All quasi-experimental (QE) designs. Results When designing a systematic review of QE studies, potential sources of heterogeneity—both theory-based and methodological—must be identified. We outline key components of inclusion criteria for syntheses of quasi-experimental studies. We provide recommendations for coding content-relevant and methodological variables and outlined the distinction between bivariate effect sizes and partial (i.e., adjusted) effect sizes. Designs used and controls used are viewed as of greatest importance. Potential sources of bias and confounding are also addressed. Conclusion Careful consideration must be given to inclusion criteria and the coding of theoretical and methodological variables during the design phase of a synthesis of quasi-experimental studies. The success of the meta-regression analysis relies on the data available to the meta-analyst. Omission of critical moderator variables (i.e., effect modifiers) will undermine the conclusions of a meta-analysis.

The question of whether the range of recommended body weights should remain the same throughout adulthood or should be more liberal (higher) in older adults has been extremely controversial. Much of the debate has centered on methodologic issues including sources of confounding and the precision of risk estimates. These methodologic issues are reviewed, and studies that compare the body mass index (BMI)-mortality relationship within appropriate age strata are summarized. Several studies have attempted to control for major confounders, but were too small in size to produce comparisons across age groups. Only three studies had as many as 400 deaths in each age group, and only one of those attempted to control for confounding variables. This study, a recent analysis of the Cancer Prevention Study-I, combined attention to possible confounding variables with the statistical power that comes from a large sample. The analysis indicated that the optimum 12-year survival was associated with a lean BMI (19-21.9) between the ages of 30 and 74 years. After 74 years, the optimum BMI was considerably higher. This study had several limitations and its generalizability is restricted, but nevertheless, it offers the best evidence to date on the impact of age on the BMI-mortality association. New studies are needed that control for possible confounding factors and include repeated measures of weight and height, and perhaps other indicators of body composition and fat distribution, over a long period of follow-up.

In a randomized prospective study of 86 metastatic breast cancer patients by Spiegel et al. in 1989, the 50 who took part in a group psychosocial intervention survived on average 18 months longer than the 36 controls who did not. Because the control survival curve looked unusually steep, lacking an expected right-skewed tail, both curves were compared with that of a population from the same region having metastatic breast cancer. When transformed to life-table format, the curves of the control sample and the regional population, neither group having had an intervention, were almost identical for a year, and differed strikingly after 20 months. This led to the hypothesis that the 12 control patients surviving for more than 20 months were an extremely aberrant sample, being subject to the strong biasing influence of possible confounders, of which a considerable number are known, but not including those accounted for in the study. Corollaries to the hypothesis are that the intervention had no effect; that the intervention curve was in fact equivalent to a control curve with mild sampling departure from that of the regional population; and that, therefore, the repetition of the study now under way would not yield confirmation of the 1989 study, but rather, would support the hypothesis and the first two corollaries. © 1998 John Wiley & Sons, Ltd.

Increased preoperative instrument exposure has been hypothesized to contribute to a greater risk of postoperative surgical site infection (SSI) in spine surgery. As such, a quality practice intervention (QPI) was implemented to reduce the instruments uncovered time (IUT) for VEPTR implantation. Reductions in IUT were significantly associated with decreased incidence of SSIs. In a multivariate logistic regression model controlling for significant confounders, IUTs exceeding 100 minutes were associated with 8.5 times the odds of developing a SSI.

Objective: Citations by other researchers are important in the dissemination of research findings. We aimed to investigate whether preferential citation of statistically significant articles exists in the psychiatric literature. Study Design and Settings: We analyzed all original research papers published in 1996 in four psychiatric journals. Using a standardized questionnaire, from each paper, we extracted the primary outcome and its statistical significance. The number of citations, excluding authors’ ‘‘self-citations,’’ received by April 2005 was obtained. Regression analysis was used to relate citation frequency to statistical significance, adjusting for confounders. Results: Of 448 extracted papers, 368 used statistical significance testing and 287 (77.8%) reported P ! 0.05. The median number of citations for papers reporting ‘‘significant’’ and ‘‘nonsignificant’’ results was 33 vs. 16. After adjustment for journal, study design, reporting quality, whether outcome confirmed previous findings and study size, the ratio of the number of citations per article for articles reporting ‘‘P ! 0.05’’ on the primary outcome to those reporting ‘‘P O 0.05’’ was 1.63 (95% confidence interval [CI]: 1.32, 2.02, P! 0.001). Conclusion: Authors cite studies based on their P-value rather than intrinsic scientific merit. This practice skews the research evidence.

Many clinicians, medical practitioners and decision makers have no formal training in epidemiology but need to understand and sometimes evaluate results in epidemiologic studies. This paper attempts to give guidance to non-epidemiologists on how to read and evaluate the quality of epidemiologic studies and their results critically. Different methodological issues for evaluating whether the results of a study are causal or by bias, chance or confounding are given. This includes criteria for the choice of an appropriate study design followed by problems of the definition of the study population and the sample selection. We will also point to potential sources of bias in the data collection procedure and list some principles for statistical analysis. Finally, we include comments on how the results should be presented and issues which are related to public health and ethical questions. Although it is not usually possible to perform a perfect study and the correct approach to study design and analysis is often highly dependent on specific features of the population and the surroundings, our paper should help to distinguish between weak and strong investigations and papers.

The possible confounding caused by smoking was studied in connection with a mortality and cancer incidence study of 3,520 workers in the pulp and paper industry. A group 1,290 sawmill workers was used for comparison in addition to the expected numbers based on national statistics. A total of 801 questionnaires was sent to a representative sample of the workers, 537 to people still alive and 264 to the next-of-kin of decedents. The reply percentage was 86.6 for the former and 80.9 for the latter. The prevalence of smoking was determined for 1956 and 1981. The proportions of moderate and heavy smokers were assessed for 1981, and smoking indices were calculated for each occupational group. The estimated rate ratios for lung cancer in relation to smoking categories and the corresponding smoking-adjusted standardized incidence ratios were calculated for 1981, and they were compared with the observed standardized incidence ratios for lung cancer. The smoking habits could not explain the observed excess of lung cancer, nor the increased mortality from coronary heart disease found among the pulp and paper workers. Postal questionnaires may be a feasible tool for assessing smoking habits in retrospective cohort studies.

The abundance of literature related to nutritional support reflects its recently recognised role in preventing metabolic complications and gut dysfunction during critical illness. However, some published studies lack relevance to critically ill patients, as a result of the selection of subjects and outcome variables, or flaws in the study design, as well as in the type, composition, timing, route of administration and amount of nutritional support given. This review will highlight these confounding factors by describing two imaginary (but typical) clinical trials and by analysing some studies published. The point at issue is that basic quality requirements, such as the formulation of a prospective hypothesis and the delineation of the effects of the reference treatment, are often lacking in many studies published. Data analysis was often found to be biased by the absence of statistical power calculation and intention-to-treat analysis. Globally, studies designed to assess the effects of nutritional support on the outcome of critically ill patients, rarely fulfil basic quality requirements and should therefore be interpreted cautiously. We suggest simple strategies or study design that will allow important questions to be answered by future clinical trials.

Introduction Control selection is a crucial step at the study designing phase. Although, the concept of the different types of matching for control selection has been discussed in the context of case-control studies, here we targeted analytical cross-sectional studies to explore the effects of each type of control selection on the amount of bias and precision of the OR. Methods 41 coronary atherosclerotic patients and 92 disease-free hospital controls were recruited to assess the relationship between opium consumption (OpiumHx) and coronary atherosclerosis (Outcome). Considering the OpiumHx as the main independent factor and age as the confounder, we calculated point estimate and the CI for OR in different scenarios of matching for control selection, namely exact, stratified, frequency and propensity matching. Syntaxes were developed by STATA 10. Results The crude OR was 3.4 (95% CI 1.5 to 7.9). By exact matching on age, 21 pairs remained for the analysis and the OR was equal to 3.3 (0.8 to 18.8). Stratified matching on age group kept 41 pairs and gave us the OR of 0.9 (0.5 to 1.7). Frequency matching kept 88 subjects for the analysis and led to the OR of 3.0 (1.2 to 7.4). By propensity matching, 27 pairs remained which gave the OR of 3.5 (1.1 to 14.6). Conclusion Matching techniques influence effect size and precision, seriously. Although the most bias reduction happened in pair matched techniques, a large reservoir of controls would be needed to prohibit immense decrease in precision. These findings should be considered at both protocol development and analysis phases of observational studies with caution.

BACKGROUND CONTEXT: There are differing views as to the causes of disc degeneration. One view, referred to as a cumulative or repetitive injury model, explains disc degeneration, in great part, as the result of ‘‘wear and tear’’ on the disc from routine daily exposures to physical loading or biomechanical forces. Perhaps paradoxically, such physical loading has been thought to be generally beneficial for other structures of the musculoskeletal system. PURPOSE: The goal of this study was to investigate the associations of physical loading from body weight using quantitative measures of disc degeneration. STUDY DESIGN: This is an exposure-discordant twin study. METHODS: We selected 44 pairs of healthy male monozygotic (MZ) twins with 8 kg or more discordance in body weight (mean 13 kg), with lumbar magnetic resonance imaging and bone density measurements available. The main outcome of disc degeneration was assessed through quantitatively measured disc height, and disc signal intensity and signal variation adjusted by the intrabody reference, adjacent cerebrospinal fluid signal, at L1–L4. Data on suspected constitutional and environmental risk factors were available for control of possible confounding factors. RESULTS: Higher body weight was associated with 6.2% higher bone density in the lumbar spine, confirming an effective discordance (p!.0001). Disc signal variation was 5.4% higher (‘‘better’’) among the heavier MZ co-twins (p5.005), but the 2.6% higher disc heights and 2.9% higher adjusted disc signals were not statistically significant. CONCLUSIONS: Contrary to common beliefs, our findings suggest that cumulative or repetitive loading because of higher body mass (nearly 30 pounds on average) was not harmful to the discs. In fact, a slight delay in L1–L4 disc desiccation was observed in the heavier men, as compared with their lighter twin brothers. 2010 Elsevier Inc. All rights reserved.

Estimating the effectiveness of protected areas (PAs) in reducing deforestation is useful to support decisions on whether to invest in better management of areas already protected or to create new ones. Statistical matching is commonly used to assess this effectiveness, but spatial autocorrelation and regional differences in protection effectiveness are frequently overlooked. Using Colombia as a case study, we employed statistical matching to account for confounding factors in park location and accounted for for spatial autocorrelation to determine statistical significance. We compared the performance of different matching procedures—ways of generating matching pairs at different scales—in estimating PA effectiveness. Differences in matching procedures affected covariate similarity between matched pairs (balance) and estimates of PA effectiveness in reducing deforestation. Independent matching yielded the greatest balance. On average 95% of variables in each region were balanced with independent matching, whereas 33% of variables were balanced when using the method that performed worst. The best estimates suggested that average deforestation inside protected areas in Colombia was 40% lower than in matched sites. Protection significantly reduced deforestation, but PA effectiveness differed among regions. Protected areas in Caribe were the most effective, whereas those in Orinoco and Pacific were least effective. Our results demonstrate that accounting for spatial autocorrelation and using independent matching for each subset of data is needed to infer the effectiveness of protection in reducing deforestation. Not accounting for spatial autocorrelation can distort the assessment of protection effectiveness, increasing type I and II errors and inflating effect size. Our method allowed improved estimates of protection effectiveness across scales and under different conditions and can be applied to other regions to effectively assess PA performance., Article impact statement: Spatial autocorrelation and sampling design can distort estimations of protected area effectiveness at reducing forest loss.

Trace elements such as cadmium, arsenic, zinc or selenium increase or decrease risk of a wide range of human diseases. Their levels in toenails may provide a measure of mid-term intake of trace elements for studies in humans. However, in biologically and clinically aggressive diseases as pancreatic cancer, the progression of the disease could modify such concentrations and produce reverse causation bias. The aim was to analyze the influence of specific time intervals between several clinical events and the collection of toenails upon concentrations of trace elements in patients with pancreatic cancer. Subjects were 118 incident cases of pancreatic adenocarcinoma prospectively recruited in eastern Spain. Toenails were collected at cancer diagnosis, and soon thereafter interviews were conducted. Information on cancer signs and symptoms was obtained from medical records and patient interviews. Levels of 12 trace elements were determined in toenail samples by inductively coupled plasma mass spectrometry. General linear models adjusting for potential confounders were applied to analyze relations between log concentrations of trace elements and the time intervals, including the interval from first symptom of cancer to toenail collection (iST). Toenail concentrations of the 12 trace elements were weakly or not influenced by the progression of the disease or the diagnostic procedures. Concentrations of aluminum were slightly higher in subjects with a longer iST (age, sex and stage adjusted geometric means: 11.44 vs. 7.75 µg/g for iST > 120 days vs. ≤ 40 days). There was a weak inverse relation of iST with concentrations of zinc and selenium (maximum differences of about 20 and 0.08 µg/g, respectively). Conclusions: concentrations of the trace elements were weakly or not influenced by the development of the disease before toenail collection. Only concentrations of aluminum increased slightly with increasing iST, whereas levels of zinc and selenium decreased weakly. Even in an aggressive disease as pancreatic cancer, toenail concentrations of trace elements may provide a valid measure of mid-term intake of trace elements, unaffected by clinical events and disease progression. Supplementary Information The online version contains supplementary material available at 10.1007/s12403-021-00436-2.

Introduction The effect of valve type on outcomes in transfemoral transcatheter aortic valve replacement (TF-TAVR) has recently been subject of debate. We investigate outcomes of patients treated with balloon-expanding (BE) vs. self-expanding (SE) valves in in a cohort of all these procedures performed in Germany in 2018. Methods All patients receiving TF-TAVR with either BE (N = 9,882) or SE (N = 7,413) valves in Germany in 2018 were identified. In-hospital outcomes were analyzed for the endpoints in-hospital mortality, major bleeding, stroke, acute kidney injury, postoperative delirium, permanent pacemaker implantation, mechanical ventilation > 48 h, length of hospital stay, and reimbursement. Since patients were not randomized to the two treatment options, logistic or linear regression models were used with 22 baseline patient characteristics and center-specific variables as potential confounders. As a sensitivity analysis, the same confounding factors were taken into account using the propensity score methods (inverse probability of treatment weighting). Results Baseline characteristics differed substantially, with higher EuroSCORE (p p p p = 0.617] major bleeding [aOR 0.91 (0.73;1.14), p = 0.400], stroke [aOR 1.13 (0.88;1.46), p = 0.347], acute kidney injury [OR 0.97 (0.85;1.10), p = 0.621], postoperative delirium [aOR 1.09 (0.96;1.24), p = 0.184], mechanical ventilation > 48 h [aOR 0.98 (0.77;1.25), p = 0.893], length of hospital stay (risk adjusted difference in days of hospitalization (SE instead of BE): − 0.05 [− 0.34;0.25], p = 0.762) and reimbursement [risk adjusted difference in reimbursement (SE instead of BE): − €72 (− €291;€147), p = 0.519)] There is, however, an increased risk of PPI for SE valves (aOR 1.27 [1.15;1.41], p Conclusions We find broadly equivalent outcomes in contemporary TF-TAVR procedures, regardless of the valve type used. Incidence of major complications is very low for both types of valve.

We investigated the relationship between low lean mass (LLM) and lower urinary tract symptoms (LUTS) using the 2005–2006 National Health and Nutrition Examination Survey (NHANES) dataset. We enrolled 959 men with an average age of 52.08 ± 7.91 years and performed weighted multiple regression analysis to determine the independent relationship between exposure variables (LLM, alternate LLM) and outcomes variables (urinary hesitancy, incomplete emptying, urinary frequency, nocturia, daytime LUTS, clinical LUTS) after adjusting for confounding factors. The prevalence of urinary hesitancy (OR = 7.76, P < 0.0001), incomplete emptying (OR = 2.49, P = 0.0070), urinary frequency (OR = 3.28, P < 0.0001), daytime LUTS (OR = 3.88, P < 0.0001) and clinical LUTS (OR = 8.11, P < 0.0001) was significantly higher among men with LLM compared to men without LLM. Moreover, alternate LLM (ALLM) was positively associated with urinary hesitancy (OR = 17.97, P < 0.0001), incomplete emptying (OR = 4.68, P = 0.0003), daytime LUTS (OR = 2.47, P = 0.0136) and clinical LUTS (OR = 12.18, P < 0.0001). These findings demonstrate that both LLM and ALLM were associated with a higher risk of LUTS in men aged ≥ 40 years, which suggested that early management and treatment of lean mass loss may improve or alleviate LUTS.

Background: Infection with multiple pathogens may play a key role in the pathogenesis of dementia. Whether Helicobacter pylori (H. pylori) infection is associated causally with dementia is controversial. Objective: We conduct a meta-analysis of case-control and cohort studies on the association between H. pylori infection and the risk for all-cause and Alzheimer’s disease (AD) dementia. Methods: Two independent reviewers searched the PubMed, Cochrane Library, and Embase databases with English language restrictions from the date of conception to September 18, 2020. The primary analysis was as follows: the exposure variable was H. pylori infection, and the outcome was incident all-cause and AD dementia. Pooled odds ratios (OR), relative risk (RR), and corresponding 95% confidence intervals (CI) were obtained using the fixed-or random-effect model. Forest plots were generated to summarize the results. Results: Ten studies involving 96,561 participants were included in the meta-analysis: 5 case-control studies and 5 cohort studies. The overall pooled cohort studies showed a significant positive association between H. pylori infection and all-cause dementia with pooled RR of 1.36 (95% CI, 1.11-1.67). There was no association between H. pylori infection and risk for developing AD: RR of 1.33 (95% CI, 0.86-2.05) in cohort studies, and OR of 1.72 (95% CI, 0.97-3.04) in case-control studies. Significant heterogeneity was showed in each comparison group. Conclusion: This meta-analysis supports a positive association between H. pylori infection and the risk of all-cause dementia, but not AD dementia. Due to the interference of confounding factors, randomized controlled trials are needed to prove their causality.

Background Recent randomized controlled trials have sparked debate about the optimal treatment of patients suffering from left main coronary artery disease. The present study analyzes outcomes of left main stenting versus coronary bypass grafting (CABG) in a nationwide registry in patients with chronic coronary syndrome (CCS). Methods All cases suffering from CCS and left main coronary artery disease treated either with CABG or stent, were identified within the database of the German bureau of statistics. Logistic or linear regression models were used with 20 baseline patient characteristics as potential confounders to compare both regimens. Results In 2018, 1318 cases with left main stenosis were treated with CABG and 8,920 with stent. Patients assigned for stenting were older (72.58 ± 9.87 vs. 68.63 ± 9.40, p p p = 0.748; aOR stroke: 0.59 [0.27; 1.32], p = 0.199). Stent implantation was associated with a reduced risk of relevant bleeding (aOR 0.38 [0.24; 0.61], p p = 0.002), and postoperative delirium (aOR 0.16 [0.11; 0.22], p p p Conclusion Left main stenting is a safe and effective treatment option for CCS-patients suffering from left main coronary artery disease at reasonable economic cost. Graphic abstract

Previous studies have provided limited evidence for the effect of tomato intake on bladder cancer incidence. This study aimed to evaluate the association between dietary tomato or lycopene consumption and bladder cancer risk in the Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) Screening study. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox regression model adjusting for confounders. After a median of 12.5 years of follow-up, 774 incident bladder cancer cases were identified. We found no statistically significant association between dietary intake of raw tomatoes and bladder cancer risk (Adjusted model: HRQ5 VS Q1 = 1.20, 95% CI: 0.95-1.52; P for trend = 0.243). Dietary intakes of tomato catsup, tomato salsa and tomato juice were also not associated with the risk of bladder cancer (all P for trend > 0.05). There was no statistically significant association between dietary consumption of lycopene and bladder cancer risk (Adjusted model: HRQ5 vs. Q1 = 1.04, 95% CI 0.82-1.33; P for trend = 0.590). In summary, analysis of the PLCO study suggested that dietary consumption of tomato or lycopene was not associated with the risk of bladder cancer.

Prostate adenocarcinoma is one of the leading adult malignancies. Identification of multiple causative biomarkers is necessary and helpful for determining the occurrence and prognosis of prostate adenocarcinoma. We aimed to identify the potential prognostic genes in the prostate adenocarcinoma microenvironment and to estimate the causal effects simultaneously. We obtained the gene expression data of prostate adenocarcinoma from TCGA project and identified the differentially expressed genes based on immune-stromal components. Among these genes, 68 were associated with biochemical recurrence at 3 years after prostatectomy in prostate adenocarcinoma. After adjusting for the minimal sets of confounding covariates, 14 genes (TNFRSF4, ZAP70, ERMN, CXCL5, SPINK6, SLC6A18, CHRM2, TG, CLLU1OS, POSTN, CTSG, NETO1, CEACAM7, and IGLV3-22) related to the microenvironment were identified as prognostic biomarkers using the targeted maximum likelihood estimation. Both the average and individual causal effects were obtained to measure the magnitude of the effect. CIBERSORT and gene set enrichment analyses showed that these prognostic genes were mainly associated with immune responses. POSTN and NETO1 were correlated with androgen receptor expression, a main driver of prostate adenocarcinoma progression. Finally, five genes were validated in another prostate adenocarcinoma cohort (GEO: GSE70770). These findings might lead to the improved prognosis of prostate adenocarcinoma.

Purpose Selenium has been suggested to be protective against breast cancer, but the evidence remains inconclusive. Hence, it is important to further examine the potential protective effect. This prospective cohort study investigates pre-diagnostic selenium intake in relation to breast cancer risk. In addition, we analyze serum selenium as a marker of dietary intake. Methods This study includes 17,035 women in the Malmö Diet and Cancer cohort. Dietary assessment and serum samples were collected at baseline (1991–1996). During 344,584 person-years of follow-up, 1,427 incident cases were retrieved. Cox regression analysis examined breast cancer risks adjusted for potential confounding factors. In addition, odds ratios (ORs) were estimated for 1186 cases and an equal number of controls in relation to quartiles (Q) of selenium intake and groups consisting of a combination of intake and serum selenium levels. Results No overall association between selenium intake, or a combination of intake and serum levels, and breast cancer risk was found. The adjusted relative risk for breast cancer in selenium intake Q4 versus Q1 was 0.96 (0.83–1.12) (Ptrend = 0.65). Similarly, adjusted the OR for breast cancer in selenium intake for Q4 versus Q1 was 0.97 (0.76–1.23). The kappa value, 0.096 (p = 0.001), showed poor agreement between serum selenium and selenium intake. Conclusion Our findings suggest that there is no overall association between selenium intake, or a combination of intake and serum levels, and breast cancer risk. Finally, our results showed a poor correlation between estimated selenium intake and serum selenium.

Objective:To examine the associations between the level of adherence to the Mediterranean diet (MedDiet) with obesity, insulin resistance (IR), metabolic syndrome (MetS) and its components in schoolchildren.Design:The Healthy Growth Study was a large epidemiological cross-sectional study.Setting:School children who were enrolled in primary schools in four counties covering the northern, southern, western and central part of Greece were invited to participate.Participants:The study was conducted with a representative sample of 9–13-year-old schoolchildren (n 1972) with complete data. This study applied the KIDMed score to determine ‘poor’ (≤3), ‘medium’ (4-7) and ‘high’ (≥8) adherence of children to the MedDiet. The research hypothesis was examined using multivariate logistic regression models, controlling for potential confounders.Results:The percentage of children with ‘poor’, ‘medium’ and ‘high’ adherence to the MedDiet was 64·8 %, 34·2 % and 1 %, respectively. Furthermore, the prevalence of obesity, IR and MetS was 11·6 %, 28·8 % and 3·4 %, respectively. Logistic regression analyses revealed that ‘poor’ adherence to the MedDiet was associated with an increased likelihood for central obesity (OR 1·31; 95 % CI 1·01, 1·73), hypertriglyceridaemia (OR 2·80; 95 % CI 1·05, 7·46) and IR (OR 1·31; 95 % CI 1·05, 1·64), even after adjusting for several potential confounders.Conclusions:The present study showed that approximately two-thirds of the examined sample of schoolchildren in Greece have ‘poor’ adherence to the MedDiet, which also increases the likelihood for central obesity, hypertriglyceridaemia and IR. Prospective studies are needed to confirm whether these are cause–effect associations.

Objective:To explore the relationship between parameters of Na and K excretion using 24-h urine sample and mild cognitive impairment (MCI) in general population.Design:This is a cross-sectional study.Setting:Community-based general population in Emin China.Participants:Totally, 1147 subjects aged ≥18 years were selected to complete the study, with a multistage proportional random sampling method. Cognitive status was assessed with Mini Mental State Examination (MMSE) questionnaire and timed 24-h urine specimens were collected. Finally, 561 participants aged ≥35 years with complete urine sample and MMSE data were included for the current analysis and divided into groups by tertiles of 24-h urinary sodium to potassium ratio (24-h UNa/K) as lowest (T1), middle (T2) and highest (T3) groups.Results:The MMSE score was significantly lower in T3, compared with the T1 group (26·0 v. 25·0, P = 0·002), and the prevalent MCI was significantly higher in T3 than in T1 group (11·7 % v. 25·8 %, P < 0·001). In multiple linear regression, 24-UNa/K (β: −0·184, 95 % CI −0·319, −0·050, P = 0·007) was negatively associated with MMSE score. In multivariable logistic regression, compared with T1 group, 24-h UNa/K in the T2 and T3 groups showed 2·01 (95 % CI 1·03, 3·93, P = 0·041) and 3·38 (95 % CI 1·77, 6·44, P < 0·001) fold odds for presence of MCI, even after adjustment for confounders. More augmented results were demonstrated in sensitivity analysis by excluding individuals taking anti-hypertensive agents.Conclusions:Higher 24-h UNa/K is in an independent association with prevalent MCI.

We aim to evaluate the risk of stroke recurrence among non-diabetes mellitus (non-DM), previously diagnosed diabetes mellitus (PDDM), newly diagnosed diabetes mellitus-related hyperglycemia (NDDM-RH) and stress hyperglycemia after minor stroke or TIA. Totally, 3026 patients with baseline fasting glucose and glycated albumin from the CHANCE trial (Clopidogrel in High-Risk Patients with Acute Nondisabling Cerebrovascular Events) were included. Patients were classified as non-DM, PDDM, NDDM-RH and stress hyperglycemia according to the status of glucose metabolism. The primary outcome was stroke recurrence during 90-day follow up. Cox regression was performed to estimate the relationship between the status of glucose metabolism and risk of 90-day stroke recurrence. Compared with PDDM, NDDM-RH had a similar risk of 90-day stroke recurrence (hazard ratios [HR]1.39, 95% confidence intervals [CI] 0.94-2.04), while stress hyperglycemia had approximately a 5.3-fold increased risk of 90-day stroke recurrence after adjusted for confounding covariates (HR 5.32, 95% CI 3.43-8.26). Parallel results were found for 90-day recurrent ischemic stroke and composite events. Compared with PDDM in minor stroke or TIA, a parallel risk of 90-day stroke recurrence were observed for NDDM-RH, while stress hyperglycemia might relate to higher risk of 90-day stroke recurrence.