Breast density and risk of breast cancer: masking and detection bias.

Breast density is associated with risk of breast cancer (BC) diagnosis, impacting risk prediction tools and patient notification policies. Density affects mammography sensitivity and may influence screening intensity. Therefore, the observed association between density and BC diagnosis may not reflect the relationship between density and disease risk. We investigate the association between breast density and BC risk using data sourced from 33,542 women in the Breast Cancer Surveillance Consortium, 2000-2018. We estimated mammogram sensitivity and rates of screening mammography among dense (BI-RADS c, d) and non-dense (BI-RADS a, b) breasts. We used Kaplan-Meier estimates to summarize the relative risks of BC diagnosis (RRdx) by density and fit a natural history model to estimate the relative risks of BC onset (RRonset) given density-specific sensitivities. RRdx for dense versus non-dense breasts was 1.80 (95% CI 1.46 to 2.57). Based on estimated screening sensitivities of 0.88 and .78 for non-dense and dense breasts, respectively, RRonset was 1.73 (95% CI 1.43 to 2.25). Sensitivity analyses suggested higher breast density is robustly associated with increased risk of BC onset, similar in magnitude to the increased risk of BC diagnosis. These finding support laws requiring notifications to women with dense breasts of their increased BC risk.

An easily implemented single-visit survey method for intermittently available and imperfectly detectable wildlife applied to the Florida east coast diamondback terrapin (Malaclemys terrapin tequesta).

Single-visit surveys of plots are often used for estimating the abundance of species of conservation concern. Less-than-perfect availability and detection of individuals can bias estimates if not properly accounted for. We developed field methods and a Bayesian model that accounts for availability and detection bias during single-visit visual plot surveys. We used simulated data to test the accuracy of the method under a realistic range of generating parameters and applied the method to Florida's east coast diamondback terrapin in the Indian River Lagoon system, where they were formerly common but have declined in recent decades. Simulations demonstrated that the method produces unbiased abundance estimates under a wide range of conditions that can be expected to occur in such surveys. Using terrapins as an example we show how to include covariates and random effects to improve estimates and learn about species-habitat relationships. Our method requires only counting individuals during short replicate surveys rather than keeping track of individual identity and is simple to implement in a variety of point count settings when individuals may be temporarily unavailable for observation. We provide examples in R and JAGS for implementing the model and to simulate and evaluate data to validate the application of the method under other study conditions.

Risk of cancer versus risk of cancer diagnosis? Accounting for diagnostic bias in predictions of breast cancer risk by race and ethnicity.

OBJECTIVES: Cancer risk prediction may be subject to detection bias if utilization of screening is related to cancer risk factors. We examine detection bias when predicting breast cancer risk by race/ethnicity. METHODS: We used screening and diagnosis histories from the Breast Cancer Surveillance Consortium to estimate risk of breast cancer onset and calculated relative risk of onset and diagnosis for each racial/ethnic group compared with non-Hispanic White women. RESULTS: Of 104,073 women aged 40-54 receiving their first screening mammogram at a Breast Cancer Surveillance Consortium facility between 2000 and 2018, 10.2% (n = 10,634) identified as Asian, 10.9% (n = 11,292) as Hispanic, and 8.4% (n = 8719) as non-Hispanic Black. Hispanic and non-Hispanic Black women had slightly lower screening frequencies but biopsy rates following a positive mammogram were similar across groups. Risk of cancer diagnosis was similar for non-Hispanic Black and White women (relative risk vs non-Hispanic White = 0.90, 95% CI 0.65 to 1.14) but was lower for Asian (relative risk = 0.70, 95% CI 0.56 to 0.97) and Hispanic women (relative risk = 0.82, 95% CI 0.62 to 1.08). Relative risks of disease onset were 0.78 (95% CI 0.68 to 0.88), 0.70 (95% CI 0.59 to 0.83), and 0.95 (95% CI 0.84 to 1.09) for Asian, Hispanic, and non-Hispanic Black women, respectively. CONCLUSIONS: Racial/ethnic differences in mammography and biopsy utilization did not induce substantial detection bias; relative risks of disease onset were similar to or modestly different than relative risks of diagnosis. Asian and Hispanic women have lower risks of developing breast cancer than non-Hispanic Black and White women, who have similar risks.

Healthcare system engagement and algorithm-identified cancer incidence following initiation of a new medication.

PURPOSE: Implausibly high algorithm-identified cancer incidence within a new user study after medication initiation may result from increased healthcare utilization (HU) around initiation ("catch-up care") that increases diagnostic opportunity. Understanding the relationships between HU prior to and around initiation and subsequent cancer rates and timing is important to avoiding protopathic bias. METHODS: We identified a cohort of 417 458 Medicare beneficiaries (2007-2014) aged ≥66 initiating an antihypertensive (AHT) after ≥180 days of non-use. Initiators were stratified into groups of 0, 1, 2-3, and ≥4 outpatient visits (OV) 60-360 days before initiation. We calculated algorithm-identified colorectal cancer (aiCRC) rates stratified by OVs and time since AHT initiation: (0-90, 91-180, 181-365, 366-730, and 731+ days). We summarized HU -360/+60 days around AHT initiation by aiCRC timing: (0-29, 30-89, 90-179, and ≥180 days). RESULTS: AiCRC incidence (311 per 100 000 overall) peaked in the first 0-90 days, was inversely associated with HU before initiation, and stabilized ≥180 days after AHT initiation. Catch-up care was greatest among persons with aiCRCs identified <30 days in follow-up. Catch-up care magnitude decreased as time to the aiCRC date increased, with aiCRCs identified ≥180 days after AHT initiation exhibiting similar HU compared with the full cohort. CONCLUSION: Lower HU before-and increased HU around AHT initiation-seem to drive excess short-term aiCRC incidence. Person-time and case accrual should only begin when incidence stabilizes. When comparison groups within a study differ by HU, outcome-detection bias may exist. Similar observations may exist in other settings when typical HU is delayed (e.g., cancer screening during SARS-CoV-2).

The relationship between prior psychiatric diagnosis and brain cancer diagnosis in the U.S. military health system.

PURPOSE: Prior research suggested the increased likelihood of brain cancer diagnosis following certain psychiatric diagnoses. This association may result from detection bias or suggest an early sign for brain cancer. This study investigated whether psychiatric illness may be an early manifestation of brain cancer while considering potential effects of detection bias. METHODS: This case-control study used the data from the Department of Defense's Central Cancer Registry and the Military Health System Data Repository. Four cancer-free controls and one negative-outcome control (cancers not associated with psychiatric illness) were matched to each brain cancer case diagnosed from 1998 to 2013 by age, sex, race, and military status. The groups were compared in the likelihood of having a pre-existing psychiatric diagnosis using conditional logistic regression. RESULTS: We found a significant association of psychiatric illnesses with brain cancer (Odds Ratio (OR) = 2.63, 95% confidence interval (CI) = 2.18-3.16) and other cancers (OR = 1.80, 95% CI = 1.49-2.19), compared to non-cancer controls. The association was stronger for psychiatric diagnoses within three months before cancer (brain cancer: OR = 26.77, 95% CI = 15.40-46.53; other cancers: OR = 4.12, 95% CI = 1.96-8.65). The association with psychiatric disorders within 3 months were higher for small brain tumors (OR = 128.32, 95% CI = 17.28-952.92 compared to non-cancer controls) while the OR was 2.79 for other cancers (95% CI = 0.86-8.99 compared to non-cancer controls). CONCLUSION: Our findings suggest an association between diagnosed psychiatric illnesses and subsequent brain cancer diagnosis, which may not be solely explained by detection bias. Psychiatric illness might be a sign for early detection of brain cancer beyond the potential effects of detection bias.

Inadequate reporting of adjudicators in open-label trials of anticancer drugs between 2017 and 2021: a methodological review.

OBJECTIVES: In open-label trials, the details of the adjudicators are essential to evaluate the risk of detection bias. We aimed to describe how the adjudicators of progression-free survival (PFS) and objective response rate (ORR) have been reported in open-label trials of anticancer drugs. STUDY DESIGN AND SETTING: A literature search was conducted using MEDLINE via PubMed. We included open-label, parallel-group superiority randomized trials that investigated the PFS and ORR of anticancer drugs for solid tumors. After screening based on the titles and abstracts, 200 articles were randomly selected from 2017 to 2021. The researchers independently checked the eligibility and collected the adjudicators' information in the protocol, registry, and original article. RESULTS: One hundred fifty five studies reported the PFS and ORR. Approximately half of the studies did not report adjudicators (47.7% in PFS and 47.6% in ORR) in the published articles. The inconsistency between the protocol/registry and the published article was 31.0% for PFS and 33.5% for ORR. The prespecified outcomes were not reported in 5.2% of the studies evaluating PFS and 4.5% evaluating ORR. CONCLUSION: This methodological review found that adjudicators were poorly and inconsistently reported between the protocol/registry and the final publication in open-label trials of anticancer drugs.

Influence of abstinence on automatic detection bias to gaming cues in individuals with Internet gaming disorder: Evidence from visual mismatch negativity.

Internet gaming disorder (IGD) has been classified by the DSM-5 as a condition for further study, and many studies have shown that the occurrence and maintenance of IGD and the automatic detection bias to gaming cues of individuals with IGD may be significantly related. However, whether abstinence, a common intervention method in behavioral addiction, can adjust the automatic detection bias in individuals with IGD and its underlying neural mechanisms is unclear. In this study, we investigated the effects of 7 days of abstinence from gaming on automatic detection bias, negative affect and craving in individuals with IGD with event-related potential technology. A total of 50 IGD subjects were recruited in this study and randomly divided into abstinence and control groups. Visual mismatch negativity (vMMN) was induced using a standard-deviant reversed oddball paradigm, and differences in automatic detection bias, negative affect, and craving between the two groups were recorded and compared at baseline, day 3, and day 7. The results showed that compared with baseline, vMMN, negative affect and craving were significantly enhanced on day 3 and significantly diminished on day 7 in the abstinence group but not in the control group, and the changes in vMMN were significantly correlated with changes in negative affect and craving in both groups. This study demonstrates that abstinence can restore automatic detection bias in individuals with IGD and that abstinence has to last for 7 days to have a significant effect, while recovery may be related to the negative affect and craving.

Lung cancer symptom appraisal, help-seeking and diagnosis - rapid systematic review of differences between patients with and without a smoking history.

BACKGROUND: Lung cancer is the leading cause of cancer death in the world. A significant minority of lung cancer patients have never smoked (14% in the UK, and ranging from 10% to 25% worldwide). Current evidence suggests that never-smokers encounter delays during the diagnostic pathway, yet it is unclear how their experiences and reasons for delayed diagnoses differ from those of current and former smokers. This rapid review assessed literature about patient experiences in relation to symptom awareness and appraisal, help-seeking, and the lung cancer diagnostic pathway, comparing patients with and without a smoking history. METHODS: MEDLINE, PsychINFO and Google Scholar were searched for studies (2010-2020) that investigated experiences of the pathway to diagnosis for patients with and without a smoking history. Findings are presented using a narrative synthesis. RESULTS: Analysis of seven quantitative and three qualitative studies revealed that some delays during symptom appraisal and diagnosis are unique to never-smokers. Due to the strong link between smoking and lung cancer, and low awareness of non-smoking related lung cancer risk factors and symptoms, never-smokers do not perceive themselves to be at risk. Never-smokers are also likely to evaluate their experiences in comparison with other non-smoking related cancers, where prognosis is likely better, potentially leading to lower satisfaction with healthcare. CONCLUSION: Never-smokers appear to have different experiences in relation to symptom appraisal and diagnosis. However, evidence in relation to help-seeking, and what is driving diagnostic delays for never-smoker patients specifically is lacking.

Assessing the risk of performance and detection bias in Cochrane reviews as a joint domain is less accurate compared to two separate domains.

BACKGROUND: Initially, the Cochrane risk of bias (RoB) tool had a domain for "blinding of participants, personnel and outcome assessors". In the 2011 tool, the assessment of blinding was split into two domains: blinding of participants and personnel (performance bias) and blinding of outcome assessors (detection bias). The aims of this study were twofold; first, to analyze the frequency of usage of the joint blinding domain (a single domain for performance and detection bias), and second, to assess the proportion of adequate assessments made in the joint versus single RoB domains for blinding by comparing whether authors' RoB judgments were supported by explanatory comments in line with the Cochrane Handbook recommendations. METHODS: We extracted information about the assessment of blinding from RoB tables (judgment, comment, and whether it was specified which outcome type; e.g., objective, subjective) of 729 Cochrane reviews published in 2015-2016. In the Cochrane RoB tool, judgment (low, unclear or high risk) needs to be accompanied by a transparent comment, in which authors provide a summary justifying RoB judgment, to ensure transparency in how these judgments were reached. We reassessed RoB based on the supporting comments reported in Cochrane RoB tables, in line with instructions from the Cochrane Handbook. Then, we compared our new assessments to judgments made by Cochrane authors. We compared the frequency of adequate judgments in reviews with two separate domains for blinding versus those with a joint domain for blinding. RESULTS: The total number of assessments for performance bias was 6918, with 8656 for detection bias and 3169 for the joint domain. The frequency of adequate assessments was 74% for performance bias, 78% for detection bias, and 59% for the joint domain. The lowest frequency of adequate assessments was found when Cochrane authors judged low risk - 47% in performance bias, 62% in detection bias, and 31% in the joint domain. The joint domain and detection bias domain had a similar proportion of specified outcome types (17% and 18%, respectively). CONCLUSIONS: Splitting joint RoB assessment about blinding into two domains was justified because the frequency of adequate judgments was higher in separate domains. Specification of outcome types in RoB domains should be further scrutinized.

Tuberculosis-associated depression: a population-based cohort study of people immigrating to British Columbia, Canada, 1985-2015.

PURPOSE: To estimate the risk of tuberculosis (TB)-associated depression. A second aim was to estimate the extent to which any increased risk of depression among TB patients may be mediated by the length of hospital length stay (LOS) METHODS: Retrospective cohort study of linked healthcare claims and public health surveillance data. Our primary outcome, time-to-depression, was analyzed using Cox proportional hazards (PH) regressions. Causal mediation analysis was used to estimate the natural direct and indirect effect of TB mediated by hospital LOS. RESULTS: Among 755,836 participants (52.2% female, median age=35 years, median follow-up=8.75 years), 2295 were diagnosed with TB (exposure), and 128,963 were diagnosed with depression (outcome). We observed a covariate-adjusted hazard ratio (aHR) of 1.24 (95% CI, 1.14-1.34) for depression by TB. The total effect of TB on depression was decomposed into a natural direct effect of TB of aHR=1.11 (95% CI, 1.02-1.21) and an indirect effect through hospital LOS of aHR=1.11 (95% CI, 1.10-1.12), indicating that TB's total effect was mediated by 50% (95% CI, 35-82%) through hospital LOS. CONCLUSIONS: TB patients had a 24% higher risk of developing depression. TB's effect was mediated substantially by hospital LOS, requiring further study. Depression screening among TB patients is warranted.

A Latent Disease Model to Reduce Detection Bias in Cancer Risk Prediction Studies.

In studies of cancer risk, detection bias arises when risk factors are associated with screening patterns, affecting the likelihood and timing of diagnosis. To eliminate detection bias in a screened cohort, we propose modeling the latent onset of cancer and estimating the association between risk factors and onset rather than diagnosis. We apply this framework to estimate the increase in prostate cancer risk associated with black race and family history using data from the SELECT prostate cancer prevention trial, in which men were screened and biopsied according to community practices. A positive family history was associated with a hazard ratio (HR) of prostate cancer onset of 1.8, lower than the corresponding HR of prostate cancer diagnosis (HR = 2.2). This result comports with a finding that men in SELECT with a family history were more likely to be biopsied following a positive PSA test than men with no family history. For black race, the HRs for onset and diagnosis were similar, consistent with similar patterns of screening and biopsy by race. If individual screening and diagnosis histories are available, latent disease modeling can be used to decouple risk of disease from risk of disease diagnosis and reduce detection bias.

Use of Warfarin or Direct Oral Anticoagulants and Risk of Prostate Cancer in PCBaSe: A Nationwide Case-Control Study.

Existing literature examining warfarin's association with prostate cancer (PCa) risk provides conflicting results, while the association with direct oral anticoagulants (DOACs) has not yet been studied. We investigated the association of warfarin and DOAC use on PCa risk among men within the population-based Prostate Cancer database Sweden (PCBaSe), using a case-control design. The study population included PCa cases diagnosed 2014-2016 and five age-matched PCa-free controls. Conditional logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CI) for PCa associated with warfarin and DOAC use, adjusted for marital status, education level, other drug use, and comorbidities. Among 31,591 cases and 156,802 controls, there were 18,522 (9.8%) warfarin and 4,455 (2.4%) DOAC users. Warfarin ever-use was associated with reduced risk of PCa overall (OR 0.92 95% CI 0.88-0.96) as were both past and current use. DOAC use was not associated with PCa risk. For some warfarin exposures, decreased risk was observed for unfavorable PCa (high risk/locally advanced/distant metastatic) but not with favorable PCa (low/intermediate risk). Increased risk of favorable PCa was observed for men whose initial warfarin exposure occurred in the 12 month period before diagnosis (OR 1.39; 95% CI 1.13-1.70). Our findings are consistent with previous publications reporting decreased PCa risk with warfarin exposure. Increased risk of favorable PCa suggests detection bias due to increased prostate specific antigen testing when starting on warfarin. Decreased overall PCa risk could reflect bias due to reduced biopsy rates among long-term warfarin users.

Outcome assessment by central adjudicators in randomised stroke trials: Simulation of differential and non-differential misclassification.

INTRODUCTION: Adjudication of the primary outcome in randomised trials is thought to control misclassification. We investigated the amount of misclassification needed before adjudication changed the primary trial results.Patients (or materials) and methods: We included data from five randomised stroke trials. Differential misclassification was introduced for each primary outcome until the estimated treatment effect was altered. This was simulated 1000 times. We calculated the between-simulation mean proportion of participants that needed to be differentially misclassified to alter the treatment effect. In addition, we simulated hypothetical trials with a binary outcome and varying sample size (1000-10,000), overall event rate (10%-50%) and treatment effect (0.67-0.90). We introduced non-differential misclassification until the treatment effect was non-significant at 5% level. RESULTS: For the five trials, the range of unweighted kappa values were reduced from 0.89-0.97 to 0.65-0.85 before the treatment effect was altered. This corresponded to 2.1%-6% of participants misclassified differentially for trials with a binary outcome. For the hypothetical trials, those with a larger sample size, stronger treatment effect and overall event rate closer to 50% needed a higher proportion of events non-differentially misclassified before the treatment effect became non-significant. DISCUSSION: We found that only a small amount of differential misclassification was required before adjudication altered the primary trial results, whereas a considerable proportion of participants needed to be misclassified non-differentially before adjudication changed trial conclusions. Given that differential misclassification should not occur in trials with sufficient blinding, these results suggest that central adjudication is of most use in studies with unblinded outcome assessment. CONCLUSION: For trials without adequate blinding, central adjudication is vital to control for differential misclassification. However, for large blinded trials, adjudication is of less importance and may not be necessary.

Using Causal Diagrams to Improve the Design and Interpretation of Medical Research.

Causal directed acyclic graphs (cDAGs) have become popular tools for researchers to better examine biases related to causal questions. DAGs comprise a series of arrows connecting nodes that represent variables and in doing so can demonstrate the causal relation between different variables. cDAGs can provide researchers with a blueprint of the exposure and outcome relation and the other variables that play a role in that causal question. cDAGs can be helpful in the design and interpretation of observational studies in pulmonary, critical care, sleep, and cardiovascular medicine. They can also help clinicians and researchers to better identify the structure of different biases that can affect the validity of observational studies. Most of the available literature on cDAGs and their function use language that might be unfamiliar to clinicians. This article explains cDAG terminology and the principles behind how they work. We use cDAGs and clinical examples that are mostly focused in the area of pulmonary medicine to describe the structure of confounding, selection bias, overadjustment bias, and detection bias. These principles are then applied to a more complex published case study on the use of statins and COPD mortality. We also introduce readers to other resources for a more in-depth discussion of causal inference principles.

Assessing risk of bias judgments for blinding of outcome assessors in Cochrane reviews.

Aim: Adequate judging of risk of bias (RoB) for blinding of outcome assessors (detection bias) is important for supporting highest level of evidence. Materials & methods: Judgments and supporting comments for detection bias were retrieved from RoB tables reported in Cochrane reviews. We categorized comments, and then compared judgment and supporting comment with instructions from the Cochrane Handbook. Results: We analyzed 8656 judgments for detection bias from 7626 trials included in 575 reviews. Overall, 1909 judgments (22%) were not in line with the Cochrane Handbook. In 9% of trials, the authors split the detection bias domain according to outcomes. Here, prevalence of inadequate judgments was 19%. Conclusion: Interventions to improve RoB assessments in systematic reviews should be explored.

Deep learning COVID-19 detection bias: accuracy through artificial intelligence.

BACKGROUND: Detection of COVID-19 cases' accuracy is posing a conundrum for scientists, physicians, and policy-makers. As of April 23, 2020, 2.7 million cases have been confirmed, over 190,000 people are dead, and about 750,000 people are reported recovered. Yet, there is no publicly available data on tests that could be missing infections. Complicating matters and furthering anxiety are specific instances of false-negative tests. METHODS: We developed a deep learning model to improve accuracy of reported cases and to precisely predict the disease from chest X-ray scans. Our model relied on convolutional neural networks (CNNs) to detect structural abnormalities and disease categorization that were keys to uncovering hidden patterns. To do so, a transfer learning approach was deployed to perform detections from the chest anterior-posterior radiographs of patients. We used publicly available datasets to achieve this. RESULTS: Our results offer very high accuracy (96.3%) and loss (0.151 binary cross-entropy) using the public dataset consisting of patients from different countries worldwide. As the confusion matrix indicates, our model is able to accurately identify true negatives (74) and true positives (32); this deep learning model identified three cases of false-positive and one false-negative finding from the healthy patient scans. CONCLUSIONS: Our COVID-19 detection model minimizes manual interaction dependent on radiologists as it automates identification of structural abnormalities in patient's CXRs, and our deep learning model is likely to detect true positives and true negatives and weed out false positive and false negatives with > 96.3% accuracy.

Expected-value bias in routine third-trimester growth scans.

OBJECTIVES: Operators performing fetal growth scans are usually aware of the gestational age of the pregnancy, which may lead to expected-value bias when performing biometric measurements. We aimed to evaluate the incidence of expected-value bias in routine fetal growth scans and assess its impact on standard biometric measurements. METHODS: We collected prospectively full-length video recordings of routine ultrasound growth scans coupled with operator eye tracking. Expected value was defined as the gestational age at the time of the scan, based on the estimated due date that was established at the dating scan. Expected-value bias was defined as occurring when the operator looked at the measurement box on the screen during the process of caliper adjustment before saving a measurement. We studied the three standard biometric planes on which measurements of head circumference (HC), abdominal circumference (AC) and femur length (FL) are obtained. We evaluated the incidence of expected-value bias and quantified the impact of biased measurements. RESULTS: We analyzed 272 third-trimester growth scans, performed by 16 operators, during which a total of 1409 measurements (354 HC, 703 AC and 352 FL; including repeat measurements) were obtained. Expected-value bias occurred in 91.4% of the saved standard biometric plane measurements (85.0% for HC, 92.9% for AC and 94.9% for FL). The operators were more likely to adjust the measurements towards the expected value than away from it (47.7% vs 19.7% of measurements; P < 0.001). On average, measurements were corrected by 2.3 ± 5.6, 2.4 ± 10.4 and 3.2 ± 10.4 days of gestation towards the expected gestational age for the HC, AC, and FL measurements, respectively. Additionally, we noted a statistically significant reduction in measurement variance once the operator was biased (P = 0.026). Comparing the lowest and highest possible estimated fetal weight (using the smallest and largest biased HC, AC and FL measurements), we noted that the discordance, in percentage terms, was 10.1% ± 6.5%, and that in 17% (95% CI, 12-21%) of the scans, the fetus could be considered as small-for-gestational age or appropriate-for-gestational age if using the smallest or largest possible measurements, respectively. Similarly, in 13% (95% CI, 9-16%) of scans, the fetus could be considered as large-for-gestational age or appropriate-for-gestational age if using the largest or smallest possible measurements, respectively. CONCLUSIONS: During routine third-trimester growth scans, expected-value bias frequently occurs and significantly changes standard biometric measurements obtained. © 2019 the Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Detection bias:structural analysis of different study designs / 上海预防医学

Detection bias is an information bias.It was first proposed by Horwitz from the study investigating the association of the administration of estrogen after menopause with the occurrence of endometrial cancer, which still prevails in most epidemiological studies.We use the Directed Acyclic Graph to analyze the effect of a given exposure on a specific outcome with the association estimates between the measured exposure and outcome.Detection bias occurs when there is additional open paths irrelevant to the target path of interest which is originated from measured exposure to measured outcome.We further analyzed how the detection bias was formed under different study designs, including cohort study, randomized clinical trial and case-control study in order to further investigate its potential influence on the effect/association estimation.

Detection bias:structural analysis of different study designs / 上海预防医学

Detection bias is an information bias.It was first proposed by Horwitz from the study investigating the association of the administration of estrogen after menopause with the occurrence of endometrial cancer, which still prevails in most epidemiological studies.We use the Directed Acyclic Graph to analyze the effect of a given exposure on a specific outcome with the association estimates between the measured exposure and outcome.Detection bias occurs when there is additional open paths irrelevant to the target path of interest which is originated from measured exposure to measured outcome.We further analyzed how the detection bias was formed under different study designs, including cohort study, randomized clinical trial and case-control study in order to further investigate its potential influence on the effect/association estimation.

Evidence-based recommendations for blinding in surgical trials.

RATIONALE: Blinding reduces performance and detection bias in randomized controlled trials (RCT). There is evidence that lack of blinding leads to overestimation of treatment effects in pharmacological trials. Since surgical trials use interventions with a physical component, blinding is often complicated. The aim of this study was to analyze, in general and abdominal surgery RCT, the status of blinding, the potential for blinding, and the influence of blinding on outcomes. METHODS: A systematic search of the literature in CENTRAL, MEDLINE, and Web of Science was conducted to identify RCT with a surgical intervention, starting in 1996, the year when the first CONSORT statement was published. Information on general study characteristics and blinding methods was extracted. The presence or absence of blinding of the study contributors-patients, surgeons, data collectors, outcome assessors, and data analysts-was analyzed. The association of blinding with the trial outcome was investigated for every study contributor. RESULTS: Out of 29,119 articles, 378 RCT were included in the analysis. These investigated a total of 62,522 patients, of whom 15,025 were blinded (24.0%). Contributors could have been blinded in far more trials, as the potential for blinding measures ranged from 69% for outcome assessors to 98% for data analysts. If blinding of surgeons would have been possible but had not been performed, this was associated with more significant trial outcomes (OR 13.670; 95% CI 1.308 to 142.840; p = 0.0289). DISCUSSION: The potential of blinding, an important quality measure in surgical RCT, has not been exhausted. This study summarizes the existing evidence on blinding in surgical trials and gives evidence-based recommendations for the use of blinding in future surgical trials. SYSTEMATIC REVIEW REGISTRATION: PROSPERO 2015:CRD42015026837.