Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Data Accuracy”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62Linked to original sources
PubMed2000

Accuracy of CT angiography versus pulmonary angiography in the diagnosis of acute pulmonary embolism: evaluation of the literature with summary ROC curve analysis.

RATIONALE AND OBJECTIVES: The authors performed this study to estimate, by using published data, the sensitivity and specificity of computed tomographic (CT) angiography in the evaluation of suspected acute pulmonary embolism (PE). MATERIALS AND METHODS: Summary receiver operating characteristic (ROC) curve analysis was used to determine the sensitivity and specificity of CT angiography in the diagnosis of acute PE. Pulmonary angiography was used as the diagnostic standard of reference. The authors reviewed the results of 11 independent studies published in the English-language literature between January 1992 and June 1999. RESULTS: The sensitivity of CT angiography in the diagnosis or exclusion of PE in the central pulmonary arteries (to the level of the segmental pulmonary arteries) ranged from 0.74 to 0.81 on the basis of specificities of 0.89-0.91. The sensitivity of CT angiography in the diagnosis or exclusion of PE in all pulmonary arteries (to the level of the subsegmental pulmonary arteries) was 0.68 on the basis of a specificity of 0.91. CONCLUSION: On the basis of the studies in the current literature, most of which used 5.0-mm collimation and single-detector CT, CT angiography may be less accurate in the diagnosis of PE than previously reported. With improvements in data acquisition, particularly the use of thinner section collimation and multidetector CT, and in the increased use of workstations for data analysis, the accuracy and utility of CT angiography will require continued investigation.

Angiography↗
PubMed2001

Non-invasive methods for measuring data quality in general practice.

AIM: To develop non-invasive methods of measuring the quality of data recorded in general practice. METHODS: Laboratory and pharmaceutical claims data from fourteen practices (44 doctors) from the FirstHealth network of general practices were examined to determine the extent to which valid minimum bounds on expected rates of diagnosis coding could be established. These were compared with recorded rates in patient notes to measure completeness of diagnosis recording. Data completeness was measured for demographic data and a marker for the accuracy of gender coding was developed from diagnosis data. RESULTS: Minimum rates of diagnosis could be established for asthma, diabetes (NIDDM and IDDM), ischaemic heart disease, hypothyroidism, bipolar affective disorder and Parkinson's disease. Minimum bounds for the number of patients requiring monitoring of warfarin and digoxin levels were also established. These expected minimum rates were combined with measures of completeness of age, gender, ethnicity and smoking data, and a gender coding accuracy measure, to produce a set of fourteen data quality indicators. Pass/fail thresholds on each indicator were set and each of the fourteen practices was scored on the number of passes they achieved. The scores ranged from three to nine out of fourteen passses. CONCLUSIONS: Non-invasive data quality measures may be useful in providing feedback to general practitioners as part of a data quality improvement cycle. The sensitivity of this method will decline as data quality improves.

Data Collection↗
PubMed2006

Evaluation of three-dimensional image registration methodologies for in vivo micro-computed tomography.

The advent of in vivo micro-computed tomography (micro-CT) provides a novel approach to measure the temporal adaptation of bone micro-architecture within an individual. Spatial alignment in the scanner between serial scans is challenging, but three-dimensional image registration can be used to superimpose the resulting image data, thus ensuring consistent regions of interest (ROI) for analysis. There have been several approaches to image registration developed, yet little is known about their application to high resolution micro-CT data. The purpose of this study was to explore combinations of three image registration similarity measures and three image interpolators, in addition to multi-resolution registration configurations, for assessment of computational efficiency and accuracy on both in vitro and in vivo micro-CT data. Accuracy measures were assessed by comparison with a gold-standard reference transform based on attached fiducial markers. It was concluded that a mutual information registration similarity measure with a linear image interpolator, applied at steps of increasing image resolution, provided the best compromise between accurate and efficient results. In vivo registration of tibial bone microstructure measured in an ovariectomized rat model provided consistent ROI thus demonstrating the usefulness of three-dimensional image registration for in vivo experimental and clinical micro-CT research. It is a technique that is poised to become commonly utilized for analysis of micro-CT data to diagnose and monitor efficacy of therapy in bone diseases.

Animals↗
PubMed2000

An algorithm for the prediction of proteasomal cleavages.

Proteasomes, major proteolytic sites in eukaryotic cells, play an important part in major histocompatibility class I (MHC I) ligand generation and thus in the regulation of specific immune responses. Their cleavage specificity is of outstanding interest for this process. In order to generalize previously determined cleavage motifs of 20 S proteasomes, we developed network-based model proteasomes trained by an evolutionary algorithm with experimental cleavage data of yeast and human 20 S proteasomes. A window of ten flanking amino acid residues proved sufficient for the model proteasomes to reproduce the experimental results with 98-100 % accuracy. Actual experimental data were reproduced significantly better than randomly selected cleavage sites, suggesting that our model proteasomes were able to extract rules inherent to proteasomal cleavage data. The affinity parameters of the model, which decide for or against cleavage, correspond with the cleavage motifs determined experimentally. The predictive power of the model was verified for unknown (to the program) test conditions: the prediction of cleavage numbers in proteins and the generation of MHC I ligands from short peptides. In summary, our model proteasomes reproduce and predict proteasomal cleavages with high degree of accuracy. They present a promising approach for predicting proteasomal cleavage products in future attempts and, in combination with existing algorithms for MHC I ligand prediction, will be tested to improve cytotoxic T lymphocyte epitope prediction.

Algorithms↗
PubMed1994

Measuring diagnostic accuracy in the absence of a "gold standard".

OBJECTIVE: As the nation debates issues of national health care reform, psychiatrists seek equal status with other medical colleagues. To defend psychiatry in the national arena, the accuracy of psychiatric diagnoses must be measured. Indexes of accuracy such as sensitivity and specificity provide valuable information, yet they are rarely computed because there is no "gold standard" with which to compare them. The goal of this article is to show how this problem can be overcome and to encourage nosologists to use accuracy statistics in assessing the adequacy of psychiatric diagnoses. METHOD: The authors reviewed the literature on medical decision making to find methodological approaches to assessing diagnostic accuracy in the absence of gold standards. RESULTS: A lack of such standards is not unique to psychiatry and has been addressed with a variety of novel analytic procedures. Although these methods differ in many respects, each recognizes that the conventional 2 x 2 table of interrater agreement does not provide enough data for estimating diagnostic accuracy. After defining the data needed, each method provides a mathematical model that estimates accuracy statistics and the prevalence of a disorder. Most of these methods are variants of latent class analysis. The authors reanalyzed data from one of the reviewed papers to show that similar inferences about accuracy of diagnoses could be drawn from a conventional latent class analysis. CONCLUSIONS: There are potential pitfalls in using latent structure methods, but their cautious use would provide valuable information for psychiatric nosology. These methods supplement, but do not replace, data about outcome, family history, laboratory studies, and other validating criteria in making accurate diagnoses.

Female↗
PubMed2005

Role of feedback in the accuracy of perceived direction of motion-in-depth and control of interceptive action.

We quantified the accuracy of the perception of the absolute direction of motion-in-depth (MID) of a simulated approaching object using a perceptual task and compared those data with the accuracy of estimating the passing distance measured by means of a simulated catching task. For the simulated catching task, movements of the index finger and thumb of the observer's hand were tracked as participants tried to "catch" the simulated approaching object. A sensation of MID was created by providing monocular and/or binocular retinal image information. Visual stimuli were identical for perceptual and simulated catching tasks. We confirm previous reports that in the perceptual task, observers judged the object to pass wider of the head than indicated by the visual information provided. Although accuracy improved when auditory feedback was added to the perceptual (button pressing) task, consistent overestimates were still recorded. For the no-feedback simulated catching task, observers consistently overreached, i.e., the hand was further away from the midline than the simulated object at the time of hand closure. When auditory feedback was added to the simulated catching task successful catching was achieved. The relative accuracy in binocular and monocular conditions for individual observers could be partially explained by individual differences in sensitivity to unidirectional changes in angular size and changes in relative disparity. We conclude that catching an approaching ball requires that errors in the perceived direction of MID are corrected by feedback-driven learning in the motor system, and that this learning is more easily achieved for the catching action than for button pressing.

Auditory Perception↗
PubMed1986

Physical-chemical model for the entry of water-insoluble compounds into cells. Studies of fatty acid uptake by the liver.

The spontaneous transfer of water-insoluble substances from plasma to the interior of cells would involve a series of steps in which the substance of interest dissociates from albumin in plasma, enters the outer half of the plasma membrane of a cell, crosses the bilayer, and then dissociates from the inner half of the plasma membrane to enter cell cytosol and diffuses to sites of its metabolism. We have examined the behavior of long-chain fatty acids in the uptake process, assuming that none of these steps is facilitated by the cell during the entry of fatty acids into the liver. Comparison of the spontaneous rates for each individual step with rates of uptake of fatty acid by perfused liver leads to the conclusion that the uptake of fatty acids is not limited by kinetic factors but is determined instead by the equilibrium distribution (Keq) of fatty acids between albumin in plasma and the phospholipids of the plasma membrane. This idea was examined further by determining whether there was a relationship between the value for Keq and rates of uptake of a fatty acid and the pattern of kinetics for uptake. The data indicate that there is a linear relationship between Keq and the rate of uptake, that uptake rates can be predicted with a high degree of accuracy from thermodynamic data, and that the pattern of kinetics of uptake is compatible with the idea that the uptake rate is determined by the relative affinity of a fatty acid for albumin and membranes.

Animals↗
PubMed2004

The influence of clinical information on the accuracy of diagnostic mammography.

OBJECTIVE: This study examined the influence of knowledge of clinical information on the accuracy of mammography in women referred for investigation of breast symptoms. METHODS: Subjects were sampled from all women consecutively attending a symptomatic breast clinic and aged 25-55 years. This included all 240 women shown to have breast cancer and 240 age-matched women shown not to have cancer. Mammography films were prospectively reported by two radiologists independently of each other in a blinded manner and without knowledge of any clinical information. The films were then re-read with information about the type and site of symptoms (and without knowledge of the level of suspicion of cancer on clinical examination). The accuracy of reading with and without information on symptoms was compared using sensitivity and specificity and receiver operating characteristic (ROC) curves. The effect of age on changes in test accuracy was examined. RESULTS: Reporting the mammogram with knowledge of clinical information (compared to without any information) significantly improved sensitivity (75.8 vs. 71.3%, P = 0.003) for one radiologist, with a non-significant reduction in specificity (85.4 vs. 87.1%, P = 0.22). For the other radiologist, it resulted in non-significant improvement in both sensitivity (75.4 vs. 73.8%, P = 0.13) and specificity (89.2 vs. 87.9%, P = 0.25). Age did not have a statistically significant effect on changes in test accuracy in our data. ROC curves for both radiologists showed that reporting mammography with knowledge of clinical information resulted in small (about 2%) but significant improvement in overall test accuracy. CONCLUSION: Our findings support the provision of clinical information relating to patients' presentation to radiologists reporting diagnostic mammography.

Adult↗
PubMed2000

Reducing DCO registrations through electronic matching of cancer registry data and routine hospital data.

The Thames Cancer Registry (TCR) has registered a high proportion of tumours from death certificate information only (DCO) registrations. This paper describes the results of a study set up to establish whether this proportion could be reduced by linking cancer registrations with routine hospital data from the Hospital Episodes Statistics (HES) data set using computerized matching. A total of 67752 registrations were identified from the TCR. Matches were found in the HES data set for 66%. The proportion of cases retrieved for each tumour site was: 72% for colorectal cancer; 62% for cancer of the lung, trachea or bronchus; and 65% for female breast cancer. For all three tumour sites the proportion of matches found for patients registered from hospital case notes was higher than the proportion found for patients registered as DCOs (P < 0.0001 for all three tumour sites). Among matched DCO cases, 58% had at least one procedure recorded. DCO rates might be reduced by as much as 43% (from 17% of total registrations to less than 10%) for the three most common cancers if the method of electronic matching outlined here was used. Younger age groups, prognosis of tumour site and residence in North Thames region were all positively associated with successful matching (P < 0.0001 in all three cases). Many matched DCO cases were found to have had more than one admission for cancer. Among ordinary in-patient admissions, admissions to patients ratios of 1.5, 1.4 and 1.9 were found for colorectal, lung and breast cancers respectively. Of 5190 matched DCOs a procedure was recorded for 3013 (58%). HES data offer a useful aid to follow-up of case notes on patients identified to the registry by death certificates. Doubts about the completeness and accuracy of HES data mean case notes must remain the 'gold standard'.

Data Collection↗
PubMed1980

Spouse-subject interviews and the reliability of diet studies.

Recently published data suggest relationships between ingestion of a number of food items and risk of cancer of the stomach, bowel and mouth. This engenders concern over the accuracy of such data. To study this, dietary interviews of 158 males in the Western New York Study of Cancer Epidemiology are compared to their spouses' estimates of the husband's dietary histories as taken in separate interviews. Respondents were asked to estimate the frequency with which the males consumed each of several foods. For analysis, the frequency estimates were assembled into categories similar to those used in recent epidemiologic analysis: 5-7 times per week, 1-4 times per week, 1-3 times per month and less than once per month. Generally, 60-80% of respondents pairs agree exactly on the frequency of consumption for individual food items. Over 90% of spouse and respondent food frequency estimates are within one frequency category of each other. Radical disagreements between spouses are rare; generally, in less than 2% of spouse pairs does one member estimate the maximum frequency category, and the other minimum frequency category. The authors suggest therefore that, although diet histories are not precisely replicable, they must be adequate to reveal gross differences between cases and controls.

Adult↗
PubMed1999

Ten-year follow-up evaluation of a school screening program for scoliosis. Is the forward-bending test an accurate diagnostic criterion for the screening of scoliosis?

STUDY DESIGN: A 10-year follow-up evaluation of the effectiveness of school screening for scoliosis performed in a closed island population. OBJECTIVES: To evaluate the diagnostic accuracy of methods used for screening scoliosis and to re-examine the long-term effectiveness of the school scoliosis screening program. SUMMARY OF BACKGROUND DATA: The diagnostic accuracy of the forward-bending test and the long-term efficacy of the screening programs have not been clearly established. METHODS: In 1987, 2700 pupils aged 8 to 16 years from the island of Samos were screened for scoliosis. The Adams forward-bending test, Moire topography, the scoliometer, and the humpometer were used. Radiologic evaluation of the spine was available for each pupil and the number of false-negative and false-positive results of the screening methods was calculated. Subsequently, sensitivity, specificity, and positive and negative predictive values were estimated for each screening technique. Pupils found positive for spinal deformity were then followed up regularly at yearly intervals. In 1997, all positive subjects attended a 10-year clinical and radiologic follow-up, and the remaining subjects were re-evaluated by a postal questionnaire and were clinically examined if necessary. RESULTS: Spinal deformity was found in 153 (5.66%) pupils. Scoliosis (defined as a spinal curvature > or = 10 degrees) was found in 32 pupils, for a prevalence of 1.18%. For scoliosis, the Adams forward-bending test showed a number of false-negative results (in five cases), for a sensitivity of 84.37% and specificity of 93.44%. The sensitivities of Moire topography, the humpometer, and the scoliometer were 100%, 93.75%, and 90.62%, respectively, and specificity was 85.38%, 78.11%, and 79.76% respectively. The negative predictive value of the forward-bending test was inferior to those of the other methods. During this scoliosis screening program, if cutoff limits for referral had been used, such as the asymmetry of two Moire fringes, a humpogram deformity of (D + H) = 10 mm, and 8 degrees of scoliometer angle, it would have been possible to reduce radiologic examination by 89.4%. Three (0.11%) pupils aged between 12 and 14 years with scoliotic deformities greater than 20 degrees underwent satisfactory nonoperative treatment with Boston braces. One pupil with a 40 degrees thoracic curvature, underwent satisfactory surgical treatment because of progression 1 year later. Of the 121 spinal deformities with an initial Cobb angle less than 10 degrees, 44 (35.8%), and of the 29 scoliotic deformities with an initial Cobb angle between 10 degrees and 20 degrees, 14 (48.3%) progressed (a Cobb angle difference of at least 5 degrees in more than one examination). Observation and physiotherapy were the only treatments applied to all except one of the pupils in these groups. CONCLUSIONS: The Adams forward-bending test cannot be considered a safe diagnostic criterion for the early detection of scoliosis (especially when it is used as the only screening tool) because it results in an unacceptable number of false-negative findings. For the early detection of scoliosis, a combination of back-shape analysis methods can be safely used with the introduction of cutoff limits for referral being a useful procedure. The incidence of significant scoliosis is low, and its natural history seems to be independent of early detection. The wide-spread use of school scoliosis screening with the use of the forward-bending test must be questioned.

Adolescent↗
PubMed2006

The effects of mild traumatic brain injury on confrontation naming in adults.

PRIMARY OBJECTIVE: To compare confrontation-naming in adults with MTBI to a group of normal adults under increased processing load conditions. RESEARCH DESIGN: A randomized block, repeated measures design was used to examine confrontation-naming response latency and accuracy using a computerized experimental program. METHODS AND PROCEDURES: Twenty-four adults having sustained a MTBI (aged 18-53) and 24 age-matched controls named pictures from three levels of vocabulary as quickly and accurately as possible. All MTBI participants were assessed with the Scales of Cognitive Ability for Traumatic Brain Injury (SCATBI) for later comparison. MAIN OUTCOMES AND RESULTS: The results revealed a main effect of group ( p < or = 0.001) for the latency data and a group by vocabulary level interaction ( p = 0.043) for the accuracy data. No significant correlations were found between response latency and accuracy with performance on the SCATBI. Reaction time measures may reveal inefficiencies not tapped by traditional measures.

Adolescent↗
PubMed1993

Diagnostic accuracy of pancreatic enzymes evaluated by use of multivariate data analysis.

We analyzed pancreatic enzyme data from 508 patients with suspected pancreatitis by neural network analysis, by an Expert multirule generation protocol, and by receiver-operator characteristic (ROC) curve analysis of a single test result. Neural network analysis showed that use of lipase provided the best means for diagnosing pancreatitis. Diagnostic accuracies achieved by using amylase only, lipase only, and amylase and lipase in combination were 76%, 82%, and 84%, respectively. Use of the Expert rule generation protocol provided a diagnostic accuracy of 92% when rules for single and multiple samplings were combined. ROC curve analysis for initial enzyme activities showed the maximal diagnostic accuracy to be 82% and 85% for amylase and lipase, respectively; use of peak enzyme activities yielded accuracies of 81% and 88%, respectively. The evaluation of laboratory test data should include analysis of the diagnostic accuracy of laboratory tests by multivariate techniques such as neural network analysis or an Expert systems approach. Multivariate analysis should allow for a more realistic assessment of the diagnosis accuracy of laboratory tests because all the available data are included in the evaluation.

Amylases↗
PubMed1999

Diagnostic accuracy of physician review, expert algorithms and data-derived algorithms in adult verbal autopsies.

BACKGROUND: The verbal autopsy (VA) is used to collect information on cause-specific mortality from bereaved relatives. A cause of death may be assigned by physician review of the questionnaires, or by an algorithm. We compared the diagnostic accuracy of physician review, an expert algorithm, and data-derived algorithms. METHODS: Data were drawn from a multicentre validation study of 796 adult deaths that occurred in hospitals in Tanzania, Ethiopia, and Ghana. A 'gold standard' cause of death was assigned using hospital records and death certificates. The VA interviews were carried out by trained fieldworkers 1-21 months after the subject's death. A cause of death was assigned by physician review and an expert algorithm. Data-derived algorithms that most accurately estimated the cause-specific mortality fraction (CSMF) for each cause of death were identified using logistic regression. RESULTS: The most common causes of death were tuberculosis/AIDS (CSMF = 18.6%), malaria (CSMF = 10.7%), meningitis (CSMF = 8.3%), and cardiovascular disorders (CSMF = 8.2%). The CSMF obtained using physician review was within +/-20% of the gold standard value for 12 causes of death including the four common causes. The CSMF obtained using the expert algorithm was within +/-20% of the gold standard for eight causes of death, including tuberculosis/AIDS, malaria, and meningitis. The CSMF obtained using the data-derived algorithms was within +/-20% of the gold standard for seven causes of death, including tuberculosis/ AIDS, meningitis, and cardiovascular disorders. All three methods yielded a specificity of at least 80% for all causes of death, and a sensitivity of at least 80% for deaths due to injuries and rabies. CONCLUSIONS: For those settings where physician review is not feasible, expert and data-derived algorithms provide an alternative approach for assigning many causes of death. We recommend that the algorithms proposed herein are validated further.

Acquired Immunodeficiency Syndrome↗
PubMed2004

Comparative evaluation of digoxin concentrations determined by three assay systems: TDx, IMx and OPUS.

Digoxin concentrations measured by three automated immunoassay systems, i.e. OPUS, TDx and IMx assays, were compared in order to evaluate precision and accuracy performance, and data compatibility. Coefficients of variation for all methods in within-run and between-run precision were less than 10% at weighed-in concentrations of 0.545, 1.090 and 2.180 ng/ml. The accuracy relative to the three weighed-in concentrations ranged from 97% to 123% for all methods. One hundred and three plasma samples from 60 patients receiving digoxin were used to evaluate the data compatibility. Digoxin concentrations measured by the three immunoassay systems correlated well with one another. These results suggest that there are few problems when switching between digoxin assay methods, and that IMx and OPUS are more useful than TDx because they do not require sample pretreatment. The digoxin concentrations of the plasma samples from one patient receiving both digoxin and potassium canrenoate were investigated as a case report. The digoxin concentrations measured by TDx and IMx became higher than those measured by OPUS after starting the combination treatment. In another patient suffering from bilirubinaemia, the digoxin concentrations measured by TDx or IMx were higher than those measured by OPUS. These results suggest that OPUS has a higher specificity for measuring the plasma digoxin concentrations compared with TDx or IMx.

Analysis of Variance↗
PubMed1984

Improving the quality of data in a computerised patient master index: implication for costs and patient care.

This study was designed to test the accuracy and completeness of nominal identifying data recorded on a hospital patient master index (PMI) and to estimate the cost-effectiveness of using patient questionnaires for improving quality of data. The study showed that the design of any PMI should include a method of obtaining and storing the last date at which a patient's details were confirmed as correct, a prompt system for identifying missing data and a routine report on completeness and inconsistencies within the data. Both accuracy and completeness of patient identification can be improved by the use of patient questionnaires. The cost of their use is more than matched by the savings obtained from complete accurate data. The use of computer produced questionnaires also reduces the need for peak time activity on busy computer systems.

Computers↗
PubMed2004

Prospects for inferring very large phylogenies by using the neighbor-joining method.

Current efforts to reconstruct the tree of life and histories of multigene families demand the inference of phylogenies consisting of thousands of gene sequences. However, for such large data sets even a moderate exploration of the tree space needed to identify the optimal tree is virtually impossible. For these cases the neighbor-joining (NJ) method is frequently used because of its demonstrated accuracy for smaller data sets and its computational speed. As data sets grow, however, the fraction of the tree space examined by the NJ algorithm becomes minuscule. Here, we report the results of our computer simulation for examining the accuracy of NJ trees for inferring very large phylogenies. First we present a likelihood method for the simultaneous estimation of all pairwise distances by using biologically realistic models of nucleotide substitution. Use of this method corrects up to 60% of NJ tree errors. Our simulation results show that the accuracy of NJ trees decline only by approximately 5% when the number of sequences used increases from 32 to 4,096 (128 times) even in the presence of extensive variation in the evolutionary rate among lineages or significant biases in the nucleotide composition and transition/transversion ratio. Our results encourage the use of complex models of nucleotide substitution for estimating evolutionary distances and hint at bright prospects for the application of the NJ and related methods in inferring large phylogenies.

Computer Simulation↗
PubMed2002

Retrospective smoking history data collection for deceased workers: completeness and accuracy of surrogate reports.

Few studies have considered the completeness and accuracy of smoking histories from surrogates for deceased workers. We surveyed 68 surrogates for uranium plant workers who completed smoking histories before their deaths. Completeness: 96% of surrogates answered definitively (yes or no) about overall smoking status (e.g., used any tobacco type), and 79 to 100% of those answered definitively about specific tobacco types. Of those reporting positive history, 50% to 67% answered detailed questions. Accuracy: Surrogates were accurate about overall smoking status (Kappa (kappa) = 0.75) compared with index self-report, but tended to underreport (P = 0.10). Spouses and nonspouses performed similarly. Accuracy by tobacco type was moderate (kappa = 0.21 to 0.56), with cigar smoking underreported. Surrogates for cigarette smokers underreported duration and amount smoked. Surrogates for deceased workers can report accurately on overall smoking status, but underreport other smoking habits.

Cause of Death↗