SEARCH · PubMed Health

Results for “measurement error”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 37 records · Page 2Linked to original sources

PubMed1996

Measurement error models for ordinal exposure variables measured with error.

In dietary epidemiology, the key nutrient variables are often expressed in the quintile scale. The nutrients are often measured with error and it is of interest to consider estimates of relative risk for exposures in the quintile scale corrected for measurement error. In this paper, I propose a measurement error model that relates diet record (true) nutrient values to food frequency (noisy) nutrient values when expressed in the quintile scale. I estimate this measurement error model from validation study data and then apply it to obtain corrected estimates of breast cancer risk in relation to intake of fat and alcohol with use of data from the Nurses' Health Study.

Adult↗

PubMed2003

The detection of gene-environment interaction for continuous traits: should we deal with measurement error by bigger studies or better measurement?

BACKGROUND: The search for biologically relevant gene-environment interactions has been facilitated by technological advances in genotyping. The design of studies to detect interactions on continuous traits such as blood pressure and insulin sensitivity is attracting increasing attention. We have previously described power calculations for such studies, and this paper describes the extension of those calculations to take account of measurement error. METHODS: The model considered in this paper is a simple linear regression relating a continuous outcome to a continuously distributed exposure variable in which the ratio of slopes for each genotype is considered as the interaction parameter. The classical measurement error model is used to describe the uncertainty in measurement in the outcome and the exposure. The sample size to detect differing magnitudes of interaction with varying frequencies of the minor allele are calculated for a given main effect observed with error both in the exposure and the outcome. The sample size to detect a given interaction for a given minor allele frequency is calculated for differing degrees of measurement error in the assessment of the exposure and the outcome. RESULTS: The required sample size is dependent upon the magnitude of the interaction, the allele frequency and the strength of the association in those with the common allele. As an example, we take the situation in which the effect size in those with the common allele was a quarter of a standard deviation change in the outcome for a standard deviation change in the exposure. If a minor allele with a frequency of 20% leads to a doubling of that effect size, then the sample size is highly dependent upon the precision with which the exposure and outcome are measured. rho(Tx) and rho(Ty) are the correlation between the measured exposure and outcome, respectively and the true value. If poor measures of the exposure and outcome are used, (e.g. rho(Tx) = 0.3, rho(Ty) = 0.4), then a study size of 150 989 people would be required to detect the interaction with 95% power at a significance level of 10(-4). Such an interaction could be detected in study samples of under 10 000 people if more precise measurements of exposure and outcome were made (e.g. rho(Tx) = 0.7, rho(Ty) = 0.7), and possibly in samples of under 5000 if the precision of estimation were enhanced by taking repeated measurements. CONCLUSIONS: The formulae for calculating the sample size required to study the interaction between a continuous exposure and a genetic factor on a continuous outcome variable in the face of measurement error will be of considerable utility in designing studies with appropriate power. These calculations suggest that smaller studies with repeated and more precise measurement of the exposure and outcome will be as powerful as studies even 20 times bigger, which necessarily employ less precise measures because of their size. Even though the cost of genotyping is falling, the magnitude of the effect of measurement error on the power to detect interaction on continuous traits suggests that investment in studies with better measurement may be a more appropriate strategy than attempting to deal with error by increasing sample sizes.

Environmental Exposure↗

PubMed2005

Triphasic MRI of pelvic organ descent: sources of measurement error.

PURPOSE: To identify sources of error when measuring pelvic organ displacement during straining using triphasic dynamic magnetic resonance imaging (MRI). MATERIALS AND METHODS: Ten healthy nulliparous woman underwent triphasic dynamic 1.5 T pelvic MRI twice with 1 week between studies. The bladder was filled with 200 ml of a saline solution, the vagina and rectum were opacified with ultrasound gel. T2 weighted images in the sagittal plane were analysed twice by each of the two observers in a blinded fashion. Horizontal and vertical displacement of the bladder neck, bladder base, introitus vaginae, posterior fornix, cul-de sac, pouch of Douglas, anterior rectal wall, anorectal junction and change of the vaginal axis were measured eight times in each volunteer (two images, each read twice by two observers). Variance components were calculated for subject, observer, week, interactions of these three factors, and pure error. An overall standard error of measurement was calculated for a single observation by one observer on a film from one woman at one visit. RESULTS: For the majority of anatomical reference points, the range of displacements measured was wide and the overall measurement error was large. Intra-observer error and week-to-week variation within a subject were important sources of measurement error. CONCLUSION: Important sources of measurement error when using triphasic dynamic MRI to measure pelvic organ displacement during straining were identified. Recommendations to minimize those errors are made.

Adolescent↗

PubMed2006

Exposure-measurement error is frequently ignored when interpreting epidemiologic study results.

INTRODUCTION: One important source of error in study results is error in measuring exposures. When interpreting study results, one should consider the impact that exposure-measurement error (EME) might have had on study results. METHODS: To assess how often this consideration is made and the form it takes, journal articles were randomly sampled from original articles appearing in the American Journal of Epidemiology and Epidemiology in 2001, and the International Journal of Epidemiology between December 2000 and October 2001. RESULTS: Twenty-two (39%) of the 57 articles surveyed mentioned nothing about EME. Of the 35 articles that mentioned something about EME, 16 articles described qualitatively the effect EME could have had on study results. Only one study quantified the impact of EME on study results; the investigators used a sensitivity analysis. Few authors discussed the measurement error in their study in any detail. CONCLUSIONS: Overall, the potential impact of EME on error in epidemiologic study results appears to be ignored frequently in practice.

Bias↗

PubMed2002

Measurement error associated with spinal mobility measures in children with and without low-back pain.

AIM: Evidence of the reliability of measurements in children is scarce, particularly in children with low-back pain. The aim of this investigation was to evaluate the measurement error associated with repeated measures of spinal mobility measures in children with and without low-back pain by establishing 95% limits of agreement. METHODS: A repeated measures study was performed involving 119 children aged 11-16 y. Of this sample 30 subjects reported recurrent low-back pain and were classified as symptomatic, the remaining 89 subjects were asymptomatic. Standardized measures were taken, including the sit-and-reach test, hip range of motion (Leighton flexometer), lumbar flexibility (modified Schöber test) and lateral flexion of the spine. The same experimenter performed all testing, with 1 wk between the repeated measures. RESULTS: Correlation coefficients suggested that all measures exhibited good reliability in both the symptomatic (r = 0.80-0.95) and asymptomatic groups (r = 0.88-0.99). In contrast, the limits of agreement showed that all measures exhibited random error. The magnitude of random error was typically greater in the symptomatic subjects, suggesting that low-back pain may influence the reliability of typical measures used in this population. CONCLUSION: The magnitude of error must be interpreted in relation to analytical goals and the expected magnitude of change. In the authors' opinion the error presented appears acceptable for the serial monitoring of patients, although this will depend on the differences in mobility typically found.

Adolescent↗

PubMed2005

Measurement error of percent diameter carotid stenosis determined by conventional angiography: implications for noninvasive evaluation.

BACKGROUND AND PURPOSE: Our aim was to quantitate the range of error in measurement of percent diameter carotid stenosis using conventional angiography in order to estimate the misclassification rate in selecting patients for carotid endarterectomy. METHODS: A secondary or meta-analysis of published data regarding carotid stenosis measurement error associated with angiography yielded a narrow range of standard deviation across several studies using different measurement protocols. A simulation of the measurement process was developed to model this error and allow estimation of misclassification rates compared with true values for stenosis. RESULTS: A standard deviation of 8% describes the average error observed at 60% diameter stenosis across a number of studies. Using the measurement model, for a population with 30% prevalence of 60% stenosis, this finding implies a misclassification rate of approximately 4% overall for conventional angiography. CONCLUSION: Like all diagnostic tests, angiography is associated with measurement error. This does not affect the results of carotid stenosis therapy trials using angiography but does need to be taken into account when evaluating noninvasive alternative tests for carotid stenosis.

Carotid Stenosis↗

PubMed2005

Impact of measurement error in the analysis of bone tunnel enlargement after anterior cruciate ligament reconstruction.

BACKGROUND: Radiographic bone tunnel enlargement after anterior cruciate ligament reconstruction is frequently reported. The error associated with measuring tunnels on radiographs has not been established and may be substantial. HYPOTHESIS: Measurement error influences interpretation of bone tunnel enlargement. STUDY DESIGN: Cohort study (Diagnosis); Level of evidence, 2. METHODS: Radiographs and clinical outcomes were obtained from 76 subjects who had undergone hamstring tendon anterior cruciate ligament reconstruction. Tunnel enlargement was determined by measuring the widths of the femoral and tibial tunnels with a digital caliper in lateral and anteroposterior radiographs. Measurements were corrected for magnification; the percentage change in tunnel widths was recorded relative to the diameters drilled at surgery. One examiner measured all radiographs and repeated the procedure for 20 randomly selected radiographs 12 months later. A second examiner measured the same 20 radiographs. Intraclass correlation coefficients and 95% confidence intervals for intrarater and interrater measurement errors were calculated. Confidence intervals were used to divide subjects into 2 groups according to whether bone tunnel enlargement was within (unenlarged tunnel group) or greater than (enlarged tunnel group) the 95% confidence interval for measurement error; clinical comparisons between these groups were made for each view of tibial and femoral tunnels. RESULTS: Measurement errors for tunnel enlargement ranged from 17% to 26% for intrarater measurements and 24% to 38% for interrater measurements. For the lateral view of the tibial tunnel, there was a significant positive association between tunnel enlargement and anterior knee laxity. There was no relationship between femoral tunnel enlargement and anterior knee laxity. CONCLUSION: There was considerable error associated with bone tunnel measurement. A relationship between tibial tunnel enlargement on the lateral radiograph and anterior knee laxity was evident once enlargement was defined as a change that exceeded measurement error. CLINICAL RELEVANCE: Measurement of bone tunnels on radiographs is useful for interpreting large changes; caution should be used when small changes are observed.

Adult↗

PubMed2000

Measurement error in prenatal care utilization: evidence of attenuation bias in the estimation of impact on birth weight.

OBJECTIVE: Errors in the measurement of the timing and number of prenatal care visits may produce downward bias in estimates of the impact of prenatal care use on birth outcomes. This paper examines the extent of attenuation bias from measurement error in the estimation of the effect of prenatal care use on birth weight. METHODS: Data were analyzed from the 1980 National Natality Survey, a nationally representative sample of live births with information on prenatal care utilization from three sources: birth certificates, medical provider surveys, and maternal surveys. The extent of attenuation bias in estimates of the impact of different measures of prenatal care use on birth weight was examined by comparing estimates robust to measurement error (including instrumental variables) with ordinary least squares results. RESULTS: There is considerable disagreement in measures of prenatal care across the three data sources, with correlations in the utilization measures computed from different sources around 0.5. The results also show evidence of attenuation bias from measurement error in estimates of the impact of prenatal care on birth weight for both White and Black mothers. Attenuation bias was least severe for information from the birth certificate report of prenatal care. CONCLUSIONS: Because of measurement error, previous studies may have underestimated the effect of prenatal care utilization on birth weight. Corrected estimates, however, do not suggest that prenatal care is a major predictor of birth weight. In addition, part of what previous analyses have interpreted as adverse selection bias may in fact be attenuation bias due to measurement error.

Birth Weight↗

PubMed1999

Instrument-induced measurement errors during strabismus surgery.

PURPOSE: The purpose of this study was to determine the clinical setting where errors in measurements of muscle position during strabismus surgery made by the Scott curved ruler or by calipers become important and to characterize the magnitude of those errors. METHODS: Geometric analysis was used to determine the measurement error between true arc lengths of 3.0 to 20.0 mm versus Scott curved ruler measurements and caliper measurements for axial lengths ranging from 18 to 30 mm. RESULTS: For measurements less than 9.0 mm, neither the Scott curved ruler nor calipers had any clinically important measurement error for any axial length. For axial lengths substantially smaller than 21 mm or larger than 24 mm, the Scott curved ruler, although more accurate than calipers, caused clinically important measurement errors with arc length measurements as small as 12 mm in very small eyes and 14 mm in large eyes. For axial lengths of 30 mm or more, both calipers and the Scott curved ruler had similar accuracy for measuring long arc lengths. CONCLUSIONS: Both the Scott curved ruler and calipers are accurate in measuring arc lengths 9.0 mm or less. For longer arc length measurements, accuracy becomes dependent on axial length. The Scott curved ruler, although substantially more accurate than calipers for most common axial lengths, can introduce clinically important measurement errors when measuring arc lengths as small as 12 mm. Axial length should be considered when measuring muscle position during strabismus surgery.

Anthropometry↗

PubMed1988

Measurement error and its impact on partial correlation and multiple linear regression analyses.

In studies examining associations between dietary factors and biomedical risk factors, the relations, if they exist, are frequently attenuated by measurement error. Measurement error may be due to a large intraindividual variation and an inadequate number of measurements or to an inaccurate measuring instrument. This paper evaluates the impact of measurement error on partial correlation and multiple linear regression analyses. Quantitative methods are derived to estimate the potential attenuation of associations. The results indicate that when the controlled variables do not have measurement error, but the correlated variables do, the attenuation of the partial correlation coefficient (or multiple regression coefficient) is greater than that of the simple correlation (or regression) coefficient. When both the correlated variables and the controlled variables have measurement error, the partial correlation (or the regression) coefficients can be either increased or decreased.

Cholesterol, Dietary↗

PubMed2005

A resampling approach for adjustment in prediction models for covariate measurement error.

Recent works on covariate measurement errors focus on the possible biases in model coefficient estimates. Usually, measurement error in a covariate tends to attenuate the coefficient estimate for the covariate, i.e., a bias toward the null occurs. Measurement error in another confounding or interacting variable typically results in incomplete adjustment for that variable. Hence, the coefficient for the covariate of interest may be biased either toward or away from the null. This paper presents a new method based on a resampling technique to deal with covariate measurement errors in the context of prediction modeling. Prediction accuracy is our primary parameter of interest. Prediction accuracy of a model is defined as the success rate of prediction when the model predicts new response. We call our method bootstrap regression calibration (BRC). We study logistic regression with interacting covariates as our prediction model. We measure the prediction accuracy of a model by receiver operating characteristic (ROC) method. Results from simulations show that bootstrap regression calibration offers consistent enhancement over the commonly used regression calibration (RC) method in terms of improving prediction accuracy of the model and reducing bias in the estimated coefficients.

Calibration↗

PubMed1994

Reversal of vertebral deformities in osteoporosis: measurement error or "rebound"?

Digitized morphometry of vertebral bodies on lateral spine films is used to identify and quantify vertebral deformities or fractures. One problem associated with this method is the phenomenon of "disappearing fractures," which results from the apparent increase in vertebral body heights of previously deformed vertebrae on subsequent radiographs. These have been considered biologically implausible and therefore a result of measurement error. Measurement error is unlikely to be unidirectional, so that a proportion of fractures identified by morphometry is also the result of measurement error. Since some vertebral deformities are real events, some disappearances of deformities detected by morphometry may be real events. In this report, we examine the data from our clinical trial of sodium fluoride in spinal osteoporosis to assess critically the plausibility of two hypotheses: (1) The "rebound" phenomenon results from measurement error. If this is the case, then some fractures of the same magnitude as the rebound must also represent measurement error. (2) Some deformed vertebrae in fact rebound toward their original shape and size, displaying an elastic response to deformation. If this occurs, then some vertebral deformities are transient events, not true fractures. We conclude that the variability inherent in morphometric data obtained from serial spine x-rays results in both disappearing fractures and a high false positive fracture rate. The use of more stringent criteria for defining significant deformities, or true fractures, will minimize these problems. We cannot exclude the second hypothesis, that some vertebral deformities may be transient events, but this needs further study.

Aged↗

PubMed2005

Correlated errors in biased surrogates: study designs and methods for measurement error correction.

A measurement error model proposed previously allows for correlations between subject-specific biases and between random within-subject errors in the surrogates obtained from two modes of measurement. However, most of these model parameters are not identifiable from the standard validation study design, including, importantly, the attenuation factor needed to correct for bias in relative risk estimates due to measurement error. We propose validation study designs that permit estimation and inference for the attenuation factor and other parameters of interest when these correlations are present. We use an estimating equations framework to develop semi-parametric estimators for these parameters, exploiting instrumental variables techniques. The methods are illustrated through application to data from the Nurses' Health Study and Health Professionals' Follow-up Study, and comparisons are made to more restrictive models.

Bias↗

PubMed1995

Exposure measurement errors, risk estimate and statistical power in case-control studies using dichotomous analysis of a continuous exposure variable.

BACKGROUND: Non-differential errors in exposure measurements have been shown to lead to differential misclassification of exposure. As a consequence, the common tenet that, in absence of bias, imprecise exposure assessment can only bias the risk estimates conservatively does not necessarily hold. We investigate the effects of exposure measurement errors on the risk estimate and on statistical power. METHODS: We used a computer model that simulates a case-control study. We used both hypothetical data and data modelled on empirical measurements of environmental magnetic fields exposure. RESULTS: Measurement errors are found to have a lesser impact on risk estimates and statistical power than would have been the case had misclassification been truly non-differential. However, for a given cutpoint, a bias away from the null cannot be excluded. The predominant direction of the errors is found to have important consequences on both the study power and the risk estimates. CONCLUSION: When sufficient empirical data are available, computer modelling may give a more accurate estimate of the effects of measurement errors than algebraic corrections.

Bias↗

PubMed2005

Quantitative analysis of variability and uncertainty with known measurement error: methodology and case study.

The appearance of measurement error in exposure and risk factor data potentially affects any inferences regarding variability and uncertainty because the distribution representing the observed data set deviates from the distribution that represents an error-free data set. A methodology for improving the characterization of variability and uncertainty with known measurement errors in data is demonstrated in this article based on an observed data set, known measurement error, and a measurement-error model. A practical method for constructing an error-free data set is presented and a numerical method based upon bootstrap pairs, incorporating two-dimensional Monte Carlo simulation, is introduced to address uncertainty arising from measurement error in selected statistics. When measurement error is a large source of uncertainty, substantial differences between the distribution representing variability of the observed data set and the distribution representing variability of the error-free data set will occur. Furthermore, the shape and range of the probability bands for uncertainty differ between the observed and error-free data set. Failure to separately characterize contributions from random sampling error and measurement error will lead to bias in the variability and uncertainty estimates. However, a key finding is that total uncertainty in mean can be properly quantified even if measurement and random sampling errors cannot be separated. An empirical case study is used to illustrate the application of the methodology.

Journal Article↗

PubMed2004

Estimation of magnitude in gene-environment interactions in the presence of measurement error.

The design of studies aimed at identifying the interaction between genetic and environmental determinants in disease risk is attracting increased attention. In this paper, we study the effect of errors on measuring exposures and the effect of assessing genotype at one locus on the association of a continuous outcome measure with continuous exposures and genotype. The estimation of misclassification errors in assessing genotypes from a separate study is proposed. If the exposure measurement error is substantial, then even relatively small errors in genotyping within limits that are often quoted can have an appreciable effect on interaction estimates. We, thus, consider a method for correcting the measurement errors when the interaction between the exposures and the genetic factor is significant. Finally, we present an epidemiological example to demonstrate the importance of correcting measurement errors.

Alleles↗

PubMed2002

Estimation of an errors-in-variables regression model when the variances of the measurement errors vary between the observations.

It is common in the analysis of aggregate data in epidemiology that the variances of the aggregate observations are available. The analysis of such data leads to a measurement error situation, where the known variances of the measurement errors vary between the observations. Assuming multivariate normal distribution for the 'true' observations and normal distributions for the measurement errors, we derive a simple EM algorithm for obtaining maximum likelihood estimates of the parameters of the multivariate normal distributions. The results also facilitate the estimation of regression parameters between the variables as well as the 'true' values of the observations. The approach is applied to re-estimate recent results of the WHO MONICA Project on cardiovascular disease and its risk factors, where the original estimation of the regression coefficients did not adjust for the regression attenuation caused by the measurement errors.

Algorithms↗

PubMed1999

Measurement error in epidemiology: the design of validation studies II: bivariate situation.

The bias in relative risk estimates caused by errors in measurement of the relevant exposure is being increasingly recognized in epidemiology. Estimation of the necessary correction factor to remove this bias for univariate exposure has been considered in an earlier paper. We consider here the multivariate situation in which non-differential errors in measurement can lead to incorrect identification of the variable most closely associated with disease. Estimation of the necessary correction factor when the true exposure is unobservable necessarily requires assumptions. We explore the robustness of the estimation to departures from a range of assumptions. The value of good biomarkers is demonstrated. We present a bivariate example in which failure to take account of measurement error leads to the incorrect exposure being identified as the important determinant of disease risk.

Adult↗