PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “REGRESSION”

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 19 recordsLinked to original sources

A comparison of regression trees, logistic regression, generalized additive models, and multivariate adaptive regression splines for predicting AMI mortality.

Clinicians and health service researchers are frequently interested in predicting patient-specific probabilities of adverse events (e.g. death, disease recurrence, post-operative complications, hospital readmission). There is an increasing interest in the use of classification and regression trees (CART) for predicting outcomes in clinical studies. We compared the predictive accuracy of logistic regression with that of regression trees for predicting mortality after hospitalization with an acute myocardial infarction (AMI). We also examined the predictive ability of two other types of data-driven models: generalized additive models (GAMs) and multivariate adaptive regression splines (MARS). We used data on 9484 patients admitted to hospital with an AMI in Ontario. We used repeated split-sample validation: the data were randomly divided into derivation and validation samples. Predictive models were estimated using the derivation sample and the predictive accuracy of the resultant model was assessed using the area under the receiver operating characteristic (ROC) curve in the validation sample. This process was repeated 1000 times-the initial data set was randomly divided into derivation and validation samples 1000 times, and the predictive accuracy of each method was assessed each time. The mean ROC curve area for the regression tree models in the 1000 derivation samples was 0.762, while the mean ROC curve area of a simple logistic regression model was 0.845. The mean ROC curve areas for the other methods ranged from a low of 0.831 to a high of 0.851. Our study shows that regression trees do not perform as well as logistic regression for predicting mortality following AMI. However, the logistic regression model had performance comparable to that of more flexible, data-driven models such as GAMs and MARS.

Data Interpretation, Statistical↗

Ordinal regression model and the linear regression model were superior to the logistic regression models.

OBJECTIVE: Ordinal scales often generate scores with skewed data distributions. The optimal method of analyzing such data is not entirely clear. The objective was to compare four statistical multivariable strategies for analyzing skewed health-related quality of life (HRQOL) outcome data. HRQOL data were collected at 1 year following catheterization using the Seattle Angina Questionnaire (SAQ), a disease-specific quality of life and symptom rating scale. STUDY DESIGN AND SETTING: In this methodological study, four regression models were constructed. The first model used linear regression. The second and third models used logistic regression with two different cutpoints and the fourth model used ordinal regression. To compare the results of these four models, odds ratios, 95% confidence intervals, and 95% confidence interval widths (i.e., ratios of upper to lower confidence interval endpoints) were assessed. RESULTS: Relative to the two logistic regression analysis, the linear regression model and the ordinal regression model produced more stable parameter estimates with smaller confidence interval widths. CONCLUSION: A combination of analysis results from both of these models (adjusted SAQ scores and odds ratios) provides the most comprehensive interpretation of the data.

Adolescent↗

Spontaneous regression of merkel cell carcinoma: a comparative study of TUNEL index and tumor-infiltrating lymphocytes between spontaneous regression and non-regression group.

Some Merkel cell carcinomas (MCC) have been reported to regress spontaneously. To clarify the mechanisms of spontaneous regression (SR) of MCC, we analyzed the TUNEL index, the labeling index of proliferating cell nuclear antigen (PCNA), the labeling index of bcl-2 protein, and the expression of p53 of the tumor cells. We also evaluated the number of infiltrating lymphocytes surrounding the tumor in the tissue specimens. Among seven patients with MCC (SR: n=4; non-regression (NR): n=3), the TUNEL index in the SR group was significantly higher than that in NR group (5.2 and 2.0%, respectively). In addition, the number of lymphocytes around the tumor nests was also significantly increased in the SR group compared to NR group (1576 and 663 cells/mm(2), respectively). Most of the infiltrating lymphocytes were UCHL-1 positive T-cells. There were no significant differences of the PCNA labeling index, the bcl-2 protein labeling index, and the expression p53 between SR and NR group. These results indicate that apoptosis and local T-cell mediated immune response might be involved in spontaneous regression of MCC.

Aged↗

Effects of lipid lowering by pravastatin on progression and regression of coronary artery disease in symptomatic men with normal to moderately elevated serum cholesterol levels. The Regression Growth Evaluation Statin Study (REGRESS).

BACKGROUND: Intensive lowering of serum cholesterol may retard progression of coronary atherosclerosis in selected groups of patients. However, few data are available on the potential benefit of serum cholesterol reduction in the broad range of patients with coronary atherosclerosis and normal to moderately elevated serum cholesterol levels who undergo various forms of treatment. The Regression Growth Evaluation Statin Study (REGRESS) addresses this group of patients. METHODS AND RESULTS: REGRESS is a double-blind, placebo-controlled multicenter study to assess the effects of 2 years of treatment with the 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor pravastatin on progression and regression of coronary atherosclerosis in 885 male patients with a serum cholesterol level between 4 and 8 mmol/L (155 and 310 mg/dL) by quantitative coronary arteriography. Primary end points were (1) change in average mean segment diameter per patient and (2) change in average minimum obstruction diameter per patient. Clinical events were also analyzed. Of the 885 patients, 778 (88%) had an evaluable final angiogram. Mean segment diameter decreased 0.10 mm in the placebo group versus 0.06 mm in the pravastatin group (P = .019): The mean difference between treatment groups was 0.04 mm, with a 95% CI of 0.01 to 0.07 mm. The median minimum obstruction diameter decreased 0.09 mm in the placebo group versus 0.03 mm in the pravastatin group (P = .001): The difference of the medians between the treatment groups was 0.06 mm, with a CI of 0.02 to 0.08 mm. At the end of the follow-up period, 89% (CI, 86% to 92%) of the pravastatin patients and 81% (CI, 77% to 85%) of the placebo patients were without new cardiovascular events (P = .002). CONCLUSIONS: In symptomatic men with significant coronary atherosclerosis and normal to moderately elevated serum cholesterol, less progression of coronary atherosclerosis and fewer new cardiovascular events were observed in the group of patients treated with pravastatin than in the placebo group.

Angioplasty, Balloon, Coronary↗

Regression growth evaluation statin study (REGRESS): study design and baseline characteristics in 600 patients. The REGRESS Research Group.

OBJECTIVE: To assess the effects of lipid lowering, using a hydroxymethyl glutaryl coenzyme A reductase inhibitor (pravastatin) in symptomatic men with coronary artery disease. To assess follow-up intervention modes and baseline cardiovascular risk factors in men after routine diagnostic coronary arteriography which showed a significant stenosis (at least 50%) in a major segment. DESIGN: Baseline characteristics are stated in a multicentre, prospective, double-blind, randomized, placebo-controlled trial entitled 'regression growth evaluation statin study' (REGRESS). SETTING: Seven university hospitals and four referral hospitals in The Netherlands. PATIENTS: The baseline characteristics are given for the initial 600 patients who entered the REGRESS trial. All patients were 'normocholesterolemic' and suffered from symptomatic coronary artery disease. INTERVENTION: Initial follow-up treatment was analyzed after coronary arteriography, in accordance with standardized clinical practice (angioplasty, bypass grafting or medical management only). MAIN RESULTS: In contrast to routine procedures, 44% of the patient population was found in the medical management only stratum, whereas 32% was found in the coronary bypass graft and 24% in the percutaneous transluminal coronary angioplasty stratum. No major coronary cardiovascular risk factor could explain the decision to perform a certain follow-up intervention. Overall patients showed a low serum high density lipoprotein (HDL) cholesterol level. A higher New York Heart Association classification for anginal complaints tended to favour a more aggressive follow-up approach. CONCLUSIONS: The total REGRESS population showed relatively low HDL cholesterol blood levels compared with the general Dutch population. The rationale to perform a specific follow-up intervention by the treating cardiologist is unclear. Severity of anginal complaints may lead to a more active intervention procedure.

Aged↗

Comparison of performance of partial least squares regression, secured principal component regression, and modified secured principal component regression for determination of human serum albumin, gamma-globulin and glucose in buffer solutions and in vivo blood glucose quantification by near-infrared spectroscopy.

The performances of three multivariate analysis methods--partial least squares (PLS) regression, secured principal component regression (sPCR) and modified secured principal component regression (msPCR)--are compared and tested for the determination of human serum albumin (HSA), gamma-globulin, and glucose in phosphate buffer solutions and blood glucose quantification by near-infrared (NIR) spectroscopy. Results from the application of PLS, sPCR and msPCR are presented, showing that the three methods can determine the concentrations of HSA, gamma-globulin and glucose in phosphate buffer solutions almost equally well provided that the prediction samples contain the same spectral information as the calibration samples. On the other hand, when some potential spectral features appear in new measurements, sPCR and msPCR outperform PLS significantly. The reason for this is that such spectral features are not included during calibration, which leads to a degradation in PLS prediction performance, while sPCR and msPCR can improve their predictions for the concentrations of the analytes by removing the uncalibrated features from the original spectra. This point is demonstrated by successfully applying sPCR and msPCR to in vivo blood glucose measurements. This work therefore shows that sPCR and msPCR may provide possible alternatives to PLS in cases where some uncalibrated spectral features are present in measurements used for concentration prediction.

Blood Glucose↗

A study of atherosclerosis regression in Macaca mulatta: III. Chemical changes in arteries from animals with atherosclerosis induced for 19 months and regressed for 48 months at plasma cholesterol concentrations of 300 or 200 mg/dl.

The influence of two levels of plasma cholesterol concentration on the long-term regression of atherosclerotic plaques in rhesus monkeys (Macaca mulatta) was studied. Forty-eight young adult male rhesus monkeys were fed an atherogenic diet for 19 months, then diets designed to maintain plasma cholesterol concentrations at 280-320 mg/dl (actual 316 +/- 10 mg/dl; mean +/- SEM) of 180-220 mg/dl (actual 204 +/- 4 mg/dl). Twelve animals were killed after 19 months to evaluate the atherosclerosis produced. The remaining 36 monkeys were studied after 48 months of atherosclerosis regression. Significantly greater amounts of accumulated nonesterififed and esterified cholesterol were lost from the arteries of monkeys that underwent regression at about 200 mg/dl, in comparison with 300 mg/dl. Regression at both plasma cholesterol concentrations resulted in increased collagen and decreased elastin content in the thoracic aorta but not in the other arteries studied. After regression at 300 mg/dl there was no change (83%) in the frequency of calcification of the thoracic aorta, while at 200 mg/dl the frequency of calcification was reduced to 50%. When compared with monkeys whose athersclerotic lesions were produced in an identical manner but were regressed for only 24 months, after 48 months had increased cholesterol concenrations in the thoracic aorta and carotid artery but not the abdominal aorta of iliaco-femoral artery, regardless of the plasma cholesterol concentration. The lipid composition of the regressed plaque suggested a change in plaque lipid toward the character of extracellular deposits described physically as crystalline in nature. The physical state was influenced by the length of the regression period and the plasma cholesterol concentration during regeression. In the thoracic aorta, principally as a result of changes in elastin content, the collagen-to-elastin ratio increased between 24 and 48 months of regression. It is proposed that the differences in removal of cholesterol from the thoracic and abdominal aorta after 24 and 48 months of regression in animals at 300 mg/dl may be influenced by the rearrangement of connective tissue. On the basis of the lipid and mineral content of uncomplicated atherosclerotic plaques after 4 yers of regression, it appears that the plasma cholesterol concentration during the regression period is a signigicant factor influencing the extent of plaque regression as well as the potential for further progression.

Animals↗

Differences between regressive eyes and non-regressive eyes after LASIK for myopia in the time course of corneal changes assessed with the Orbscan.

The aim of this study was to compare regressive eyes with non-regressive eyes after laser in situ keratomileusis (LASIK) for myopia with regard to the time course of biomechanical changes of the cornea and to evaluate the possible roles of these changes in refractive regression following surgery. 15 eyes of 9 patients with refractive regression and 15 eyes of 11 patients with no refractive regression after LASIK for myopia were enrolled in this study. The time courses of central corneal thickness (CCT) and anterior/posterior corneal curvature radiuses (CCR) were measured with Orbscan slit scanning before and 1 week, 1 month and 3 months after LASIK for myopia. The regressive and non-regressive eyes had similar time courses of CCT before and after LASIK, whereas the 2 groups differed in the time courses of corneal shifting movements. In the regression group, the posterior CCR after LASIK was relatively stable (all p > 0.05). By contrast, in the non-regression group, the postoperative posterior CCR at week 1 had a significantly steeper curvature than that at month 3 (p = 0.000). The anterior CCR in the regression group at week 1 had a significantly flatter curvature than that at month 3 (p = 0.002) postoperatively. In contrast, there was no significant change of anterior CCR postoperatively (all p > 0.05) in the non-regression group. Taken together, these data suggest that refractive regression after LASIK might be mainly induced by corneal protrusion rather than central corneal thickening.

Adult↗

Adjustment for regression dilution in epidemiological regression analyses.

PURPOSE: The term "regression dilution" describes the dilution/attenuation in a regression coefficient that occurs when a single measured value of a covariate is used instead of the usual or average value over a period of time. This paper reviews the current knowledge concerning a simple method of adjusting for regression dilution in single and multiple covariate situations and illustrates the adjustment procedure. METHODS: Formulation of the regression dilution problem as a measurement error problem allows existing measurement error theory to be applied to developing methods of adjustment for regression dilution. This theory leads to a precise method of adjustment for linear regression and approximate methods for logistic and Cox proportional hazards regression. The method involves obtaining the naive estimates of coefficients by assuming that covariates are not measured with error, and then adjusting these coefficients using reliability estimates for the covariates. Methods for estimating the reliability of covariates from the reliability and main study data and a method for the calculation of standard errors and confidence intervals for adjusted coefficients are described. RESULTS: An illustration involving logistic regression analysis of risk factors for death from cardiovascular disease based on cohort and reliability data from the Busselton Health Study shows that the different methods for estimating the adjustment factors give very similar adjusted estimates of coefficients, that univariate adjustment procedures may lead to inappropriate adjustments in multiple covariate situations, whether or not other covariates have intra-individual variation, and when the reliability study is moderate to large, the precision of the estimates of reliability coefficients has little impact on the standard errors of adjusted regression coefficients. CONCLUSIONS: The simple method of adjusting regression coefficients for "regression dilution" that arises out of measurement error theory is applicable to many epidemiological settings and is easily implemented. The choice of method to estimate the reliability coefficient has little impact on the results. The practice of applying univariate adjustments in multiple covariate situations is not recommended.

Analysis of Variance↗

A simple approach to power and sample size calculations in logistic regression and Cox regression models.

For a given regression problem it is possible to identify a suitably defined equivalent two-sample problem such that the power or sample size obtained for the two-sample problem also applies to the regression problem. For a standard linear regression model the equivalent two-sample problem is easily identified, but for generalized linear models and for Cox regression models the situation is more complicated. An approximately equivalent two-sample problem may, however, also be identified here. In particular, we show that for logistic regression and Cox regression models the equivalent two-sample problem is obtained by selecting two equally sized samples for which the parameters differ by a value equal to the slope times twice the standard deviation of the independent variable and further requiring that the overall expected number of events is unchanged. In a simulation study we examine the validity of this approach to power calculations in logistic regression and Cox regression models. Several different covariate distributions are considered for selected values of the overall response probability and a range of alternatives. For the Cox regression model we consider both constant and non-constant hazard rates. The results show that in general the approach is remarkably accurate even in relatively small samples. Some discrepancies are, however, found in small samples with few events and a highly skewed covariate distribution. Comparison with results based on alternative methods for logistic regression models with a single continuous covariate indicates that the proposed method is at least as good as its competitors. The method is easy to implement and therefore provides a simple way to extend the range of problems that can be covered by the usual formulas for power and sample size determination.

Breast Neoplasms↗

Regression of papillomas induced by cottontail rabbit papillomavirus is associated with infiltration of CD8+ cells and persistence of viral DNA after regression.

Cottontail rabbit papillomavirus (CRPV) is a highly oncogenic papillomavirus and has been successfully used as a model to develop protective vaccines against papillomaviruses. Papillomas induced by the virus may spontaneously regress, suggesting that CRPV can also serve as a model to develop therapeutic vaccines. As a first step toward this goal, we have analyzed immunologic and viral aspects associated with papilloma regression and have identified several features unique to regression. Immunohistochemical staining of biopsies from growing and regressing papillomas and from sites after complete regression showed infiltration of CD8+ cells into the basal and suprabasal layers of the epidermis only during active regression. In situ hybridizations with mRNA-specific probes were strongly positive for E6 and E7 mRNAs during regression, but no late mRNA was present. Viral DNA was detected by in situ hybridization during regression but not after regression. However, analysis by PCR revealed persistence of viral DNA for several months at the majority of regression sites. The results suggest that stimulation of a strong CD8+ response to virus-infected cells is important for an effective therapeutic vaccine and that special attention should be given to the suppression of latent infection.

Animals↗

[Research of myopia development in children with self-regression of retinopathy of prematurity and in children with regression after cryotherapy].

PURPOSE: The aim of the paper is to estimate the influence of self-regression of retinopathy of prematurity (ROP) and regression after cryotherapy for development of axial myopia in premature babies. MATERIAL AND METHODS: The researches concerned 4 groups of premature babies in the 6th, 12th and 24th month of life. Group 1--without ROP, group 2--with self-regression of ROP, group 3--with regression of ROP after cryotherapy and group 4--with one eye with self-regression of ROP and second eye with regression after cryotherapy. RESULTS: In the 6th and 24th month of life mean axial length of eyeball in children from group 3 was statistically significantly higher than mean axial length of eyeball in children from group 1 and 2. In the 12th month of life mean axial length of eyeball in children from group 3 was statistically significantly higher than mean axial length of eyeball in children from group 1. In group 4 mean axial length of eyeball with regression of ROP after cryotherapy was statistically significantly higher than mean axial length of eyeball with self-regression of ROP. CONCLUSIONS: Self-regression of ROP and regression of ROP after cryotherapy may cause development of axial myopia in premature babies.

Child, Preschool↗

Cyclic angiogenesis and blood vessel regression in the ovary: blood vessel regression during luteolysis involves endothelial cell detachment and vessel occlusion.

Angiogenesis occurs as a cyclically regulated process in the ovary and the uterus. After ovulation, there is massive sprouting of blood vessels in the growing corpus luteum (CL) during the first third of the ovarian cycle. During luteolysis and for several weeks thereafter, all newly formed vessels regress. Here we have systematically analyzed regression of blood vessels during luteolysis to identify mechanisms of blood vessel regression. Blood vessel counts are highest in the midcycle CL and drop rapidly after the onset of luteolysis. After a rapid phase of tissue dissociation, blood vessel regression proceeds slowly over several weeks in the residual CL. Endothelial cells in regressing vessels acquire a distinctly rounded and condensed phenotype. Ultrastructural analysis of blood vessel regression processes in the cyclic CL suggests two major mechanisms of blood vessel regression: a) detachment of rounded endothelial cells from their basement membrane, leaving areas devoid of covering endothelial cell monolayer, and b) contraction and occlusion of arterioles and small arteries with pronounced proliferation of smooth muscle cells. In situ detection of nucleosomal fragmentation products demonstrates numerous apoptotic luteal cells, but only a few apoptotic endothelial cells in the regressing CL. Induction of apoptosis in cultured endothelial cell monolayers by RGD peptides demonstrated that endothelial cells detach from their adhesive surface before fully becoming positive for nucleosomal fragmentation products. These data indicate that cyclic angiogenic processes in the ovary offer a suitable experimental system to analyze mechanisms of blood vessel growth and regression, and suggest that detachment of endothelial cells before apoptosis as well as contractive occlusion of blood vessels may be critical determinants of blood vessel regression.

Animals↗

Spontaneous regression of Friend-virus-induced erythroleukemia. VII. The genetic control of regression.

The genetic control of spontaneous regression of erythroleukemia was studied in parental and hybrid mice in which leukemia was induced by the regressing strain of Friend virus (RFV). Because in previous studies parental regressor mouse strains tested (N/PLCR, SIM, NIH Swiss) all had the FV-1n/n genotype and the progressor mouse strains (SIM.R, BALB/c) had the Fv-1b/b genotype, we detemined the influence of Fv-1 alleles on regression. Genes which influence regression were dominant and had partial penetrance in (progressor x regressor) F1 mice. Regression occurred in hybrid mice which inherited the Fv-1b/b genes of each of the progressor mouse strains. Regression and Fv-1 alleles also segregated independently in (N/PLCR x BALB/c) F2 mice, in random-bred Swiss mice heterozygous for the Fv-1 gene, in partially inbred Swiss recombinant progressor and regressor mouse lines, and in hybrid mice carrying the Fv-1b/b gene of SIM.R mice. Regressor SIM and progressor SIM.R mice, which were bred to be congeneic at the Fv-1 locus, also differ with respect to recovery from viremia, suggesting that their Rfv-3 genes differ and influence regression. Crosses of SIM and SIM.R mice with the A.BY (Rfv-3s/s) mouse strain confirmed that SIM and SIM.R carry Rfv-3r/r and Rfv-3s/s, respectively. We conclude that Fv-1b/b is not inhibitory to regression nor is the Fv-1 gene a genetic deteminant in the process. The data suggest that regression is influenced by several genes, including those (Rfv-1, Rfv-2, Rfv-3) shown to affect recovery from leukemia in other systems.

Alleles↗

Regression versus no regression in the autistic disorder: developmental trajectories.

Developmental regression is a complex phenomenon which occurs in 20-49% of the autistic population. Aim of the study was to assess possible differences in the development of regressed and non-regressed autistic preschoolers. We longitudinally studied 40 autistic children (18 regressed, 22 non-regressed) aged 2-6 years. The following developmental areas were considered fundamental in the first years of life, and were assessed at ages 2, 3, 4, 5, and 6: receptive and expressive language, communicative and request modalities, play activities, and mental age. Children who regressed showed lower mean performances than those who did not regress and, in the time intervals considered, non-regressed children improved their ratings in the above mentioned variables significantly more than regressed children.

Autistic Disorder↗

A case study found that a regression tree outperformed multiple linear regression in predicting the relationship between impairments and Social and Productive Activities scores.

OBJECTIVE: Many important physiologic and clinical predictors are continuous. Clinical investigators and epidemiologists' interest in these predictors lies, in part, in the risk they pose for adverse outcomes, which may be continuous as well. The relationship between continuous predictors and a continuous outcome may be complex and difficult to interpret. Therefore, methods to detect levels of a predictor variable that predict the outcome and determine the threshold for clinical intervention would provide a beneficial tool for clinical investigators and epidemiologists. STUDY DESIGN AND SETTING: We present a case study using regression tree methodology to predict Social and Productive Activities score at 3 years using five modifiable impairments. The predictive ability of regression tree methodology was compared with multiple linear regression using two independent data sets, one for development and one for validation. RESULTS: The regression tree approach and the multiple linear regression model provided similar fit (model deviances) on the development cohort. In the validation cohort, the deviance of the multiple linear regression model was 31% greater than the regression tree approach. CONCLUSION: Regression tree analysis developed a better model of impairments predicting Social and Productive Activities score that may be more easily applied in research settings than multiple linear regression alone.

Aged↗

Detecting outliers when fitting data with nonlinear regression - a new method based on robust nonlinear regression and the false discovery rate.

BACKGROUND: Nonlinear regression, like linear regression, assumes that the scatter of data around the ideal curve follows a Gaussian or normal distribution. This assumption leads to the familiar goal of regression: to minimize the sum of the squares of the vertical or Y-value distances between the points and the curve. Outliers can dominate the sum-of-the-squares calculation, and lead to misleading results. However, we know of no practical method for routinely identifying outliers when fitting curves with nonlinear regression. RESULTS: We describe a new method for identifying outliers when fitting data with nonlinear regression. We first fit the data using a robust form of nonlinear regression, based on the assumption that scatter follows a Lorentzian distribution. We devised a new adaptive method that gradually becomes more robust as the method proceeds. To define outliers, we adapted the false discovery rate approach to handling multiple comparisons. We then remove the outliers, and analyze the data using ordinary least-squares regression. Because the method combines robust regression and outlier removal, we call it the ROUT method. When analyzing simulated data, where all scatter is Gaussian, our method detects (falsely) one or more outlier in only about 1-3% of experiments. When analyzing data contaminated with one or several outliers, the ROUT method performs well at outlier identification, with an average False Discovery Rate less than 1%. CONCLUSION: Our method, which combines a new method of robust nonlinear regression with a new method of outlier identification, identifies outliers from nonlinear curve fits with reasonable power and few false positives.

Computer Simulation↗

Regression analysis when covariates are regression parameters of a random effects model for observed longitudinal measurements.

We consider regression analysis when covariate variables are the underlying regression coefficients of another linear mixed model. A naive approach is to use each subject's repeated measurements, which are assumed to follow a linear mixed model, and obtain subject-specific estimated coefficients to replace the covariate variables. However, directly replacing the unobserved covariates in the primary regression by these estimated coefficients may result in a significantly biased estimator. The aforementioned problem can be evaluated as a generalization of the classical additive error model where repeated measures are considered as replicates. To correct for these biases, we investigate a pseudo-expected estimating equation (EEE) estimator, a regression calibration (RC) estimator, and a refined version of the RC estimator. For linear regression, the first two estimators are identical under certain conditions. However, when the primary regression model is a nonlinear model, the RC estimator is usually biased. We thus consider a refined regression calibration estimator whose performance is close to that of the pseudo-EEE estimator but does not require numerical integration. The RC estimator is also extended to the proportional hazards regression model. In addition to the distribution theory, we evaluate the methods through simulation studies. The methods are applied to analyze a real dataset from a child growth study.

Bias↗