SEARCH · Search PubMed

Results for “Imputation”

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.

SourceResearch connectionFrom yearThrough year

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

At least 181 records · Page 10Linked to original sources

PubMed2006

A simple imputation method for longitudinal studies with non-ignorable non-responses.

Missing data are a common problem in longitudinal studies in the health sciences. Motivated by data from the Muscatine Coronary Risk Factor (MCRF) study, a longitudinal study of obesity, we propose a simple imputation method for handling non-ignorable non-responses (i.e., when non-response is related to the specific values that should have been obtained) in longitudinal studies with either discrete or continuous outcomes. In the proposed approach, two regression models are specified; one for the marginal mean of the response, the other for the conditional mean of the response given non-response patterns. Statistical inference for the model parameters is based on the generalized estimating equations (GEE) approach. An appealing feature of the proposed method is that it can be readily implemented using existing, widely-available statistical software. The method is illustrated using longitudinal data on obesity from the MCRF study.

Algorithms↗

PubMed2004

Imputing gene-treatment interactions when the genotype distribution is unknown using case-only and putative placebo analyses--a new method for the Genetics of Hypertension Associated Treatment (GenHAT) study.

There is a sizeable literature on methods for detecting gene-environment interaction in the framework of case-control studies, particularly with reference to the assumption of independence of genotype and exposure. In the context of a clinical trial, wherein gene-drug interactions with regard to outcomes are examined, these methods may be readily applied, as gene and drug are independent by randomization. In an active-controlled trial (experimental treatment vs standard) that has collected genotype information, gene-drug interactions can be estimated. In addition, the effect of the experimental treatment vs placebo can be imputed by using data from a historical placebo-controlled trial (standard vs placebo) if either (a) genotype information is available from the historical trial or (b) assumptions are made about the prevalence of genotype and the odds ratios of genotype and disease in the historical trial using information from other studies. Motivation for these procedures is provided by the Genetics of Hypertension Associated Treatment, a large pharmacogenetics, ancillary study of a hypertension clinical trial, and examples from published hypertension trials will be used to illustrate the methods.

Controlled Clinical Trials as Topic↗

PubMed2006

Analysis of incomplete longitudinal binary data using multiple imputation.

We propose a propensity score-based multiple imputation (MI) method to tackle incomplete missing data resulting from drop-outs and/or intermittent skipped visits in longitudinal clinical trials with binary responses. The estimation and inferential properties of the proposed method are contrasted via simulation with those of the commonly used complete-case (CC) and generalized estimating equations (GEE) methods. Three key results are noted. First, if data are missing completely at random, MI can be notably more efficient than the CC and GEE methods. Second, with small samples, GEE often fails due to 'convergence problems', but MI is free of that problem. Finally, if the data are missing at random, while the CC and GEE methods yield results with moderate to large bias, MI generally yields results with negligible bias. A numerical example with real data is provided for illustration.

Clinical Trials as Topic↗

PubMed1993

Improved imputation of non-responses to mailback questionnaires.

Poor response rates, perhaps as low as 10 to 30 per cent, certainly bring into question the validity of any survey suffering such a malady. Repeated mailouts are capable of providing information on those persons who are resistant to responding. If one is willing to assume an exponential drop-off in response rates, then imputation is possible for the entire group of non-responders. Furthermore, a second exponential drop-off in favourable (or non-favourable) responses to particular items within the questionnaire leads to closed form estimates of population parameters and their associated standard errors. Three or more mailouts also provide a test of validity of the model assumed.

Bias↗

PubMedDate not supplied

The effect of covariates on the induction time of AIDS using improved imputation of exact seroconversion times.

This paper explores the effects of covariates such as age, gender, chronological time of seroconversion, and mode of virus transmission on the interval between HIV seroconversion and the development of AIDS. For many cohorts which are available to provide information about such effects, both the time of seroconversion and the time of onset of AIDS may be right- or interval-censored. To analyse these doubly-censored data, we propose a method in which we first estimate the marginal distribution of the chronological time of seroconversion and use this distribution to impute the seroconversion time for each subject. Subsequently, we estimate the distribution of induction times conditional on the seroconversion distribution. We apply this technique to a cohort of HIV seropositive subjects recruited from various outpatient facilities throughout Italy and show that age at seroconversion has the greatest effect on induction time.

Acquired Immunodeficiency Syndrome↗

PubMed2000

Cadmium-dependent enzyme activity alteration is not imputable to lipid peroxidation.

The effect of cadmium on the liver-specific activities of NADPH-cytochrome P450 reductase (CPR), malic dehydrogenase (MDH), glyceraldehyde-3-phosphate dehydrogenase (GADPH), and sorbitol dehydrogenase (SDH) was assessed 6, 24, and 48 h after administration of the metal to rats (2.5 mg/kg of body weight, as CdCl2, single ip injection). CPR specific activity increased after 6 h and afterward decreased significantly, while MDH specific activity increased up to 24 h and then remained unchanged. Both SDH and GADPH specific activities reduced after 6 h, the former only a little but the latter much more, and after 24 and 48 h were strongly inhibited. In vitro experiments, by incubating rat liver microsomes, mitochondria, or cytosol with CdCl2 in the pH range 6.0-8.0, excluded cadmium-induced lipid peroxidation as the cause of the reduction in enzyme activity. In addition, from these experiments, we obtained indications on the type of interactions between cadmium and the enzymes studied. In the case of CPR, the inhibitory effect is probably due to Cd2+ binding to the histidine residue of the apoenzyme, which, at physiological pH, acts as a nucleophilic group. In vitro, mitochondrial MDH was not significantly affected by cadmium at any pH, indicating that this enzyme is probably not involved in the decrease in mitochondrial respiration caused by this metal. As for GADPH specific activity, its inhibition at pH 7.4 and above is imputable to the binding of cadmium to the SH groups present in the enzyme active site, since in the presence of dithiothreitol this inhibition was removed. SDH was subjected to a dual effect when cytosol was exposed to cadmium. At pH 6.0 and 6.5, its activity was strongly stimulated up to 75 microM CdCl2 while at higher metal concentrations it was reduced. At pH 7.4 and 8.0, a stimulation up to 50 microM CdCl2 occurred but above this concentration, a reduction was found. These data seem to indicate that cadmium can bind to different enzyme sites. One, at low cadmium concentration, stimulates the SDH activity while the other, at higher metal concentrations, substitutes for zinc, thus causing inhibition. This last possibility seems to occur in vivo essentially at least 24 h after intoxication. The cadmium-induced alterations of the investigated enzymes are discussed in terms of the metabolic disorders produced which are responsible for several pathological conditions.

Animals↗

PubMed1996

A computer algorithm to impute interrupted heart rate data for the spectral analysis of heart rate variability--the ARIC study.

The shorter term beat-to-beat heart rate data collected from the general population are often interrupted by artifacts, and an arbitrary exclusion of such individuals from analysis may significantly reduce the sample size and/or introduce selection bias. A computer algorithm was developed to label as artifacts any data points outside the upper and lower limits generated by a 5-beat moving average +/- 25% (or set manually by an operator using a mouse) and to impute beat-to-beat heart rate throughout an artifact period to preserve the timing relationships of the adjacent, uncorrupted heart rate data. The algorithm applies Fast Fourier Transformation to the smoothed data to estimate low-frequency (LF; 0.025-0.15 Hz) and high-frequency (HF; 0.16-0.35 Hz) spectral powers and the HF/LF ratio as conventional indices of sympathetic, vagal, and vagal-sympathetic balance components, respectively. We applied this algorithm to resting, supine, 2-min beat-to-beat heart rate data collected in the population-based Atherosclerosis Risk in Communities study to assess the performance (success rate) of the algorithm (N = 526) and the inter-and intra-data-operator repeatability of using this computer algorithm (N = 108). Eighty-eight percent (88%) of the records could be smoothed by the computer-generated limits, an additional 4.8% by manually set limits, and 7.4% of the data could not be processed due to a large number of artifacts in the beginning or the end of the records. For the repeatability study, 108 records were selected at random, and two trained data operators applied this algorithm to the same records twice within a 6-month interval of each process (blinded to each other's results and their own prior results). The inter-data-operator reliability coefficients were 0.86, 0.92, and 0.90 for the HF, LF, and HF/LF components, respectively. The average intra-data-operator reliability coefficients were 0.99, 0.99, and 0.98 for the HF, LF, and HF/LF components, respectively. These results indicate that this computer algorithm is efficient and highly repeatable in processing short-term beat-to-beat heart rate data collected from the general population, given that the data operators are trained according to standardized protocol.

Algorithms↗

PubMed2005

Ximelagatran compared with warfarin for the prevention of systemic embolism and stroke. An imputed placebo analysis.

The active control trials, SPORTIF III and SPORTIF V, compared the direct thrombin inhibitor ximelagatran to warfarin, where each was given as a treatment to prevent systemic embolism and stroke in patients with atrial fibrillation. Because warfarin has previously been compared to placebo in similar patients and ximelagatran has now been compared to warfarin, an indirect comparison between ximelagatran and placebo is possible (imputed placebo analysis). In this analysis, ximelagatran reduces the risk of stroke and systemic embolism by 66% (hazard ratio 0.338; 95% confidence interval [CI] 0.204-0.560). Ximelagatran preserves 102% (95% CI 72-132%) of the benefit of warfarin. Based on these data, ximelagatran may be an effective alternative to warfarin for the prevention of stroke and systemic embolism in high-risk patients with atrial fibrillation.

Aged↗

PubMed2004

A method for imputing missing data in longitudinal studies.

PURPOSE: In a cohort in which racial data are unknown for some persons, race-specific persons and person-years are imputed using a model-based iterative allocation algorithm (IAA). METHODS: An EM algorithm-based approach to address misclassification in a censored data regression setting can be adapted to estimate the probability that a person of unknown race is white. The corresponding race-specific person-years are obtained as a by-product of the estimation procedure. Variance estimates are computed using the bootstrap. The proposed approach is compared with the proportional allocation method (PAM). RESULTS: In an occupational cohort where racial data were missing for 41% of the workers, the age-time-race-specific person-years were estimated within a relative variation of approximately 20%, using the IAA. The deaths were less reliably estimated. The standardized mortality ratios (SMRs) for all-cause mortality estimated using the IAA and the PAM were more similar for the non-white workers than for a smaller subgroup of white workers. CONCLUSIONS: The IAA provides a method to reliably estimate race-specific person-year denominators in cohort studies with missing racial data. This method is applicable to other incompletely observed non-time-dependent categorical covariates. Internal cohort rates or SMRs can be computed and modeled, with bootstrap confidence intervals that account for the uncertainty in the determination of race.

Adolescent↗

PubMed2001

Imputation of missing dates of death or institutionalization for time-to-event analyses in the Canadian Study of Health and Aging.

Data from the Canadian Study of Health and Aging (CSHA) allow investigators to study patterns and predictors of mortality and institutional placement in a well characterized, population-based cohort of elderly Canadians. However, it is impossible to study the timing of these events if the date of occurrence is missing. This technical article describes a procedure for imputing missing dates of death or institutionalization. The first step consists in identifying and correcting dates that are inconsistent with other available dates on which we know the event has or has not occurred. A missing date for an event is then replaced by the middle of a range of plausible dates for its occurrence. This constitutes a valuable addition to the CSHA data since it precludes the loss of information that results from discarding subjects with missing occurrence dates in time-to-event analyses.

Aged↗

PubMed2005

Coparenting conflict, nonacceptance, and depression among divorced adults: results from a 12-year follow-up study of child custody mediation using multiple imputation.

Using statistically imputed data to increase available power, this article reevaluated the long-term effects of divorce mediation on adults' psychological adjustment and investigated the relations among coparenting custody conflict, nonacceptance of marital termination, and depression at 2 occasions over a decade apart following marital dissolution. Group comparisons revealed that fathers and parents who mediated their custody disputes reported significantly more nonacceptance at the 12-year follow-up assessment. Significant interactions were observed by gender in regression models predicting nonacceptance at the follow-up; mothers' nonacceptance was positively associated with concurrent depression, whereas fathers' nonacceptance was positively associated with early nonacceptance and negatively associated with concurrent conflict.

Adaptation, Psychological↗

PubMed2005

[Missing data in the somatisation subscale of the scl-90-R due to a 'checklist effect': occurrence in different surveys and results of various imputation methods].

Item non-response is a potential threat to the validity of study results. Taking the somatisation subscale of the SCL-90-R as an example, we hypothesise a specific response pattern ("checklist-effect") that is characterised by symptom-free persons not checking the "not at all"-category. The present study analyses the extent and relevance of this postulated "checklist-effect". Our data is derived from a survey of n = 228 blue-collar workers who previously had filed applications for medical rehabilitation benefits (A1-study), and two additional surveys as well. We defined the "checklist-effect" by the following response pattern: (1) at least one missing value and (2) at least one valid item response and (3) no "not at all"-responses. Occurrence of the "checklist-effect" in the three datasets differed widely. 75 % of the responders in the A1-Study had complete data, 16.2 % a postulated "checklist-effect". Imputation of missing values under the assumption of a "checklist-effect" led to a reduction of missing data in the somatisation-subscale from 12.3 % to 0.4 %. Ignoring the "checklist-effect" would overestimate the symptom level. Possible explanations for the effect are discussed. However, the validity of this effect has yet to be proven via methodological studies.

Bias↗

PubMed1994

Sequential imputation for multilocus linkage analysis.

A Monte Carlo method called sequential imputation is proposed for multilocus likelihood computations. This method is most useful in mapping situations where the data consist of large pedigrees with substantial missing information and it is desirable to perform linkage analysis utilizing data from many polymorphic markers simultaneously. A pedigree example with 155 individuals, 9 loci, and 155,520 haplotypes is used for illustration.

Diabetes Mellitus, Type 2↗

PubMed1996

Functional characterization of DNA-binding domains of the subunits of the heterodimeric aryl hydrocarbon receptor complex imputing novel and canonical basic helix-loop-helix protein-DNA interactions.

The aryl hydrocarbon receptor (AHR) and the aryl hydrocarbon receptor nuclear translocator (ARNT) belong to a novel subclass of basic helix-loop-helix transcription factors. The AHR.ARNT heterodimer binds to the xenobiotic responsive element (XRE). Substitution of each of four amino acids in the basic region of ARNT with alanine severely diminishes or abolishes XRE binding, intimating that these amino acids contact DNA bases. Three of these amino acids are conserved among basic helix-loop-helix proteins, and the corresponding amino acids of Max and USF are known to contact DNA bases. Alanine scanning mutagenesis of the basic domain of AHR and substitution with conservative amino acids at particular positions in this domain and in a more amino-proximal AHR segment previously shown to be required for XRE binding (Fukunaga, B. N., and Hankinson, O. (1996) J. Biol. Chem. 271, 3743-3749) demonstrate that the most carboxyl-proximal amino acid position of the basic domain and a position within the amino-proximal segment are intolerant to amino acid substitution with regard to XRE binding, suggesting that these two amino acids make base contacts. Amino acid positions in these AHR regions and in the ARNT basic region less adversely affected by substitution are also identified. The amino acids at these positions may contact the phosphodiester backbone. The apparent bipartite nature of the DNA binding region of AHR and the identity of those of its amino acids that apparently make DNA contacts impute a novel protein-DNA binding behavior for AHR.

Amino Acid Sequence↗

PubMed2006

Multiple-imputation for measurement-error correction.

BACKGROUND: There are many methods for measurement-error correction. These methods remain rarely used despite the ubiquity of measurement error. METHODS: Treating measurement error as a missing-data problem, the authors show how multiple-imputation for measurement-error (MIME) correction can be done using SAS software and evaluate the approach with a simulation experiment. RESULTS: Based on hypothetical data from a planned cohort study of 600 children with chronic kidney disease, the estimated hazard ratio for end-stage renal disease from the complete data was 2.0 [95% confidence limits (95% CL) 1.4, 2.8] and was reduced to 1.5 (95% CL 1.1, 2.1) using a misclassified exposure of low glomerular filtration rate at study entry (sensitivity of 0.9 and specificity of 0.7). The MIME correction hazard ratio was 2.0 (95% CL 1.2, 3.3), the regression calibration (RC) hazard ratio was 2.0 (95% CL 1.1, 3.7), and restriction to a 25% validation substudy yielded a hazard ratio of 2.0 (95% CL 1.0, 3.7). Based on Monte Carlo simulations across eight scenarios, MIME was approximately unbiased, had approximately correct coverage, and was sometimes more powerful than misclassified or RC analyses. Using root mean squared error as a criterion, the MIME bias correction is sometimes outweighed by added imprecision. CONCLUSION: The choice between MIME and RC depends on performance, ease, and objectives. The usefulness of MIME correction in specific applications will depend upon the sample size or the proportion validated. MIME correction may be valuable in interpreting imperfectly measured epidemiological data.

Bias↗

PubMed2002

Recurrent fatal drug-induced toxic epidermal necrolysis (Lyell's syndrome) after putative beta-lactam cross-reactivity: Case report and scrutiny of antibiotic imputability.

OBJECTIVE: A series of antibiotics may be responsible for toxic epidermal necrolysis. We report two successive episodes of toxic epidermal necrolysis in the same patient. Drug imputability criteria designate a cross-reactivity between two antibiotics of different chemical classes but sharing the beta-lactam ring in common. DESIGN: Descriptive case report and review of the literature. SETTING: Medical intensive care unit in a university medical center. PATIENT AND MAIN RESULTS: A 75-yr-old woman developed a first episode of toxic epidermal necrolysis (involving 40% of the body surface) after intake of cefotaxime, a third-generation cephalosporin. Perfusions of high-dose immunoglobulins rapidly improved the lesions, followed by partial reepithelialization in 5 days. Sepsis required the administration of meropenem, which is a carbapenem antibiotic. The epidermal destruction immediately recurred, with extension to previously uninvolved skin areas and fatal consequences. CONCLUSIONS: The beta-lactam ring present in cephalosporins and carbapenems represents the putative chemical structure responsible for the presently reported cross-reactivity to two antibiotics of different classes. Drugs having any chemical similarity to the initial culprit compound should be strictly avoided when possible in the management of toxic epidermal necrolysis.

Aged↗

PubMed2007

Impact of supported housing on clinical outcomes: analysis of a randomized trial using multiple imputation technique.

In 1992, the US Department of Housing and Urban Development (HUD) and the US Department of Veterans Affairs (VA) established the HUD-VA Supported Housing (HUD-VASH) Program to provide integrated clinical and housing services to homeless veterans with psychiatric and/or substance abuse disorders at 19 sites. At four sites, 460 subjects were randomly assigned to one of the three groups: (1) HUD-VASH, with both Section 8 vouchers and intensive case management; (2) case management only; and (3) standard VA care. A previous publication found HUD-VASH resulted in superior housing outcomes but yielded no benefits on clinical outcomes. Since many participants missed prescheduled visits during the follow-up period and follow-up rates were quite different across the groups, we reanalyzed these data using multiple imputation statistical methods to account for the missing observations. Significant benefits were found for HUD-VASH in drug and alcohol abuse outcomes that had not previously been identified.

Alcoholism↗

PubMedDate not supplied

Children's ability to impute inferentially based knowledge.

Do young children appreciate the importance of access to premises when judging what another person knows? In Experiment 1, 5-year-olds (N=31) were sensitive to another person's access to premises when predicting that person's ability to point to a target after eliminating alternatives in a set of 3 cartoon characters. Experiment 2 replicated the finding when 5- to 6-year-olds (N=102) judged who the other person thought the target was, and whether the other person knew who the target was. Experiment 3 demonstrated that children aged 5-7 years (N=107) more successfully imputed inference by elimination than syllogistical inferential knowledge. Findings suggest that an early understanding of inference by elimination offers a route into understanding that people can sometimes gain knowledge without direct perceptual access.

Aptitude↗