Search PubMed⌕ Search

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.

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

At least 451 records · Page 25Linked to original sources
PubMed1995

Facial expressions of patients with dementia: a comparison of two methods of interpretation.

Two methods of interpreting the videotaped facial expressions of four patients with severe dementia of the Alzheimer type were compared. Interpretations of facial expressions performed by means of unstructured naturalistic judgments revealed episodes when the four patients exhibited anger, disgust, happiness, sadness, and surprise. When these episodes were assessed by use of a modified version of the Facial Action Coding System, there was, in total, 48% agreement between the two methods. The highest agreement, 98%, occurred for happiness shown by one patient. It was concluded that more emotions could be judged by means of the unstructured naturalistic method, which is based on an awareness of the total situation that facilitates imputing meaning into the patients' cues. It is a difficult task to find a balance between imputing too much meaning into the severely demented patients' sparse and unclear cues and ignoring the possibility that there is some meaning to be interpreted.

Aged↗
PubMed1999

Evaluation of a simplified vitamin supplement inventory developed for the Women's Health Initiative.

OBJECTIVE: To evaluate the accuracy of a simplified inventory procedure for assessing nutrient intake from vitamin and mineral supplements. DESIGN: Participants brought their supplements to a clinic. An interviewer conducted the supplement inventory procedure, which consisted of recording data on the type of multiple vitamin and single supplements used. For the multiple vitamins, the interviewer recorded the exact dose for a subset of nutrients (vitamin C, calcium, selenium). For other nutrients, we imputed the dose in multiple vitamins. The dose of all single supplements was recorded. Labels of the supplements were photocopied and we transcribed the exact nutrient label data for the criterion measure. Spearman correlation coefficients were used to assess precision of nutrient intakes from the simplified inventory compared to the criterion measure. SETTING/SUBJECTS: Data are from 104 adult vitamin supplement users in Washington state. RESULTS: Correlation coefficients between nutrient intake estimated from the simplified inventory compared to the criterion measure were high (0.8-1.0) for those nutrients (vitamin C, calcium, selenium) for which the interviewer recorded the exact dose contained in multiple vitamins. However, for nutrients for which imputations were made regarding dose in multiple vitamins, correlation coefficients ranged from good (0.8 for vitamin E) to poor (0.3 for iron). CONCLUSIONS: The simplified inventory is rapid (4-5 min) and practical for large-scale studies. The precision of nutrient estimates using this procedure was variable, although excellent for the subset of nutrients for which the dose was recorded exactly. This study illustrates many of the challenges of collecting high quality supplement data.

Adult↗
PubMed1998

Estimating the effect of treatment on quality of life in the presence of missing data due to drop-out and death.

Quality of life was measured in a study of two multidrug regimens for mycobacterium avium complex MAC bactaeremia using the MOS-HIV questionnaire. The effect of treatment on quality of life was estimated at each follow-up time in three ways: (1) using only the observed data, (2) after assigning the worst possible quality of life scores for individuals who died, and (3) after imputing missing scores for patients who either died or dropped out of the study. The overall quality of life scores were also compared between treatment groups with categorical generalized estimating equation models and three-dimensional graphs. Of the 179 patients included in these analyses, 84 (47%) either died or dropped out during the 16 week study period. When the quality of life scores were compared between the treatment groups with the Wilcoxon rank sum test using only the observed data, there was no significant difference between the groups at 16 weeks of follow-up. When the worst possible quality of life scores were assumed for patients who had died, both the magnitude and the statistical significance of the difference in the quality of life scores between the groups increased. Imputing missing data for patients who either dropped out or died resulted in even larger differences in quality of life between the treatment groups. We conclude that ignoring missing data due to drop-outs and death can result in an underestimation of the treatment effect and overly optimistic statements about the outcome of the participants on both treatment arms due to the selective drop-out of participants with poorer quality of life. To obtain a valid assessment of the effect of treatment on quality of life, the experience of the patients who died or dropped out of the study must be considered.

AIDS-Related Opportunistic Infections↗
PubMedDate not supplied

Worker decisions to purchase health insurance.

Studying worker health insurance choices is usually limited by the absence of price data for workers who decline their employer's offer. This paper uses a new Medical Expenditure Panel Survey file which links household and employer survey respondents, supplying data for both employer insurance takers and declines. We test for whether out-of-pocket or total premium better explains worker behavior, estimate price elasticities with observed prices and with imputed prices, and test for worker sorting among jobs with and without health insurance. We find that out-of-pocket price dominates, that there is some upward bias from estimating elasticities with imputed premiums rather than observed premiums, and that workers do sort among jobs but this does not affect elasticity estimates appreciably. Like earlier studies with less representative worker samples, we find worker price elasticity of demand to be quite low. This suggests that any premium subsidies must be large to elicit much change in worker take-up behavior.

Adult↗
PubMed2004

How well do parents know their children? Implications for proxy reporting of child health-related quality of life.

OBJECTIVES: This study examined parental knowledge of their children's oral-health-related quality of life (OHRQoL) (Objective 1), and the effects of different analytical techniques to manage 'Don't know' (DK) responses on the validity and reliability of the questionnaire (Objective 2) and the level of agreement between parental and child reports (Objective 3). METHODS: The parental (PPQ) and child (CPQ11-14) components of the Child Oral Health Quality of Life Questionnaire were used. Objectives 1 and 2 were addressed in the study that involved 221 parents and Objective 3 in the study that involved 63 pairs of parents and children. Four methods for treating DK responses in the PPQ were tested: listwise deletion, item mean imputation, imputation of the value zero and adjustment of scores to account for items with DK responses. RESULTS: Respectively, 26 and 11% of the parents gave > or = 3 and 6 > or = DK responses to 33 items comprising the PPQ. DK responses were associated with child's age and clinical condition, and parental gender. The methods of managing DK responses did not have differing effects on the measurement properties of the PPQ and the level of agreement between parents and children. CONCLUSION: Some parents have limited knowledge concerning their children's OHRQoL. However, given that parental and child reports are measuring different realities, information provided by parents is useful even if it is incomplete.

Adolescent↗
PubMed2005

Maternal employment and child development: a fresh look using newer methods.

The employment rate for mothers with young children has increased dramatically over the past 25 years. Estimating the effects of maternal employment on children's development is challenged by selection bias and the missing data endemic to most policy research. To address these issues, this study uses propensity score matching and multiple imputation. The authors compare outcomes across 4 maternal employment patterns: no work in first 3 years postbirth, work only after 1st year, part-time work in 1st year, and full-time work in 1st year. Our results demonstrate small but significant negative effects of maternal employment on children's cognitive outcomes for full-time employment in the 1st year postbirth as compared with employment postponed until after the 1st year. Multiple imputation yields noticeably different estimates as compared with a complete case approach for many measures. Differences between results from propensity score approaches and regression modeling are often minimal.

Child↗
PubMed2024

Rare variant analyses in 51,256 type 2 diabetes cases and 370,487 controls reveal the pathogenicity spectrum of monogenic diabetes genes.

Type 2 diabetes (T2D) genome-wide association studies (GWASs) often overlook rare variants as a result of previous imputation panels' limitations and scarce whole-genome sequencing (WGS) data. We used TOPMed imputation and WGS to conduct the largest T2D GWAS meta-analysis involving 51,256 cases of T2D and 370,487 controls, targeting variants with a minor allele frequency as low as 5 × 10-5. We identified 12 new variants, including a rare African/African American-enriched enhancer variant near the LEP gene (rs147287548), associated with fourfold increased T2D risk. We also identified a rare missense variant in HNF4A (p.Arg114Trp), associated with eightfold increased T2D risk, previously reported in maturity-onset diabetes of the young with reduced penetrance, but observed here in a T2D GWAS. We further leveraged these data to analyze 1,634 ClinVar variants in 22 genes related to monogenic diabetes, identifying two additional rare variants in HNF1A and GCK associated with fivefold and eightfold increased T2D risk, respectively, the effects of which were modified by the individual's polygenic risk score. For 21% of the variants with conflicting interpretations or uncertain significance in ClinVar, we provided support of being benign based on their lack of association with T2D. Our work provides a framework for using rare variant GWASs to identify large-effect variants and assess variant pathogenicity in monogenic diabetes genes.

Diabetes Mellitus, Type 2↗
PubMed2002

Vitamin K and childhood cancer: analysis of individual patient data from six case-control studies.

To investigate the hypothesis that neonates who receive intramuscular vitamin K are at an increased risk of developing cancer, particularly leukaemia, a pooled analysis of individual patient data from six case-control studies conducted in Great Britain and Germany has been undertaken. Subjects comprised 2431 case children diagnosed with cancer before 15 years of age and 6338 control children. The retrospective assessment of whether or not an individual baby received vitamin K is not straightforward. In many cases no record was found in stored medical notes and two types of analysis were therefore conducted; in the first it was assumed that where no written record of vitamin K was found it had not been given, and in the second, where no written record of administration was found, information on hospital policy and perinatal morbidity was used to 'impute' whether or not vitamin K had been given. In the first analysis, no association was found between neonatal administration of intramuscular. vitamin K and childhood cancer: odds ratios adjusted for mode of delivery, admission to special care baby unit and low birth weight were 1.09 (95% confidence interval 0.92-1.28) for leukaemia and 1.05 (0.92-1.20) for other cancers. In the second analysis, the adjusted odds ratios increased to 1.21 (1.02-1.44) for leukaemia and 1.10 (0.95-1.26) for other cancers. This shift did not occur in all studies, and when data from the hypothesis generating Bristol study were excluded, the adjusted odds ratios for leukaemia became 1.06 (0.89-1.25) in the first analysis and 1.16 (0.97-1.39) when data on prophylaxis imputed from hospital policy and perinatal morbidity were used. We conclude that whilst the broad nature of the diagnostic groups and the poor quality of some of the vitamin K data mean that small effects cannot be entirely ruled out, our analysis provides no convincing evidence that intramuscular vitamin K is associated with childhood leukaemia.

Adolescent↗
PubMed2004

Vitamin A deficiency and xerophthalmia among school-aged children in Southeastern Asia.

OBJECTIVE: To determine provisional estimates of the extent of vitamin A (VA) deficiency and xerophthalmia among school-aged children. DESIGN: Literature search of published, unpublished and website-based population survey and study reports, with country-specific imputation of prevalence rates and numbers of children affected by: (1) VA deficiency based on measured or imputed distributions of serum retinol concentration < 0.70 micromol/l (equivalent to < 20 microg/dl) and (2) xerophthalmia, by country. SETTING: Countries within the WHO South-East Asian Region. SUBJECTS: The target group for estimation was children 5-15 y of age. INTERVENTIONS: None. RESULTS: The estimated prevalence of VA deficiency is 23.4%, suggesting that there are approximately 83 million VA-deficient school-aged children in the region, of whom 10.9% (9 million, at an overall prevalence of 2.6%) have mild xerophthalmia (night blindness or Bitot's spot). Potentially blinding corneal xerophthalmia appears to be negligible at this age. CONCLUSIONS: VA deficiency, including mild xerophthalmia, appears to affect large numbers of school-aged children in South-East Asia. However, nationally representative data on the prevalence, risk factors and health consequences of VA deficiency among school-aged children are lacking within the region and globally, representing a future public health research priority.

Adolescent↗
PubMed1998

Serotonin neural adaptations to ontogenetic loss of dopamine neurons in rat brain.

In rat, the neonatal destruction of nigrostriatal dopamine (DA) neurons by intracerebral administration of 6-hydroxydopamine entails dramatic changes in serotonin (5-hydroxytryptamine, 5-HT) as well as DA function. Most striking is the 5-HT hyperinnervation of the adult neostriatum, associated with increases in density of various 5-HT receptor subtypes and enhanced neuronal responsiveness to the iontophoretic application of 5-HT and its 5-HT(1B/2C) and 5-HT(2A/2C) receptor agonists, m-chlorophenylpiperazine and iododimethoxyphenylaminopropane. The topographical distribution of these changes is consistent with up-regulation and/or increased production and transport of 5-HT1B and 5-HT2A receptors by the neostriatal projection neurons, as confirmed for the 5-HT2A receptor in a recent in situ hybridization study. It is interesting that this study has also shown that increases in both 5-HT2A binding and mRNA level were abolished by chronic pretreatment with the DA agonists, apomorphine and SKF 38393, suggesting a regulatory influence of DA in the expression of this 5-HT receptor. D1 receptor binding is known to be slightly reduced in the rostral neostriatum of these rats, a down-regulation apparently imputable to a reduced rate of synthesis of the receptor. In contrast, D2 receptor binding is increased throughout the DA-denervated and 5-HT-hyperinnervated neostriatum, perhaps due to some posttranscriptional modifications. Stereotyped and motor behaviors induced by systemic treatment with D1 and D2 agonists are markedly enhanced in these rats (behavioral supersensitivity), although priming is commonly required to unmask a latent D1 supersensitivity. In the case of oral activity, however, overt behavioral supersensitivity is induced by D1 as well as D2 agonists. Moreover, there is overt supersensitivity of oral activity in response to the 5-HT receptor agonist m-chlorophenylpiperazine, which is presumably imputable to 5-HT2C receptors and may be demonstrated even in the absence of supersensitivity to D1 receptor agonist. 5-HT adaptations, therefore, seem to play a role not only in the abnormal spontaneous behavior, but also in the behavioral supersensitivity to 5-HT as well as DA receptor agonists in these rats.

Adaptation, Physiological↗
PubMed2004

Severe adverse drug reactions of antidepressants: results of the German multicenter drug surveillance program AMSP.

The goal of the German drug safety program in psychiatry AMSP (Arzneimittelsicherheit in der Psychiatrie) is the assessment of severe or new adverse drug reactions (ADRs). Here we report on 53,042 of 122,562 patients treated with antidepressants who were monitored from 1993 to 2000 in 35 psychiatric hospitals in German-speaking countries. The overall incidence of severe ADRs of antidepressants was 1.4 % of exposed patients; when only ADRs rated as probable or definite were considered, a rate of 0.9 % in patients treated with antidepressants was observed. ADR rates were higher for TCAs (imputed in 1.0 % of patients overall, respectively in 0.6 % of patients when only ADs were imputed) and lower for MAO inhibitors and SSRIs (0.7 % for both, respectively 0.3 % and 0.4 %). Within the TCA group there was a difference among clomipramine (2.1 %, respectively 1.0 %), amitriptyline (1.0 %, respectively 0.6 %), and doxepin or trimipramine (both 0.6 %, respectively 0.3 %). With regard to single SSRI, similar rates were observed for paroxetine (0.8 %, respectively 0.5 %) and for citalopram (0.7 %, respectively 0.4 %). Of the new dual-acting antidepressants, venlafaxine ranged at 0.9 %, (respectively 0.5 %) and mirtazapine at 0.6 % (respectively 0.5 %). In particular, TCAs were associated with known risks, such as toxic delirium, grand mal seizures, and hepatic (i. e., increased liver enzymes), urologic (i. e., urinary retention), allergic (i. e., exanthema), or cardiovascular (i. e., mainly orthostatic collapse) reactions. In SSRI-treated patients (non-delirious) psychic and neurological ADRs were most prominent, followed by gastrointestinal, dermatologic, and endocrinological/electrolyte reactions, with agitation, hyponatremia (probably as part of the SIADH syndrome and associated with severe neurologic or psychiatric symptoms in 64 % of all cases), increased liver enzymes, nausea, and the serotonin syndrome as leading unwanted symptoms. Venlafaxine (in the immediate-release formulation) was associated with adverse CNS and somatic symptoms such as severe agitation, diarrhea, increased liver enzymes, hypertension, and hyponatremia. Mirtazapine was mostly connected with increased liver enzymes, cutaneous edema, and collapse, but with no case of significant hyponatremia. For drugs that potently inhibit serotonin uptake, serum sodium concentration should be controlled when applied in high-dose therapy or in vulnerable patients.

Adult↗
PubMed2004

Galactorrhea due to psychotropic drugs.

Within the drug safety program in psychiatry AMSP ( Arzneimittelsicherheit in der Psychiatrie), severe adverse drug reactions (ADRs) are assessed. Currently 35 psychiatric hospitals and departments are participating in detecting severe ADRs. This paper focuses on prolactin-dependent ADRs such as gynecomastia and galactorrhea due to psychotropic medications. Related to the number of patients surveyed (122,562 from 1993 to 2000), these are rare events (0.03 % or 35 cases). Imputed drugs were mostly antipsychotics, but antidepressants were also imputed in single cases. In the group of antipsychotics, relative frequencies of galactorrhea were highest for amisulpride and risperidone and corresponded to the degree of D2 binding. Galactorrhea assessed as "severe" was accompanied by distressing symptoms such as pain, tension, enlargement of breasts, or soaked clothing. The AMSP data contribute to the knowledge on endocrine ADRs by the large number of patients examined and help clinicians select the appropriate drug if their patients have been prone to for these ADRs in the past.

Adult↗
PubMed2001

Formal analysis of the optimal duration of tilt testing for the diagnosis of neurally mediated syncope.

BACKGROUND: Although tilt testing has emerged as the test of choice for assessing patients with suspected neurally mediated syncope, the optimum duration of tilt testing is poorly defined. This in part relates to the absence of a gold standard to assess test performance. OBJECTIVE: Our purpose was to formally estimate the effects of varying duration of drug-free tilt testing on test performance in diagnosing neurally mediated syncope. DESIGN: If a test's specificity is known, then in the absence of a gold standard an imputed (estimated) sensitivity may be calculated on the basis of the observed diagnostic yield in a given population as a function of assumed population prevalence. We determined the relationship of specificity to drug-free tilt test duration by use of data from 11 previous studies reporting the results of drug-free tilt testing in a total of 435 control subjects (60 to 80 degrees of tilt, footboard support, 15- to 60-minute duration). Data (weighted for study size) were fit to an exponential function relating specificity to tilt duration. Test yield was evaluated as a function of tilt duration in 213 consecutive patients referred to our laboratory for the evaluation of suspected neurally mediated syncope who underwent passive tilt testing for up to 30 to 60 minutes. RESULTS: The estimated specificity of tilt testing was 94% at 30 minutes, 92% at 40 minutes, and 88% after 60 minutes of passive tilt. The cumulative yield of tilt testing was only 17% at 30 minutes, 22% at 40 minutes, and 28% after 60 minutes. On the basis of an estimated population prevalence of 25% to 50% in this referral population, imputed sensitivity is 27% to 48% at 30 minutes, 36% to 64% at 40 minutes, and 43% to 74% after 60 minutes of passive tilt. The overall diagnostic accuracy was not strongly influenced by tilt duration beyond 30 minutes and ranged from 60% to 84%. CONCLUSIONS: Passive tilt testing (ie, tilt testing without pharmacologic provocation) for durations of up to 60 minutes has limited sensitivity for diagnosing neurally mediated syncope. For populations with a pretest likelihood of 25% to 50%, test results are inaccurate in one to two fifths of patients.

Data Interpretation, Statistical↗
PubMed1994

Repeated measures for two within-subject factors: analysis and missing data solutions.

The statistical analysis of the repeated measures design with two factors within and no factors between subjects, which is popular in clinical pharmacology, is discussed. Use of restricted maximum likelihood (REML) methodology is compared to an imputation procedure in small sample situations with missing data and is illustrated by simulations. While imputation leads to undesirable results that are not easily corrected, REML estimation in which test statistics are compared to an F-distribution provides an elegant tool for the analysis of these designs.

Algorithms↗
PubMed1998

A comparison of the random-effects pattern mixture model with last-observation-carried-forward (LOCF) analysis in longitudinal clinical trials with dropouts.

The last-observation-carried-forward imputation method is commonly used for imputting data missing due to dropouts in longitudinal clinical trials. The method assumes that outcome remains constant at the last observed value after dropout, which is unlikely in many clinical trials. Recently, random-effects regression models have become popular for analysis of longitudinal clinical trial data with dropouts. However, inference obtained from random-effects regression models is valid when the missing-at-random dropout process is present. The random-effects pattern-mixture model, on the other hand, provides an approach that is valid under more general missingness mechanisms. In this article we describe the use of random-effects pattern-mixture models under different patterns for dropouts. First, subjects are divided into groups depending on their missing-data patterns, and then model parameters are estimated for each pattern. Finally, overall estimates are obtained by averaging over the missing-data patterns and corresponding standard errors are obtained using the delta method. A typical longitudinal clinical trial data set is used to illustrate and compare the above methods of data analyses in the presence of missing data due to dropouts.

Clinical Trials as Topic↗
PubMed2006

A fast and flexible statistical model for large-scale population genotype data: applications to inferring missing genotypes and haplotypic phase.

We present a statistical model for patterns of genetic variation in samples of unrelated individuals from natural populations. This model is based on the idea that, over short regions, haplotypes in a population tend to cluster into groups of similar haplotypes. To capture the fact that, because of recombination, this clustering tends to be local in nature, our model allows cluster memberships to change continuously along the chromosome according to a hidden Markov model. This approach is flexible, allowing for both "block-like" patterns of linkage disequilibrium (LD) and gradual decline in LD with distance. The resulting model is also fast and, as a result, is practicable for large data sets (e.g., thousands of individuals typed at hundreds of thousands of markers). We illustrate the utility of the model by applying it to dense single-nucleotide-polymorphism genotype data for the tasks of imputing missing genotypes and estimating haplotypic phase. For imputing missing genotypes, methods based on this model are as accurate or more accurate than existing methods. For haplotype estimation, the point estimates are slightly less accurate than those from the best existing methods (e.g., for unrelated Centre d'Etude du Polymorphisme Humain individuals from the HapMap project, switch error was 0.055 for our method vs. 0.051 for PHASE) but require a small fraction of the computational cost. In addition, we demonstrate that the model accurately reflects uncertainty in its estimates, in that probabilities computed using the model are approximately well calibrated. The methods described in this article are implemented in a software package, fastPHASE, which is available from the Stephens Lab Web site.

Calibration↗
PubMed2001

Missing value estimation methods for DNA microarrays.

MOTIVATION: Gene expression microarray experiments can generate data sets with multiple missing expression values. Unfortunately, many algorithms for gene expression analysis require a complete matrix of gene array values as input. For example, methods such as hierarchical clustering and K-means clustering are not robust to missing data, and may lose effectiveness even with a few missing values. Methods for imputing missing data are needed, therefore, to minimize the effect of incomplete data sets on analyses, and to increase the range of data sets to which these algorithms can be applied. In this report, we investigate automated methods for estimating missing data. RESULTS: We present a comparative study of several methods for the estimation of missing values in gene microarray data. We implemented and evaluated three methods: a Singular Value Decomposition (SVD) based method (SVDimpute), weighted K-nearest neighbors (KNNimpute), and row average. We evaluated the methods using a variety of parameter settings and over different real data sets, and assessed the robustness of the imputation methods to the amount of missing data over the range of 1--20% missing values. We show that KNNimpute appears to provide a more robust and sensitive method for missing value estimation than SVDimpute, and both SVDimpute and KNNimpute surpass the commonly used row average method (as well as filling missing values with zeros). We report results of the comparative experiments and provide recommendations and tools for accurate estimation of missing microarray data under a variety of conditions.

Algorithms↗
PubMed2003

Estimating disease prevalence in two-phase studies.

Disease prevalence is ideally estimated using a 'gold standard' to ascertain true disease status on all subjects in a population of interest. In practice, however, the gold standard may be too costly or invasive to be applied to all subjects, in which case a two-phase design is often employed. Phase 1 data consisting of inexpensive and non-invasive screening tests on all study subjects are used to determine the subjects that receive the gold standard in the second phase. Naive estimates of prevalence in two-phase studies can be biased (verification bias). Imputation and re-weighting estimators are often used to avoid this bias. We contrast the forms and attributes of the various prevalence estimators. Distribution theory and simulation studies are used to investigate their bias and efficiency. We conclude that the semiparametric efficient approach is the preferred method for prevalence estimation in two-phase studies. It is more robust and comparable in its efficiency to imputation and other re-weighting estimators. It is also easy to implement. We use this approach to examine the prevalence of depression in adolescents with data from the Great Smoky Mountain Study.

Adolescent↗