Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “missing data”

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 613 records · Page 34Linked to original sources
PubMed2002

Joint modeling of event time and nonignorable missing longitudinal data.

Survival studies usually collect on each participant, both duration until some terminal event and repeated measures of a time-dependent covariate. Such a covariate is referred to as an internal time-dependent covariate. Usually, some subjects drop out of the study before occurrence of the terminal event of interest. One may then wish to evaluate the relationship between time to dropout and the internal covariate. The Cox model is a standard framework for that purpose. Here, we address this problem in situations where the value of the covariate at dropout is unobserved. We suggest a joint model which combines a first-order Markov model for the longitudinally measured covariate with a time-dependent Cox model for the dropout process. We consider maximum likelihood estimation in this model and show how estimation can be carried out via the EM-algorithm. We state that the suggested joint model may have applications in the context of longitudinal data with nonignorable dropout. Indeed, it can be viewed as generalizing Diggle and Kenward's model (1994) to situations where dropout may occur at any point in time and may be censored. Hence we apply both models and compare their results on a data set concerning longitudinal measurements among patients in a cancer clinical trial.

Algorithms↗
PubMed1996

Malaria vaccine trials: the missing qualitative data.

Recent population-based efficacy trials of the synthetic malaria vaccine SPf66 have shown restricted, if any, clinical protection against Plasmodium falciparum infection. Despite the well-established role of antibodies in effector responses against asexual blood-stage malaria parasites, the titres of anti-SPf66 IgG antibodies do not correlate with the ability of sera from vaccine recipients to inhibit parasite growth in vitro nor with partial clinical protection which could be detected in some trials. Qualitative or functional parameters of SP66-induced antibody responses, such as IgG subclass composition and affinity, may be more predictive of clinical protection against malaria than quantitative estimates of antibody concentration or titre. Since these parameters are readily estimated by laboratory techniques currently available, and may be modulated by changes in vaccination protocols and by the use of different adjuvants, a better understanding of qualitative antibody responses induced by SPf66 and other asexual blood-stage malaria vaccine candidates, and of their relationship with clinical protection in vivo, is urgently needed for the improvement of currently used immunization schedules.

Animals↗
PubMed2005

Survey on haematopoietic stem cell transplantation for children in Europe.

A recent report, prepared in March 2003, regarding the paediatric transplantation activity registered between 1970 and 2002 in the European Bone Marrow Transplantation (EBMT) database showed a decrease in the number of registrations in 2001 and in 2002. In order to validate this observation, the Paediatric Diseases Working Party (PDsWP) secretariat distributed a questionnaire to 395 institutions participating in the EBMT Registry. Each institution was requested to check the number of transplants they reported and to confirm or to correct the figures. As of 15 March 2004, replies had been received from 135 centres reporting a median of 48 transplants per centre over the study period, total 17 891 (58% of the total number). Among them, 55 confirmed their original figures, while 80 corrected the numbers. The overall number of autologous and allogeneic SCTs performed and not reported were 461 and 692, respectively. Most of the teams that corrected their figures stated that their data managers could provide missing data to the EBMT; 260 other teams, each reporting a median of 15 transplants during the study period, total 12 866 (42% of the total number) chose not to reply. A report prepared in March 2004, following the PDsWP survey, showed an increasing number of transplants performed on patients below 18 years of age between 1973 and 2002 and reported to the EBMT Registry (328 autologous and 628 allogeneic) as compared to the 2003 report. This first PDsWP survey, reaching more than 50% of activity in the field, illustrates that the decrease in activity we observed in the 2003 report does not correspond to a decrease in the number of transplants that were actually performed. It demonstrates the compliance of most major paediatric institutions and confirms the important role of cooperation between National Registries and EBMT Registries.

Adolescent↗
PubMedDate not supplied

[Ambulant training programs for obese children. Criterions of comparison for the development of valid therapy recommendations].

Valid knowledge concerning structure and contents of an ambulant training program for obese children and adolescents suggested by experts is still missing. In 1999/2000, we assessed a survey based on defined criterions in the "Arbeitsgemeinschaft für pädiatrische Diäthetik (APD)" and investigated the literature to cover and to compare based on defined criterions the spectrum of ambulant treatment models in Germany to advance guidelines for therapy in obese children. 13 ambulant training programs were detected. Training of children aged 8 to 12 years and their parents based on diet, behavioural treatment and sports therapy over couple of months without great experiences and without complete financing by health insurance were in common. Time volume of the training programs widely varies. In the seven evaluated training programs, up to one third of the patients broke off, while more than two third of the children successfully finished the training programs. The comparison of the programs failed on missing data concerning the collective, missing or different evaluation criteria, sometimes low number of patients and not standardised calculation of overweight. The survey reflects the often low experiences and the heterogeneous, insufficient evaluation of ambulant treatment models in Germany and demonstrates the great requirement of research. The schedule of criterions facilitates the comparison of programs to develop valid ambulant training programs for obese children.

Adolescent↗
PubMed2006

The ancestral distance test: what relatedness can reveal about correlated evolution in large lineages with missing character data and incomplete phylogenies.

The ancestral distance test is introduced to detect correlated evolution between two binary traits in large phylogenies that may lack resolved subclades, branch lengths, and/or comparative data. We define the ancestral distance as the time separating a randomly sampled taxon from its most recent ancestor (MRA) with extant descendants that have an independent trait. The sampled taxon either has (target sample) or lacks (nontarget sample) a dependent trait. Modeled as a Markov process, we show that the distribution of ancestral distances for the target sample is identical to that of the nontarget sample when characters are uncorrelated, whereas ancestral distances are smaller on average for the target sample when characters are correlated. Simulations suggest that the ancestral distance can be estimated using the time, total branch length, taxonomic rank, or number of speciation events between a sampled taxon and the MRA. These results are shown to be robust to deviations from Markov assumptions. A Monte Carlo technique estimates P-values when fully resolved phylogenies with branch lengths are available, and we evaluate the Monte Carlo approach using a data set with known correlation. Measures of relatedness were found to provide a robust means to test hypotheses of correlated character evolution.

Classification↗
PubMed2003

Analysis of survival data with missing measurements of a time-dependent binary covariate.

The objective of this study was to investigate the influence of the number and timing of a binary time-dependent covariate on the bias using the last-observation carried forward in the proportional hazards model. Under various assumptions of censoring rates, transition probabilities of the time-dependent covariate, sample size, and the log hazard-ratio for the covariate, we empirically examined the impact that the number and timing have on the bias of the estimator of the covariate. An example from the Systolic Hypertension in the Elderly Program was used. Inference on the effect of systolic blood pressure on survival is strongly affected by the number and timing of systolic blood measurements.

Humans↗
PubMed2001

The Multiple Sclerosis Impact Scale (MSIS-29): a new patient-based outcome measure.

Changes in health policy have underlined the importance of evidence-based clinical practice and rigorous evaluation of patient-based outcomes. As patient-based outcome measurement is particularly important in treatment trials of multiple sclerosis, a number of disease-specific instruments have been developed recently. One limitation of these instruments is that none was developed using the standard psychometric approach of reducing a large item pool generated from people with multiple sclerosis. Consequently, an outcome measure for clinical trials of multiple sclerosis that is disease specific and combines patient perspective with rigorous psychometric methods will complement existing instruments. The aim of this study was to develop such a measure. Standard psychometric methods were used. A pool of 129 questionnaire items was generated from interviews with 30 people with multiple sclerosis, expert opinion and literature review. The questionnaire was administered by postal survey to 1530 people selected randomly from the Multiple Sclerosis Society membership database. Redundant items and those with limited measurement properties were removed. The remaining items (n = 41) were grouped into scales using factor analysis, and then refined to form the Multiple Sclerosis Impact Scale (MSIS-29), an instrument measuring the physical (20 items) and psychological (nine items) impact of multiple sclerosis. Five psychometric properties of the MSIS-29 (data quality, scaling assumptions, acceptability, reliability and validity) were examined in a separate postal survey of 1250 Multiple Sclerosis Society members. A preliminary responsiveness study of the MSIS-29 was undertaken in 55 people admitted for rehabilitation and intravenous steroid treatment of relapses. The MSIS-29 satisfied all psychometric criteria. Data quality was excellent, missing data were low (maximum 3.9%), item test-re-test reliability was high (r = 0.65-0.90) and scale scores could be generated for >98% of respondents. Item descriptive statistics, item convergent and discriminant validity, and factor analysis indicated that it was legitimate to generate scores for MSIS-29 scales by summing items. MSIS-29 scales showed good variability, small floor and ceiling effects, high internal consistency (Cronbach's alpha <or=0.91) and high test-re-test reliability (intraclass correlation <or=0.87). Correlations with other measures and the analysis of group differences provided evidence that the MSIS-29 measures the physical and psychological impact of multiple sclerosis. Effect sizes (physical scale = 0.82, psychological scale = 0.66) demonstrated preliminary evidence of good responsiveness. These results indicate the MSIS-29 is a clinically useful and scientifically sound patient-based outcome measure of the impact of multiple sclerosis suitable for clinical trials and epidemiological studies.

Evidence-Based Medicine↗
PubMed1990

Research in physical medicine and rehabilitation. VIII. Preliminary data analysis.

This paper describes important aspects of preliminary data analysis to be taken after data are checked for clerical entry errors and before the primary statistical analysis is performed. These include description and graphic display of each variable, recoding categorical data, transforming continuous data into another continuous variable and recoding continuous to categorical data. Missing values and outlying data points are identified and several techniques are recommended to minimize mistakes in variable recoding. Related variables measured with different units may be combined by using the z transformation and converted back to one of the original units for ease of interpretation. Finally, both categorical and continuous variables are checked for reliability by using kappa or the intraclass R.

Data Collection↗
PubMed2000

X-ray CT metal artifact reduction using wavelets: an application for imaging total hip prostheses.

Traditional computed tomography (CT) reconstructions of total joint prostheses are limited by metal artifacts from corrupted projection data. Published metal artifact reduction methods are based on the assumption that severe attenuation of X-rays by prostheses renders corresponding portions of projection data unavailable, hence the "missing" data are either avoided (in iterative reconstruction) or interpolated (in filtered backprojection with data completion; typically, with filling data "gaps" via linear functions). In this paper, we propose a wavelet-based multiresolution analysis method for metal artifact reduction, in which information is extracted from corrupted projection data. The wavelet method improves image quality by a successive interpolation in the wavelet domain. Theoretical analysis and experimental results demonstrate that the metal artifacts due to both photon starving and beam hardening can be effectively suppressed using our method. As compared to the filtered backprojection after linear interpolation, the wavelet-based reconstruction is significantly more accurate for depiction of anatomical structures, especially in the immediate neighborhood of the prostheses. This superior imaging precision is highly advantageous in geometric modeling for fitting hip prostheses.

Artifacts↗
PubMed2003

Generalized additive selection models for the analysis of studies with potentially nonignorable missing outcome data.

Rotnitzky, Robins, and Scharfstein (1998, Journal of the American Statistical Association 93, 1321-1339) developed a methodology for conducting sensitivity analysis of studies in which longitudinal outcome data are subject to potentially nonignorable missingness. In their approach, they specify a class of fully parametric selection models, indexed by a non- or weakly identified selection bias function that indicates the degree to which missingness depends on potentially unobservable outcomes. Estimation of the parameters of interest proceeds by varying the selection bias function over a range considered plausible by subject-matter experts. In this article, we focus on cross-sectional, univariate outcome data and extend their approach to a class of semiparametric selection models, using generalized additive restrictions. We propose a backfitting algorithm to estimate the parameters of the generalized additive selection model. For estimation of the mean outcome, we propose three types of estimating functions: simple inverse weighted, doubly robust, and orthogonal. We present the results of a data analysis and a simulation study.

Acquired Immunodeficiency Syndrome↗
PubMed2005

Challenges in arranging interhospital transfers from a small regional hospital: an observational study.

OBJECTIVE: To describe the process of interhospital transfer from a small regional hospital ED. METHODS: Prospective observational case series of patients transferred from the ED of West Gippsland Hospital (WGH), Victoria from February to August 2003. Outcome measures were expressed as the number of phone calls required to facilitate interhospital transfer, the time taken for a proposed transfer to be accepted and the total time a transferred patient spent in the ED. RESULTS: Eighty-one of 129 transferred patients had all data available. Those with missing data had similar ED length of stays, and were of similar age and acuity to those with complete data available. Mean (95% CI) values for the study population were number of phone calls: 4.7 (95% CI 3.96-5.43); time to transfer acceptance: 56.7 min (95% CI 19.1-94.8 min); and total time in ED: 307 min (95% CI 32.9-372.6 min). Men (P = 0.01), older patients (P = 0.02) and those for whom a centralized referral service were consulted (P = 0.006) were more likely to require four or more phone calls to facilitate transfer. The time until a transfer was accepted was more likely to be longer than an hour for patients who were transferred to critical care areas (P = 0.02) and those for whom a centralized referral service was consulted (P = 0.002). Patients referred to another ED spent less time in WGH ED than other transferred patients. CONCLUSION: There is wide variation in the time taken and the number of phone calls to arrange an interhospital transfer, with some patient groups requiring more effort to organize a transfer.

Adolescent↗
PubMed2003

Estimating the prevalence of breast cancer using a disease model: data problems and trends.

BACKGROUND: Health policy and planning depend on quantitative data of disease epidemiology. However, empirical data are often incomplete or are of questionable validity. Disease models describing the relationship between incidence, prevalence and mortality are used to detect data problems or supplement missing data. Because time trends in the data affect their outcome, we compared the extent to which trends and known data problems affected model outcome for breast cancer. METHODS: We calculated breast cancer prevalence from Dutch incidence and mortality data (the Netherlands Cancer Registry and Statistics Netherlands) and compared this to regionally available prevalence data (Eindhoven Cancer Registry, IKZ). Subsequently, we recalculated the model adjusting for 1) limitations of the prevalence data, 2) a trend in incidence, 3) secondary primaries, and 4) excess mortality due to non-breast cancer deaths. RESULTS: There was a large discrepancy between calculated and IKZ prevalence, which could be explained for 60% by the limitations of the prevalence data plus the trend in incidence. Secondary primaries and excess mortality had relatively small effects only (explaining 17% and 6%, respectively), leaving a smaller part of the difference unexplained. CONCLUSION: IPM models can be useful both for checking data inconsistencies and for supplementing incomplete data, but their results should be interpreted with caution. Unknown data problems and trends may affect the outcome and in the absence of additional data, expert opinion is the only available judge.

Journal Article↗
PubMed2004

Improving the evaluation of therapeutic interventions in multiple sclerosis: development of a patient-based measure of outcome.

OBJECTIVES: To develop a patient-based, disease-specific measure of the health impact of multiple sclerosis (MS) for use in clinical trials and clinical practice. DATA SOURCES: People with MS. Members of the MS Society of Great Britain and Northern Ireland. METHODS: Standard psychometric methods were used to develop the Multiple Sclerosis Impact Scale (MSIS-29) in three stages. Stage 1 (item generation): questionnaire items were generated from 30 patient interviews on the impact of MS on their lives, expert opinion and literature review. Stage 2 (item reduction and scale generation): the questionnaire developed in stage 1 was administered by postal survey to 1530 randomly selected members of the MS Society. Standard item reduction techniques were used to develop a rating scale from the pool of questionnaire items. Stage 3 (psychometric evaluation): the questionnaire was evaluated for data quality, scaling assumptions, acceptability, reliability and validity in a separate postal survey of 1250 MS Society members. Responsiveness was evaluated in 55 people admitted to hospital for rehabilitation and intravenous steroid treatment of MS relapses. RESULTS: Stage 1 resulted in a 129-item questionnaire. Stage 2 resulted in a 29-item rating scale measuring the physical and psychological impact of MS. The MSIS-29 satisfied all recommended psychometric criteria for rigorous measurement. Data quality was excellent: missing data were low, item test-retest reliability was high and scale scores could be generated for over 98% of respondents. Item descriptive statistics, item convergent and discriminant validity, and factor analysis supported summing items to produce two summary scores. MSIS-29 physical and psychological scale scores showed good variability, low floor and ceiling effects, good internal consistency and test-retest reliability. Correlations with other measures and confirmation of hypotheses about group differences provided evidence for the validity of the MSIS-29 as a measure of the physical and psychological impact of multiple sclerosis. Effect sizes provided preliminary evidence for responsiveness. CONCLUSIONS: The 29-item MSIS-29 is a rigorous new measure of the physical and psychological impact of MS. All psychometric criteria were satisfied and there is preliminary evidence of responsiveness. The MSIS-29 is particularly appropriate for use in clinical trials to evaluate therapeutic effectiveness from the patient's perspective. Further critical evaluations of the MSIS-29 completed by people with neurologist-confirmed MS in different settings are suggested. Head-to-head comparisons of the psychometric properties of the MSIS-29 and other outcome measures for MS will help to determine the relative advantages of different instruments so that the choice of measures for studies can be evidence based.

Adult↗
PubMed2004

Psychometric evaluation of the SF-36 health survey in Medicare managed care.

Data quality and scoring assumptions for the SF-36 Health Survey were evaluated among the elderly and disabled, using 1998 Cohort I baseline Medicare HOS data (n=177,714). Missing data rates were low, and scoring assumptions were met. Internal consistency reliability was 0.83 to 0.93 for the eight scales and 0.94 and 0.89, respectively, for the physical (PCS) and mental (MCS) component summary measures. Results declined with increased risk factors (e.g., older age, more chronic conditions), but were well above accepted standards for all subgroups. These findings support using standard algorithms for scoring the SF-36 in the HOS and subgroup analyses of HOS data.

Adult↗