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 685 records · Page 38Linked to original sources
PubMedDate not supplied

Repeated measures designs in behavioral toxicology: application to chronic marijuana smoke exposure.

This paper discusses the application of repeated measures methods in the statistical analysis of an experiment in behavioral toxicology. The chronic marijuana smoke exposure study conducted at the National Center for Toxicological Research is used for an example of the types of problems that one encounters in analyzing these types of studies. In particular, the standard univariate analysis most frequently used for repeated measures analyses has some very restrictive assumptions on the form of the covariance matrices. These assumptions are not met in the example discussed and are rarely met in many other problems. Other possible models for analyzing repeated measures when these assumptions are not met are presented and discussed. Other problems specific to the chronic marijuana smoke exposure study that may occur in similar type studies are presented. These include pooling the experimental units into groups with comparable baselines, choosing a function of the measures to be analyzed, dealing with a large data set with many observation times and missing data, unequal group sizes and different designs for different subsets of the experimental animals. The standard univariate repeated measures analysis was chosen to analyze the data even though the violations of the covariance assumptions may lead to finding differences that do not exist (Type I or false-positive errors), since the other methods presented also had covariance assumptions that were not met or had low power. Use of Bonferroni-type multiple comparisons on the single degree of freedom contrasts of interest hopefully reduced the chances of these false-positive results.

Analysis of Variance↗
PubMedDate not supplied

Using large data bases in nursing and health policy research.

Concern about the quality, cost, and outcomes of health care has become a driving force in health policy research. The growing accessibility of large clinical and administrative health care data bases has led to an interest in using such data in health policy research. Clinical data bases are created by providers of care and contain data about episodes and outcomes of care, usually organized as patient records. Administrative data bases contain data about indirect care processes such as insurance claims processing, vital event recording, and quality assurance. Clinical and administrative data bases may contain millions of records, consist of data from multiple sites, and often have missing data issues that must be considered by researchers. These and other characteristics of large data bases require special data manipulation and analytic techniques. Large data bases have been used in epidemiological studies, risk assessment, and technology assessment and to study variations in caregiver practice patterns. Because the use of large data bases by nurse researchers has been constrained by the lack of nursing-relevant data in them, there is a need to reach consensus on useful and feasible nursing data elements and to include those data in ongoing data collection efforts by government agencies and private organizations.

Bias↗
PubMed1997

A dilemma in analysis: issues in the serial measurement of quality of life in patients with advanced lung cancer.

Despite the availability of several instruments to evaluate quality of life (QL) over time in patients with lung cancer, barriers in measurement remain. This methodological study used LCSS data (Lung Cancer Symptom Scale, a disease- and site-specific QL measure) to examine analysis methods to quantify QL where data needed for serial evaluation may be missing. Data from two large randomized trials, conducted at 30 centers, of a new combination chemotherapy regimen incorporating a new agent for patients (n = 673) with Stage III and IV non-small cell lung cancer were obtained for this study. QL had been prospectively measured at baseline, day 29, and every six weeks thereafter using the LCSS. For the slope analysis (SA) and area under the curve (AUC) analyses, an adjustment score of zero was used to indicate QL on the day of death (mortality adjustment) and each subsequent day until the end of the assessment period. Significant differences in QL, symptom scores and known prognostic factors at baseline were found in the attrition group. SA and AUC analysis allowed inclusion of 581 patients, giving an adequacy rate of 86%. By using a mortality adjustment, an additional 45 patients were included, increasing the inclusion rate to 93%. With the use of the mortality adjustment, QL was shown to decline over the interval, as opposed to rise if the adjustment had not been performed. The conclusions of the study were: (1) analysis for serial data using SA and AUC provides useful, but differing information; (2) when attrition (caused by death) is a factor, a mortality adjustment presented a more accurate assessment of QL as an endpoint; (3) more frequent evaluations of QL will capture rapid changes in patient status and reduce the attrition bias; (4) all patients should be followed until they die; and (5) QL should be given full consideration as a primary endpoint separate from survival.

Aged↗
PubMed2006

The implementation of an anesthesia information management system.

BACKGROUND AND OBJECTIVE: Anaesthesia information management systems, though still not used widely, will inevitably replace handwritten records and may eventually serve as a core for the development of computerized decision support. We investigated staff expectations and the accuracy of data entry in a recently implemented commercially available anaesthesia information management system. METHODS: A structured questionnaire was administered to the staff before and 1 week and 3 months after implementation in order to assess their opinion. The quality of manual data entry, and of automatic data record was evaluated by looking for missing data and the prevalence of artefacts. RESULTS: Despite initial fears the users quickly accepted the system. Both automatic and manual data entry were found to be accurate and reliable while the prevalence of artefacts was relatively low. CONCLUSIONS: A commercially available anaesthesia information management system can be easily implemented and used instead of paper charts.

Adult↗
PubMed2001

A review of gynaecological cancer management: a large West Midlands general hospital experience.

The records of 372 women with gynaecological cancer were reviewed to assess the quality of care which has been given and to compare the standards achieved with those suggested by The West Midlands Gynaecological Oncology Group. The median primary referral interval was 18 days; 54% of patients were seen within 2 weeks of referral. The median interval between first hospital visit and surgery was 32 days and 77.5% of patients had their operation within 8 weeks. The median interval between histology report and starting adjuvant therapy was 19 days and 81% of patients commenced treatment within 6 weeks. The histological diagnosis was available within 2 weeks of surgery in 77% of patients. Less than one-third of patients in our series were referred to the lead consultant of gynaecological oncology, who performed 50% of the operations for gynaecological malignancy. There was no record of the primary referral interval, date of surgery and date of commencement of adjuvant therapy in 35%, 25% and 20% of patients, respectively. More attention must be given to redistribution of caseload with more involvement of the gynaecological cancer team and to improvement in clinical cancer data collection to decrease the extent of missing data. The delays in referral of patients with suspected gynaecological malignancy could be minimised by improving communication between the general practitioners and the hospitals.

Journal Article↗
PubMed1994

Validity of parental report of a child's medical history in otitis media research.

The authors compared parental reports with medical records for 157 children enrolled in a prospective study of chronic otitis media with effusion between 1987 and 1991. Parents completed a questionnaire about the child's past health history, and the research nurse abstracted history information from the clinic's medical record. Previous insertion of a tympanostomy tube (kappa = 0.96) and premature birth (kappa = 0.68) were accurately reported, but there was a substantial proportion of missing data for age at first episode of otitis media, occurrence of otitis media the previous summer, and number of episodes in the previous 18 months. Data were significantly more likely to be missing for male children, children with siblings, and those with more episodes. Parents who reported six or more previous episodes for their child overestimated the number compared with the medical record (8.7 vs. 7.4, respectively; p = 0.01), while those who reported fewer episodes underestimated the number (3.1 vs. 4.6, respectively; p = 0.01). Episodes of otitis media during the 3 months between study visits were also accurately reported (kappa = 0.94). The accuracy and completeness of parental report of the child's health history was influenced by the chronicity of otitis media, the duration of recall, and the seriousness of the event being recalled.

Child, Preschool↗
PubMed2006

Differential item functioning (DIF) and the Mini-Mental State Examination (MMSE). Overview, sample, and issues of translation.

BACKGROUND: Various forms of differential item functioning (DIF) in the Mini-Mental State Examination (MMSE) have been identified. Items have been found to perform differently for individuals of different educational levels, racial/ethnic groups, and/or of groups whose first language is not English. The articles in this section illustrate the use of different methods to examine DIF in relation to English and Spanish language administration of the MMSE. OBJECTIVES: The aim of this article is to provide a context for interpretation of the findings contained in the following set of papers examining DIF in the MMSE. METHODS: The performance of the MMSE, when administered in English and Spanish, was reviewed. "Translation" has been discussed in the context of measurement bias, illustrating the variability in Spanish translations. Presented are the readability of the MMSE, description of the translation method, the study design and sample for the data set used, together with treatment of missing data, and model assumptions related to the analyses described in the accompanying set of papers examining DIF. CONCLUSIONS: The examination of item bias in cognitive impairment assessment instruments has practical and theoretical implications in the context of health disparities. Considerable DIF has been identified in the MMSE. A critical factor that may contribute to measurement bias is language translation and conversion. Once DIF has been established consistently in a measure, decisions regarding adjustments proceed. Perhaps the development of guidelines for appropriate adjustments for DIF correction in self-reported measures represents the next challenge in addressing measurement equivalence in crosscultural research.

Aged↗
PubMed2004

Treatment patterns and cost-of-illness of severe haemophilia in patients with inhibitors in Germany.

To evaluate current treatment patterns and resource utilization as well as related cost in the management of severe haemophilia patients with inhibitors in Germany, a cost-of-illness study was conducted. Generally, data were generated by structured literature search. Missing data were collected by expert interviews. All data were validated by a panel of German experts in haemophilia care. In Germany, immune tolerance therapy (ITT) is first-line therapy in inhibitor management for children in the initial year after inhibitor development, particularly for high responders (HR). In adult HR patients ITT is applied but to a remarkably lower extent than in children. To treat bleeding episodes, factor VIII (FVIII) is first-line therapy in low responders (LR). For paediatric HR patients, bleeds are mainly treated with recombinant FVIIa (rFVIIa). In adult HR patients, activated prothrombin complex concentrate (aPCC) and rFVIIa are more equally distributed as treatment options. Treatment costs were calculated for paediatric patients (15 kg) and adult patients (75 kg) from third party payers' perspective. Cost for ITT ranges from Euro 70,290 (2 months; LR) to Euro 3 812,400 (24 months; with aPCC; HR) in a paediatric patient. For an adult patient ITT cost ranges from Euro 287,500 (6 months; LR) to Euro 17,253,000 (36 months; HR). For on average 12.5 acute bleeds, average annual treatment costs amount to Euro 77,000 for a child and Euro 354,000 for an adult. Assessing the results it has been taken into consideration that ITT can last longer and annual number of bleeds can be extremely higher than on average 12.5 episodes. This indicates more health care resource consumption in some patients.

Adult↗
PubMed2005

Prediction of death for extremely low birth weight neonates.

OBJECTIVE: To compare multiple logistic regression and neural network models in predicting death for extremely low birth weight neonates at 5 time points with cumulative data sets, as follows: scenario A, limited prenatal data; scenario B, scenario A plus additional prenatal data; scenario C, scenario B plus data from the first 5 minutes after birth; scenario D, scenario C plus data from the first 24 hours after birth; scenario E, scenario D plus data from the first 1 week after birth. METHODS: Data for all infants with birth weights of 401 to 1000 g who were born between January 1998 and April 2003 in 19 National Institute of Child Health and Human Development Neonatal Research Network centers were used (n = 8608). Twenty-eight variables were selected for analysis (3 for scenario A, 15 for scenario B, 20 for scenario C, 25 for scenario D, and 28 for scenario E) from those collected routinely. Data sets censored for prior death or missing data were created for each scenario and divided randomly into training (70%) and test (30%) data sets. Logistic regression and neural network models for predicting subsequent death were created with training data sets and evaluated with test data sets. The predictive abilities of the models were evaluated with the area under the curve of the receiver operating characteristic curves. RESULTS: The data sets for scenarios A, B, and C were similar, and prediction was best with scenario C (area under the curve: 0.85 for regression; 0.84 for neural networks), compared with scenarios A and B. The logistic regression and neural network models performed similarly well for scenarios A, B, D, and E, but the regression model was superior for scenario C. CONCLUSIONS: Prediction of death is limited even with sophisticated statistical methods such as logistic regression and nonlinear modeling techniques such as neural networks. The difficulty of predicting death should be acknowledged in discussions with families and caregivers about decisions regarding initiation or continuation of care.

Female↗
PubMedDate not supplied

[Quality of life at the end of life. Analysis of the quality of life of oncologic patients treated with palliative care. Results of a multicenter observational study (staging)].

Outcome in palliative care can be defined as patients' quality of life, quality of death and satisfaction with care. In an Italian multicentre prospectic study ('Staging') the quality of life of 571 palliative care patients with advanced cancer disease was assessed since the beginning of palliative care till the end of the study. We analyzed the tissue of quality of life missing data and the possibility to input the missing quality of life evaluation through the quality of life evaluation made by a proxy (doctor, nurse). The greatest functional impairment and an increasing level of some symptoms (fatigue, general malaise, emotional status) were observed during the last two weeks of life, whereas for other symptoms (gastro-intestinal, pain) some degree of control was possible. The quality of life analysis for palliative care patients should consider the different response of different quality of life components to the palliative care intervention.

Aged↗
PubMed2005

Data quality in the outpatient setting: impact on clinical decision support systems.

Clinical decision support system (CDSS) performance may vary with the quality of the input data. We assessed the impact of medical record completeness and accuracy on a CDSS that provides risk assessment for gastrointestinal bleeding and recommends therapy when prescribing NSAIDs. We examined the documentation of six data elements in the medical record and the impact on the performance of the CDSS. We reviewed 178 transcribed clinical encounters from standardized patients with predefined clinical histories. Results showed that the mean completeness score across all encounters was .34. The mean correctness score for those elements present was .94. When the available data was input into the CDSS, the missing data elements resulted in inappropriate and unsafe recommendations in almost 77% of the encounters. The results show that important gaps in the medical record can affect the accuracy of a CDSS designed to improve safe prescribing.

Adult↗
PubMed2005

Analysis of incomplete quality of life data in advanced stage cancer: a practical application of multiple imputation.

This paper presents a practical approach to analyzing incomplete quality of life (QOL) data that contains non-ignorable dropouts in patients with advanced non-small-cell lung cancer (NSCLC). QOL scores for the physical domain at baseline and at the end of the first and second courses of chemotherapy were compared between two treatment groups in a phase III trial. One hundred and 103 eligible patients were randomized to receive cisplatin and irinotecan (CPT-P) or cisplatin and vindesine, respectively; of those two groups, 83 and 85, respectively, completed a QOL questionnaire at least at baseline. A multiple imputation incorporating auxiliary QOL variables was implemented as one of alternatives of sensitivity analyses; these were complete case, available case, and pattern mixture analyses. Although larger sensitivity to missing data was found for CPT-P treatment, none of the alternative analyses demonstrated a significant difference in estimated slopes over time between the groups. This study presents an analytical approach for dealing with the complex problem of missing QOL data. It must be noted, however, that the validity of the multiple imputation method we present is not certain unless we can specify sufficiently informative auxiliary variables to ensure the conversion of non-ignorable missingness to ignorable.

Aged↗
PubMed1998

Creating a general practice national minimum data set: present possibility or future plan?

AIM: To assess the feasibility of implementing the recommendations of the New Zealand National Minimum Data Set working party in computerised general practices. METHOD: Doctors from 12 computerised general practices belonging to the Royal New Zealand College of General Practitioners' Dunedin Research Unit Computer Network participated in the study (five Dunedin practices, four in rural Otago and Southland, and three in Christchurch). A three-month sample of data was extracted from practice computers and evaluated for completeness and compliance to the national minimum data set structure. Rates of recording practice identifier, provider, patient identifiers, sex, ethnicity, government subsidy eligibility, consultation identifier and date, prescriptions and Read codes were calculated for each practice. RESULTS: Apart from data recorded automatically by computers, there was a wide range in the extent of missing data. Of the data requiring manual computer entry, patient demography and subsidy eligibility were most comprehensively recorded (date of birth 99.9%, sex 99.6%, eligibility to subsidies 98.5%). Data with little immediate clinical or management relevance were poorly recorded (Read codes 32.4% and ethnicity 5.0%). CONCLUSIONS: It is possible to derive a common minimum data set from different computerised general practices. However some data elements will be missing unless suitable education and support are provided for the doctors and other staff members who record patient information.

Data Collection↗
PubMedDate not supplied

Missing quality of life data in cancer clinical trials: serious problems and challenges.

Measurement of quality of life (QOL) in cancer clinical trials has increased in recent years as more groups realize the importance of such endpoints. A key problem has been missing data. Some QOL data may unavoidably be missing, as for example when patients are too ill to complete forms. Other important sources are potentially avoidable and can broadly be divided into three categories: (i) methodological factors; (ii) logistic and administrative factors; (iii) patient-related factors. Logistic and administrative factors, for example, staff oversights, have proven to be most important. Since most QOL measurements require patient self-report, it is usually not possible to rectify the failure to collect baseline data or any follow-up assessments. There is strong evidence that such data are not 'missing at random', and cannot be ignored without introducing bias. Although several approaches to the analysis of partly missing data have been described, none is entirely satisfactory. Prevention of avoidable missing data is better than attempted cure. In July 1996, an international conference on missing QOL data in cancer clinical trials reported the experience of most major groups involved. This paper will serve as an introduction to the problem and provide an estimation of its magnitude, and approaches to its prevention and solution.

Bias↗
PubMedDate not supplied

Full conjugate analysis of normal multiple traits with missing records using a generalized inverted Wishart distribution.

A Markov chain Monte Carlo (MCMC) algorithm to sample an exchangeable covariance matrix, such as the one of the error terms (R0) in a multiple trait animal model with missing records under normal-inverted Wishart priors is presented. The algorithm (FCG) is based on a conjugate form of the inverted Wishart density that avoids sampling the missing error terms. Normal prior densities are assumed for the 'fixed' effects and breeding values, whereas the covariance matrices are assumed to follow inverted Wishart distributions. The inverted Wishart prior for the environmental covariance matrix is a product density of all patterns of missing data. The resulting MCMC scheme eliminates the correlation between the sampled missing residuals and the sampled R0, which in turn has the effect of decreasing the total amount of samples needed to reach convergence. The use of the FCG algorithm in a multiple trait data set with an extreme pattern of missing records produced a dramatic reduction in the size of the autocorrelations among samples for all lags from 1 to 50, and this increased the effective sample size from 2.5 to 7 times and reduced the number of samples needed to attain convergence, when compared with the 'data augmentation' algorithm.

Algorithms↗
PubMed2005

[Analysis with the propensity score of the association between likelihood of treatment and event of interest in observational studies. An example with myocardial reperfusion].

INTRODUCTION AND OBJECTIVES: Analysis of the effect of treatment in observational studies is complex due to differences between treated and nontreated patients. Calculating the probability of receiving treatment conditioned on relevant covariates (propensity score [PS]) has been proposed as a method to control for these differences. We report an application of PS to assess the association between reperfusion treatment and 28-day case fatality in patients with acute myocardial infarction (AMI). METHOD: We describe the procedure used to calculate PS for receiving reperfusion treatment, and different strategies to analyze the association between PS and case fatality with regression modeling and matching. Data were from a population-based registry of 6307 patients with AMI in Spain during 1997-98. RESULTS: The PS for reperfusion was calculated in 5622 patients. In the multivariate analysis, reperfusion was associated with lower case fatality (OR = 0.59; 95% confidence interval [95% CI]: 0.46-0.77). When PS was included as a covariate, this association became non- significant (OR = 0.76; 95% CI: 0.57-1.01). In the subgroup of matched patients with a similar PS (n = 3138), treatment was not associated with case fatality (OR = 0.95; 95% CI: 0.72-1.26). When the influence of cases with missing data on PS was controlled for, reperfusion treatment was associated with lower fatality (OR = 0.66; 95% CI: 0.55-0.80). CONCLUSIONS: Calculating propensity score is a method that controls for differences between treated and nontreated patients. This score has limitations when matching is incomplete and when data are missing. Results of the present example suggest that reperfusion treatment reduces AMI case fatality.

Adult↗
PubMed2002

Assessing probability of ancestry using simple sequence repeat profiles: applications to maize hybrids and inbreds.

Determination of parentage is fundamental to the study of biology and to applications such as the identification of pedigrees. Limitations to studies of parentage have stemmed from the use of an insufficient number of hypervariable loci and mismatches of alleles that can be caused by mutation or by laboratory error and that can generate false exclusions. Furthermore, most studies of parentage have been limited to comparisons of small numbers of specific parent-progeny triplets thereby precluding large-scale surveys of candidates where there may be no prior knowledge of parentage. We present an algorithm that can determine probability of parentage in circumstances where there is no prior knowledge of pedigree and that is robust in the face of missing data or mistyped data. We present data from 54 maize hybrids and 586 maize inbreds that were profiled using 195 SSR loci including simulations of additional levels of missing and mistyped data to demonstrate the utility and flexibility of this algorithm.

Algorithms↗
PubMedDate not supplied

Quality of life as an endpoint in EORTC clinical trials. European Organization for Research and Treatment for Cancer.

For more than 30 years the European Organization for Research and Treatment for Cancer (EORTC) has conducted, co-ordinated, and stimulated research on the experimental and clinical bases of treatment of cancer and related problems. For more than a decade the EORTC has included quality of life as an outcome measure in some of its trials. The number of clinical studies that include QOL as an evaluation endpoint has increased rapidly in the last few years, and is still increasing steadily. This necessitated a careful and critical evaluation of procedures and results so far in order to generate appropriate guidelines and procedures for incorporating QOL issues in all stages of the clinical trial process, including protocol writing, data collection, data analysis, and reporting of results. This paper provides an overview of the types and the design of studies, data management of quality of life assessment, compliance, missing data and lessons learned during the past years with respect to QOL assessments in the EORTC studies.

Adult↗