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 1,045 records · Page 58Linked to original sources
PubMedDate not supplied

Health-related quality of life among persons living in Japan with a permanent colostomy.

OBJECTIVE: To compare the health-related quality of life (HRQOL) of persons with a permanent colostomy to that of the general Japanese population and explore the factors influencing HRQOL. DESIGN: A cross-sectional survey. SETTING AND SUBJECTS: A total of 255 persons with an ostomy who attended a meeting of the Japan Ostomy Association in the Kanto region of Japan. INSTRUMENTS: The HRQOL was assessed using Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), Japanese version 1.2. Factors potentially influencing the HRQOL were age, sex, marital status, employment status, number of people in the household, time since colostomy, and diagnosis. METHODS: Questionnaires were distributed to the subjects and SF-36 scores were determined and compared with the corresponding national-norm data by Wilcoxon signed rank sum test. A logistic regression analysis was used to explore the influencing factors. RESULTS: The questionnaire response rate was 66.7%. Subjects with a urostomy or an ileostomy and those with missing data were excluded. Data from 102 subjects with a permanent colostomy were analyzed. The subjects' scores were significantly lower than the national-norm scores in the role-physical and social functioning scales. Being employed was associated with significantly lower scores or associated with a tendency toward lower scores. CONCLUSIONS: Scores in two scales in these subjects were lower than those of national-norm scores. Being employed had a negative impact on the HRQOL of subjects with a permanent colostomy. The results of this study provide reference data for future research and underscore the importance of support for persons with a colostomy.

Activities of Daily Living↗
PubMed2004

Economies of scale in institutional review boards.

BACKGROUND: Research with human subjects is essential for most clinical and social science research. As such, the ethical treatment of subjects, including the role of Institutional Review Boards (IRBs), is of paramount concern. The prevailing system of IRBs in the United States reflects an integrated approach in which research organizations have their own local IRB. Recent regulatory changes and a few high-profile problems have prompted proposals for greater investments in IRBs. OBJECTIVES: We conducted regression analyses, looking at how IRB size was associated with IRB costs (economies of scale). RESEARCH DESIGN: We studied data from a cross-sectional survey. SUBJECTS: We studied IRBs at Veterans Affairs (VA) and VA-affiliated medical centers (n = 109); 81 (73%) IRB administrators completed the survey. Fourteen of the administrators had missing data and were excluded from final analysis, leaving a sample of 67. MEASURES: The primary dependent variable was IRB costs in 2001, which we estimated from the survey. Independent variables included IRB size measured as the number of actions (ie, number of initial reviews, amendments, continuing/annual reviews, and harms/adverse event reports) reviewed by the IRB in the last year. RESULTS: The results indicate that very large economies of scale exist, especially for IRBs that handle fewer than 150 actions per year. CONCLUSIONS: A discussion of the costs of benefits of having 3000 to 5000 local IRBs in the United States is warranted because other organizational arrangements could be economically and socially advantageous.

Clinical Protocols↗
PubMed2001

Nonparametric models and methods for designs with dependent censored data: part I.

We consider a nonparametric (NP) approach to the analysis of repeated measures designs with censored data. Using the NP model of Akritas and Arnold (1994, Journal of the American Statistical Association 89, 336-343) for marginal distributions, we present test procedures for the NP hypotheses of no main effects, no interaction, and no simple effects. This extends the existing NP methodology for such designs (Wei and Lachin, 1984, Journal of the American Statistical Association 79, 653-661). The procedures do not require any modeling assumptions and should be useful in cases where the assumptions of proportional hazards or location shift fail to be satisfied. The large-sample distribution of the test statistics is based on an i.i.d. representation for Kaplan-Meier integrals. The testing procedures apply also to ordinal data and to data with ties. Useful small-sample approximations are presented, and their performance is examined in a simulation study. Finally, the methodology is illustrated with two real life examples, one with censored and one with missing data. It is indicated that one of the data sets does not conform to any set of assumptions underlying the available methods and also that the present method provides a useful additional analysis even when data sets conform to modeling assumptions.

Biometry↗
PubMed2006

CARMA: A platform for analyzing microarray datasets that incorporate replicate measures.

BACKGROUND: The incorporation of statistical models that account for experimental variability provides a necessary framework for the interpretation of microarray data. A robust experimental design coupled with an analysis of variance (ANOVA) incorporating a model that accounts for known sources of experimental variability can significantly improve the determination of differences in gene expression and estimations of their significance. RESULTS: To realize the full benefits of performing analysis of variance on microarray data we have developed CARMA, a microarray analysis platform that reads data files generated by most microarray image processing software packages, performs ANOVA using a user-defined linear model, and produces easily interpretable graphical and numeric results. No pre-processing of the data is required and user-specified parameters control most aspects of the analysis including statistical significance criterion. The software also performs location and intensity dependent lowess normalization, automatic outlier detection and removal, and accommodates missing data. CONCLUSION: CARMA provides a clear quantitative and statistical characterization of each measured gene that can be used to assess marginally acceptable measures and improve confidence in the interpretation of microarray results. Overall, applying CARMA to microarray datasets incorporating repeated measures effectively reduces the number of gene incorrectly identified as differentially expressed and results in a more robust and reliable analysis.

Analysis of Variance↗
PubMed1998

The effect of hospital volume on the in-hospital complication rate in knee replacement patients.

OBJECTIVE: To examine the effect of hospital volume on in-hospital surgical outcomes for knee replacement using six years of Medicare claims data. DATA SOURCES/STUDY SETTING: The data include inpatient claims for a 100 percent sample of Medicare patients who underwent primary knee replacement during 1985-1990. We supplemented these data with information from HCFA's denominator files, the Area Resource File, and the American Hospital Association survey files. STUDY DESIGN: We estimated the probability that a patient has an in-hospital complication in the initial hospitalization for the first primary knee replacement, using a Logit model, for three definitions of complication. The models controlled for hospital volume, other hospital characteristics, patient demographics, and patient health status. We tested for the endogeneity of hospital volume. DATA COLLECTION/EXTRACTION METHODS: A panel of two orthopaedic surgeons and two internists reviewed diagnosis codes to determine whether a complication was likely, possible, or due to anemia. After removing the few observations with bad or missing data, the final population has 295,473 observations. PRINCIPAL FINDINGS: The probability of a likely in-hospital complication declines rapidly from 53 through 107 operations per year, then levels off. Statistical tests imply that hospital volume is exogenous in this patient-level data. Complication rates increased steadily through the study period. Although obesity appeared to lower the probability of a complication, a counterintuitive result, further investigation revealed this to be an artifact of the claims data limit of listing no more than five diagnoses. Controlling for this restriction reversed the effect of obesity. CONCLUSIONS: Rather than uncontrolled expansion of knee surgery to small hospitals, decentralization to regional centers where at least about 50, and preferably about 100, operations per year are assured appears to be the optimal policy to reduce in-hospital complications.

Arthroplasty, Replacement, Knee↗
PubMed1989

RECPAM: a computer program for recursive partition amalgamation for censored survival data and other situations frequently occurring in biostatistics. II. Applications to data on small cell carcinoma of the lung (SCCL).

The RECPAM methodology previously presented in part I (A. Ciampi et al., Comput. Methods Programs Biomed. 26 (1988) 239-256) is applied to the analysis of survival data on small cell carcinoma of the lung (SCCL). It is shown how RECPAM can help answer the following questions which occur frequently in the analysis of clinical data: Is it possible to find a classification of patients with a certain disease into distinct prognostic groups? Given a covariate of special interest, does it have an independent prognostic significance even after confounding is taken into account? Does the prognostic significance of a covariate of special interest vary across patient subgroups? For the SCCL data, a prognostic classification is obtained and the tumor marker LDH is treated as a variable of special interest. Many features of RECPAM are illustrated, including, among others, Forward and Backward (Pruning) Stopping Rules, treatment of missing data, and use of several dissimilarity measures.

Biomarkers, Tumor↗
PubMed2002

An assessment of non-randomized medical treatment of long-term schizophrenia relapse using bivariate binary-response transition models.

The analyses of observational longitudinal studies involving concurrent changes in treatment and medical conditions present difficulties because of the multitude of directions of potential relationships: past medication influences current symptoms; past symptoms influence current medication; and current medication is associated with current symptoms. In the context of a long-term study of non-randomized pharmacological treatment of schizophrenic relapse, we present an analysis of bivariate discrete-time transitional data with binary responses in an attempt to understand the transitional and concurrent relationships between schizophrenia relapse and medication use. A naive analysis does not show any association between previous medication and current relapse. However, we provide evidence suggesting that current treatment may impact current relapse for those who have previously taken medication, but not for those who haven't taken medication in the past. When univariate models are specified to assess these associations, the bivariate nature of the problem requires a choice of which response, relapse or medication, should be the dependent variable. In this case, the choice of relapse or medication as a dependent variable does matter. Hence, our results derive from models where both relapse and medication are treated as dependent variables. Specifically, we specify a bivariate log odds ratio for current relapse and current medication use and a separate univariate logit component for each of these outcomes. Each of these components contains transitional associations with previous relapse and medication. Such models represent extensions of univariate transitional association models (e.g. Diggle et al. (1994)) and correspond to bivariate transitional models (e.g. Zeger and Liang (1991)). We incorporate changes in transitional associations into the full-data parametric model for final inference, and investigate if these temporal changes are due to learning effects or the impact of drop-out. We also perform residual analyses and sensitivity analyses in the context of missing data patterns.

Journal Article↗
PubMedDate not supplied

An electronic menstrual cycle calendar: comparison of data quality with a paper version.

OBJECTIVES: This pilot study compared a prototype electronic menstrual calendar on a handheld computer with a paper calendar for data quality and participants' perceptions. DESIGN: Twenty-three women completed identical information about menstrual bleeding and symptoms using paper and electronic calendars for 1 month each. RESULTS: Use of the paper calendar resulted in more missing data than the electronic calendar for bleeding characteristics (13% vs. 4%) and symptoms (35% vs. 4%). The electronic calendar's ability to log data entries revealed retrospective entry for 61% of the data. Total data entry and cleaning time was reduced by 81% with the electronic calendar. Overall, participants preferred the electronic (70%) to the paper (22%) calendar. CONCLUSIONS: Data quality with conventional paper calendars may be poorer than recognized. The data-logging feature, unique to the electronic calendar, is critical for assessing data quality. Electronic menstrual calendars can be useful data collection tools for research in women's health.

Computers↗
PubMed2004

Why ethnic designation matters for stroke rehabilitation: comparing VA administrative data and clinical records.

Using existing administrative data to look at issues of ethnic disparities in rehabilitation-related outcomes may lead to misleading results. Problems can emerge from apparently small issues of reliability that are magnified by reclassification of ethnic designation and missing data in complete-subject analyses. We compared the reliability of ethnic assignment in Department of Veterans Affairs (VA) medical rehabilitation records for stroke patients with administrative records; reclassified the racial identifier from the administrative data in two ways; and examined the different sources of ethnic information in relation to severity, length-of-stay, disability assessment, and discharge disposition. Our results show how small changes increase the potential for Type II error when describing ethnic differences in outcomes or using ethnicity as a predictor with dichotomous response variables. We discuss our results with reference to the literature on ethnic classification and underline the importance of initiatives for improved data collection on ethnicity in VA data sources and in rehabilitation research.

Databases, Factual↗
PubMed1984

Improving the quality of data in a computerised patient master index: implication for costs and patient care.

This study was designed to test the accuracy and completeness of nominal identifying data recorded on a hospital patient master index (PMI) and to estimate the cost-effectiveness of using patient questionnaires for improving quality of data. The study showed that the design of any PMI should include a method of obtaining and storing the last date at which a patient's details were confirmed as correct, a prompt system for identifying missing data and a routine report on completeness and inconsistencies within the data. Both accuracy and completeness of patient identification can be improved by the use of patient questionnaires. The cost of their use is more than matched by the savings obtained from complete accurate data. The use of computer produced questionnaires also reduces the need for peak time activity on busy computer systems.

Computers↗
PubMed1980

Why data bases should not replace randomized clinical trials.

Advances in computer technology have made it possible to store large amounts of observational data concerning treatment of patients for medical disorders. It has been suggested that these data banks might replace randomized clinical trials as a means of evaluating the efficacy of therapies. A review of the methodological problems likely to arise in analyzing such data for the purpose of comparing treatments suggests that sound inferences would not generally be possible because of difficulties with bias in treatment assignment, nonstandard definitions, definitions changing in time, specification of groups to be compared, missing data, and multiple comparisons.

Clinical Trials as Topic↗
PubMed2005

Glucocorticosteroids for primary biliary cirrhosis.

BACKGROUND: Primary biliary cirrhosis is a chronic progressive cholestatic liver disease of presumed autoimmune etiology, characterised by the destruction of small intrahepatic bile ducts and the eventual development of cirrhosis and liver failure. Its progression may be influenced by immunosuppression. Glucocorticosteroids are potent immunosuppressive agents, but they are associated with significant adverse effects, including osteoporosis. OBJECTIVES: To systematically evaluate the beneficial and harmful effects of glucocorticosteroids versus placebo or no intervention for patients with primary biliary cirrhosis. SEARCH STRATEGY: The Cochrane Hepato-Biliary Controlled Trials Register,The Cochrane Library, MEDLINE, EMBASE, and the full text of the identified studies were searched until June 2004. The search strategy included terms for primary biliary cirrhosis and glucocorticosteroids (including the names of frequently used preparations). Previous research groups and manufacturers were contacted for additional references. No language restrictions were applied. SELECTION CRITERIA: Double-blind, single-blind, or unblinded randomised clinical trials evaluating any preparation of glucocorticosteroids versus placebo or no intervention in patients with primary biliary cirrhosis diagnosed by abnormal liver function tests and either anti-mitochondrial antibodies or histology were included. Additional agents were allowed if they were administered to both groups equally. DATA COLLECTION AND ANALYSIS: The quality of the randomised clinical trials was evaluated by methodology components (generation of allocation sequence; allocation concealment; blinding; follow up). Analyses were performed according to the intention-to-treat method with missing data being accounted for by imputation. MAIN RESULTS: Only two underpowered trials (reporting 36 and 40 patients) were identified. These differed markedly in their inclusion criteria and treatment protocols. Both stated that they used placebo. However, allocation concealment was unclear. Only one trial reported any patient deaths. No significant improvement in mortality was identified (odds ratio (OR) 0.42, 95% confidence interval (CI) 0.10 to 1.76). Improvements in serum markers of liver inflammation and liver histology were identified. Potentially prognostically linked markers such as bilirubin and albumin were incompletely reported. Bone mineral density (weighted mean difference -2.84%, 95% CI -4.16 to -1.53) and the number of patients with any adverse event (OR 8.99, 95% CI 2.15 to 37.58) were significantly increased in the glucocorticosteroid group. AUTHORS' CONCLUSIONS: There is insufficient data to support or reject the use of glucocorticosteroids for patients with primary biliary cirrhosis. It may be appropriate to consider a large prospective randomised clinical trial on this topic.

Budesonide↗
PubMed1989

Repeated-measures model for the investigation of temporal trends using longitudinal family studies: application to systolic blood pressure.

A contemporary path model for the analysis of familial resemblance is extended to incorporate repeated measurements on the entire pedigree over time, in order to assess age-related changes in familiality. The parameters of the model can be defined as arbitrary functions of the ages, age differences, or cohabitation times of the family members at the exact time of measurement. Tracking of the phenotypes is decomposed into a familial and a nonfamilial component, which varies with both the time span between measurements and the ages at measurement. Some of the family members may have data missing on one or more visits, and the visits may be unequally spaced both within and across families. The method incorporates all measurements available from all visits into a single model. The model is applied to longitudinal data on systolic blood pressure in 490 East Boston families measured two times at 3-year intervals. Evidence for some nonfamilial tracking is found. Additionally, significant temporal trends are demonstrated in the familiality as a function of age, t2(A), which appears to be near zero at birth, grow to a maximum of about 40% at around age 30, and then appears to monotonically decrease again. No evidence was found for temporal trends in marital resemblance or residual sibling environmental effects. This model provides an objective method of investigating developmental changes in the correlational structure of families over time using repeated-measures and of estimating continuous changes in familiality with age.

Age Factors↗
PubMed2006

2D-GRAPPA-operator for faster 3D parallel MRI.

When using parallel MRI (pMRI) methods in combination with three-dimensional (3D) imaging, it is beneficial to subsample the k-space along both phase-encoding directions because one can then take advantage of coil sensitivity variations along two spatial dimensions. This results in an improved reconstruction quality and therefore allows greater scan time reductions as compared to subsampling along one dimension. In this work we present a new approach based on the generalized autocalibrating partially parallel acquisitions (GRAPPA) technique that allows Fourier-domain reconstructions of data sets that are subsampled along two dimensions. The method works by splitting the 2D reconstruction process into two separate 1D reconstructions. This approach is compared with an extension of the conventional GRAPPA method that directly regenerates missing data points of a 2D subsampled k-space by performing a linear combination of acquired data points. In this paper we describe the theoretical background and present computer simulations and in vivo experiments.

Adult↗
PubMed2002

The impact of prenatal care in the United States on preterm births in the presence and absence of antenatal high-risk conditions.

OBJECTIVE: This study was undertaken to determine the association between prenatal care in the United States and preterm birth rate in the presence, as well as absence, of high-risk pregnancy conditions for African American and white women. STUDY DESIGN: Data were derived from the natality data set for the years 1995 to 1998 provided by the National Center for Health Statistics. Analyses were restricted to singleton live births that occurred at >/=20 weeks' gestation. Multiple births, fetal deaths, congenital malformations, chromosomal abnormalities, missing data on gestational age, and birth weight less than 500 g were excluded. Multivariable logistic regression analyses were used to adjust for the presence or absence of various antenatal high-risk conditions, maternal age, gravidity, marital status, smoking, alcohol, and education. Prenatal care was considered present if there was one or more prenatal visits. Preterm delivery was defined as delivery at less than 37 completed weeks of gestation. RESULTS: For 14,071,757 births analyzed, 1,348,643 (9.6%) resulted in preterm birth. Preterm birth rates were higher for African American women than white women in the presence (15.1% vs 8.3%) and absence (34.9% vs 21.9%) of prenatal care. The absence of prenatal care increased the relative risk for preterm birth 2.8-fold in both African American and white women. There was an inverse dose-response relationship between the number of prenatal visits and the gestational age at delivery both among African American and white women. Lack of prenatal care was associated with increased preterm birth rates to a similar degree in the presence of pregnancy complications for both African American and white women, ranging from 1.6-fold to 5.5-fold for the various antenatal high-risk conditions. CONCLUSION: In the United States, prenatal care is associated with fewer preterm births in the presence, as well as absence of high-risk conditions for both African American and white women. Strategies to increase prenatal care participation may decrease preterm birth rates.

Black or African American↗
PubMed2002

The impact of prenatal care on postneonatal deaths in the presence and absence of antenatal high-risk conditions.

OBJECTIVE: This study was undertaken to determine the association, if any, between prenatal care and postneonatal death in the presence and absence of high-risk pregnancy conditions. STUDY DESIGN: Data were derived from the national linked birth/infant death data set for the years 1995 to 1997 provided by the National Center for Health Statistics. Analyses were restricted to singleton live births that occurred after 23 completed weeks of gestation. Multiple births, congenital malformations, chromosomal abnormalities, missing data on gestational age, and birth weight less than 500 g were excluded. Multivariable logistic regression analyses were used to adjust for various antenatal high-risk conditions, maternal age, gravidity, gestational age at delivery, birth weight, maternal education, marital status, smoking, and alcohol use. Postneonatal death rate was defined as the number of deaths between 28 and 365 days of life per 1,000 neonatal survivors. RESULTS: For 10,512,269 singleton live births analyzed, 21,962 (2.1 per 1,000) resulted in postneonatal death. Postneonatal death rates were higher for African American women than white women in the presence (3.8 vs 1.7 per 1,000) and absence (11.2 vs 5.3 per 1,000) of prenatal care. Lack of prenatal care was associated with increased relative risk (RR) for postneonatal death, 1.8-fold in African American women and 1.6-fold in white women. Lack of prenatal care was associated with increased postneonatal death rates to a similar degree for the individual high-risk pregnancy conditions for both African American and white women. Lack of prenatal care was associated with increased postneonatal death rates, especially in the presence of postterm pregnancy (RR 2.3, 95% CI 1.6, 3.1), pregnancy-induced hypertension (RR 2.2, 95% CI 1.5, 3.4), intrapartum fever (RR 2.1, 95% CI 1.2, 3.5), and small-for-gestational-age infant (RR 1.6, 95% CI 1.3, 2.0). CONCLUSION: Lack of prenatal care should be considered as a high-risk factor for postneonatal death for both African American and white women, especially if the pregnancy has been complicated by postdates, pregnancy-induced hypertension, intrapartum fever or small-for-gestational-age infant.

Black or African American↗
PubMed1989

Comparison of questionnaire and diary methods in acute childhood respiratory illness surveillance.

We compared two prospective survey methods, an interviewer-administered questionnaire and a daily diary, used concurrently to record acute respiratory illness experience over a 2-yr period in 422 children 5 to 11 yr of age from East Boston, Massachusetts. Respondents contributed more months of data with the questionnaire than with the diary method. Respiratory symptom and illness rates, as determined for the first year by each of the methods, were compared for 277 children who had less than 4 months of missing data. Respondents from families with more children tended to report a lower total respiratory illness rate by the diary than by the questionnaire method (p = 0.006). Although upper respiratory illness rates did not differ by method, lower respiratory illnesses were reported more frequently (p = 0.0001) by questionnaire than by diary. In the group of 49 children who were identified as having had greater than one lower respiratory illness, 25% of the illnesses reported as having been lower respiratory by questionnaire were reported as having been another form of respiratory illness by diary. For this group the ratio of 3:1 of boys to girls for the diary as compared with 1.5:1 for the questionnaire suggests the presence of reporting bias and no comparability of methods. Standardization of an acute respiratory illness questionnaire would provide greater opportunity than use of diaries for synthesis of prospective data from different epidemiologic studies.

Acute Disease↗
PubMed1986

Problems in setting up an executing large-scale psychiatric epidemiological studies.

This paper focuses on problems that can be encountered in conceptualizing, executing and writing up large-scale psychiatric epidemiological studies. It makes no attempt to cover fundamental issues of design and analysis, rather it centers on problems associated with projects of considerable size. In the conceptual area, it discusses the prerequisites to be considered before deciding to launch such a study. It notes the administrative and scientific uses of epidemiological studies and considers the strengths and weaknesses of large-scale studies to address those concerns. Issues in carrying out such studies are discussed including decisions about study design, sampling method and instrumentation. All are dependent on the central purpose of the study but trade-offs between feasibility and scientific rigor are always present. Data collection and analysis problems highlighted in large-scale studies are examined. They include the difficulty, in the former, of adequately motivating and supervising field personnel and, in the latter, of dealing with problems that accompany missing data and complicated sampling strategies. Potential problems in data access and use and writing up the results are seen as arising from the presence of a large investigative team with diverse interests. Lastly, the comparative worth of these studies is considered.

Data Collection↗