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 649 records · Page 36Linked to original sources
PubMed2005

Antibiotics for trachoma.

BACKGROUND: Trachoma is the world's leading cause of preventable blindness. In 1997 the World Health Organization launched an initiative on trachoma control based on the 'SAFE' strategy (surgery, antibiotics, facial cleanliness and environmental improvement). OBJECTIVES: To assess the evidence supporting the antibiotic arm of the SAFE strategy by assessing the effects of antibiotics on both active trachoma (primary objective) and on Chlamydia trachomatis infection of the conjunctiva (secondary objective). SEARCH STRATEGY: We searched the Cochrane Central Register of Controlled Trials - CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (The Cochrane Library Issue 1, 2005), MEDLINE (1966 to February 2005), and EMBASE (1980 to February 2005). We used the Science Citation Index to look for articles that cited the included studies. We searched the reference lists of identified articles and we contacted authors and experts for details of further relevant studies. SELECTION CRITERIA: We included only randomised trials that satisfied either of two criteria: (a) trials in which topical or oral administration of an antibiotic was compared to placebo or no treatment in people with trachoma, (b) trials in which a topical antibiotic was compared with an oral antibiotic in people with trachoma. A subdivision of particular interest was of trials in which topical tetracycline/chlortetracycline was compared with oral azithromycin, as these are the two World Health Organization recommended treatments. DATA COLLECTION AND ANALYSIS: Two authors independently assessed trial quality and extracted data. We contacted investigators for missing data. MAIN RESULTS: We found 15 studies that randomised a total of 8678 participants. For both outcomes (active trachoma and laboratory evidence of infection) the results of the chi squared tests suggested that there was significant statistical heterogeneity among the trials. There was also marked clinical heterogeneity. No summary statistics were calculated and we therefore present a narrative summary of the results. For the comparisons of oral or topical antibiotic against placebo/no treatment, the data are consistent with there being no effect of antibiotics but are suggestive of a lowering of the point prevalence of relative risk of both active disease and laboratory evidence of infection at three and 12 months after treatment. For the comparison of oral against topical antibiotics the results suggest that oral treatment is neither more nor less effective than topical treatment. AUTHORS' CONCLUSIONS: There is some evidence that antibiotics reduce active trachoma but results are not consistent and cannot be pooled.

Administration, Oral↗
PubMed2001

Continuing education meetings and workshops: effects on professional practice and health care outcomes.

BACKGROUND: Educational meetings and printed educational materials are the two most common types of continuing education for health professionals. An important aim of continuing education is to improve professional practice so that patients can receive improved health care. OBJECTIVES: To assess the effects of educational meetings on professional practice and health care outcomes. SEARCH STRATEGY: We searched the Cochrane Effective Practice and Organisation of Care Group specialised register, MEDLINE (from 1966), the Research and Development Resource Base in Continuing Medical Education in January 1999 and reference lists of articles. SELECTION CRITERIA: Randomised trials or well designed quasi-experimental studies examining the effect of continuing education meetings (including lectures, workshops, and courses) on the clinical practice of health professionals or health care outcomes. DATA COLLECTION AND ANALYSIS: Two reviewers independently applied inclusion criteria, assessed the quality of each study, and extracted study data. We attempted to collect missing data from investigators. We conducted both qualitative and quantitative analyses. MAIN RESULTS: Thirty-two studies were included with a total of 36 comparisons. The studies involved from 13 to 411 health professionals (total N= 2995) and were judged to be of moderate or high quality, although methods were generally poorly reported. There was substantial variation in the complexity of the targeted behaviours, baseline compliance, the characteristics of the interventions and the results. The heterogeneity of the results was best explained by differences in the interventions. For 10 comparisons of interactive workshops, there were moderate or moderately large effects in six (all of which were statistically significant) and small effects in four (one of which was statistically significant). For interventions that combined workshops and didactic presentations, there were moderate or moderately large effects in 12 comparisons (eleven of which were statistically significant) and small effects in seven comparisons (one of which was statistically significant). In seven comparisons of didactic presentations, there were no statistically significant effects, with the exception of one out of four outcome measures in one study. REVIEWER'S CONCLUSIONS: Interactive workshops can result in moderately large changes in professional practice. Didactic sessions alone are unlikely to change professional practice.

Congresses as Topic↗
PubMed2004

Assessment of inpatient paediatric care in first referral level hospitals in 13 districts in Kenya.

BACKGROUND: The district hospital is considered essential for delivering basic, cost-effective health care to children in resource poor countries. We aimed to investigate the performance of these facilities in Kenya. METHODS: Government hospitals providing first referral level care were prospectively sampled from 13 Kenyan districts. Workload statistics and data documenting the management and care of admitted children were obtained by specially trained health workers. FINDINGS: Data from 14 hospitals were surveyed with routine statistics showing considerable variation in inpatient paediatric mortality (range 4-15%) and specific case fatality rates (eg, anaemia 3-46%). The value of these routine data is seriously undermined by missing data, apparent avoidance of a diagnosis of HIV/AIDS, and absence of standard definitions. Case management practices are often not in line with national or international guidelines. For malaria, signs defining severity such as the level of consciousness and degree of respiratory distress are often not documented (range per hospital 0-100% and 9-77%, respectively), loading doses of quinine are rarely given (3% of cases) and dose errors are not uncommon. Resource constraints such as a lack of nutritional supplements for malnourished children also restrict the provision of basic, effective care. INTERPRETATION: Even crude performance measures suggest there is a great need to improve care and data quality, and to identify and tackle key health system constraints at the first referral level in Kenya. Appropriate intervention might lead to more effective use of health workers' efforts in such hospitals.

Anemia↗
PubMedDate not supplied

Long-term survival data from a clinical trial on resin-bonded bridges.

OBJECTIVES: A clinical trial, involving 203 resin-bonded bridges (RBBs) was undertaken to investigate the influence of retainer-type and luting material on the survival of these restorations. METHODS: For this evaluation, 157 patients were available (14% of the original sample was lost to follow-up or excluded from the study following the stopping criteria). Fifty per cent of the patients were questioned concerning the fate of the RBBs and 59% of questioned patients were examined clinically. The patients that were seen for examination were representatives of the experimental groups. The findings from the clinical examination were compared with the data obtained from the questionnaire. Missing data were censored at the date of the last available information. Kaplan-Meier estimates were calculated to assess the survivals at the endpoints and compared using Cox's proportional hazards procedure. RESULTS: A significant difference was found between perforated (P-type) and etched (E-type) RBBs (P = 0.05) for original bonded restorations but not when rebonded RBBs were taken into account. The results of the survival analysis were: anterior P-type, 49 +/- 7% after 10.5 years: anterior E-type, 57 +/- 7% after 10.5 years; posterior P-type, 18 +/- 11% after 6.8 years; posterior E-type, 37 +/- 13% after 10.2 years. Survivals of RBBs that were rebonded once during the evaluation period were 62 +/- 9% (11.0 years) for anterior RBBs and 51 +/- 11% (10.2 years) for posterior RBBs. CONCLUSIONS: The factor location (anterior versus posterior) was as in previous analyses, highly significant. Differences in survival between cementation materials were not significant.

Acid Etching, Dental↗
PubMed1991

Poor agreement of occupational data between a hospital-based cancer registry and interview.

With occupation recognized as a risk factor for various cancers, collecting occupation and industry data by a number of vital registries, including cancer registries, has developed. Registries may be data sources for cancer etiology research and occupational disease surveillance, despite concerns that their data are fragmentary and may lack validity. To improve completeness and validity of occupational information in a hospital-based cancer registry, this study compared information obtained through abstracting medical records for the registry with information obtained through lung-cancer patient interviews. Employing the kappa statistic, agreement was generally poor, largely due to data missing in the medical record. Data quality of hospital-based cancer registries can be improved by employing trained cancer registrars to elicit occupational histories from patients.

Data Collection↗
PubMed2004

Mixtures of varying coefficient models for longitudinal data with discrete or continuous nonignorable dropout.

The analysis of longitudinal repeated measures data is frequently complicated by missing data due to informative dropout. We describe a mixture model for joint distribution for longitudinal repeated measures, where the dropout distribution may be continuous and the dependence between response and dropout is semiparametric. Specifically, we assume that responses follow a varying coefficient random effects model conditional on dropout time, where the regression coefficients depend on dropout time through unspecified nonparametric functions that are estimated using step functions when dropout time is discrete (e.g., for panel data) and using smoothing splines when dropout time is continuous. Inference under the proposed semiparametric model is hence more robust than the parametric conditional linear model. The unconditional distribution of the repeated measures is a mixture over the dropout distribution. We show that estimation in the semiparametric varying coefficient mixture model can proceed by fitting a parametric mixed effects model and can be carried out on standard software platforms such as SAS. The model is used to analyze data from a recent AIDS clinical trial and its performance is evaluated using simulations.

Anti-HIV Agents↗
PubMed2002

Data quality in evaluation of an alcohol-related harm prevention program.

The authors report the reliability and convergent validity in a sample of college students for 27 composite scales and two items covering alcohol use, cigarette smoking, marijuana use, and other drug use; beliefs relating to alcohol use; perceived norms for alcohol-related behavior; harm prevention skills; intentions to take prevention action; harm prevention action taken; risk taken; experienced harm; and other health-related behaviors and person characteristics. Data quality assessment strategies and missing data procedures were illustrated for large, multivariate, longitudinal data sets. Results indicate 23 of the 27 composite scales had at least acceptable reliability, and the remaining 4 composite scales had at least marginally acceptable reliability. At least moderate construct validity was demonstrated for 25 scales.

Adult↗
PubMed1990

Time-series analysis--cosinor analysis: a special case.

Cosinor analysis provides an accessible means of evaluating and estimating the parameter of a cyclic phenomenon. Cosinor analysis does not require that the data be equal intervals without missing data. Cosinor analysis does require that the data can reasonably be considered to take the form of a deterministic cycle with a known period.

Humans↗
PubMed1999

Searching for biological rhythms: peak detection in the periodogram of unequally spaced data.

The classical power spectrum, computed in the frequency domain, outranks traditionally used periodograms derived in the time domain (such as the chi2 periodogram) regarding the search for biological rhythms. Unfortunately, classical power spectral analysis is not possible with unequally spaced data (e.g., time series with missing data). The Lomb-Scargle periodogram fixes this shortcoming. However, peak detection in the Lomb-Scargle periodogram of unequally spaced data requires some careful consideration. To guide researchers in the proper evaluation of detected peaks, therefore, a novel procedure and a computer program have recently become available. It is recommended that the Lomb-Scargle periodogram be the default method of periodogram analysis in future biomedical applications of rhythm investigation.

Computer Simulation↗
PubMed2007

Monitoring health outcomes following cytoreductive surgery plus intraperitoneal hyperthermic chemotherapy for peritoneal carcinomatosis.

BACKGROUND: Cytoreductive surgery with intraperitoneal hyperthermic chemotherapy for peritoneal carcinomitosis (PC) is an aggressive treatment designed to alleviate symptoms and prolong life. It is associated with physical and psychological morbidity. The purpose of this study was to monitor health outcomes related to treatment. METHODS: Patients completed a questionnaire before and after surgery at 3, 6 and 12 months. The questionnaire consisted of basic demographic items as well as the Functional Assessment of Cancer Therapy-Colon Scale (FACT-C), SF-36 Medical Outcomes Study Health Survey, Center for Epidemiologic Studies-Depression Scale (CES-D), Brief Pain Inventory-Short Form, and ECOG Performance Status Rating. Time trends were assessed with mixed models (SAS PROC MIXED) so as to use all data and to account for missing data. RESULTS: Ninety-six patients (49% females, 9% African Americans) of an average age of 52.9 (SD = 12.5) years were assessed before surgery. PC originated in primary lesions of the appendix (n = 36); colon/rectum (n = 24); mesothelium (n = 9); ovary (n = 5); stomach (n = 4); and miscellaneous (n = 18). Quality of life (QOL) and pain scores improved from baseline to 12 months. Physical functioning changed over the 12-month study period with improvement recorded at 6 months. The percentage of patients reporting significant depressive symptom at each time point was; baseline = 32%; 3 months = 19%; 6 & 12 months = 24%. CONCLUSIONS: Acceptable QOL, return of functional status, and reduced pain can be attained between 3 and 6 months following treatment although some deficits in general health remain. Depressive symptoms are common and should be monitored.

Activities of Daily Living↗
PubMed1995

Use of a spreadsheet program for circadian analysis of biological/physiological data.

Biological/physiological data sampled over a period of 24 h can be subjected to a mathematical analysis to determine the presence of circadian rhythmicity. Several procedures have been proposed, most being complex. To render such an analysis simpler and easy to use by non-mathematicians, we developed and tested the cosinor technique using a commonly available commercial spreadsheet (Excel). It can be used to analyze equally or unequally time-spaced data over 24 h with missing data, as well as to calculate the significance and the main limit of the resultant circadian rhythm (mesor, amplitude, acrophase and their confidence limits). Examples of its application to hourly samples of plasma cortisol and minute-by-minute rectal temperatures are shown.

Body Temperature↗
PubMed1997

Regression calibration in failure time regression.

In this paper we study a regression calibration method for failure time regression analysis when data on some covariates are missing or mismeasured. The method estimates the missing data based on the data structure estimated from a validation data set, a random subsample of the study cohort in which covariates are always observed. Ordinary Cox (1972; Journal of the Royal Statistical Society, Series B 34, 187-220) regression is then applied to estimate the regression coefficients, using the observed covariates in the validation data set and the estimated covariates in the nonvalidation data set. The method can be easily implemented. We present the asymptotic theory of the proposed estimator. Finite sample performance is examined and compared with an estimated partial likelihood estimator and other related methods via simulation studies, where the proposed method performs well even though it is technically inconsistent. Finally, we illustrate the method with a mouse leukemia data set.

Animals↗
PubMed2003

Simple approaches to assess the possible impact of missing outcome information on estimates of risk ratios, odds ratios, and risk differences.

Often in clinical trials, the primary outcome is binary and the impact of an intervention is summarized using risk ratios (RRs), odds ratios (ORs), or risk differences (RDs). It is typical that in such studies, the binary outcome variable is not observed for some study participants. When there is missing data, it is well known that analyses based on those participants with complete data can be biased unless it can be assumed that the probability of a missing outcome is unrelated to the value of the missing binary outcome (i.e., missing at random). Unfortunately, this assumption cannot be assessed with the data since the missing outcomes, by definition, are not observed. One approach to this problem is to perform a sensitivity analysis to see the degree to which conclusions based only on the complete data would be affected given various degrees of departure from the missing at random assumption. In this paper we provide researchers formulae for doing such a sensitivity analysis. We quantify the departure from the missing at random assumption with a parameter we call the "response probability ratio" (RPR). This is the ratio between the probability of a nonmissing outcome among those with one value of the binary outcome and the probability of a nonmissing outcome among those with the other value of the outcome. Then we provide simple formulae for the estimation of the RRs, ORs, and RDs given any specific values of the RPRs. In addition to being useful for sensitivity analyses, these formulae provide some insight into the conditions that are necessary for bias to occur. In particular, it can be seen that, under certain plausible assumptions, OR estimates based on participants with complete data will be asymptotically unbiased, even if the probability of missing outcome depends on both the treatment and the outcome.

Bias↗
PubMed2007

Impact of nonignorable coarsening on Bayesian inference.

The coarse data model of Heitjan and Rubin (1991) generalizes the missing data model of Rubin (1976) to cover other forms of incompleteness such as censoring and grouping. The model has 2 components: an ideal data model describing the distribution of the quantity of interest and a coarsening mechanism that describes a distribution over degrees of coarsening given the ideal data. The coarsening mechanism is said to be nonignorable when the degree of coarsening depends on an incompletely observed ideal outcome, in which case failure to properly account for it can spoil inferences. A theme in recent research is to measure sensitivity to nonignorability by evaluating the effect of a small departure from ignorability on the maximum likelihood estimate (MLE) of a parameter of the ideal data model. One such construct is the "index of local sensitivity to nonignorability" (ISNI) (Troxel and others, 2004), which is the derivative of the MLE with respect to a nonignorability parameter evaluated at the ignorable model. In this paper, we adapt ISNI to Bayesian modeling by instead defining it as the derivative of the posterior expectation. We propose the application of ISNI as a first step in judging the robustness of a Bayesian analysis to nonignorable coarsening. We derive formulas for a range of models and apply the method to evaluate sensitivity to nonignorable coarsening in 2 real data examples, one involving missing CD4 counts in an HIV trial and the other involving potentially informatively censored relapse times in a leukemia trial.

Bayes Theorem↗
PubMed1993

Direct standardization of incidence rates in the presence of incomplete data.

When comparing the disease incidence rates for several subpopulations, epidemiologists often use direct standardization to adjust for potential confounding variables. In population-based studies, however, the data are often incompletely classified with respect to membership in the subpopulations of interest. In such a situation, one often assumes that the cases with missing data have the same distribution as the complete cases, that is the data are missing completely at random. In this setting, we derive variance estimates for the directly standardized rates which account for the use of incomplete data. We illustrate the use of these methods with data from a study of the incidence of gastrointestinal cancer by immigrant status where birthplace data are often incomplete.

Bias↗
PubMed2001

[Application of diagnosis related groups (DRGs) in ophthalmology in Germany].

Starting on 01.01.2003 the hospital reimbursement system in Germany will be transferred completely to a diagnosis-related groups (DRG) system based on the data of 2001. As the reimbursement system is of vital importance for the development of a speciality, current information is given about the impact of the new payment system on ophthalmology. The Australian AR-DRG system to be adopted in Germany distinguishes 20 DRGs for ophthalmology. The patients are grouped into the DRGs according to their principal diagnoses and comorbidities and the procedures coded by electronic data processing. Errors in coding and missing data may cause reductions in the hospital budget as do economically inhomogeneous DRGs. The tight timetable given by the legislation requires completion of determining DRG groups and weightings by 31.12.2001. However, important prerequisites such as the mapping tables of the Australian DRGs are still missing, which means that hospitals have to prepare for the new system like a pilot flying blind with no instruments.

Cost Control↗
PubMed2006

Semiparametric models for missing covariate and response data in regression models.

We consider a class of semiparametric models for the covariate distribution and missing data mechanism for missing covariate and/or response data for general classes of regression models including generalized linear models and generalized linear mixed models. Ignorable and nonignorable missing covariate and/or response data are considered. The proposed semiparametric model can be viewed as a sensitivity analysis for model misspecification of the missing covariate distribution and/or missing data mechanism. The semiparametric model consists of a generalized additive model (GAM) for the covariate distribution and/or missing data mechanism. Penalized regression splines are used to express the GAMs as a generalized linear mixed effects model, in which the variance of the corresponding random effects provides an intuitive index for choosing between the semiparametric and parametric model. Maximum likelihood estimates are then obtained via the EM algorithm. Simulations are given to demonstrate the methodology, and a real data set from a melanoma cancer clinical trial is analyzed using the proposed methods.

Algorithms↗
PubMed1990

Validation of malaria surveillance case reports: implications for studies of malaria risk.

STUDY OBJECTIVE: The aim of the study was to investigate the quality of national malaria surveillance reports in the United Kingdom. DESIGN: Persons with malaria reported to the Malaria Reference Laboratory (MRL) in 1987 were contacted by post to verify existing records with respect to key variables. The MRL data set was then analysed for inaccuracies. SETTING: The study was confined to UK residents. PARTICIPANTS: 602 persons with malaria in 1987 responded (53%). MEASUREMENTS AND MAIN RESULTS: Review of case reports showed few missing data except for duration of residence in the UK, detailed chemoprophylactic regimens, and compliance. There were more missing surveillance data in reports of ethnic minority groups, principally in dates of travel (p = 0.008) and chemoprophylaxis use (p less than 0.0001). Patient recall in the survey was at variance with the surveillance reports in dates of travel and onset of infection, chemoprophylaxis use, and in compliance. Surveillance reports overestimated the number of days between leaving a malarious area and onset of symptoms (by 9 d for P falciparum and by 24 d for P vivax), and underestimated the delay between onset and diagnosis of P falciparum by 3 d. Over 50% of patients who had recalled the use of chloroquine, proguanil, pyrimethamine/dapsone, and pyrimethamine had not been recorded as having taken these drugs on the surveillance reports. Reported compliance also differed between the two data sets. CONCLUSIONS: It is recommended that research units test the quality of their surveillance data before embarking on analytical studies used to generate health policy guidelines.

Adult↗