Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Imputation”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 325 records · Page 18Linked to original sources
PubMedDate not supplied

Estimating the cost-effectiveness of fluticasone propionate for treating chronic obstructive pulmonary disease in the presence of missing data.

OBJECTIVES: To explore the cost-effectiveness of fluticasone propionate (FP) for the treatment of chronic obstructive pulmonary disease (COPD), we estimated costs and quality-adjusted life-years (QALYs) over 3 years, based on an economic appraisal of a previously reported clinical trial (Inhaled Steroids in Obstructive Lung Disease in Europe [ISOLDE]). METHODS: Seven hundred forty-two patients enrolled in the ISOLDE trial who received either FP or placebo had data available on health-care costs and quality of life over the period of the study. The SF-36-based utility scores for quality of life were used to calculate QALYs. A combined imputation and bootstrapping procedure was employed to handle missing data and to estimate statistical uncertainty in the estimated cumulative costs and QALYs over the study period. The imputation approach was based on propensity scoring and nesting this approach within the bootstrap ensured that multiple imputations were performed such that statistical estimates included imputation uncertainty. RESULTS: Complete data were available on mortality within the follow-up period of the study and a nonsignificant trend toward improved survival of 0.06 (95% confidence interval [CI]-0.01 to 0.15) life-years was observed. In an analysis based on a propensity scoring approach to missing data we estimated the incremental costs of FP versus placebo to be 1021 sterling pound(95% CI 619-1338 sterling pound) with an additional effect of 0.11 QALYs (CI 0.04-0.20). Cost-effectiveness estimates for the within-trial period of 17,700 sterling pound per life-year gained (6900 sterling pound to infinity) and 9500 sterling pound per QALY gained (CI 4300-26,500 sterling pound) were generated that include uncertainty due to the imputation process. An alternative imputation approach did not materially affect these estimates. CONCLUSIONS: Previous analyses of the ISOLDE study showed significant improvement on disease-specific health status measures and a trend toward a survival advantage for treatment with FP. This analysis shows that joint considerations of quality of life and survival result in a substantial increase in QALYs favoring treatment with FP. Based on these data, the inhaled corticosteroid FP appears cost-effective for the treatment of COPD. Confirmation or refutation of this result may be achieved once the Towards a Revolution in COPD Health (TORCH) study reports, a large randomized controlled trial powered to detect mortality changes associated with the use of FP alone, or in combination with salmeterol, which is also collecting resource use and utility data suitable for estimating cost-effectiveness.

Administration, Inhalation↗
PubMed2005

Handling missing data in patient-level cost-effectiveness analysis alongside randomised clinical trials.

BACKGROUND: Missing data are potentially an extensive problem in cost-effectiveness analyses conducted alongside randomised clinical trials, where prospective collection of both resource use and health outcome information is required. There are several possible reasons for the presence of incomplete records, and the validity of the analysis in the presence of data with missing values is dependent upon the mechanism generating the missing data phenomenon. In the past, the most commonly used methods for analysing datasets with incomplete observations were relatively ad hoc (e.g. case deletion, mean imputation) and suffered from potential limitations. Recently, several alternative and more sophisticated approaches (e.g. multiple imputation) have been proposed that attempt to correct the flaws of the simple imputation methods. OBJECTIVES: The objectives are to provide a concise and accessible description of the quantitative methods most commonly used in trial-based cost-effectiveness analysis for handling missing data, and also to demonstrate the potential impact of these alternative approaches on the cost-effectiveness results reported in two case studies. METHODS: Data from two recently conducted, trial-based economic evaluations are used to explore the sensitivity of the study results to the technique used to deal with incomplete observations. A statistical framework for representing the uncertainty in the alternative methods is outlined using an approach based on net benefits and cost-effectiveness acceptability curves. RESULTS: The case studies demonstrate the potential importance of the approach used to handle missing data. Although the analytical strategy did not appear to alter the results of one of the studies, the other case study showed that that the results of the cost-effectiveness analysis were sensitive to both the decision to impute and also the imputation strategy adopted. CONCLUSIONS: Analysts should be more explicit in reporting the analytical strategies applied in the presence of missing data. The use of a multiple imputation approach is recommended in the majority of cases, so as to adequately reflect the uncertainty in the study results due to the presence of missing data.

Bias↗
PubMedDate not supplied

[Alpha-interferon and mental disorders].

UNLABELLED: The interferon alpha stands as a reference both in oncology and virology. But its efficiency is limited by frequent somatic as well as neuropsychic side effects. As a matter of fact, the reduction or the ending of a chemotherapy treatment come chiefly from the psychiatric complications caused by the use of interferon. For about 30% of patients, various psychic disorders are noticed: personality disorders, mood disorders, anxiety states, suicidal tendencies, manic and psychotic symptoms. We thus propose a review which shall be completed by a discussion on wether the interferon is responsible or not of the appearance of the described mental disorders. We shall conclude with a synthesis of the proposed practical management when confronted with such disorders. Psychiatric complications under interferon-Alpha. The appearance of psychiatric complications caused by interferon has been the subject of many publications. They have also raised the question of the toxicity mechanism which is still misunderstood today. This toxicity appears to be dose-dependent with variations depending on the daily dose given, the mode of administration, the combination with other chemotherapy treatments, the concomitance with a cerebral radiotherapy or a medical history of psychiatric disorders. Most of these effects occur after three weeks of treatment but non specific neuropsychic symptoms can be observed earlier. Non specific symptoms. They appear early but are difficult to detect, though they bring together a whole lot of clinical signs: asthenia, irritability, psychomotor slowdown, depressive mood or even a real "subsyndromic" depressive syndrome, anorexia, decline of the libido, concentration and attention problems, dizzy spells and headaches. Some authors have described intense and fluctuating of personality, mixing anxiety, irritability and disorder of drive control. Depression. Depression is the most frequently found psychiatric pathology in studies but the real frequency of clear cases of depressive problems is difficult to determine through lack of serious studies. So the incidence of depressive disorders usually varies from 5 to 15%. The depressive syndrome can settle as soon as the first week treatment, with a peak in the frequency during the first and third months. The seriousness and the incidence of this syndrome seem to be dose-dependent. The gravity of this complication lies in the suicidal risk, a risk all the more dreadful since there is not any identified risk factor. Suicides and suicidal behaviours. Serious complications, because they act directly on the vital prognosis. However fortunately, suicidal behaviours only represent a minority within all the side effects attributed to the interferon-alpha. These actions fit into three main clinical dimensions: complication of a severe depressive syndrome, confusional context and disorder of the impulses control. In practical terms, prevention proves to be difficult without identified predictive factors. Nevertheless, some authors point out the importance of aggravating comorbid disorders like alcoholism or the coinfection by the HIV. Manic syndrome. The appearance of a manic state under a chemotherapy treatment seems to be rare, given that there have been only a dozen cases published around the world. But these observations are interesting as far as both the study of imputability and the understanding of the toxicity mechanisms are concerned. Most of the cases deal with patients without a family or personal history of psychiatric disorders, and whose symptomatology disappears with the end of the treatment, which is an argument in favour of the imputation of the interferon in the appearance of manic disorders. In addition, some authors introduce the notion of tertiary mania: the appearance of an autoimmune hypothyroidism in relation with interferon and leading to athymic elation. Eventually, the appearance of manic problems at the end of the treatment makes it possible to speculate about the physiopathological mechanisms that are at issue. Anxiety disorders. These disorders are not much described: they generally are already existing disorders (like phobic or obsessive compulsive disorders), reactivated or aggravated by the interferon-alpha molecule. Adaptation disorders. It deals with adaptation disorders along with anxious temper coming at the beginning of the treatment. These problems are more concerned with the announcement of the diagnosis and its seriousness than with the toxicity of the interferon-alpha molecule. Psychotic states. There are less papers on the prevalence of psychotic disorders during the treatment, or at the end of it. But they can be found in both viral and malignant pathologies. A large retrospective study has shown ten cases of psychotic disorders and that in the absence of history of psychiatry or of a HIV co-infection. In every case the psychiatric aspect is stopped by the ending of the treatment or by an appropriated treatment. Usually, the few cases of paranoïd delusion described in papers seem to appear between one and three months of treatment, with patients having a history of psychiatric disorders. Aggravation of pre-existing mental disorders. Numerous authors have reported the recurrence of addictive behaviours (alcohol or other psychoactive matter) by weaned patients. Imputability to interferon-alpha in psychiatric disorders. It is difficult to draw the relationship between the chemotherapy with the interferon-alpha treatment and neuropsychiatric complications because there is a lack in specific studies. Nevertheless, it seems to be causal relations between the prescription of interferon and the appearance of psychic disorders. As a matter of fact, even if there is neither predictive criterion nor diagnosis of clinical type (set apart a dose effect), it is clear that there are diagnostic criteria of chronological kind: delay of appearance and disappearance of side effects compatible with the kinetics of the molecule and test of positive reintroduction. The imputability is thus most likely towards, given the reported clinical observations and signs of direct cerebral toxicity described for interferon: induction of neurophysiological changes among healthy volunteers, reversible EEG impairments the second week of treatment, direct vascular and neurological toxicity. Eventually, authors have shown that the psychiatric morbidity could be more important among patients under treatment than in a control group. In conclusion, the imputability of interferon appears to be very likely, more particularly in the appearance of mood disorders, mainly depressive ones, of manic syndromes and of certain psychotic episodes. MANAGEMENT: The most numerous therapeutic propositions naturally concern the depressive syndromes, because of their high frequency. In a recent article, the authors have detailed the pharmacological criteria of the ideal molecule: limited hepatic metabolism, low rate of proteinic fixation, long half-life and absence of active metabolite. So they advise not to prescribe imipraminic molecules and recommend the use of some SRI in first intention: citalopram and sertraline mainly, paroxetine to avoid given its pharmacological features that do not seem adapted. Only the minalcipram seems to show all the theoretical advantages described above. If there is an indication in the introduction of an anxiolytic medication, we shall prefer a benzodiazepine with short half-life like loxazepam and alprazolam. Besides, all the publications point out the importance of a specific clinical observation during the treatment as well as in the six months following its end. The agreement must bear full medical costs, above all including psychotherapic and social aspects. The proposed treatments for the other disorders are conventional: haloperidol and lithium for bipolar disorders, fluvoxamine for obsessive compulsive disorders and neuroleptics for psychotic disorders. CONCLUSION: The appearance of neuropsychiatric side effects during a chemotherapy using the interferon-alpha molecule is a frequent complication, the consequences of which can prove tragic: involvement of the vital prognosis, family and professional relation disturbances, compliance problems, risks of psychiatric morbidity at short and middle terms.... In spite of the absence of rigorous controlled studies, the imputability to the interferon of the appearance of psychological disorders appears very likely. So the role of the psychiatrist seems to be determining in the follow-up care of these patients who must be considered at high risk to develop a psychiatric pathology. The agreement to bear medical costs has to be made in narrow collaboration with clinical practitioners and must be part of a clinical continuity, from the pre-therapeutical evaluation to the remote follow-up care. Finally, it seems important to implement controlled studies, resting on a great diagnostic and methodological rigour, in order to clarify the toxicity mechanisms of interferon and to optimise the agreement to bear medical cost for the patients.

Adaptation, Psychological↗
PubMed2026

Predicted brain-regional gene expression patterns in individuals living with Alzheimer's disease.

Studying brain gene expression in Alzheimer's Disease (AD) remains difficult as postmortem brain is difficult to access, cannot be used to guide donor treatment, may be confounded by environmental factors before and after death, and is difficult to link to early AD states or disease progression. To circumvent these limitations, several studies have tested blood transcriptome biomarkers for AD. However, gene-expression levels in the blood have limited correlation with those in the brain. To evaluate the potential of monitoring Alzheimer's progression with peripheral data, we used transcriptome-imputation to identify brain-region-specific AD-associated gene-expression differences in cohorts with blood-based transcriptome data. This approach provides a high-resolution image of AD-associated molecular differences in the brains of individuals actively living with disease. We analyzed eight AD studies (777 AD cases, 779 cognitively unimpaired controls), imputing transcriptomes in 10 brain regions via the Brain Gene Expression and Network Imputation Engine (BrainGENIE). Hundreds of differentially expressed genes (DEGs) associated with AD were identified in nine brain regions, with anterior cingulate cortex and amygdala showing the most differential expression. AD-associated genes were enriched in pathways such as proteostasis, mitochondrial dysfunction, and immune activation. We observed significant yet moderate concordance between imputed AD-associated changes and those directly measured in the dorsolateral prefrontal cortex and cerebellum. These transcriptomic changes can guide future in vitro studies focused on pathogenesis or be targets of novel therapeutic development. In conclusion, we demonstrated the scope and utility of brain expression imputation from the peripheral transcriptome, laying the groundwork for biomarker discovery and prospective AD studies.

Alzheimer Disease↗
PubMed2007

A residuals-based transition model for longitudinal analysis with estimation in the presence of missing data.

We propose a transition model for analysing data from complex longitudinal studies. Because missing values are practically unavoidable in large longitudinal studies, we also present a two-stage imputation method for handling general patterns of missing values on both the outcome and the covariates by combining multiple imputation with stochastic regression imputation. Our model is a time-varying auto-regression on the past innovations (residuals), and it can be used in cases where general dynamics must be taken into account, and where the model selection is important. The entire estimation process was carried out using available procedures in statistical packages such as SAS and S-PLUS. To illustrate the viability of the proposed model and the two-stage imputation method, we analyse data collected in an epidemiological study that focused on various factors relating to childhood growth. Finally, we present a simulation study to investigate the behaviour of our two-stage imputation procedure.

Bias↗
PubMed2005

Methods to account for attrition in longitudinal data: do they work? A simulation study.

Attrition threatens the internal validity of cohort studies. Epidemiologists use various imputation and weighting methods to limit bias due to attrition. However, the ability of these methods to correct for attrition bias has not been tested. We simulated a cohort of 300 subjects using 500 computer replications to determine whether regression imputation, individual weighting, or multiple imputation is useful to reduce attrition bias. We compared these results to a complete subject analysis. Our logistic regression model included a binary exposure and two confounders. We generated 10, 25, and 40% attrition through three missing data mechanisms: missing completely at random (MCAR), missing at random (MAR) and missing not at random (MNAR), and used four covariance matrices to vary attrition. We compared true and estimated mean odds ratios (ORs), standard deviations (SDs), and coverage. With data MCAR and MAR for all attrition rates, the complete subject analysis produced results at least as valid as those from the imputation and weighting methods. With data MNAR, no method provided unbiased estimates of the OR at attrition rates of 25 or 40%. When observations are not MAR or MCAR, imputation and weighting methods may not effectively reduce attrition bias.

Bias↗
PubMed2005

Handling of dioxin measurement data in the presence of non-detectable values: overview of available methods and their application in the Seveso chloracne study.

Exposure measurements of concentrations that are non-detectable or near the detection limit (DL) are common in environmental research. Proper statistical treatment of non-detects is critical to avoid bias and unnecessary loss of information. In the present work, we present an overview of possible statistical strategies for handling non-detectable values, including deletion, simple substitution, distributional methods, and distribution-based imputation. Simple substitution methods (e.g., substituting 0, DL/2, DL/ radical2, or DL for the non-detects) are the most commonly applied, even though the EPA Guidance for Data Quality Assessment discouraged their use when the percentage of non-detects is >15%. Distribution-based multiple imputation methods, also known as robust or "fill-in" procedures, may produce dependable results even when 50-70% of the observations are non-detects and can be performed using commonly available statistical software. Any statistical analysis can be conducted on the imputed datasets. Results properly reflect the presence of non-detectable values and produce valid statistical inference. We describe the use of distribution-based multiple imputation in a recent investigation conducted on subjects from the Seveso population exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), in which 55.6% of plasma TCDD measurements were non-detects. We suggest that distribution-based multiple imputation be the preferred method to analyze environmental data when substantial proportions of observations are non-detects.

Acne Vulgaris↗
PubMed2006

Effects of replacing the unreliable cDNA microarray measurements on the disease classification based on gene expression profiles and functional modules.

MOTIVATION: Microarrays datasets frequently contain a large number of missing values (MVs), which need to be estimated and replaced for subsequent data mining. The focus of the paper is to study the effects of different MV treatments for cDNA microarray data on disease classification analysis. RESULTS: By analyzing five datasets, we demonstrate that among three kinds of classifiers evaluated in this study, support vector machine (SVM) classifiers are robust to varied MV imputation methods [e.g. replacing MVs by zero, K nearest-neighbor (KNN) imputation algorithm, local least square imputation and Bayesian principal component analysis], while the classification and regression tree classifiers are sensitive in terms of classification accuracy. The KNNclassifiers built on differentially expressed genes (DEGs) are robust to the varied MV treatments, but the performances of the KNN classifiers based on all measured genes can be significantly deteriorated when imputing MVs for genes with larger missing rate (MR) (e.g. MR > 5%). Generally, while replacing MVs by zero performs relatively poor, the other imputation algorithms have little difference in affecting classification performances of the SVM or KNN classifiers. We further demonstrate the power and feasibility of our recently proposed functional expression profile (FEP) approach as means to handle microarray data with MVs. The FEPs, which are derived from the functional modules that are enriched with sets of DEGs and thus can be consistently identified under varied MV treatments, achieve precise disease classification with better biological interpretation. We conclude that the choice of MV treatments should be determined in context of the later approaches used for disease classification. The suggested exclusion criterion of ignoring the genes with larger MR (e.g. >5%), while justifiable for some classifiers such as KNN classifiers, might not be considered as a general rule for all classifiers.

Algorithms↗
PubMed1996

Impact of missing data due to dropouts on estimates of the treatment effect in a randomized trial of antiretroviral therapy for HIV-infected individuals. Canadian HIV Trials Network A002 Study Group.

PURPOSE: To evaluate the impact of missing data due to nonrandom dropout on estimates of the effect of treatment on the CD4 count in a clinical trial of antiretroviral therapy for HIV infected individuals. METHODS: The effect of treatment on CD4 counts in a recent study of continued ZDV versus ddI in HIV-infected individuals was estimated from the observed data and after imputing missing CD4 counts for patients who dropped out of the study. Imputation methods studied were (a) carrying forward the last observed CD4 count, (b) predicting missing CD4 counts from regression models, and (c) assuming that CD4 counts of patients who dropped out declined at a rate of 100 cells per year. RESULTS: Of the 245 patients enrolled in the study, 52% completed the planned 48 weeks of follow-up. Patients with lower CD4 counts were more likely to drop out of the study (RR = 1.77; p = 0.0001). Patients receiving ZDV had a greater tendency to drop out than patients receiving ddI (p = 0.07). Mean CD4 counts calculated after imputing missing data were lower than those obtained from the observed data at all follow-up times for both treatment groups. Imputing CD4 counts with regression models yielded higher estimates of the effect of treatment than were obtained using the observed data. CONCLUSION: Missing outcome data due to dropouts can result in an underestimation of the treatment effect and overly optimistic statements about the outcome of participants on both treatment arms due to the selective dropout of participants with lower or decreasing CD4 counts. When there are significant dropout rates in randomized trials, imputation is a useful technique to assess the range of plausible values of the treatment effect.

Antiviral Agents↗
PubMed2004

Recovering the missing components in a large noisy low-rank matrix: application to SFM.

In computer vision, it is common to require operations on matrices with "missing data," for example, because of occlusion or tracking failures in the Structure from Motion (SFM) problem. Such a problem can be tackled, allowing the recovery of the missing values, if the matrix should be of low rank (when noise free). The filling in of missing values is known as imputation. Imputation can also be applied in the various subspace techniques for face and shape classification, online "recommender" systems, and a wide variety of other applications. However, iterative imputation can lead to the "recovery" of data that is seriously in error. In this paper, we provide a method to recover the most reliable imputation, in terms of deciding when the inclusion of extra rows or columns, containing significant numbers of missing entries, is likely to lead to poor recovery of the missing parts. Although the proposed approach can be equally applied to a wide range of imputation methods, this paper addresses only the SFM problem. The performance of the proposed method is compared with Jacobs' and Shum's methods for SFM.

Algorithms↗
PubMed2003

A 6-month prospective survey of cutaneous drug reactions in a hospital setting.

BACKGROUND: Few prospective studies are available on the incidence and analysis of the characteristics of adverse cutaneous drug reactions in hospital settings. OBJECTIVES: A 6-month prospective study was managed in our hospital among hospitalized patients to: (i) evaluate the incidence of cutaneous allergic reactions from systemic drugs; (ii) study characteristics of patients with cutaneous drug reactions; (iii) describe the adverse cutaneous reactions; and (iv) evaluate drug reaction imputability and preventability. METHODS: All suspected allergic cutaneous reactions to systemic drugs were collected during a 6-month period (November 2000 to May 2001). Exhaustivity of recording was ensured by regular dissemination of information about this study to the practitioners; a simple method for inclusion by fax was established. Inclusion criteria were suspected cutaneous allergic reactions induced by systemic drugs responsible for hospitalization or developed during hospitalization. A physical examination was done by a dermatologist who completed a standardized questionnaire. Requested information included patient characteristics (associated disorders, severity score), drug intake (list and chronology of the drug intake during the 3 weeks preceding the adverse reaction) and characteristics of the skin reaction (type, course). A group comprising dermatologists and pharmacologists evaluated the drug imputability and preventability. RESULTS: Forty-eight cases were collected. A prevalence of 3.6/1000 among hospitalized patients was estimated. The prevalence rate was higher in patients hospitalized in medical departments (0.5%) than in surgical departments (0.01%) (P < 0.001). The cases were mostly recruited in departments of infectious diseases and dermatology. The most frequent associated disorders were: human immunodeficiency virus (HIV) infection (19%), connective tissue disease (10%) and viral or autoimmune hepatitis (12%). Of these patients, 31% had had a previous immunological drug reaction. Adverse cutaneous drug reactions were principally exanthematous (56%). Reactions were considered severe in 34% of cases because they were responsible for hospitalization (18%), increased the duration of hospitalization (14%) or were life threatening (2%). Principal imputable drugs were antibiotics, mainly penicillins. An imputability score was likely in 56% of cases, but it was only possible to conclude definitively in 44% of patients. In 15% of the cases, the side-effect was considered to be preventable. CONCLUSIONS: This study finds a lower incidence than other studies that reported an incidence of 2% of cutaneous drug reactions in hospitalized patients, but only allergic adverse cutaneous reactions induced by systemic drugs were collected in this study. The previous studies were principally done in selected patients hospitalized only in general internal medicine or medical divisions. Our results confirm some data already known about skin drug reactions: HIV infection as a risk factor (P < 0.0001), high prevalence of adverse cutaneous reactions due to antibiotics, and difficulty in ascertaining the imputability of a drug. A high proportion (34%) of these reactions was severe and 15% were avoidable; these two facts justify the development of an intensive programme of clinical pharmacology.

Adult↗
PubMed2003

Genome-wide linkage analysis of systolic blood pressure slope using the Genetic Analysis Workshop 13 data sets.

Systolic blood pressure (SBP) is an age-dependent complex trait for which both environmental and genetic factors may play a role in explaining variability among individuals. We performed a genome-wide scan of the rate of change in SBP over time on the Framingham Heart Study data and one randomly selected replicate of the simulated data from the Genetic Analysis Workshop 13. We used a variance-component model to carry out linkage analysis and a Markov chain Monte Carlo-based multiple imputation approach to recover missing information. Furthermore, we adopted two selection strategies along with the multiple imputation to deal with subjects taking antihypertensive treatment. The simulated data were used to compare these two strategies, to explore the effectiveness of the multiple imputation in recovering varying degrees of missing information, and its impact on linkage analysis results. For the Framingham data, the marker with the highest LOD score for SBP slope was found on chromosome 7. Interestingly, we found that SBP slopes were not heritable in males but were for females; the marker with the highest LOD score was found on chromosome 18. Using the simulated data, we found that handling treated subjects using the multiple imputation improved the linkage results. We conclude that multiple imputation is a promising approach in recovering missing information in longitudinal genetic studies and hence in improving subsequent linkage analyses.

Age Factors↗
PubMed2005

Home injuries and built form--methodological issues and developments in database linkage.

BACKGROUND: The aim of this body of research is to determine whether injuries in the home are more common in particular types of housing. Previous home injuries research has tended to focus on behaviours or the provision of safety equipment to families with young children. There has been little consideration of the physical environment. This study reports methodological developments in database linkage and analysis to improve researchers abilities to utilise large administrative and clinical databases to carry out health and health services research. METHODS: The study involved linking a database of home injuries obtained from an emergency department surveillance system with an external survey of all homes in an area and population denominators for home types derived from a health service administrative database. Analysis of injury incidence by housing type was adjusted for potential biases due to deprivation and distance to hospital. For non-injured individuals data confidentiality considerations required the deprivation and distance measures be imputed. The process of randomly imputing these variables and the testing of the validity of this approach is detailed. RESULTS: There were 14,081 first injuries in 112,248 residents living in 54,081 homes over a two-year period. The imputation method worked well with imputed and observed measures in the injured group being very similar. Re-randomisation and a repeated analysis gave identical results to the first analysis. One particular housing type had a substantially elevated odds ratio for injury occurrence, OR = 2.07 (95% CI: 1.87 to 2.30). CONCLUSIONS: The method of data linkage, imputation and statistical analysis used provides a basis for improved analysis of database linkage studies.

Accidents, Home↗
PubMed2004

Practical methods for dealing with 'not applicable' item responses in the AMC Linear Disability Score project.

BACKGROUND: Whenever questionnaires are used to collect data on constructs, such as functional status or health related quality of life, it is unlikely that all respondents will respond to all items. This paper examines ways of dealing with responses in a 'not applicable' category to items included in the AMC Linear Disability Score (ALDS) project item bank. METHODS: The data examined in this paper come from the responses of 392 respondents to 32 items and form part of the calibration sample for the ALDS item bank. The data are analysed using the one-parameter logistic item response theory model. The four practical strategies for dealing with this type of response are: cold deck imputation; hot deck imputation; treating the missing responses as if these items had never been offered to those individual patients; and using a model which takes account of the 'tendency to respond to items'. RESULTS: The item and respondent population parameter estimates were very similar for the strategies involving hot deck imputation; treating the missing responses as if these items had never been offered to those individual patients; and using a model which takes account of the 'tendency to respond to items'. The estimates obtained using the cold deck imputation method were substantially different. CONCLUSIONS: The cold deck imputation method was not considered suitable for use in the ALDS item bank. The other three methods described can be usefully implemented in the ALDS item bank, depending on the purpose of the data analysis to be carried out. These three methods may be useful for other data sets examining similar constructs, when item response theory based methods are used.

Activities of Daily Living↗
PubMed2003

Methods to analyse cost data of patients who withdraw in a clinical trial setting.

BACKGROUND: Missing data resulting from premature study withdrawal are a common problem in the analysis of longitudinal data in clinical trials. To date, this subject has received little attention in the context of economic evaluations and with regard to the analysis of cost data. OBJECTIVES: To (i) demonstrate the impact of patients who drop out during the study on the outcomes of an economic evaluation, and (ii) to compare the mean and variation in costs after applying five different methods to deal with incomplete data: multiple imputation, complete cases analysis, linear extrapolation, predicted mean and hot decking. STUDY DESIGN: The study was performed using cost data collected in two randomised clinical trials comparing patients with chronic obstructive pulmonary disease receiving either tiotropium bromide or ipratropium bromide. The overall dropout rate was 17%, with the daily costs of the dropouts approximately 4 times higher than the costs of the completers. METHODS: Multiple imputation is a principled method that deals with missing observations by replacing each missing observation with a set of multiple plausible values. The variance between the resulting multiple datasets is combined with the variance between the datasets to take account of the extra uncertainty that results from missing data. The outcomes after multiple imputation were compared with the results of four naive methods to deal with missing observations: complete cases analysis, linear extrapolation, predicted mean and hot decking. All costs were expressed in 2001 euros. RESULTS: In the tiotropium bromide group, mean (standard error) costs varied from Euro 955 (137) after complete cases analysis to Euro 1298 (198) after linear extrapolation. The corresponding estimates in the ipratropium bromide group were Euro 970 (125) and Euro 1561 (244), respectively. The difference in costs between treatment groups varied from -Euro 15 (95% CI: -379 to 349) after complete cases analysis to -Euro 402 (95% CI: -883 to 79) after predicted mean, in favour of the tiotropium bromide group. The difference in costs according to the other methods varied from -Euro 263 (95% CI: -878 to 353) after linear extrapolation to -Euro 265 (95% CI: -709 to 180) after multiple imputation to -Euro 359 (95% CI: -771 to 54) after hot decking. CONCLUSION: This study showed that the method of dealing with the data of the dropouts had a large impact on the outcomes of an economic evaluation. Information about the rate of patient withdrawal and the way data of dropouts are treated is of vital importance in assessing the results of economic evaluations and should always be reported. Multiple imputation is a principled method that can be used to deal with the data of these patients.

Bronchodilator Agents↗
PubMed1998

Issues related to using a short-form of the Center for Epidemiological Studies-Depression Scale.

The study examined the reliability and validity, including the factor structure, of a 10-item abbreviated version of the Center for Epidemiological Studies-Depression (CES-Depression) scale when administered as part of a mail questionnaire. It also examined patterns of nonresponse to items and the effects of imputation of data for missing items on the factor structure of the inventory. A problem of missing data has been reported even with interview administration of the CES-Depression. Researchers have varied considerably in the amount of imputation used to replace missing datapoints. In this study, factor structures varied when items were imputed. In addition, those subjects with complete data were compared with those with up to two imputed datapoints. Those subjects with imputed data were more likely to be female, have lower functional status scores, lower self-reported health status, more advanced age, and a greater number of depressive symptoms than those with no missing data. While the estimate of coefficient alpha of .78 indicated the inventory was reliable, the effects of missing data on construct validity were problematic.

Aged↗
PubMed2006

Correcting for numerator/denominator bias when assessing changing inequalities in occupational class mortality, Australia 1981 -2002.

OBJECTIVE: Comparisons of the changing patterns of inequalities in occupational mortality provide one way to monitor the achievement of equity goals. However, previous comparisons have not corrected for numerator/denominator bias, which is a consequence of the different ways in which occupational details are recorded on death certificates and on census forms. The objective of this study was to measure the impact of this bias on mortality rates and ratios over time. METHODS: Using data provided by the Australian Bureau of Statistics, we examined the evidence for bias over the period 1981 -2002, and used imputation methods to adjust for this bias. We compared unadjusted with imputed rates of mortality for manual/non-manual workers. FINDINGS: Unadjusted data indicate increasing inequality in the age-adjusted rates of mortality for manual/non-manual workers during 1981 -2002. Imputed data suggest that there have been modest fluctuations in the ratios of mortality for manual/non-manual workers during this time, but with evidence that inequalities have increased only in recent years and are now at historic highs. CONCLUSION: We found that imputation for missing data leads to changes in estimates of inequalities related to social class in mortality for some years but not for others. Occupational class comparisons should be imputed or otherwise adjusted for missing data on census or death certificates.

Adolescent↗
PubMed2005

Handling missing data in self-report measures.

Self-report measures are extensively used in nursing research. Data derived from such reports can be compromised by the problem of missing data. To help ensure accurate parameter estimates and valid research results, the problem of missing data needs to be appropriately addressed. However, a review of nursing research literature revealed that issues such as the extent and pattern of missingness, and the approach used to handle missing data are seldom reported. The purpose of this article is to provide researchers with a conceptual overview of the issues associated with missing data, procedures used in determining the pattern of missingness, and techniques for handling missing data. The article also highlights the advantages and disadvantages of these techniques, and makes distinctions between data that are missing at the item versus variable levels. Missing data handling techniques addressed in this article include deletion approaches, mean substitution, regression-based imputation, hot-deck imputation, multiple imputation, and maximum likelihood imputation.

Data Collection↗