SEARCH · PubMed Health

Results for “imputation of missing not at random values”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 73 records · Page 4Linked to original sources

PubMed2003

Memantine in moderate-to-severe Alzheimer's disease.

BACKGROUND: Overstimulation of the N-methyl-D-aspartate (NMDA) receptor by glutamate is implicated in neurodegenerative disorders. Accordingly, we investigated memantine, an NMDA antagonist, for the treatment of Alzheimer's disease. METHODS: Patients with moderate-to-severe Alzheimer's disease were randomly assigned to receive placebo or 20 mg of memantine daily for 28 weeks. The primary efficacy variables were the Clinician's Interview-Based Impression of Change Plus Caregiver Input (CIBIC-Plus) and the Alzheimer's Disease Cooperative Study Activities of Daily Living Inventory modified for severe dementia (ADCS-ADLsev). The secondary efficacy end points included the Severe Impairment Battery and other measures of cognition, function, and behavior. Treatment differences between base line and the end point were assessed. Missing observations were imputed by using the most recent previous observation (the last observation carried forward). The results were also analyzed with only the observed values included, without replacing the missing values (observed-cases analysis). RESULTS: Two hundred fifty-two patients (67 percent women; mean age, 76 years) from 32 U.S. centers were enrolled. Of these, 181 (72 percent) completed the study and were evaluated at week 28. Seventy-one patients discontinued treatment prematurely (42 taking placebo and 29 taking memantine). Patients receiving memantine had a better outcome than those receiving placebo, according to the results of the CIBIC-Plus (P=0.06 with the last observation carried forward, P=0.03 for observed cases), the ADCS-ADLsev (P=0.02 with the last observation carried forward, P=0.003 for observed cases), and the Severe Impairment Battery (P<0.001 with the last observation carried forward, P=0.002 for observed cases). Memantine was not associated with a significant frequency of adverse events. CONCLUSIONS: Antiglutamatergic treatment reduced clinical deterioration in moderate-to-severe Alzheimer's disease, a phase associated with distress for patients and burden on caregivers, for which other treatments are not available.

Activities of Daily Living↗

PubMed1995

A critical look at methods for handling missing covariates in epidemiologic regression analyses.

Epidemiologic studies often encounter missing covariate values. While simple methods such as stratification on missing-data status, conditional-mean imputation, and complete-subject analysis are commonly employed for handling this problem, several studies have shown that these methods can be biased under reasonable circumstances. The authors review these results in the context of logistic regression and present simulation experiments showing the limitations of the methods. The method based on missing-data indicators can exhibit severe bias even when the data are missing completely at random, and regression (conditional-mean) imputation can be inordinately sensitive to model misspecification. Even complete-subject analysis can outperform these methods. More sophisticated methods, such as maximum likelihood, multiple imputation, and weighted estimating equations, have been given extensive attention in the statistics literature. While these methods are superior to simple methods, they are not commonly used in epidemiology, no doubt due to their complexity and the lack of packaged software to apply these methods. The authors contrast the results of multiple imputation to simple methods in the analysis of a case-control study of endometrial cancer, and they find a meaningful difference in results for age at menarche. In general, the authors recommend that epidemiologists avoid using the missing-indicator method and use more sophisticated methods whenever a large proportion of data are missing.

Case-Control Studies↗

PubMed2026

Impact of N-PEP-12 Supplementation on Attentional Performance and Mental Wellbeing in Healthy Adults with Subjective Cognitive Complaints: A Randomized, Placebo-Controlled Trial.

Background: Subjective cognitive complaints (SCCs) are common in middle-aged and older adults and may reflect early cognitive changes, alongside alterations in stress, mood, sleep, and quality of life. N-PEP-12 is a peptide-based nutritional supplement with potential neuroprotective effects, but evidence of its benefits in healthy adults with SCCs remains limited. Objective: To evaluate the effects of N-PEP-12 supplementation on attention, cognitive function, and mental wellbeing in healthy middle-aged and older adults with SCCs. Methods: In this prospective, randomized, double-blind, placebo-controlled trial, 276 participants aged 50-75 years with SCCs and no clinically significant cognitive impairment were randomized to placebo, N-PEP-12 45 mg, or N-PEP-12 90 mg. Assessments were performed at baseline and after 30, 90, and 180 days. Primary outcomes included Test of Attentional Performance measures. Secondary outcomes included WAIS-IV Digit Span Forward and Backward, perceived stress, mood, sleep quality, and EQ-5D-5L visual analog scale. Results: Across the three primary attention outcomes analyzed jointly in a multivariate repeated-measures model, there was a significant group-by-visit interaction (p = 0.003) and a significant effect of visit (p < 0.001), without a significant main effect of treatment arm (p = 0.133), indicating a time-dependent treatment effect. This result was obtained in a sensitivity analysis population in which missing values were imputed under assumptions least favorable to the active arms. In exploratory endpoint-specific comparisons at 90 days, both N-PEP-12 groups showed greater improvements than placebo in alertness, attention omissions and memory omissions, and Digit Span Forward and Backward scores also improved. In a subsequent uncontrolled extension phase, in which all participants received active treatment, participants initially assigned to placebo showed comparable improvements after switching to N-PEP-12 90 mg. These observations are exploratory. Adverse events were infrequent and similarly distributed across groups. Conclusions: N-PEP-12 supplementation was associated with improvements in attention, working memory, and mental wellbeing in healthy middle-aged and older adults with subjective cognitive complaints. These findings support further investigation of N-PEP-12 as a nutritional intervention for early subjective cognitive changes associated with aging.

Humans↗

PubMed1995

Efficacy and safety of lowering dietary intake of fat and cholesterol in children with elevated low-density lipoprotein cholesterol. The Dietary Intervention Study in Children (DISC). The Writing Group for the DISC Collaborative Research Group.

OBJECTIVE: To assess the efficacy and safety of lowering dietary intake of total fat, saturated fat, and cholesterol to decrease low-density lipoprotein cholesterol (LDL-C) levels in children. DESIGN: Six-center randomized controlled clinical trial. PARTICIPANTS: Prepubertal boys (n = 362) and girls (n = 301) aged 8 to 10 years with LDL-C levels greater than or equal to the 80th and less than the 98th percentiles for age and sex were randomized into an intervention group (n = 334) and a usual care group (n = 329). INTERVENTION: Behavioral intervention to promote adherence to a diet providing 28% of energy from total fat, less than 8% from saturated fat, up to 9% from polyunsaturated fat, and less than 75 mg/4200 kJ (1000 kcal) per day of cholesterol (not to exceed 150 mg/d). MAIN OUTCOME MEASURES: The primary efficacy measure was the mean LDL-C level at 3 years. Primary safety measures were mean height and serum ferritin levels at 3 years. Secondary efficacy outcomes were mean LDL-C levels at 1 year and mean total cholesterol levels at 1 and 3 years. Secondary safety outcomes included red blood cell folate values; serum zinc, retinol, and albumin levels; serum high-density lipoprotein cholesterol (HDL-C) values, LDL-C:HDL-C ratio, and total triglyceride levels; sexual maturation; and psychosocial health. RESULTS: At 3 years, dietary total fat, saturated fat, and cholesterol levels decreased significantly in the intervention group compared with the usual care group (all P < .001). Levels of LDL-C decreased in the intervention and usual care groups by 0.40 mmol/L (15.4 mg/dL) and 0.31 mmol/L (11.9 mg/dL), respectively. Adjusting for baseline level and sex and imputting values for missing data, the mean difference between the groups was -0.08 mmol/L (-3.23 mg/dL) (95% confidence interval [CI], -0.15 to -0.01 mmol/L [-5.6 to -0.5 mg/dL]), which was significant (P = .02). There were no significant differences between the groups in adjusted mean height or serum ferritin levels (P > .05) or other safety outcomes. CONCLUSIONS: The dietary intervention achieved modest lowering of LDL-C levels over 3 years while maintaining adequate growth, iron stores, nutritional adequacy, and psychological well-being during the critical growth period of adolescence.

Analysis of Variance↗

PubMed2004

Sensitivity analysis for pattern mixture models.

Incomplete series of data is a common feature in quality-of-life studies, in particular in chronic diseases where attrition of patients is high. Two alternative approaches to modeling longitudinal data with incomplete measurements have frequently been proposed in the literature, selection models and pattern-mixture models. In this paper we focus on, by way of sensitivity analysis, extrapolating incomplete patterns using identifying restrictions. Perhaps the best known ones are so-called complete case missing value restrictions (CCMV), where for a given pattern, the conditional distribution of the missing data, given the observed data, is equated to its counterpart in the completers. Available case missing value (ACMV) restrictions equate this conditional density to the one calculated from the subgroup of all patterns for which all required components have been observed. Neighboring case missing value restrictions (NCMV) equate this conditional density to the one calculated from the the pattern with one additional measurement obtained. In this paper, these three identifying restriction strategies are used to multiply impute missing data in a study in metastatic prostate cancer. Multiple imputation is employed to reduce the uncertainty of single imputation. It is shown how hypothesis testing and sensitivity analyses are carried out in this setting.

Humans↗

PubMed2005

Imputing response rates from means and standard deviations in meta-analyses.

The principle of intention-to-treat analysis must be strictly applied to both individual randomized controlled trial and meta-analysis but, in doing so, would involve imputation of some missing data. There is little literature on how to perform this in the case of meta-analysis. For dichotomous outcome measures, one possible strategy is to carry out a sensitivity analysis based on the so-called best case/worst case analyses. For continuous outcomes, it may be possible to achieve this if we can dichotomise the continuous outcomes. Here, we empirically examined the appropriateness of converting continuous outcomes (expressed as mean+/-SD) into dichotomous outcomes (expressed as response rates) in four completed meta-analyses of depression and anxiety, assuming normal distribution of the continuous outcome measures. The agreement between the actually observed versus the imputed raw numbers of responders was indicated by an intraclass correlation coefficient of 0.97 (95% confidence interval 0.95-0.98). The pooled relative risks of the four meta-analyses based on the imputed values were virtually identical to those based on the actually observed values. When individual trials report the means+/-SDs of their outcome measures but fail to report response rates, it may therefore be possible to impute the response rates based on the means+/-SDs, and then submit the meta-analysis to worst case/best case analyses. This would allow a more robust and clinically interpretable estimation of the true, underlying treatment effect to be made.

Data Interpretation, Statistical↗

PubMed2004

The methods for handling missing data in clinical trials influence sample size requirements.

OBJECTIVE: Results of studies estimating osteoarthritis progression may be affected by missing values. In clinical trials assessing disease-modifying osteoarthritis drugs, sample sizes should be calculated using close estimates of outcome variables. STUDY DESIGN AND SETTING: Supposing a two-parallel group design in hip osteoarthritis clinical trials, we estimated sample sizes using the joint space width (JSW), number of patients with JSW progression >0.5 mm (JSN), time to total hip arthroplasty (THA), and time to JSN or THA using several approaches to deal with missing data. RESULTS: Three-year clinical trials testing a treatment effect of 50%, with a power of 80%, could require sample sizes of 121 patients for JSW, 57 for JS progression using multiple imputation for handling missing values; 200 for THA; and 47 for JSN or THA. These numbers vary greatly depending on the approach chosen for handling missing data. CONCLUSIONS: These results can help investigators plan clinical trials to select the primary outcome and a priori specify the way missing data will be handled.

Aged↗

PubMed2003

Assessing response profiles from incomplete longitudinal clinical trial data under regulatory considerations.

Treatment effects are often evaluated by comparing change over time in outcome measures. However, valid analyses of longitudinal data can be problematic, particularly when some data are missing for reasons related to the outcome. In choosing the primary analysis for confirmatory clinical trials, regulatory agencies have for decades favored the last observation carried forward (LOCF) approach for imputing missing values. Many advances in statistical methodology, and also in our ability to implement those methods, have been made in recent years. The characteristics of data from acute phase clinical trials can be exploited to develop an appropriate analysis for assessing response profiles in a regulatory setting. These data characteristics and regulatory considerations will be reviewed. Approaches for handling missing data are compared along with options for modeling time effects and correlations between repeated measurements. Theory and empirical evidence are utilized to support the proposal that likelihood-based mixed-effects model repeated measures (MMRM) approaches, based on the missing at random assumption, provide superior control of Type I and Type II errors when compared with the traditional LOCF approach, which is based on the more restrictive missing completely at random assumption. It is further reasoned that in acute phase clinical trials, unstructured modeling of time trends and within-subject error correlations may be preferred.

Clinical Trials as Topic↗

PubMedDate not supplied

Suggestions for the presentation of quality of life data from clinical trials.

Quality of life (QOL) data is complex since it is both multidimensional and longitudinal. This complexity is compounded with its unbalanced nature through missing observations as a consequence of patient non-compliance with assessment schedules, and, for example, in cancer clinical trials data absence due to patient attrition often through death. QOL data poses difficulties for presentation and analysis and hence interpretation. This paper illustrates, using data from a randomized trial of the United Kingdom Medical Research Council Lung Cancer Working Party, a step-by-step approach to presentation of QOL data. This begins with a description of compliance and its relationship with patient attrition caused by death, to a final summary profile to indicate change over time. We recognize that no single summary statistic is likely to be able to encapsulate all the subtleties of QOL data. We stress the importance of examining data graphically before performing detailed analysis and also to facilitate interpretation in the final clinical report. Although a description of analytical methods is not the purpose of this paper, we draw attention to the need for imputing missing values and to the (multi-level) modelling approach to summarizing the data, both essential adjuncts to the less formal methods described here.

Authorship↗

PubMed2003

Effects of coaching by community pharmacists on psychological symptoms of antidepressant users; a randomised controlled trial.

BACKGROUND: Community pharmacists strive to deliver pharmaceutical care to patients. At the moment, coaching of depressive primary care patients on taking their antidepressants (ADs) is not yet part of their standard care package. AIMS: To investigate the effects of coaching by community pharmacists on psychological symptoms. METHOD: A randomised controlled trial with a 6-month follow-up. OUTCOMES: psychological symptoms with the Hopkins Symptom Checklist (SCL). Intention-to-treat (ITT) was performed with (1) last observation carried forward and (2) with group mean imputation (GMI). RESULTS: Analyses with LOCF and GMI resulted in different findings. The LOCF method revealed that at the 6-month follow-up, the intervention patients were less depressed and less anxious than the controls. The intervention was particularly effective in patients with lower levels of education who received pharmacist's coaching. However, ITT with the GMI method showed no differences in psychological symptoms. Differences between LOCF and GMI were explained by the selective attrition in the intervention arm (attrition intervention patients had lower initial SCL-item scores on depression and anxiety than the completers) and by the higher attrition rate in controls. CONCLUSIONS: Our study indicates that the interpretation of the effects of an intervention on psychological symptoms can differ substantially by the way missing values are imputed. If both LOCF and GMI produce significant differences, efficacy can be concluded. If not, the effects based on ITT analyses with LOCF are based on artefacts. We recommend that positive intervention effects should only be reported when findings with LOCF and GMI are in accordance.

Adult↗

PubMed1993

Regression imputation of missing values in longitudinal data sets.

A stand-alone, menu-driven PC program, written in GAUSS, which can be used to estimate missing observations in longitudinal data sets is described and male available to interested readers. The program is limited to the situation in which we have complete data on N cases at each of the planned times of measurement t1, t2,..., tT; and we wish to use this information, together with the non-missing values for n additional cases, to estimate the missing values for those cases. The augmented data matrix may be saved in an ASCII file and subsequently imported into programs requiring complete data. The use of the program is illustrated. Ten percent of the observations in a data set consisting of mandibular ramus height measurements for N = 12 young male rhesus monkeys measured at T = 5 time points are randomly discarded. The augmented data matrix is used to determine the lowest degree polynomial adequate to fit the average growth curve (AGC); the regression coefficients are estimated and confidence intervals for them are determined; and confidence bands for the AGC are constructed. The results are compared with those obtained when the original complete data set is used.

Animals↗

PubMed2004

LSimpute: accurate estimation of missing values in microarray data with least squares methods.

Microarray experiments generate data sets with information on the expression levels of thousands of genes in a set of biological samples. Unfortunately, such experiments often produce multiple missing expression values, normally due to various experimental problems. As many algorithms for gene expression analysis require a complete data matrix as input, the missing values have to be estimated in order to analyze the available data. Alternatively, genes and arrays can be removed until no missing values remain. However, for genes or arrays with only a small number of missing values, it is desirable to impute those values. For the subsequent analysis to be as informative as possible, it is essential that the estimates for the missing gene expression values are accurate. A small amount of badly estimated missing values in the data might be enough for clustering methods, such as hierachical clustering or K-means clustering, to produce misleading results. Thus, accurate methods for missing value estimation are needed. We present novel methods for estimation of missing values in microarray data sets that are based on the least squares principle, and that utilize correlations between both genes and arrays. For this set of methods, we use the common reference name LSimpute. We compare the estimation accuracy of our methods with the widely used KNNimpute on three complete data matrices from public data sets by randomly knocking out data (labeling as missing). From these tests, we conclude that our LSimpute methods produce estimates that consistently are more accurate than those obtained using KNNimpute. Additionally, we examine a more classic approach to missing value estimation based on expectation maximization (EM). We refer to our EM implementations as EMimpute, and the estimate errors using the EMimpute methods are compared with those our novel methods produce. The results indicate that on average, the estimates from our best performing LSimpute method are at least as accurate as those from the best EMimpute algorithm.

Algorithms↗

PubMed1989

Estimation of parameters and missing values under a regression model with non-normally distributed and non-randomly incomplete data.

We carried out a simulation study to compare the performance of three algorithms (complete cases, ALLVALUE, and expectation maximization, EM) in estimating regression parameters and missing values for situations that have varying amounts of missing data, distributions (normal, mixture of normals and lognormal), patterns of incomplete data (random, related and censored), and degrees of correlational structure among the dependent and independent variables. We found that the EM and complete cases algorithms performed equally well regardless of the correlational structure, when the percentage of incomplete data was only 5 per cent. When this percentage increased to 25 per cent, the EM algorithm was generally best for estimation, but the complete cases algorithm was safe and conservative. This finding may be attributed to the study design, which required that the slopes be the same in the population of all cases, and in the population of complete cases. In addition, the one-step imputing method (ALLVALUE) was competitive only for situations with weak correlational structure and/or little missing data. In that situation the bias caused with use of all available information was less than that caused with use of only complete cases. On the other hand, for imputation, the EM algorithm performed optimally, even in situations of censored or log-normally distributed data.

Algorithms↗

PubMed2007

Analysis of antiretroviral immunotherapy trials with potentially non-normal and incomplete longitudinal data.

For many HIV-infected patients, use of antiretroviral therapy (ART) results in a sustained suppression of plasma viral load to undetectable levels. However, due to lack of antigenic stimulation, this may also result in a gradual loss of cell-mediated immune (CMI) responses that help control HIV infection. In concept, augmenting ART with periodic administrations of an HIV vaccine that boosts CMI responses could enhance control of viral replication. Researchers are designing 'antiretroviral immunotherapy' (ARI) trials to test this hypothesis. In a typical ARI trial, HIV-infected patients with sustained viral suppression will receive inoculations of an experimental HIV vaccine or a placebo, and subsequently stop taking their antiretroviral drugs. The goal is to assess whether plasma viral loads during the ART interruption phase are generally lower in the vaccine group. Assessment of a vaccine effect will be challenging if some subjects resume ART or drop out before the end of the treatment interruption phase. To tackle this 'missing' data problem and potential non-normality of the viral loads in ARI trials, we propose a two-step approach: multiple imputation of the missing values followed by use of the Wei-Lachin method with Wilcoxon scores. We use a numerical example and extensive simulations to illustrate the robustness and power advantages of our proposed method compared with other methods for incomplete longitudinal data, including REML, weighted GEE, last observation carried forward, and 'worst-rank' methods. Our proposed method is general enough for the robust analysis of longitudinal data in other therapeutic areas as well.

AIDS Vaccines↗

PubMed2004

Epidemiologic evaluation of measurement data in the presence of detection limits.

Quantitative measurements of environmental factors greatly improve the quality of epidemiologic studies but can pose challenges because of the presence of upper or lower detection limits or interfering compounds, which do not allow for precise measured values. We consider the regression of an environmental measurement (dependent variable) on several covariates (independent variables). Various strategies are commonly employed to impute values for interval-measured data, including assignment of one-half the detection limit to nondetected values or of "fill-in" values randomly selected from an appropriate distribution. On the basis of a limited simulation study, we found that the former approach can be biased unless the percentage of measurements below detection limits is small (5-10%). The fill-in approach generally produces unbiased parameter estimates but may produce biased variance estimates and thereby distort inference when 30% or more of the data are below detection limits. Truncated data methods (e.g., Tobit regression) and multiple imputation offer two unbiased approaches for analyzing measurement data with detection limits. If interest resides solely on regression parameters, then Tobit regression can be used. If individualized values for measurements below detection limits are needed for additional analysis, such as relative risk regression or graphical display, then multiple imputation produces unbiased estimates and nominal confidence intervals unless the proportion of missing data is extreme. We illustrate various approaches using measurements of pesticide residues in carpet dust in control subjects from a case-control study of non-Hodgkin lymphoma.

Bias↗

PubMedDate not supplied

Options for handling missing data in the Health Utilities Index Mark 3.

BACKGROUND: The Health Utilities Index Mark 3 (HUI3) is a tool composed of 41 questions, covering 8 attributes: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain. Responses to these questions can define more than 972,000 health situations. This tool allows respondents to answer "Don't Know," for which there is no scoring instruction, to any given question. This situation creates a break in the scoring algorithm and leads to considerable amounts of missing data. The goal of this study is to develop strategies to deal with HUI3 scores for participants who have missing data. METHODS: The authors used data from 248 individuals enrolled in the Cataract Management Trial, focusing on the HUI3 vision and ambulation attributes, which had 19% and 10% of attribute levels missing, respectively. Inspection and deduction were used to fill in values independent of the value of the missing data, then alternative analytic techniques were compared, including mean substitution, model scoring, hot deck, multiple imputation, and regression imputation. RESULTS: Inspection and logical deduction reduced the percentage of missing information in the HUI3 by 49% to 87%. A comparison of analytic techniques used for the remaining HUI3 vision data missing demonstrated the value of building models based on internal response patterns and that simple analytic techniques fare as well as more complicated ones when the number of missing cases is small. CONCLUSION: Analyzing the pattern of responses in cases where the attribute level score is missing reduces the amount of missing data and can simplify the analytic process for the remaining missing data.

Activities of Daily Living↗

PubMed2026

Soft Tissue Volume Augmentation at Single Implant Sites Applying Collagen Matrices or Connective Tissue Grafts: 10-Year Follow-Up of a Randomized Controlled Trial.

AIM: To compare up to 10 years clinical, profilometric and patient-reported outcomes of implant sites previously augmented using a volume-stable collagen matrix (VCMX) or connective tissue graft (SCTG) in the aesthetic zone. METHODS: The original non-inferiority randomized controlled trial (RCT) enrolled 20 patients who received soft tissue volume augmentation with VCMX or SCTG at single implant sites. Clinical assessments and standardized measurements were performed at baseline after crown insertion and at 6 months, 1, 3, 5, 7.5, and 10 years. The primary outcome was mucosal thickness. Secondary outcomes included marginal bone levels (MBL), probing depth (PD), bleeding on probing (BOP), plaque control record, Pink Aesthetic Score (PES), OHIP-14 and buccal profilometric changes. Group comparisons were performed using mixed-effects and generalized estimating equation (GEE) models, which account for within-patient correlations due to repeated measurements and allow inclusion of all available data without requiring imputation for missing observations. RESULTS: Of the 20 originally enrolled patients, 10 (5 in the SCTG group and 5 in the VCMX group) were available for re-examination at 10 years. The adjusted between-group difference in mucosal thickness was -0.02 mm (95% CI -0.99 to 0.96). As the lower bound of the confidence interval remained above the prespecified non-inferiority margin of -1 mm, non-inferiority of VCMX was shown. Buccal contour changes were comparable during the early follow-up, while a trend toward a greater long-term contour decrease was observed in group VCMX (-0.31 mm [95% CI, -0.65 to 0.03]; p = 0.07). Mean PES values were 10.6 in the SCTG group and 9.6 in the VCMX group, with no significant between-group differences (p = 0.45). Both groups revealed high levels of oral health-related quality of life, with low median OHIP-14 scores (SCTG, 0.0; VCMX, 1.0; p = 0.26). CONCLUSION: These preliminary long-term findings showed no clinically relevant differences between SCTG and VCMX in terms of clinical, profilometric and patient-reported outcomes. While SCTG remains the reference standard, VCMX represents a less invasive alternative but with a slight tendency toward greater long-term contour reduction. CLINICAL SIGNIFICANCE: Volume-stable collagen matrices serve as a viable alternative to autogenous connective tissue grafts for peri-implant soft tissue volume augmentation, particularly in patients seeking a reduced morbidity, without compromising long-term clinical or aesthetic outcomes. TRIAL REGISTRATION: German Clinical Trials Register: DRKS00017484.

Humans↗

PubMed2001

Group comparisons involving missing data in clinical trials: a comparison of estimates and power (size) for some simple approaches.

When using 'intent-to-treat' approaches to compare outcomes between groups in clinical trials, analysts face a decision regarding how to account for missing observations. Most model-based approaches can be summarized as a process whereby the analyst makes assumptions about the distribution of the missing data in an attempt to obtain unbiased estimates that are based on functions of the observed data. Although pointed out by Rubin as often leading to biased estimates of variances, an alternative approach that continues to appear in the applied literature is to use fixed-value imputation of means for missing observations. The purpose of this paper is to provide illustrations of how several fixed-value mean imputation schemes can be formulated in terms of general linear models that characterize the means of distributions of missing observations in terms of the means of the distributions of observed data. We show that several fixed-value imputation strategies will result in estimated intervention effects that correspond to maximum likelihood estimates obtained under analogous assumptions. If the missing data process has been correctly characterized, hypothesis tests based on variances estimated using maximum likelihood techniques asymptotically have the correct size. In contrast, hypothesis tests performed using the uncorrected variance, obtained by applying standard complete data formula to singly imputed data, can provide either conservative or anticonservative results. Surprisingly, under several non-ignorable non-response scenarios, maximum likelihood based analyses can yield equivalent hypothesis tests to those obtained when analysing only the observed data.

Aged↗