The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies

Much biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalisability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case-control, and cross-sectional studies. We convened a 2-day workshop in September, 2004, with methodologists, researchers, and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE statement) that relate to the title, abstract, introduction, methods, results, and discussion sections of articles.18 items are common to all three study designs and four are specific for cohort, case-control, or cross-sectional studies.A detailed explanation and elaboration document is published separately and is freely available on the websites of PLoS Medicine, Annals of Internal Medicine, and Epidemiology. We hope that the STROBE statement will contribute to improving the quality of reporting of observational studies

Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness

Qualitative content analysis as described in published literature shows conflicting opinions and unsolved issues regarding meaning and use of concepts, procedures and interpretation. This paper provides an overview of important concepts (manifest and latent content, unit of analysis, meaning unit, condensation, abstraction, content area, code, category and theme) related to qualitative content analysis; illustrates the use of concepts related to the research procedure; and proposes measures to achieve trustworthiness (credibility, dependability and transferability) throughout the steps of the research procedure. Interpretation in qualitative content analysis is discussed in light of Watzlawick et al.'s [Pragmatics of Human Communication. A Study of Interactional Patterns, Pathologies and Paradoxes. W.W. Norton & Company, New York, London] theory of communication.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies

Much of biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalizability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case-control, and cross-sectional studies. We convened a 2-day workshop in September 2004, with methodologists, researchers, and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE Statement) that relate to the title, abstract, introduction, methods, results, and discussion sections of articles. Eighteen items are common to all three study designs and four are specific for cohort, case-control, or cross-sectional studies. A detailed Explanation and Elaboration document is published separately and is freely available on the web sites of PLoS Medicine, Annals of Internal Medicine, and Epidemiology. We hope that the STROBE Statement will contribute to improving the quality of reporting of observational studies.

Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration

Much medical research is observational. The reporting of observational studies is often of insufficient quality. Poor reporting hampers the assessment of the strengths and weaknesses of a study and the generalisability of its results. Taking into account empirical evidence and theoretical considerations, a group of methodologists, researchers, and editors developed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations to improve the quality of reporting of observational studies. The STROBE Statement consists of a checklist of 22 items, which relate to the title, abstract, introduction, methods, results and discussion sections of articles. Eighteen items are common to cohort studies, case-control studies and cross-sectional studies and four are specific to each of the three study designs. The STROBE Statement provides guidance to authors about how to improve the reporting of observational studies and facilitates critical appraisal and interpretation of studies by reviewers, journal editors and readers. This explanatory and elaboration document is intended to enhance the use, understanding, and dissemination of the STROBE Statement. The meaning and rationale for each checklist item are presented. For each item, one or several published examples and, where possible, references to relevant empirical studies and methodological literature are provided. Examples of useful flow diagrams are also included. The STROBE Statement, this document, and the associated Web site (http://www.strobe-statement.org/) should be helpful resources to improve reporting of observational research.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies

Much biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalisability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case-control, and cross-sectional studies. We convened a 2-day workshop in September 2004, with methodologists, researchers, and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE Statement) that relate to the title, abstract, introduction, methods, results, and discussion sections of articles. 18 items are common to all three study designs and four are specific for cohort, case-control, or cross-sectional studies. A detailed Explanation and Elaboration document is published separately and is freely available on the Web sites of PLoS Medicine, Annals of Internal Medicine, and Epidemiology. We hope that the STROBE Statement will contribute to improving the quality of reporting of observational studies.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies

Much biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalizability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case-control and cross-sectional studies. We convened a two-day workshop, in September 2004, with methodologists, researchers and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE Statement) that relate to the title, abstract, introduction, methods, results and discussion sections of articles. Eighteen items are common to all three study designs and four are specific for cohort, case-control, or cross-sectional studies. A detailed Explanation and Elaboration document is published separately and is freely available on the web sites of PLoS Medicine, Annals of Internal Medicine and Epidemiology. We hope that the STROBE Statement will contribute to improving the quality of reporting of observational studies.

Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease

BACKGROUND

The Women's Health Initiative randomized trial found greater coronary heart disease (CHD) risk in women assigned to estrogen/progestin therapy than in those assigned to placebo. Observational studies had previously suggested reduced CHD risk in hormone users.

METHODS

Using data from the observational Nurses' Health Study, we emulated the design and intention-to-treat (ITT) analysis of the randomized trial. The observational study was conceptualized as a sequence of "trials," in which eligible women were classified as initiators or noninitiators of estrogen/progestin therapy.

RESULTS

The ITT hazard ratios (HRs) (95% confidence intervals) of CHD for initiators versus noninitiators were 1.42 (0.92-2.20) for the first 2 years, and 0.96 (0.78-1.18) for the entire follow-up. The ITT HRs were 0.84 (0.61-1.14) in women within 10 years of menopause, and 1.12 (0.84-1.48) in the others (P value for interaction = 0.08). These ITT estimates are similar to those from the Women's Health Initiative. Because the ITT approach causes severe treatment misclassification, we also estimated adherence-adjusted effects by inverse probability weighting. The HRs were 1.61 (0.97-2.66) for the first 2 years, and 0.98 (0.66-1.49) for the entire follow-up. The HRs were 0.54 (0.19-1.51) in women within 10 years after menopause, and 1.20 (0.78-1.84) in others (P value for interaction = 0.01). We also present comparisons between these estimates and previously reported Nurses' Health Study estimates.

CONCLUSIONS

Our findings suggest that the discrepancies between the Women's Health Initiative and Nurses' Health Study ITT estimates could be largely explained by differences in the distribution of time since menopause and length of follow-up.

Physical activity assessment in children and adolescents

Chronic disease risk factors, including a sedentary lifestyle, may be present even in young children, suggesting that early prevention programmes may be critical to reducing the rates of chronic disease. Accurate assessment of physical activity in children is necessary to identify current levels of activity and to assess the effectiveness of intervention programmes designed to increase physical activity. This article summarises the strengths and limitations of the methods used to evaluate physical activity in children and adolescents. MEDLINE searches and journal article citations were used to locate 59 articles that validated physical activity measurement methods in children and adolescents. Only those methods that were validated against a more stringent measure were included in the review. Based on the definition of physical activity as any bodily movement resulting in energy expenditure (EE), direct observation of the individual's movement should be used as the gold standard for physical activity research. The doubly labelled water technique and indirect calorimetry can also be considered criterion measures for physical activity research, because they measure EE, a physiologic consequence closely associated with physical activity. Devices such as heart rate monitors, pedometers and accelerometers have become increasingly popular as measurement tools for physical activity. These devices reduce the subjectivity inherent in survey methods and can be used with large groups of individuals. Heart rate monitoring is sufficiently valid to use in creating broad physical activity categories (e.g. highly active, somewhat active, sedentary) but lacks the specificity needed to estimate physical activity in individuals. Laboratory and field validations of pedometers and accelerometers yield relatively high correlations using oxygen consumption (r = 0.62 to 0.93) or direct observation (r = 0.80 to 0.97) as criterion measures, although, they may not be able to capture all physical activity. Physical activity has traditionally been measured with surveys and recall instruments. These techniques must be used cautiously in a paediatric population that has difficulty recalling such information. Still, some studies have reported 73.4% to 86.3% agreement between these instruments and direct observation. Future investigations of physical activity instruments should validate the novel instrument against a higher standard. Additional studies are needed to investigate the possibility of improving the accuracy of measurement by combining 2 or more techniques. The accurate measurement of physical activity is critical for determining current levels of physical activity, monitoring compliance with physical activity guidelines, understanding the dose-response relationship between physical activity and health and determining the effectiveness of intervention programmes designed to improve physical activity.

Hospitalisations caused by adverse drug reactions (ADR): a meta-analysis of observational studies

AIM

To establish the percentage hospital admission related to adverse drug reactions (ADRs) from the data available in the literature.

METHOD

Literature search in the Medline database, meta-analysis.

RESULTS

From the literature it is revealed that a considerable part of all hospital admissions are related to adverse drug reactions. However, these data are not homogenous, i.e. larger studies display a lower percentage of ADR related hospital admission, while smaller studies display a higher percentage. Subgroup analysis showed that for elderly people the odds of being hospitalised by ADR related problems is 4 times higher than for younger ones (16.6% vs. 4.1%). A considerable part of these hospitalisations can be prevented. Subgroup analysis revealed that in the elderly up to 88% of the ADR related hospitalisations are preventable; for the non-elderly this is only 24%. Comparatively more elderly people are hospitalised than younger ones. Combining these findings, twice as much elderly people are hospitalised by ADR related problems than non-elderly, while preventability of ADR related hospitalisation might yield 7 times more people in the elderly than in the non-elderly. The estimation of the costs of ADR related hospitalisations in the Health Care system in The Netherlands is discussed.

CONCLUSION

Many elderly people are hospitalised by ADR related problems; an important part of these hospitalisations can be avoided.

The Delirium Observation Screening Scale: a screening instrument for delirium

The Delirium Observation Screening (DOS) scale, a 25-item scale, was developed to facilitate early recognition of delirium, according to the Diagnostic and Statistical Manual-IV criteria, based on nurses' observations during regular care. The scale was tested for content validity by a group of seven experts in the field of delirium. Internal consistency, predictive validity, and concurrent and construct validity were tested in two prospective studies with high risk groups of patients: geriatric medicine patients and elderly hip fracture patients. Among the patients admitted to a geriatric department (N = 82), 4 became delirious; among the elderly hip fracture patients (N = 92), 18 became delirious. The DOS scale was determined to be content valid and showed high internal consistency, alpha = 0.93 and alpha = 0.96. Predictive validity against the Diagnostic and Statistical Manual-IV diagnosis of delirium made by a geriatrician was good in both studies. Correlations of the DOS scale with the Mini Mental State Examination (MMSE) were Rs -0.79 (p < or = 0.001) in the hip fracture patients and Rs -0.66 (p < or = 0.001) in the geriatric medicine patients. Concurrent validity, as tested by comparison of the research nurse's ratings of the DOS scale and the Confusion Assessment Method (CAM), for the group of hip fracture patients was 0.63 (p < or = 0.001). Construct validity of the DOS was tested against the Informant Questionnaire of Cognitive Decline in Elderly (IQCODE), a preexisting psychiatric diagnosis and the Barthel Index. Correlation with the IQCODE was 0.74 (p < or = 0.001) in the study with the hip fracture patients and 0.33 (p < or = 0.05) in the study with the geriatric medicine patients. Correlation with the Barthel Index was -0.26 (p < or = 0.05) in the geriatric medicine patients and -0.55 (p < or = 0.001) in the hip fracture patients. The overall conclusion of these studies is that the DOS scale shows satisfactory validity and reliability, to guide early recognition of delirium by nurses' observation.

High-Density Lipoprotein Subpopulation Profile and Coronary Heart Disease Prevalence in Male Participants of the Framingham Offspring Study

OBJECTIVE

High-density lipoprotein (HDL) is a heterogeneous lipoprotein class and there is no consensus on the value of HDL subspecies in coronary heart disease (CHD) risk assessment. We tested the hypothesis whether specific HDL subpopulations are significantly associated with CHD-prevalence.

METHODS AND RESULTS

ApoA-I concentrations (mg/dL) in HDL subpopulations were quantitatively determined by native 2d gel electrophoresis, immunoblotting, and image analysis in male participants in the Framingham Offspring Study (FOS). CHD cases (n=169) had higher prebeta-1 and alpha-3 particle and lower alpha-1, prealpha-3, and prealpha-1 particle levels than either all (n=1277) or HDL cholesterol-matched (n=358) controls. alpha-1 and prealpha-3 levels had an inverse association, whereas alpha-3 and prealpha-1 particle levels had a positive association with CHD prevalence after adjusting the data for established CHD risk factors. Standardized logit coefficients indicated that alpha-1 HDL was most significantly associated with CHD prevalence. Moreover, each mg/dL increase in alpha-1 particle level decreased odds of CHD by 26% (P<0.0001), whereas each mg/dL increase in HDL cholesterol decreased odds of CHD by 2% in a model including all established CHD risk factors.

CONCLUSIONS

Specific HDL subpopulations were positively correlated, whereas others were inversely correlated with CHD prevalence in male subject in the FOS, indicating that the various HDL particles might have different roles in the cause of CHD.

A theoretical framework for understanding self-report and observational measures of pain: a communications model

Self-report and observational measures of pain are examined from the perspective of a model of human communication. This model examines the experience of pain as affected by intrapersonal and contextual factors, the process whereby it is encoded into expressive behaviour, and the process of decoding by observers prior to their engaging in action. Self-report measures primarily capture expressive pain behaviour that is under the control of higher mental processes, whereas observational measures capture behaviour that is less subject to voluntary control and more automatic. Automatic expressive behaviours are subject to less purposeful distortion than are behaviours dependent upon higher mental processes. Consequently, observational measures can be used and have clinical utility as indices of pain when self-report is not available, for example, in infants, young children, people with intellectual disabilities or brain damage, and seniors with dementia. These measures are also useful when the credibility of self-report is questioned and even when credible self-report is available. However, automatic behaviours may be more difficult for observers to decode. The model outlined herein takes into account the role of various human developmental stages in pain experience and expression and in understanding the utility of self-report and observational measures. We conclude that both observational and self-report measures are essential in the assessment of pain because of the unique information that each type contributes.

Assessment and control for confounding by indication in observational studies

In the evaluation of pharmacologic therapies, the controlled clinical trial is the preferred design. When clinical trial results are not available, the alternative designs are observational epidemiologic studies. A traditional concern about the validity of findings from epidemiologic studies is the possibility of bias from uncontrolled confounding. In studies of pharmacologic therapies, confounding by indication may arise when a drug treatment serves as a marker for a clinical characteristic or medical condition that triggers the use of the treatment and that, at the same time, increases the risk of the outcome under study. Confounding by indication is not conceptually different from confounding by other factors, and the approaches to detect and control for confounding--matching, stratification, restriction, and multivariate adjustment--are the same. Even after adjustment for known risk factors, residual confounding may occur because of measurement error or unmeasured or unknown risk factors. Although residual confounding is difficult to exclude in observational studies, there are limits to what this "unknown" confounding can explain. The degree of confounding depends on the prevalence of the putative confounding factor, the level of its association with the disease, and the level of its association with the exposure. For example, a confounding factor with a prevalence of 20% would have to increase the relative odds of both outcome and exposure by factors of 4 to 5 before the relative risk of 1.57 would be reduced to 1.00. Observational studies have provided important scientific evidence about the risks associated with several risk factors, including drug therapies, and they are often the only option for assessing safety. Understanding the methods to detect and control for confounding makes it possible to assess the plausibility of claims that confounding is an alternative explanation for the findings of particular studies.

In the field: notes on observation in qualitative research

BACKGROUND

Observation is used in research in two ways - structured and unstructured. Which of these methods to choose depends on the research question but will be defined predominantly by the paradigm underlying each study. Positivistic research generally uses structured observation and interpretist/naturalistic paradigms use unstructured observation. This paper is concerned with unstructured observation.

AIM

To discuss the importance of unstructured observation as a research method and to examine critically the problems associated with both access and field notes.

APPLICATION TO NURSING

Although unstructured observation has been used in nursing research, it is less favoured than the interview. Furthermore, although some of the dilemmas concerned with observation such as the role of the researcher, covert research and informed consent have received attention in the nursing literature, other aspects remain relatively unexplored. In particular little has been written concerning access to the field and field notes. Both of these aspects have practical and theoretical considerations that require the attention of the researcher. This omission may partly explain nurses' apparent reluctance to use unstructured observation in their studies. This paper seeks to redress this situation.

Symptoms experience: a concept analysis

PURPOSE/OBJECTIVES

To provide a clearly constructed definition of the concept of symptoms experience.

DATA SOURCES

Articles and book chapters.

DATA SYNTHESIS

Symptoms experience has not been explored previously. Several approaches to the concept of symptoms have been addressed, including symptom occurrence, symptom distress, and unpleasant symptoms. Limitations of these approaches may include the lack of focus on symptoms as occurring concurrently or in clusters or the multiplicative nature of symptoms. In addition, situational and existential meaning often is not explored.

CONCLUSIONS

Symptoms experience is the perception of the frequency, intensity, distress, and meaning of symptoms as they are produced and expressed. Symptoms are multiplicative in nature and may act as catalysts for the occurrence of other symptoms. Antecedents to the symptoms experience include demographic, disease, and individual factors. Consequences include the impact on mood state, psychological status, functional status, quality of life, disease progression, and survival.

IMPLICATIONS FOR NURSING

Evaluation of symptoms in patients with cancer should include a meaning-centered approach, in which symptoms are evaluated not only for occurrence characteristics and perceived distress but also for the meaning of the symptoms experience to individuals.

Motor learning is optimally tuned to the properties of motor noise

In motor learning, our brain uses movement errors to adjust planning of future movements. This process has traditionally been studied by examining how motor planning is adjusted in response to visuomotor or dynamic perturbations. Here, I show that the learning strategy can be better identified from the statistics of movements made in the absence of perturbations. The strategy identified this way differs from the learning mechanism assumed in mainstream models for motor learning. Crucial for this strategy is that motor noise arises partly centrally, in movement planning, and partly peripherally, in movement execution. Corrections are made by modification of central planning signals from the previous movement, which include the effects of planning but not execution noise. The size of the corrections is such that the movement variability is minimized. This physiologically plausible strategy is optimally tuned to the properties of motor noise, and likely underlies learning in many motor tasks.

Interpretation/analysis methods in hermeneutic interpretive phenomenology

BACKGROUND

Although hermeneutical interpretive phenomenological methodology has been used in many nursing research studies, reports of findings are generally not specific regarding the analysis, or "interpretive process," of this methodology.

APPROACH

The purpose of this article is to assist researchers in analyzing interviews and observations, as part of hermeneutical interpretive phenomenology. In this article the interpretive process is explained, illustrated by examples from a recent research study.

RESULTS

Other facets of the methodology including research question development, sampling issues, and interview and observation methods are briefly reviewed.

DISCUSSION

Approaching the interpretive process as systematically as possible within a nonlinear methodology streamlines and clarifies interpretations of the interview data.

Missing in action: teacher and health worker absence in developing countries

In this paper, we report results from surveys in which enumerators make unannounced visits to primary schools and health clinics in Bangladesh, Ecuador, India, Indonesia, Peru and Uganda and recorded whether they found teachers and health workers in the facilities.

Current techniques for assessing physical exposure to work-related musculoskeletal risks, with emphasis on posture-based methods

Physical exposure to risks for potential work-related musculoskeletal injuries has been assessed using a variety of methods, including pen and paper based observation methods, videotaping and computer-aided analysis, direct or instrumental techniques, and various approaches to self-report assessment. These methods are critically reviewed in this paper. The applications of these techniques in ergonomic and epidemiologic studies are considered, and their advantages and shortcomings are highlighted. Finally, a strategy that considers both the ergonomics experts' view and the practitioners' needs for developing a practical exposure assessment tool is then discussed.

Risk factors, confounding, and the illusion of statistical control

When experimental designs are premature, impractical, or impossible, researchers must rely on statistical methods to adjust for potentially confounding effects. Such procedures, however, are quite fallible. We examine several errors that often follow the use of statistical adjustment. The first is inferring a factor is causal because it predicts an outcome even after "statistical control" for other factors. This inference is fallacious when (as usual) such control involves removing the linear contribution of imperfectly measured variables, or when some confounders remain unmeasured. The converse fallacy is inferring a factor is not causally important because its association with the outcome is attenuated or eliminated by the inclusion of covariates in the adjustment process. This attenuation may only reflect that the covariates treated as confounders are actually mediators (intermediates) and critical to the causal chain from the study factor to the study outcome. Other problems arise due to mismeasurement of the study factor or outcome, or because these study variables are only proxies for underlying constructs. Statistical adjustment serves a useful function, but it cannot transform observational studies into natural experiments, and involves far more subjective judgment than many users realize.

Sleep disturbance and melatonin levels following traumatic brain injury

OBJECTIVES

Sleep disturbances commonly follow traumatic brain injury (TBI) and contribute to ongoing disability. However, there are no conclusive findings regarding specific changes to sleep quality and sleep architecture measured using polysomnography. Possible causes of the sleep disturbances include disruption of circadian regulation of sleep-wakefulness, psychological distress, and a neuronal response to injury. We investigated sleep-wake disturbances and their underlying mechanisms in a TBI patient sample.

METHODS

This was an observational study comparing 23 patients with TBI (429.7 +/- 287.6 days post injury) and 23 age- and gender-matched healthy volunteers on polysomnographic sleep measures, salivary dim light melatonin onset (DLMO) time, and self-reported sleep quality, anxiety, and depression.

RESULTS

Patients with TBI reported higher anxiety and depressive symptoms and sleep disturbance than controls. Patients with TBI showed decreased sleep efficiency (SE) and increased wake after sleep onset (WASO). Although no significant group differences were found in sleep architecture, when anxiety and depression scores were controlled, patients with TBI showed higher amount of slow wave sleep. No differences in self-reported sleep timing or salivary DLMO time were found. However, patients with TBI showed significantly lower levels of evening melatonin production. Melatonin level was significantly correlated with REM sleep but not SE or WASO.

CONCLUSIONS

Reduced evening melatonin production may indicate disruption to circadian regulation of melatonin synthesis. The results suggest that there are at least 2 factors contributing to sleep disturbances in patients with traumatic brain injury. We propose that elevated depression is associated with reduced sleep quality, and increased slow wave sleep is attributed to the effects of mechanical brain damage.

Last observation carry-forward and last observation analysis

Drop-out often occurs in clinical trials with multiple visits and drop-out is often informative in the sense that the population of patients who dropped out is different from the population of patients who completed the study. To handle data with informative drop-out, an intention-to-treat analysis, which evaluates treatment effects over the population of all randomized patients with at least one post-treatment evaluation, is often required by the regulatory agencies. As a popular and simple intention-to-treat analysis, the last observation carry-forward (LOCF) analysis of variance (ANOVA) performs a statistical test for treatment effects by treating the last observation prior to drop-out as the observation from the last visit. Although discussions, examples and limited empirical results about the LOCF analysis can be found, its theoretical property is unclear. We find that the LOCF one-way ANOVA test is actually asymptotically valid (that is, its asymptotic size is equal to the nominal size) in the special but important case where only two treatments are compared and the two treatment groups have the same number of patients, regardless of whether drop-out is informative or not. In other cases, however, the asymptotic size of the LOCF test is different from the nominal size and is often too small when drop-out is informative, which results in a loss in power of detecting treatment effects, a disadvantage to drug companies. We propose an asymptotically valid test for comparing the global means over subpopulations, where each subpopulation contains patients dropping out after a particular visit. Some simulation results are presented to study the finite sample performance of the LOCF test and our proposed test.