Journal of Clinical Epidemiology - Articles in Press

Validation testing of a three-component model of Short Form-36 scores - Corrected Proof

2010-09-01

Abstract: Objective: The two-component factor structure underlying Short Form-36 (SF-36) summary scores may not be valid worldwide. We studied a three-component model of SF-36 scores in Japan.Study Design and Setting: The SF-36 scores came from representative samples of the population of Japan. Factor analysis and structural equation modeling were used. The two-component model gave physical component summary (PCS) scores and mental component summary (MCS) scores. The three-component model gave scores on the PCS, the MCS, and also on the third component, which we call the role component summary (RCS) score. These were evaluated with external criteria.Results: In the three-component model, the RCS was strongly associated with the role-physical, social functioning, and role-emotional subscales, whereas the PCS and MCS were associated with the physical functioning and mental health subscales, as expected. The goodness-of-fit index was 0.945 for the three-component model and 0.935 for the two-component model. The PCS discriminated between groups stratified by comorbid conditions, and the MCS discriminated between groups stratified by psychological depression. Absence from work was associated with both PCS and RCS.Conclusion: The three-component model is better than the two-component model, and it provides more useful PCS and MCS scores. Criteria for validation testing of the RCS are needed.

Covariate adjustment increases statistical power in randomized controlled trials - Corrected Proof

2010-09-01

In the January issue of this journal, Peter Austin et al. presented a systematic review on how baseline covariates are handled in randomized controlled trials (RCTs) published in four leading medical journals. In this well-performed study, the authors reviewed 114 RCTs published in the BMJ, the JAMA, The Lancet, and the NEJM, between January 1, 2007 and June 30, 2007.

Uptake of methods to deal with publication bias in systematic reviews has increased over time, but there is still much scope for improvement - Corrected Proof

2010-09-01

Abstract: Objective: To evaluate the measures taken to deal with publication bias across different categories of systematic reviews published in 2006 and to compare these with reviews published in 1996.Study Design and Setting: PubMed was searched for systematic reviews published in 2006; 100 treatment effect, 50 diagnostic accuracy, 100 risk factor, and 50 gene–disease association reviews were randomly selected.Results: The use of MEDLINE increased from 74% to 95%; checking references increased from 42% to 73%; use of Cochrane Library increased from 5% to 58%; and use of CINAHL increased from 8% in 1996 to 24% in treatment reviews, 20% in diagnostic reviews, 18% in risk factor reviews, and 0% in genetic reviews published in 2006. A 20% increase was observed for explicit searching of non–English-language studies in all reviews published in 2006. Efforts to search for unpublished studies increased to 61% from 35% in treatment reviews published in 1996. Twenty-six percent of the reviews used funnel plots or related methods to test for publication bias compared with less than 6% in earlier reviews.Conclusion: Recent reviews show a significant improvement in the measures taken to prevent publication bias. However, few methods exist to deal with publication bias in the nonquantitative findings of systematic reviews.

Routine health insurance data for scientific research: potential and limitations of the Agis Health Database - Corrected Proof

Hugo M. Smeets, Niek J. de Wit, Arno W. Hoes — 2010-09-01

Abstract: Objective: Observational studies performed within routine health care databases have the advantage of their large size and, when the aim is to assess the effect of interventions, can offer a completion to randomized controlled trials with usually small samples from experimental situations. Institutional Health Insurance Databases (HIDs) are attractive for research because of their large size, their longitudinal perspective, and their practice-based information. As they are based on financial reimbursement, the information is generally reliable.Study design: The database of one of the major insurance companies in the Netherlands, the Agis Health Database (AHD), is described in detail. Whether the AHD data sets meet the specific requirements to conduct several types of clinical studies is discussed according to the classification of the four different types of clinical research; that is, diagnostic, etiologic, prognostic, and intervention research. The potential of the AHD for these various types of research is illustrated using examples of studies recently conducted in the AHD.Conclusion: HIDs such as the AHD offer large potential for several types of clinical research, in particular etiologic and intervention studies, but at present the lack of detailed clinical information is an important limitation.

Prospectively screening for eligible patients was inaccurate in predicting patient recruitment of orthopedic randomized trials - Corrected Proof

2010-08-30

Abstract: Objective: To compare the accuracy of estimates of potential recruitment from a prospective 8-week screening study compared with a retrospective chart review across sites participating in two fracture management trials.Study Design and Setting: During the planning phase of two large, multicenter, randomized controlled fracture management trials, 74 clinical sites provided estimates of the annual recruitment rate both retrospectively (based on chart reviews) and prospectively. The prospective estimate was generated by screening, for 8 weeks, all incoming patients for eligibility in the concerning trial, without actually enrolling any patient. We compared these prospective and retrospective estimates with one another (for 74 sites in the two trials) and with actual 1-year recruitment rates in the definitive trial (for nine sites in one trial).Results: There was a median difference of four patients (interquartile range: −14 to 18 patients; P=0.89) between a center's prospective estimate and its retrospective estimate. Both predictions were overestimations of recruitment in the definitive trial; only 31% (95% confidence interval [CI]: 28, 35) of retrospectively estimated patients, and 31% (95% CI: 27, 35) of prospectively estimated patients were recruited in the definitive trials.Conclusion: Compared with relatively simple chart reviews, prospectively screening for eligible patients at clinical sites, which is associated with substantial costs, did not result in more accurate predictions of accrual in large, multicenter, randomized controlled trials.

Published norms underestimate the health-related quality of life among persons with type 2 diabetes - Corrected Proof

2010-08-30

Abstract: Objective: To assess health-related quality of life (HRQL) among adults with type 2 diabetes using the Short Form (SF)-36 and to obtain pooled estimates of HRQL for subpopulations defined by demographic characteristics, diabetes-related complications, and comorbidities.Study Design and Methods: We conducted computerized searches of multiple electronic bibliographic databases, and studies in any language were selected in which HRQL was reported among adults with type 2 diabetes using the SF-36. Estimates were combined using a random-effects model.Results: One hundred eighteen studies fulfilled the inclusion criteria. HRQL was lower in persons with type 2 diabetes, as measured by all the eight component scores of the SF-36 when compared with the existing U.S. population norms and with previously published type 2 diabetes norms. SF-36 component and summary scores were extremely heterogeneous, and subpopulation data were sparse; this precluded obtaining meaningful pooled scores for most populations of interest and made comparisons among subpopulations difficult.Conclusion: Our data suggest that previously published norms may underestimate the effect of diabetes on HRQL, and diabetes populations are extremely heterogeneous, making broad population “norms” for HRQL in type 2 diabetes of limited use. Additional research with important subpopulations and individual-level data are needed to further explore the effect of diabetes on HRQL.

Bayesian statistical method was underused despite its advantages in the assessment of implantable medical devices - Corrected Proof

Leslie Pibouleau, Sylvie Chevret — 2010-08-30

Abstract: Objective: We sought to review the use of Bayesian methods in the evaluation of the effectiveness of implantable medical devices (IMDs) to identify which areas of research need to be further investigated to improve IMD assessment.Study Design and Setting: Relevant studies were identified by searching PubMed and the Food and Drug Administration Web site. Data were extracted independently by the two authors. The quality of Bayesian analysis reporting was summarized using the ROBUST (Reporting Of Bayes Used in clinical STudies) checklist.Results: Seventeen studies met the inclusion criteria; five published meta-analyses and 12 clinical studies were reported as FDA summaries of safety and effectiveness. Reporting of data was of high quality in meta-analyses, whereas it was of poor quality in clinical studies. Bayesian methods were used in meta-analyses to model study heterogeneity. In clinical studies, the objectives of the Bayesian analyses were mostly to address the question of equivalence and to use surrogate outcome predictors. Prior information, when reported, was less informative. Clinical data external to the trial itself and expert opinions were never used to elicit prior information.Conclusion: Our review highlighted the underuse of Bayesian methods in IMD assessment. The major challenge is to provide to clinical researchers a framework that helps them use external evidence to elicit prior distributions.

Future disability projections could be improved by connecting to the theory of a dynamic equilibrium - Corrected Proof

Bart Klijs, Johan P. Mackenbach, Anton E Kunst — 2010-08-30

Abstract: Objective: Projections of future trends in the burden of disability could be guided by models linking disability to life expectancy, such as the dynamic equilibrium theory. This article tests the key assumption of this theory that severe disability is associated with proximity to death, whereas mild disability is not.Study Design and Setting: Using data from the GLOBE study (Gezondheid en Levensomstandigheden Bevolking Eindhoven en omstreken), the association of three levels of self-reported disabilities in activities of daily living with age and proximity to death was studied using logistic regression models. Regression estimates were used to estimate the number of life years with disability for life spans of 75 and 85 years.Results: Odds ratios of 0.976 (not significant) for mild disability, 1.137 for moderate disability, and 1.231 for severe disability showed a stronger effect of proximity to death for more severe levels of disability. A 10-year increase of life span was estimated to result in a substantial expansion of mild disability (4.6 years) compared with a small expansion of moderate (0.7 years) and severe (0.9 years) disability.Conclusion: These findings support the theory of a dynamic equilibrium. Projections of the future burden of disability could be substantially improved by connecting to this theory and incorporating information on proximity to death.

Covariate adjustment in RCTs results in increased power to detect conditional effects compared with the power to detect unadjusted or marginal effects - Corrected Proof

2010-08-30

We thank Dr Lingsma et al. for their letter, which expands on our recently published review of how baseline covariates are treated in reports of randomized controlled trials (RCTs). Their letter represents an important component of the discussion that we hoped to stimulate regarding the advantages of adjusted vs. unadjusted analyses in the reports of RCTs.

Predictive value of Medicare claims data for identifying revision of index hip replacement was modest - Corrected Proof

2010-08-30

Abstract: Objective: To determine the positive predictive value of Medicare claims for identifying revision of total hip replacement (THR), a frequent marker of THR quality and outcome.Study Design and Setting: We obtained Medicare Part A (Hospital) claims from seven states on patients that had primary THR from July 1995 through June 1996. We searched claims to determine whether these THR recipients had a subsequent revision THR through December 2006. We selected a sample of subjects with codes indicating both index primary and subsequent revision THR. We obtained medical records for both procedures to establish whether the revision occurred on the same side as index primary THR.Results: Three hundred seventy-four subjects had codes indicating primary THR in 1995–96 and subsequent revision. Seventy-one percent (95% confidence interval: 66, 76) of the revisions were performed on the index joint and would be correctly attributed as revisions of the index THR, using Medicare claims data.Conclusion: Claims data on revision THR that do not contain information on the side that was operated on are ambiguous with respect to whether the revision was performed on the index or contralateral side. Claims-based analyses of revisions after an index THR should acknowledge and adjust for this source of potential misclassification.

N-of-1 clinical trials should be incorporated into clinical practice - Corrected Proof

Jesse A. Berlin — 2010-08-30

Abstract: N-of-1 clinical trials have the potential to contribute to individual patient management and to the accrual of important information about populations. Incorporating these studies into clinical practice will require creative thinking so as to maintain rigor without excessive disruption of routine care.

The rhythm of chronic obstructive pulmonary disease exacerbations - Corrected Proof

John R. Hurst — 2010-08-30

Chronic obstructive pulmonary disease (COPD) is a global, prevalent, underdiagnosed, and neglected condition characterized by progressive respiratory symptoms, airflow obstruction, limitation in activity, and an inflammatory response of the lung to noxious insult . The natural history of COPD is punctuated by acute deteriorations in respiratory health termed “exacerbations,” and these exacerbations are responsible for much of the disease burden . Most exacerbations are caused by airway infection , and many effective interventions are now available to reduce exacerbation occurrence. But which patients should receive these? Using therapy widely is costly and risks overtreating some patients. Targeting therapy is, therefore, of advantage both to patients and health services. Predicting when and in whom exacerbations will occur is of considerable importance.

A threshold regression model for recurrent exacerbations in chronic obstructive pulmonary disease - Corrected Proof

S.D. Aaron, T. Ramsay, K. Vandemheen, G.A. Whitmore — 2010-08-30

Abstract: Objective: Respiratory exacerbations are a major source of morbidity in patients with chronic obstructive pulmonary disease (COPD). In this article, we model COPD health status as a formal stochastic process. A successful model will provide a suitable statistical structure for analysis of the effects of medical interventions on a patient's health status, and, possibly, offer new insights into the underlying disease process.Study Design and Setting: Our approach uses a regression methodology for time-to-event data called threshold regression (TR). We test the methodology on COPD data from a randomized clinical trial. Two TR models are studied: one based on a Poisson process and the other, a Wiener diffusion process.Results: Both models provide reasonably accurate fits to the clinical trial data. The insights offered by the fitted models are interpreted. Analysis of the clinical trial data set using these TR models revealed that patients who experienced multiple exacerbations showed a progressive acceleration in rate of exacerbations, and successive shortening of stable intervals between exacerbations.Conclusion: TR techniques allow for realistic modeling of the COPD health state. A hybrid Poisson/Wiener diffusion TR model that incorporates the causal determinants of disease operating in each patient may be preferable.

Weighted index explained more variance in physical function than an additively scored functional comorbidity scale - Corrected Proof

Linda Resnik, Pedro Gozalo, Dennis L. Hart — 2010-08-18

Abstract: Objective: 1) examine association between the Functional Comorbidity Index (FCI) and discharge functional status (FS); 2) examine impact of FCI on FS when added to comprehensive models; and 3) compare additive FCI with weighted FCI and list of condition variables (list).Study Design and Setting: Patients were drawn from Focus On Therapeutic Outcomes, Inc. (FOTO) database (1/1/06–12/31/07). FS collected using computer adaptive tests. Linear regression examined association between FCI and FS. Three methods of including functional comorbidities (FC) were compared.Results: Relationship between FCI and FS varied by group (range, 0.02–0.9). Models with weighted index or list had similar R2. Weighted FCI or list increased R2 of crude models by <0.01 for cervical, shoulder, and lumbar; by 0.01 for wrist/hand, knee, and foot/ankle; by 0.02 for hip; by 0.03 for elbow; and by 0.08 for neurological. Addition of FCI to comprehensive models added <0.01 to R2 (all groups). Weighted FCI increased R2 by <0.01 for cervical, lumbar, and shoulder; by 0.01 for wrist/hand, hip, knee, and foot/ankle; by 0.02 for elbow; and by 0.04 for neurological; whereas list increased R2 by <0.01 for cervical, shoulder, and lumbar; by 0.01 for knee and foot/ankle; by 0.02 for elbow, wrist/hand, and hip; and by 0.05 for neurological.Conclusion: List of comorbidities or weighted FCI is preferable to using additive FCI.

Standing on the promises: First wave validation reports of the Patient-Reported Outcome Measurement Information System - Corrected Proof

Maarten Boers — 2010-08-18

This issue of the journal features three articles that form the core of the so-called first wave testing of the PROMIS initiative . PROMIS, or “Patient-Reported Outcome Measurement Information System” in full, is an initiative funded by the National Institutes of Health to create “a national resource for precise and efficient measurement of patient-reported symptoms, functioning, and health-related quality of life, appropriate for patients with a wide variety of chronic disease conditions” .

Estimating prevalence of diabetes in a Congolese town was feasible - Corrected Proof

2010-08-13

Abstract: Objectives: To study prevalence, determinants, and complications at diagnosis of diabetes and intermediate hyperglycemia (IH) in Kisantu, a semirural town in Bas-Congo province, Democratic Republic of Congo.Study Design and Setting: A large-scale analytical cross-sectional population-based survey was performed in 2007 in Kisantu. After extensive sensitization, the study sample was collected using a modified World Health Organization (WHO) STEPwise strategy, taking subsequently a random sample of streets, households within streets, and inhabitants aged 20 years and older within households. After informed consent, subjects were invited to fixed sites for interview, anthropometry, clinical examination (blood pressure, monofilament, and ophthalmology), and biochemical tests (fasting capillary glucose, serum creatinine, and albuminuria). Fasting glycemia was repeated or 2-hour postload glycemia was measured the next day in subjects with an initial glycemia of 126–199mg/dL (7.0–11.1mmol/L) or 100–125mg/dL (5.6–6.9mmol/L), respectively. Hence, prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance according to both 2006 WHO/International Diabetes Federation and 2003 American Diabetes Association criteria could be evaluated. Bivariate and multivariate analyses were used for statistical analyses.Results: Response rate was 93.7% (1,898 of 2,025). Complete data were available in 1,866 (92.1%) subjects.Conclusion: Estimating the prevalence of diabetes and IH in a small Congolese town was proven to be feasible.

Systematic review highlights difficulty interpreting diverse clinical outcomes in abnormal uterine bleeding trials - Corrected Proof

2010-08-13

Abstract: Objectives: (1) To systematically collect and organize into clinical categories all outcomes reported in trials for abnormal uterine bleeding (AUB); (2) to rank the importance of outcomes for patient decision making; and (3) to improve future comparisons of effects in trials of AUB interventions.Study Design and Setting: Systematic review of English-language randomized controlled trials of AUB treatments in MEDLINE from 1950 to June 2008. All outcomes and definitions were extracted and organized into major outcome categories by an expert group. Each outcome was ranked “critically important,” “important,” or “not important” for informing patients' choices.Results: One hundred thirteen articles from 79 trials met the criteria. One hundred fourteen different outcomes were identified, only 15 (13%) of which were ranked as critically important and 29 (25%) as important. Outcomes were grouped into eight categories: (1) bleeding; (2) quality of life; (3) pain; (4) sexual health; (5) patient satisfaction; (6) bulk-related complaints; (7) need for subsequent surgical treatment; and (8) adverse events.Conclusion: To improve the quality, consistency, and utility of future AUB trials, we recommend assessing a limited number of clinical outcomes for bleeding, disease-specific quality of life, pain, sexual health, and bulk-related symptoms both before and after treatment and reporting satisfaction and adverse events. Further development of validated patient-based outcome measures and the standardization of outcome reporting are needed.

Reviews assessing the quality or the reporting of randomized controlled trials are increasing over time but raised questions about how quality is assessed - Corrected Proof

2010-08-13

Abstract: Objective: Many reviews specifically aimed to assess the quality of randomized controlled trials (RCTs). We evaluated the quality of reporting in such reviews.Study Design and Setting: PubMed and the Cochrane library were searched for all reviews assessing the quality of RCTs between 1987 and 2007, and experts in the field were also contacted.Results: We found 177 reviews published from 1987 to 2007, 58% of which were published after 2002. Of these, 131 (74%) focused on the quality of RCTs, 44 (25%) on quality of reporting, and 2 (1%) assessed both. The search strategy was well reported (92%). The criteria for assessment were reported in 97% of the reviews but were defined in only 38%. Seventy-four different items and 26 different scales were identified. Allocation sequence generation and concealment were reported in 41% and 40%, respectively, but their adequacy was assessed in 20% and 29%, respectively; scales were used in 40% and Consolidated Standards of Reporting Trials (CONSORT) checklist in 12%.Conclusion: The number of methodological reviews has dramatically increased in recent years. Despite an improved reporting of the methodology, how quality is assessed still raises important issues. Heterogeneity of criteria used and lack of definition may limit the relevance of these reviews.

Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial - Corrected Proof

Georgia Salanti, A.E. Ades, John P.A. Ioannidis — 2010-08-06

Abstract: Objective: To present some simple graphical and quantitative ways to assist interpretation and improve presentation of results from multiple-treatment meta-analysis (MTM).Study Design and Setting: We reanalyze a published network of trials comparing various antiplatelet interventions regarding the incidence of serious vascular events using Bayesian approaches for random effects MTM, and we explore the advantages and drawbacks of various traditional and new forms of quantitative displays and graphical presentations of results.Results: We present the results under various forms, conventionally based on the mean of the distribution of the effect sizes; based on predictions; based on ranking probabilities; and finally, based on probabilities to be within an acceptable range from a reference. We show how to obtain and present results on ranking of all treatments and how to appraise the overall ranks.Conclusions: Bayesian methodology offers a multitude of ways to present results from MTM models, as it enables a natural and easy estimation of all measures based on probabilities, ranks, or predictions.

Representativeness of the Patient-Reported Outcomes Measurement Information System Internet panel - Corrected Proof

2010-08-06

Abstract: Objectives: To evaluate the Patient-Reported Outcomes Measurement Information System (PROMIS), which collected data from an Internet polling panel, and to compare PROMIS with national norms.Study Design and Setting: We compared demographics and self-rated health of the PROMIS general Internet sample (N=11,796) and one of its subsamples (n=2,196) selected to approximate the joint distribution of demographics from the 2000 U.S. Census, with three national surveys and U.S. Census data. The comparisons were conducted using equivalence testing with weights created for PROMIS by raking.Results: The weighted PROMIS population and subsample had similar demographics compared with the 2000 U.S. Census, except that the subsample had a higher percentage of people with higher education than high school. Equivalence testing shows similarity between PROMIS general population and national norms with regard to body mass index, EQ-5D health index (EuroQol group defined descriptive system of health-related quality of life states consisting of five dimensions including mobility, self-care, usual activities, pain/discomfort, anxiety/depression), and self-rating of general health.Conclusion: Self-rated health of the PROMIS general population is similar to that of existing samples from the general U.S. population. The weighted PROMIS general population is more comparable to national norms than the unweighted population with regard to subject characteristics. The findings suggest that the representativeness of the Internet data is comparable to those from probability-based general population samples.

Relative to the general US population, chronic diseases are associated with poorer health-related quality of life as measured by the Patient-Reported Outcomes Measurement Information System (PROMIS) - Corrected Proof

2010-08-06

Abstract: Objectives: The Patient-Reported Outcomes Measurement Information System (PROMIS) allows assessment of the impact of chronic conditions on health-related quality of life (HRQL) across diseases. We report on the HRQL impact of individual and comorbid conditions as well as conditions that are described as limiting activity.Study Design and Setting: Data were collected through online and clinic recruitment as part of the PROMIS item calibration sample (n=21,133). Participants reported the presence or absence of 24 chronic health conditions and whether their activity was limited by each condition.Results: Across health status domains, the presence of a chronic condition was associated with poorer scores than those without a diagnosis, particularly for those individuals who reported that their condition was disabling. The magnitude of detriment in HRQL was more pronounced for individuals with two or more chronic conditions and could not be explained by sociodemographic factors. Patterns of HRQL deficits varied across disease and comorbidity status.Conclusion: The impact of chronic conditions, particularly when experienced with comorbid disease, is associated with detriments in HRQL. The negative impact on HRQL varies across symptoms and functional areas within a given condition.

Five preference-based indexes in cataract and heart failure patients were not equally responsive to change - Corrected Proof

2010-08-05

Abstract: Objective: To compare the responsiveness to clinical change of five widely used preference-based health-related quality-of-life indexes in two longitudinal cohorts.Study Design and Setting: Five generic instruments were simultaneously administered to 376 adults undergoing cataract surgery and 160 adults in heart failure management programs. Patients were assessed at baseline and reevaluated after 1 and 6 months. The measures were the Short Form (SF)-6D (based on responses scored from SF-36v2), Self-Administered Quality of Well-being Scale (QWB-SA), the EuroQol-5D developed by the EuroQol Group, the Health Utilities Indexes Mark 2 (HUI2) and Mark 3 (HUI3). Cataract patients completed the National Eye Institute Visual Functioning Questionnaire-25, and heart failure patients completed the Minnesota Living with Heart Failure Questionnaire. Responsiveness was estimated by the standardized response mean.Results: For cataract patients, mean changes between baseline and 1-month follow-up for the generic indices ranged from 0.00 (SF-6D) to 0.052 (HUI3) and were statistically significant for all indexes except the SF-6D. For heart failure patients, only the SF-6D showed significant change from baseline to 1 month, whereas only the QWB-SA change was significant between 1 and 6 months.Conclusions: Preference-based methods for measuring health outcomes are not equally responsive to change.

The Patient Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008 - Corrected Proof

2010-08-05

Abstract: Objectives: Patient-reported outcomes (PROs) are essential when evaluating many new treatments in health care; yet, current measures have been limited by a lack of precision, standardization, and comparability of scores across studies and diseases. The Patient-Reported Outcomes Measurement Information System (PROMIS) provides item banks that offer the potential for efficient (minimizes item number without compromising reliability), flexible (enables optional use of interchangeable items), and precise (has minimal error in estimate) measurement of commonly studied PROs. We report results from the first large-scale testing of PROMIS items.Study Design and Setting: Fourteen item pools were tested in the U.S. general population and clinical groups using an online panel and clinic recruitment. A scale-setting subsample was created reflecting demographics proportional to the 2000 U.S. census.Results: Using item-response theory (graded response model), 11 item banks were calibrated on a sample of 21,133, measuring components of self-reported physical, mental, and social health, along with a 10-item Global Health Scale. Short forms from each bank were developed and compared with the overall bank and with other well-validated and widely accepted (“legacy”) measures. All item banks demonstrated good reliability across most of the score distributions. Construct validity was supported by moderate to strong correlations with legacy measures.Conclusion: PROMIS item banks and their short forms provide evidence that they are reliable and precise measures of generic symptoms and functional reports comparable to legacy instruments. Further testing will continue to validate and test PROMIS items and banks in diverse clinical populations.

AGREE II: Advancing guideline development, reporting and evaluation in health care - Corrected Proof

2010-07-26

Strengthening the reliability and credibility of observational epidemiology studies by creating an Observational Studies Register - Corrected Proof

Gerard M.H. Swaen, Neil Carmichael, John Doe — 2010-07-21

Abstract: Objective: To evaluate the need for the creation of a system in which observational epidemiology studies are registered; an Observational Studies Register (OSR).Study Design and Setting: The current scientific process for observational epidemiology studies is described. Next, a parallel is made with the clinical trials area, where the creation of clinical trial registers has greatly restored and improved their credibility and reliability. Next, the advantages and disadvantages of an OSR are compared.Results: The advantages of an OSR outweigh its disadvantages.Conclusion: The creation of an OSR, similar to the existing Clinical Trials Registers, will improve the assessment of publication bias and will provide an opportunity to compare the original study protocol with the results reported in the publication. Reliability, credibility, and transparency of observational epidemiology studies are strengthened by the creation of an OSR. We propose a structured, collaborative, and coordinated approach for observational epidemiology studies that can provide solutions for existing weaknesses and will strengthen credibility and reliability, similar to the approach currently used in clinical trials, where Clinical Trials Registers have played a key role in strengthening their scientific value.

Critical appraisal of health literacy indices revealed variable underlying constructs, narrow content and psychometric weaknesses - Corrected Proof

Joanne E. Jordan, Richard H. Osborne, Rachelle Buchbinder — 2010-07-20

Abstract: Objective: Health literacy refers to an individual's ability to seek, understand, and use health information. A range of indices exist that purport to measure health literacy across individuals and populations. This study aimed to review the development and content of existing indices and to critically appraise their properties.Study Design and Setting: Using standardized search terms, published generic health literacy indices (1990–2008) were identified. Using a methodological framework, each was evaluated for purpose, validity (face, content, construct), reliability, responsiveness, feasibility, and generalizability.Results: Nineteen instruments were evaluated. Three measurement approaches were identified: direct testing of individual abilities, self-report of abilities, and population-based proxy measures. Composition of underlying constructs and content varied widely across instruments, and none appeared to fully measure a person's ability to seek, understand, and use health information. The content was focused primarily on reading comprehension and numeracy; scoring categories were poorly defined and may not be mutually exclusive, and few indices had been assessed for reliability.Conclusion: Health literacy is not consistently measured, making it difficult to interpret and compare health literacy at individual and population levels. Empirical evidence demonstrating validity and reliability of existing indices is required, and more comprehensive health literacy instruments need to be developed.

Diabetics can be identified in an electronic medical record using laboratory tests and prescriptions - Corrected Proof

Karen Tu, Doug Manuel, Kelvin Lam, Doug Kavanagh, Tezeta F. Mitiku, Helen Guo — 2010-07-20

Abstract: Objective: With the increasing use of electronic medical records (EMRs) comes the potential to efficiently evaluate and improve quality of care. We set out to determine if diabetics could be accurately identified using structured data contained within an EMR.Study Design and Setting: We used a 5% random sample of adult patients (969 patients) within a convenience sample of 17 primary care physicians using Practices Solutions EMR in Ontario. A reference standard of diabetes status was manually confirmed by reviewing each patient's record. Accuracy for identifying people with diabetes was assessed using various combinations of laboratory tests and prescriptions. EMR data was also compared with administrative data.Results: A rule of one elevated blood sugar or a prescription for an antidiabetic medication had a 83.1% sensitivity, 98.2% specificity, 80.0% positive predictive value (PPV) and 98.5% negative predictive value (NPV) compared with the reference standard of diabetes status.Conclusion: We found that the use of structured data within an EMR could be used to identify patients with diabetes. Our results have positive implications for policy makers, researchers, and clinicians as they develop registries of diabetic patients to examine quality of care using EMR data.

Departures from community equipoise may lead to incorrect inference in randomized trials - Corrected Proof

Jeffrey N. Katz, John Wright, Bruce A. Levy, John A. Baron, Elena Losina — 2010-07-20

Abstract: Objective: To assess the impact of selective enrollment on the results of randomized controlled trials (RCTs).Study Design and Setting: We simulated an RCT of arthroscopic partial meniscectomy vs. nonoperative therapy in patients with meniscal tear and osteoarthritis (OA). We estimated efficacy with the risk ratio (RR) comparing the likelihood of clinically important improvement after surgery with that after nonoperative therapy. We assumed that efficacy differs by extent of OA. We simulated four scenarios: (1) nonselective enrollment; (2) higher likelihood of enrolling subjects with mild OA; (3) higher likelihood of enrolling subjects with severe OA; (4) much higher likelihood of enrolling subjects with severe OA. For each scenario, we simulated 100 trials with sample size 340.Results: With nonselective enrollment, reflecting community equipoise, the results in 100 trials were consistent with those in the underlying population (mean RR: 1.87; 95% confidence interval [95% CI]: 1.57, 2.14). Selective enrollment of subjects with much higher likelihood of severe OA resulted in 28% lower efficacy of surgery (mean RR: 1.34; 95% CI: 0.93, 2.15), with 95% CI containing the true efficacy in just 25% of trials and empirical power of 44%.Conclusion: Selective enrollment with respect to factors associated with efficacy may affect trial results and lead to inaccurate conclusions.

Meta-analyses of small numbers of trials often agree with longer-term results - Corrected Proof

Peter Herbison, Jean Hay-Smith, William J. Gillespie — 2010-07-07

Abstract: Objective: Many systematic reviews include only a few studies. It is unclear whether recommendations based on these will be correct in the longer term; hence, this article explores whether meta-analyses give reliable results after only a few studies.Study Design and Setting: Cumulative meta-analysis of data from 65 meta-analyses from 18 Cochrane systematic reviews was carried out. Various measures of closeness to the pooled estimate from all trials after three and five trials were included. Changes during the accumulation of evidence were noted.Results: The 95% confidence interval included the final estimate in 72% of meta-analyses after three studies and in 83% after five studies. It took a median of four (interquartile range: 1.25–6) studies to get within 10% of the final point estimate. Agreement between the results at three and five studies and the final estimate was not predicted by the number of participants, the number of events, τ2, or I2. Estimates could still change substantially after many trials were included.Conclusion: Many of the conclusions drawn from systematic reviews with small numbers of included studies will be correct in the long run, but it is not possible to predict which ones.

Osteoarthritis Index delivered by mobile phone (m-WOMAC) is valid, reliable, and responsive - Corrected Proof

2010-07-07

Abstract: Objectives: To evaluate the validity, reliability, responsiveness, and mode preference of electronic data capture (EDC) using the Western Ontario and McMaster (WOMAC) numerical rating scale (NRS) 3.1 Osteoarthritis (OA) Index on Motorola V3 mobile phones.Study Design and Setting: Patients with OA undergoing hip or knee joint replacement were assessed preoperatively and 3–4 months postoperatively, completing the WOMAC Index in paper (p-WOMAC) and electronic (m-WOMAC) format in random order.Results: Data were successfully and securely transmitted from patients in Australia to a server in the United States. Pearson correlations between the summated total index scores (TISs) for the p-WOMAC and m-WOMAC pre- and postsurgery were 0.98 and 0.99 (P<0.0001). There were no clinically important or statistically significant between-method differences in the adjusted total summated scores, pre- and postsurgery (adjusted mean differences=4.44, P=0.474 and 1.73, P=0.781, respectively). Internal consistency estimates of m-WOMAC reliability were 0.87–0.98. The m-WOMAC detected clinically important, statistically significant (P<0.0001) improvements in pain, stiffness, function, and TIS. No statistically significant differences in mode preference were detected.Conclusions: There was close agreement and no significant differences between m-WOMAC and p-WOMAC scores. This study confirms the validity, reliability, and responsiveness of the Exco InTouch-engineered, Java-based m-WOMAC Index application. EDC with the m-WOMAC Index provides unique opportunities for using quantitative measurement in clinical research and practice.

Appraising a tool for guideline appraisal (the AGREE II instrument) - Corrected Proof

Antonio L. Dans, Leonila F. Dans — 2010-07-07

The AGREE II instrument has 3 goals – to assess the quality of clinical practice guidelines (CPGs), to provide a methodologic strategy for the development of guidelines, and to recommend how and what information should be reported in guidelines. Of these, the main purpose seems to be to assess guideline quality (i.e., “confidence that the potential biases of guideline development have been addressed adequately and that the recommendations are both internally and externally valid, and are feasible for practice”) . There is no doubt that this instrument is badly needed by health care providers and policy-makers around the world, who deal with a deluge of clinical practice guidelines in their day-to-day decisions.

Assessing the impact of attrition in randomized controlled trials - Corrected Proof

2010-06-23

Abstract: Objectives: A survey of randomized controlled trials found that almost a quarter of trials had more than 10% of responses missing for the primary outcome. There are a number of ways in which data could be missing: the subject is unable to provide it, or they withdraw, or become lost to follow-up. Such attrition means that balance in baseline characteristics for those randomized may not be maintained in the subsample who has outcome data. For individual trials, if the attrition is systematic and linked to outcome, then this will result in biased estimates of the overall effect. It then follows that if such trials are combined in a meta-analysis, it will result in a biased estimate of the overall effect and be misleading. The aim of this study was to investigate the impact of attrition on baseline imbalance within individual trials and across multiple trials.Study Design and Setting: In this article, we used individual patient data from a convenience sample of 10 trials evaluating interventions for the treatment of musculoskeletal disorders. Meta-analyses using the mean difference at baseline between the trial arms were carried out using individual patient data from these trials. The analyses were first carried out using all randomized participants and secondly only including participants with outcome data on the quality-of-life score. Meta-regression was carried out to evaluate whether the level of baseline imbalance was associated with the level of attrition.Results: The overall attrition rates for the quality-of-life score ranged between 4% and 28% of the total randomized patients. All trials showed some level of differential attrition between the treatment arms, ranging from 1% to 14%. Attrition within the control group ranged from 3% to 25% and within the intervention group, it ranged from 0% to 31%. For individual trials, there was no indication that attrition altered the results in favor of either the treatment or the control. Forest plots highlighted that the attrition had some impact on the baseline imbalance for the primary outcome score as more heterogeneity was introduced (I-squared value of 0.4% for the initial data set vs. I-squared value of 16.9% for the analyzed data set). However, the standardized mean difference increased only slightly (from 0.01 to 0.03 with 95% confidence interval [CI]: −0.05, 0.10). Meta-regression showed little or no evidence of a significant dose–response relationship between the level of attrition and the baseline imbalance (coefficient 0.73, 95% CI: −0.81, 2.28).Conclusion: Although, in theory, attrition can introduce selection bias in randomized trials, we did not find sufficient evidence to support this claim in our convenience sample of trials. However, the number of trials included was relatively small, which may have led to small but important differences in outcomes being missed. In addition, only 2 of 10 trials included had attrition levels greater than 15% suggesting a low level of potential bias. Meta-analyses and systematic reviews should always consider the impact of attrition on baseline imbalances and where possible any baseline imbalances in the analyzed data set and their impact on the outcomes reported.

A review of critical appraisal tools show they lack rigor: Alternative tool structure is proposed - Corrected Proof

Michael Crowe, Lorraine Sheppard — 2010-06-21

Abstract: Objectives: To evaluate critical appraisal tools (CATs) that have been through a peer-reviewed development process with the aim of analyzing well-designed, documented, and researched CATs that could be used to develop a comprehensive CAT.Study Design and Setting: A critical review of the development of CATs was undertaken.Results: Of the 44 CATs reviewed, 25 (57%) were applicable to more than one research design, 11 (25%) to true experimental studies, and the remaining 8 (18%) to individual research designs. Comprehensive explanation of how a CAT was developed and guidelines to use the CAT were available in five (11%) instances. There was no validation process reported in 11 CATs (25%) and 33 CATs (77%) had not been reliability tested. The questions and statements that made up each CAT were coded into 8 categories and 22 items such that each item was distinct from every other.Conclusions: CATs are being developed while ignoring basic research techniques, the evidence available for design, and comprehensive validation and reliability testing. The basic structure for a comprehensive CAT is suggested that requires further study to verify its overall usefulness. Meanwhile, users of CATs should be careful about which CAT they use and how they use it.

Item nonresponse to psychosocial questionnaires was associated with higher mortality after acute myocardial infarction - Corrected Proof

Elisa Candido, Paul Kurdyak, David A. Alter — 2010-06-21

Abstract: Objective: To examine the relationship between selective nonresponse to a psychosocial questionnaire and mortality after acute myocardial infarction (AMI).Study Design and Setting: Two thousand six hundred and ninety AMI survivors after AMI hospitalization were recruited to complete a 30-day follow-up interview. Patients were classified into four groups (survey nonparticipation and complete, partial, and no item nonresponse) according to their degree of response to the Medical Outcomes Study (MOS) Social Support Survey (MOS-SSS). Cox proportional hazard models, adjusted for baseline sociodemographic, clinical, and psychosocial (i.e., social isolation) characteristics, were used to examine all-cause mortality, 3 years post-AMI, across the response levels.Results: 13.9% of the eligible patients refused follow-up participation; MOS-SSS item nonresponse was present in up to 14.7% of participants and was more frequent among the elderly, socially disadvantaged, and those with higher clinical risk. A nonresponse mortality gradient existed, ranging from 8.9% (no item nonresponse) to 18.7% (complete item nonresponse) (P<0.001). After adjusting for baseline characteristics, complete item nonresponse remained significantly associated with mortality (hazard ratio: 1.33; 95% confidence interval: 1.02–1.73).Conclusions: Item nonresponse to a social support questionnaire is associated with higher mortality post-AMI. Although explanatory factors may include age and baseline clinical risk, additional psychosocial and/or unmeasured factors may account for the poorer prognosis.

Five-point scales outperform 10-point scales in a randomized comparison of item scaling for the Patient Experiences Questionnaire - Corrected Proof

Andrew M. Garratt, Jon Helgeland, Pål Gulbrandsen — 2010-06-21

Abstract: Objective: To assess the data quality of two approaches to scaling items within the Patient Experiences Questionnaire (PEQ); a five-point scale with descriptors for all scale points and a 10-point scale with descriptors only at the end points.Study Design and Setting: The two versions were pretested through cognitive interviews with 14 patients. The PEQ was then mailed to 1,000 patients after inpatient treatment at a large university hospital in Norway, randomized to receive the 5- or 10-point scale versions. Response rates, missing data, item means, floor, and ceiling effects were assessed. Regression analysis was used to examine the impact of response scale on missing data, floor, and ceiling effects after controlling for age, education level, and health status.Results: The five-point scale produced data with unimodal and fairly symmetric distributions in contrast to the highly skewed J- and U-shaped distributions for the 10-point scale. The five-point scale data had significantly lower item means, floor, and ceiling effects. Regression analysis showed that the type of scale explained a significant component of the variation in both floor and ceiling effects.Conclusion: The five-point scale performed better than the 10-point scale and is more suitable for assessing patient experiences. The revised PEQ will be used in Norwegian national surveys.

Pre-notification did not increase response rate in addition to follow-up: a randomized trial - Corrected Proof

Alice Hammink, Paul Giesen, Michel Wensing — 2010-06-21

Abstract: Objective: Although there is growing evidence on the effectiveness of pre-notification and follow-up on response rates in patient surveys, no studies report the effectiveness of pre-notification in addition to follow-up. The aim of this study was to determine the effect of a pre-notification by mail on the response rate in a patient survey with follow-up, compared with follow-up or pre-notification only.Study Design and Setting: Randomized trial that compared (1) a combination of pre-notification and follow-up with (2) pre-notification only and (3) follow-up only. The trial was integrated in a survey study among patients, which measured their experiences with general practice cooperatives for out-of-hour care.Results: Of the total number of 880 patients who received the questionnaire, 45% returned it. No significant effect was found of the combination of pre-notification and follow-up compared with the two other arms of the trial.Conclusion: Adding pre-notification to follow-up in a patient survey had no additional effect on the response rate. This finding must be interpreted with respect to the questionnaire and the study population.

Guidelines for Reporting Reliability and Agreement Studies (GRRAS) were proposed - Corrected Proof

2010-06-18

Abstract: Objective: Results of reliability and agreement studies are intended to provide information about the amount of error inherent in any diagnosis, score, or measurement. The level of reliability and agreement among users of scales, instruments, or classifications is widely unknown. Therefore, there is a need for rigorously conducted interrater and intrarater reliability and agreement studies. Information about sample selection, study design, and statistical analysis is often incomplete. Because of inadequate reporting, interpretation and synthesis of study results are often difficult. Widely accepted criteria, standards, or guidelines for reporting reliability and agreement in the health care and medical field are lacking. The objective was to develop guidelines for reporting reliability and agreement studies.Study Design and Setting: Eight experts in reliability and agreement investigation developed guidelines for reporting.Results: Fifteen issues that should be addressed when reliability and agreement are reported are proposed. The issues correspond to the headings usually used in publications.Conclusion: The proposed guidelines intend to improve the quality of reporting.

Application of Rasch analysis in health care is increasing and is applied for variable reasons in mobility instruments - Corrected Proof

Shane L. Belvedere, Natalie A. de Morton — 2010-06-18

Abstract: Objective: To identify the frequency of Rasch analysis use in health instrument development or refinement and the characteristics of Rasch application in mobility scales.Study Design and Setting: The entire databases of Medline, CINAHL, PEDro, EMBASE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews were searched until January 2009. Articles that reported the development or refinement of health instruments using Rasch analysis were included. Of the 234 articles that met inclusion, 10 were categorized as “mobility” instruments. Data were extracted relating to each instrument and the use of Rasch analysis in the development or refinement of the instruments.Results: The number of articles reporting the use of Rasch analysis of health instruments is increasing, from 1 article in 1987 to 48 articles in 2007. Of the 10 mobility instruments examined, the primary reason Rasch was used varied. Reasons included assessing instrument unidimensionality, differential item functioning, rating categories, item hierarchy, and redundant items.Conclusion: The application of Rasch analysis in health instrument development has markedly increased in recent years. However, few mobility instruments have been developed or refined using Rasch analysis. The reasons that the Rasch model was used varied across mobility instruments.

The quality of safety reporting in trials is still suboptimal: Survey of major general medical journals - Corrected Proof

2010-06-18

Abstract: Objective: To evaluate whether the quality of reporting harms improved after the publication of the Extension of the Consolidated Standards of Reporting Trials (CONSORT) statement and predictors that influence the safety reporting in randomized controlled trials (RCTs)Study Design and Setting: Systematic survey of published RCTs assessing drugs. In MEDLINE, we identified 228 RCTs published in Annals of Internal Medicine, British Medical Journal, Journal of American Medical Association, The Lancet, and The New England Journal of Medicine in 2003 and 2006.Results: The reporting of harms have improved over time both in quality and extent of space. However, the mean score as an overall measure of adequacy in reporting harms was 0.58 in 2003 and increased to 0.67 in 2006, indicating a moderate safety reporting. Safety was more adequate in trials with statistically significant results for efficacy, private funding, primary harms outcome, and anti-infective, antineoplasmatic, or immunosuppressive agents.Conclusion: The use of the Extension of the CONSORT statement may be associated with improving the quality of safety reporting in RCTs, but there are still deficiencies that need to be corrected to use quantitative objective evidence for harms in performing meta-analyses and making therapeutic decisions.

The Pragmatic-Explanatory Continuum Indicator Summary (PRECIS) instrument was useful for refining a randomized trial design: Experiences from an investigative team - Corrected Proof

2010-06-18

Abstract: Objective: A recently published instrument (PRECIS) was designed to assist investigative teams in understanding the various design decisions that need to be made regarding pragmatic vs. explanatory trials. Our team used this instrument during an investigators' meeting to organize our discussion regarding the design of a planned trial and to determine the extent of consensus among the study investigators.Study Design and Setting: The study was descriptive in nature and occurred during an investigator meeting. After reading and reviewing the 10 PRECIS criteria, the team made quantitative judgments of the planned study regarding each PRECIS criteria to reflect initial, ideal, and final study design perceptions.Results: Data indicated that the final study design was more explanatory in nature than the preliminary plan. Evidence of consensus was obtained.Conclusion: The investigative team found that applying PRECIS principles were useful for (1) detailing points of discussion related to trial design, (2) making revisions to the design to be consistent with the project goals, and (3) achieving consensus. We believe our experiences with PRECIS may prove valuable for trial researchers in much the same way that case reports can provide valuable insights for clinicians.

Charlson and Rx-Risk comorbidity indices were predictive of mortality in the Australian health care setting - Corrected Proof

Christine Y. Lu, John Barratt, Agnes Vitry, Elizabeth Roughead — 2010-06-18

Abstract: Objective: To compare the performance of Charlson index and Rx-Risk score using data from Australian Department of Veterans' Affairs.Study Design and Setting: A study of older adults (N=94,714) who had both Charlson and Rx-Risk scores based on their hospital diagnoses and prescription medication dispensings during the baseline year (January 2005–December 2005). Predictive ability of 1-year and 3-year mortality was compared by Akaike information criterion model fit statistic and c statistic in logistic regression models. We also compared the scores for identifying specific medical conditions.Results: Both indices were significant predictors of all-cause mortality (P<0.0001). Of the population identified with a condition from either score, Rx-Risk score identified more than 95% of patients with gastric, respiratory, or cardiovascular condition, compared with Charlson index only identifying 2%, 31%, and 14%, respectively. The indices were comparable regarding diabetes. The Charlson index identified 83% of patients with dementia and 67% of those with cancers, whereas Rx-Risk score identified 38% and 43%, respectively.Conclusion: Both the Charlson and Rx-Risk scores predict mortality, but neither index identified all comorbidities. Based on data availability, preferences, and research purposes, investigators can use either Charlson index or Rx-Risk score to adjust for comorbidity.

Study design attributes influenced patients' willingness to participate in clinical research: a randomized vignette-based study - Corrected Proof

Thomas Agoritsas, Marie Deom, Thomas V. Perneger — 2010-06-17

Abstract: Objective: To identify characteristics of clinical research projects that influence patients' willingness to participate in research.Study Design and Setting: We surveyed all patients discharged during 1 month from a Swiss public teaching hospital. We described four hypothetical studies and asked patients whether they would agree to participate. We randomly manipulated three study attributes in each vignette, using a factorial design.Results: All studies were not equally acceptable to the 1,277 respondents. A higher willingness to participate was found when a new drug had no side effects, no additional visit was required, balanced information was given, results were in the public domain, and the project was approved by a research ethics committee. In contrast, destruction of blood samples at the end of the project, use of placebo controls, and random allocation to study arms were associated with a lower likelihood of participation. The origin of funds, financial reward, the need to complete a questionnaire, and clinical vs. economic purpose of the study did not influence willingness to participate.Conclusion: Patients valued safety, convenience, oversight, and open communication in research. However, they were put off by some aspects that are valued by health care professionals. Educating the public about research methods may improve participation.

Two prognostic indicators of the publication rate of clinical studies were available during ethical review - Corrected Proof

Jean Philippe de Jonga, Gerben Ter Rieta, Dick Ludolf Willemsa — 2010-06-17

Abstract: Objective: To identify prognostic indicators of the publication rate of clinical studies, available to research ethics committees (RECs) during review.Study Design and Setting: Retrospective survival study of a random sample of 100 studies, approved by a Dutch academic REC, with follow-up information by questionnaire and bibliographic searches. Multivariate Cox regression analysis of the association between publication rate and seven factors available during review: six study characteristics and the number of letters sent by the committee during review representing the length of the review process.Results: Two factors were associated with publication rate: studies with possible therapeutic benefit to participants were less likely to be published than nontherapeutic studies (adjusted hazard ratio [AHR]: 0.16; 95% confidence interval [CI]: 0.03–0.54); with every letter sent, publication was less likely (AHR: 0.46 per letter; 95% CI: 0.17–0.98). Possibly, studies with more-than-minimal burdens to participants were more likely to be published than studies with minimal burdens (AHR: 3.90, 95% CI: 1.03–16.64).Conclusion: We identified two prognostic indicators of publication rate. After suitable replication, RECs might explore using prognostic indicators, such as these, to target study protocols at high risk for nonpublication. Discussing the risk of nonpublication with investigators could help prevent nonpublication.

Subjective risk vs. objective risk can lead to different post-cesarean birth decisions based on multiattribute modeling - Corrected Proof

2010-06-17

Abstract: Objective: To compare birth recommendations for pregnant women with a prior cesarean produced from a decision model using absolute risks vs. one using subjective interpretation of the same risks: (1) a multiattribute decision model based on patient prioritization of risks (subjective risk) and (2) a hybrid model that used absolute risks (objective risk).Study Design and Setting: The subjective risk multiattribute model used the Analytic Hierarchy Process to elicit priorities for maternal risks, neonatal risks, and the delivery experience from 96 postnatal women with a prior cesarean. The hybrid model combined the priorities for delivery experience obtained in the first model with the unadjusted absolute risk values.Results: The multiattribute model generated more recommendations for repeat cesarean delivery than the hybrid model: 73% vs. 18%, (P-value <0.001). The multiattribute model favored repeat cesarean because women heavily prioritized avoiding any risk (even rare risk) to the infant. The hybrid model favored the trial of labor because of lower probabilities of risk to the mother and its high success rate of vaginal birth after cesarean.Conclusion: This study highlights the importance of patients and clinicians discussing the patient's priorities regarding the risks and other nonclinical considerations that may be important to her in the birthing decision.

Are randomized trials conducted in China or India biased? A comparative empirical analysis - Corrected Proof

Dalu Zhang, Nick Freemantle, K.K. Cheng — 2010-06-16

Abstract: Context: China and India are two emerging forces in undertaking randomized clinical trials. The quality of trials from these countries may affect not just their substantial populations but also their contribution to health policy throughout the world.Objective: The objectives of this study were to describe and contrast the quality and biases in reports of trials conducted in China and India with a set of “gold standard” trials reported in leading European and North American journals.Method: A systematic review and comparative empirical analysis of randomized controlled trial reports published in selected Chinese, Indian, and European or North American medical journals were performed. Quality was assessed against a subset of criteria from the CONSORT statement. We compared the rate of reporting of positive outcomes in clinical trials to describe potential bias.Result: In total, 307 Chinese papers, 117 Indian papers, and 304 Western papers were included. Reports of Indian trials were slightly better than Chinese papers on the trial reporting quality indicators and much better than Chinese papers on reporting patients' ethical issues. However, the gold standard Western trial reports scored considerably higher on all quality criteria. Chinese papers were substantially more likely to report statistically significant results (odds ratio [OR]=2.96, 95% confidence interval [CI]=2.23–3.94; P<0.0001). Indian trials reported a similar rate of positive results to Western papers (OR=0.92, 95% CI=0.69–1.24; P=0.59).Conclusion: Reporting of trials in major Chinese and Indian journals falls short of that achieved in the gold standard Western journals we appraised and may reflect underlying inadequacies in the design and conduct of these trials. Chinese trials appear biased and may selectively report positive outcomes while ignoring neutral or negative outcomes. Trialists and journal editors in China and India should adopt the CONSORT reporting guidelines, should ensure that a primary outcome is prespecified and reported, and should ensure that analysis is conducted according to the intention-to-treat principle. Ethical questions in the conduct of trials in China must be addressed.

A randomized trial of electronic reminders showed a reduction in the time to respond to postal questionnaires - Corrected Proof

Rebecca Ashby, Gwen Turner, Ben Cross, Natasha Mitchell, David Torgerson — 2010-06-16

Abstract: Objective: To assess the effect of electronic reminders (ERs) on response rate and time to response for the return of postal questionnaires.Study Design and Setting: This open randomized controlled trial (RCT) was conducted at the University of York. Participants who were taking part in an established RCT and who provided an electronic mail address and/or mobile telephone number were eligible to take part in the study. The intervention group received ERs on the day they were expected to receive postal questionnaires.Results: One hundred forty-eight participants (19 male and 129 female) aged 47±11 (range, 19–65) years were studied. About 89.2% of participants returned postal questionnaires. There was no difference in questionnaire response rates in control (64 of 74 [86.5%]) vs. intervention (68 of 74 [91.9%]), groups (relative risk=1.063, 95% confidence interval: 0.949–1.189). Median questionnaire time to response was 4 days less in the intervention group (10.0±0.2; range, 10–14 days) compared with the control group (14.0±1.4; range, 10–23 days) (χ21df=5.27, P=0.022).Conclusion: ERs are useful tools for reducing participant time to response for postal questionnaires. We found little evidence for an effect of ERs on response rate for postal questionnaires.

A self-report home environment screening tool identified older women at risk of falls - Corrected Proof

Afsoon Hassani Mehraban, Lynette A. Mackenzie, Julie E. Byles — 2010-06-16

Abstract: Objective: To evaluate a self-report version of the Home Falls and Accidents Screening Tool (HOME FAST-SR).Study Design and Setting: The HOME FAST-SR was designed using expert review, pretesting and piloting. Agreement of self-reported home hazard ratings and ratings by health professionals was evaluated using the kappa statistic. The HOME FAST-SR was validated in a cross-sectional survey of 568 older community-living Australian women using endorsement rates for HOME FAST-SR items and the association of scores with self-reported falls.Results: The 87-item HOME FAST-SR was constructed, and a scoring system was devised to calculate equivalent scores for the health professional version of the HOME FAST (HOME FAST-HP). Endorsement rates demonstrated that no items needed to be removed. Agreement between self-reported and professional ratings was moderate with therapists under-reporting hazards compared with older people. The mean HOME FAST score for the 568 cross-sectional participants was 9.39 (95% confidence interval: 9.1, 9.7) with a range from 2 to 23 out of a possible 25. Fallers had a significantly higher HOME FAST score (P=0.02).Conclusion: The HOME FAST-SR is a viable alternative to the HOME FAST-HP, and scores were associated with falls experienced by older women in a cross-sectional study.

Considerable differences exist between prevalent and incident myocardial infarction cohorts derived from the same population - Corrected Proof

2010-05-17

Abstract: Objective: Both prevalent and incident cohorts have been used in epidemiological and prognostic studies of ischemic heart disease (IHD). This study considers the differences between the cohort types.Study Design and Setting: Using linked primary care, secondary care, and death certification data, prevalent and incident cohorts of people with a first acute myocardial infarction (AMI) were formed from the same population. They were analyzed independently in terms of baseline characteristics and survival to revascularization, another AMI, or death.Results: 55.7% of the prevalent cohort members were males, with a mean age of 71.0 years (standard deviation [SD]: 12.0). 59.0% of the incident cohort members were males, with a mean age of 64.7 years (SD: 13.3). Over 5 years, a greater proportion of prevalent cases died from any cause (31.4% [95% confidence interval(CI): 28.6–34.3]) and IHD (18.5% [95% CI: 16.2–21.0]) than incident cases (18.0% [95% CI: 15.0–21.4] and 12.2% [95% CI: 9.7–15.2], respectively). Mean time to death was shorter in prevalent cases. There was a small difference in the numbers of subsequent AMIs between cohorts. In the incident cohort, mean time to AMI was shorter. Fewer prevalent cases underwent coronary artery bypass grafting or percutaneous transluminal coronary angioplasty.Conclusion: Considerable differences existed between the two cohorts in terms of baseline characteristics and prognosis. Incident cohorts derived from whole populations should be sought for estimation of survival.

The usefulness of administrative databases for identifying disease cohorts is increased with a multivariate model - Corrected Proof

2010-05-11

Abstract: Background: Administrative databases commonly use codes to indicate diagnoses. These codes alone are often inadequate to accurately identify patients with particular conditions. In this study, we determined whether we could quantify the probability that a person has a particular disease—in this case renal failure—using other routinely collected information available in an administrative data set. This would allow the accurate identification of a disease cohort in an administrative database.Methods: We determined whether patients in a randomly selected 100,000 hospitalizations had kidney disease (defined as two or more sequential serum creatinines or the single admission creatinine indicating a calculated glomerular filtration rate less than 60mL/min/1.73m2). The independent association of patient- and hospitalization-level variables with renal failure was measured using a multivariate logistic regression model in a random 50% sample of the patients. The model was validated in the remaining patients.Results: Twenty thousand seven hundred thirteen patients had kidney disease (20.7%). A diagnostic code of kidney disease was strongly associated with kidney disease (relative risk: 34.4), but the accuracy of the code was poor (sensitivity: 37.9%; specificity: 98.9%). Twenty-nine patient- and hospitalization-level variables entered the kidney disease model. This model had excellent discrimination (c-statistic: 90.1%) and accurately predicted the probability of true renal failure. The probability threshold that maximized sensitivity and specificity for the identification of true kidney disease was 21.3% (sensitivity: 80.0%; specificity: 82.2%).Conclusion: Multiple variables available in administrative databases can be combined to quantify the probability that a person has a particular disease. This process permits accurate identification of a disease cohort in an administrative database. These methods may be extended to other diagnoses or procedures and could both facilitate and clarify the use of administrative databases for research and quality improvement.

Five questions predicted long-term, severe, back-related functional limitations: evidence from three large prospective studies - Corrected Proof

2010-05-10

Abstract: Objectives: The objectives of the study were as follows: (1) to investigate whether the predictive validity of a previously developed back pain prediction rule could be improved; (2) to determine if the rule can be shortened without loss of predictive validity; (3) to compare the rule with the physician's judgment; (4) to assess, in a different population, its 2-year predictive validity; and (5) to evaluate the clinical applicability of the rule in a first-line care setting.Study Design and Setting: One thousand two hundred and sixty-two participants were enrolled in the study (participation: 91%) before a medical consultation for nonspecific back pain in a large emergency room and were followed up for 2 years (follow-up: 92.5%). The effects of adding new items and deleting any one of the original items were evaluated. The predictions by the rule and the physicians were compared with the 2-year actual functional limitations (measured with the Roland–Morris Disability Questionnaire).Results: Although the final prediction rule included only five items (feeling everything is an effort, trouble getting breath, hot/cold spells, numbness/tingling in parts of body, and pain in heart/chest), its predictive validity was greater than that of the original 17-item version and was superior to the physician's prediction. The rule was easily applied.Conclusion: A five-item clinical prediction rule of long-term back-related functional limitations could help first-line care physicians to concentrate the clinical attention on patients at higher risk.