Journal of Clinical Epidemiology - Articles in Press

Industry influenced evidence production in collaborative research communities: A network analysis - Corrected Proof

Adam G. Dunn, Blanca Gallego, Enrico Coiera — 2012-02-03

Abstract: Objective: To measure the relative influence that industry authors have on collaborative research communities and evidence production.Study Design and Setting: Using 22 commonly prescribed drugs, 6,711 randomized controlled trials (RCTs), and 28,104 authors, 22 collaboration networks were constructed and analyzed. The directly industry-affiliated (DIA) authors were identified in the networks according to their published affiliations. Measures of influence (network centrality) and impact (citations) were determined for every author. Network-level measures of community structure and collaborative preference were used to further characterize the groups.Results: Six percent (1,741 of 28,104) of authors listed a direct affiliation with the manufacturer of a drug evaluated in the RCT. These authors received significantly more citations (P<0.05 in 19 networks) and were significantly more central in the networks (P<0.05 in 20 networks). The networks show that DIA authors tend to have greater reach in the networks and collaborate more often with non-DIA authors despite a preference toward their own group. Potential confounders include publication bias, trial sizes, and conclusions.Conclusions: Industry-based authors are more central in their networks and are deeply embedded within highly connected drug research communities. As a consequence, they have the potential to influence information flow in the production of evidence.

Clinical practice guidelines and patient decision aids. An inevitable relationship - Corrected Proof

2012-02-02

Abstract: As health professionals and patients are moving toward shared models of decision making, there is a growing need for integrated decision support tools that facilitate uptake of best evidence in routine clinical practice in a patient-centered manner. This article charts the landscape of clinical practice guidelines (CPGs) and patient decision aids.Decision support tools for medical practice can be mapped on two dimensions. (1) The target user and his or her level of decision making; either for groups of patients or for an individual patient and (2) the level of uncertainty: either supporting more directive decision making (behavior support) in the case of strong recommendations with a single best option or supporting dialog (deliberation support) on the pros and cons of different options in the case of conditional (or weak) recommendations.We conclude that it is important to establish closer links between CPGs and patient decision aids, through collaborative development of both. Such collaboration will encourage the design of decision support tools for professionals and patients who share the same evidence and the aim to increase the quality of decision making between doctor and patient. This could facilitate the implementation of CPGs and shared decision making in clinical practice.

Trial registration in Latin America and the Caribbean’s: study of randomized trials published in 2010 - Corrected Proof

2012-01-30

Abstract: Objective: To determine the prevalence of trial registration in randomized controlled trials (RCTs) published in 2010 (PUBMED/LILACS) from Latin America and the Caribbean’s (LAC) and to compare methodological characteristics between registered and nonregistered RCTs.Study Design and Setting: A search for detecting RCTs in which at least the first/contact author had a LAC’s affiliation was made. We determined if RCTs were registered in the International Clinical Trial Registry Platform (ICTRP). Data were independently extracted by two authors. The risk of bias (RoB) was assessed in all registered RCTs (n=89) and in a sample of nonregistered RCTs (n=237).Results: The search identified 1,695 references; 526 RCTs from 19 countries were included. 16.9% (89/526) of RCTs were registered in the ICTRP; however, only 21 (4.0%) were prospectively registered. A significant difference was found in the overall assessment of the RoB between registered and nonregistered RCTs. Overall, registered RCTs were multinational, had larger sample size and longer follow-up, and reported more frequently information on funding, conflict of interests, and ethic issues. No significant differences were found when analyzing prospectively registered RCTs.Conclusion: This study shows that trial registration rates are still low in LAC and the quality of reporting needs to be improved.

Response to Letter to the Editor on administrative database research infrequently uses validated diagnostic or procedural codes - Corrected Proof

Carl van Walraven — 2012-01-30

Callahan clarifies an important issue surrounding validation studies restricted to the review of cases having a particular coding algorithm to determine whether the condition represented by the codes truly exists. Disease prevalence in the population sampled for the validation is important. In our study, only two of the eight studies that solely examined patients having the code algorithm in question actually reported the prevalence of that code algorithm in the population used for the validation study . Ideally, the prevalence of the actual disease in the validation population would be reported.

Endocrine clinical practice guidelines in North America. A systematic assessment of quality - Corrected Proof

2012-01-27

Abstract: Objective: To assess the quality of endocrine guidelines developed in North America.Study Design and Setting: A systematic review of the literature was conducted to identify all endocrine clinical practice guidelines developed in North America and published between January 1, 2007 and January 13, 2010. Two independent reviewers used the Appraisal of Guidelines, Research and Evaluation instrument to evaluate the quality of the guidelines in six domains: scope and purpose, stakeholder involvement, rigor of development, clarity and presentation, applicability, and editorial independence.Results: One hundred eligible endocrine guidelines had high scores in the scope-and-purpose (mean pooled standardized score [MPSD] of 82±14) and clarity domains (MPSD=64±17) and low scores in the stakeholder-involvement (MPSD of 36±12) and editorial independence domains (MPSD=36±36). Only 29% of guidelines scored above 60% for more than three domains. Rigor-of-development domain score was significantly higher in guidelines using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) approach, nondiabetes guidelines, and in published in-print vs. online publications.Conclusions: The quality of endocrine guidelines published in 2007–2009 is moderate and can be improved by (1) using methodologically sound development frameworks, (2) increasing stakeholder involvement, and (3) paying more attention to resource implications of guideline implementation.

US general population norms for telephone administration of the SF-36v2 - Corrected Proof

Gregory A. Maglinte, Ron D. Hays, Robert M. Kaplan — 2012-01-24

Abstract: Objective: US general population norms for mail administration of the Medical Outcomes Study 36-Item Short Form Version 2 (SF-36v2) were established in 1998. This article reports SF-36v2 telephone-administered norms collected in 2005–2006 for adults aged 35–89 years.Study Design and Setting: The SF-36v2 was administered to 3,844 adults in the National Health Measurement Study (NHMS), a random-digit dial telephone survey. Scale scores and physical and mental component summary (PCS and MCS) scores were computed.Results: When compared with 1998 norms (mean=50.00, standard deviation [SD]=10.00), SF-36v2 scores for the 2005–2006 general population tended to be higher: physical functioning (mean=50.68, SD=14.48); role limitations due to physical health problems (mean=49.47, SD=14.71); bodily pain (mean=50.66, SD=16.28); general health perceptions (mean=50.10, SD=16.87); vitality (mean=53.71, SD=15.35); social functioning (mean=51.37, SD=13.93); role limitations due to emotional problems (mean=51.44, SD=13.93); mental health (mean=54.27, SD=13.28); PCS (mean=49.22, SD=15.13); MCS (mean=53.78, SD=13.14). PCS and MCS factor scoring coefficients were similar to those previously reported for the 1998 norms. SF-36v2 norms for telephone administration were created.Conclusion: The higher scores for NHMS data are likely due to the effect of telephone administration. The 2005–2006 norms can be used as a reference to interpret scale and component summary scores for telephone-administered surveys with the SF-36v2.

Trends in death rate from diabetes according to multiple-cause-of-death differed from that according to underlying-cause-of-death in Taiwan but not in the United States, 1987–2007 - Corrected Proof

Yu-Pei Lin, Tsung-Hsueh Lu — 2012-01-24

Abstract: Objective: To examine whether trends in death rate from diabetes according to multiple-cause-of-death (MCOD) data differed from those according to underlying-cause-of-death (UCOD) data in Taiwan and the United States.Study Design and Setting: We used multiple cause mortality files for the years 1987, 1992, 1997, 2002, and 2007 to calculate the age-adjusted death rates from diabetes according to MCOD and UCOD and the ratio between diabetes mortality according to UCOD and that according to MCOD (U/M ratio) in Taiwan and the United States.Results: In Taiwan, diabetes mortality according to MCOD increased persistently from 1987 to 2007, but no prominent changes were found according to UCOD in men. For women, the death rates according to MCOD did not change significantly between 1987 and 2007 but decreased drastically from 1992 to 2007 according to UCOD. In the United States, the patterns of change in diabetes mortality according to MCOD were similar to those according to UCOD in both sexes. The U/M ratio of diabetes mortality declined persistently between 1987 and 2007 in Taiwan, but no prominent change was found in the United States.Conclusion: Trends in death rate from diabetes according to MCOD differed from that according to UCOD in Taiwan but not in the United States. To properly interpret cause-specific mortality trends, it is important to provide both MCOD and UCOD data.

Robust meta-analytic conclusions mandate the provision of prediction intervals in meta-analysis summaries - Corrected Proof

Petra L. Graham, John L. Moran — 2012-01-24

Abstract: Objectives: Results of meta-analyses typically conclude that future large studies may be mandated. However, the predictive ability of these estimates is deficient. We explored meta-analytic prediction intervals as means for providing a clear and appropriate future treatment summary reflecting current estimates.Study Design: A meta-epidemiological study of binary outcome critical care meta-analyses published between 2002 and 2010. Computation of 95% DerSimonian-Laird and Bayesian random-effects meta-analytic confidence intervals (CI) and 95% credible intervals (CrI), respectively, and frequentist (PI) and Bayesian (PrI) prediction intervals for odds ratio (OR) and risk ratio (RR) were undertaken. Bayesian calculations included the probability that the OR and RR point estimates ≥1.Results: Seventy-two meta-analyses from 70 articles were identified, containing between three and 80 studies each, with median nine studies. For both frequentist and Bayesian settings, 49–69% of the meta-analyses excluded the null. All significant CrI had high probabilities of efficacy/harm. The number of PI vs. PrI excluding 1 was 25% vs. 3% (OR), 26% vs. 3% (RR) of the total meta-analyses. Unsurprisingly, PI/PrI width was greater than CI/CrI width and increased with increasing heterogeneity and combination of fewer studies.Conclusion: Robust meta-analytic conclusions and determination of studies warranting new large trials may be more appropriately signaled by consideration of initial interval estimates with prediction intervals. Substantial heterogeneity results in exceedingly wide PIs. More caution should be exercised regarding the conclusions of a meta-analysis.

Gradually implemented new biomarkers for prognostication of breast cancer: complete case analysis may introduce bias - Corrected Proof

2012-01-24

Abstract: Objective: Many recent studies investigated the prognostic value of new biomarkers in breast cancer using data from cancer registries. Some of these studies were conducted using only patients for whom biomarker status was available (or tested). Using human epidermal growth factor receptor 2 (HER2) as an example, we determined whether testing for a recently introduced biomarker was associated with the outcome of women with breast cancer.Study Design and Setting: We included 910 women with newly diagnosed breast cancer in a tertiary academic hospital in Kuala Lumpur, Malaysia, between 2005 and 2007. Individual 2-year absolute mortality risk was estimated using Cox regression analysis. Logistic regression was used to assess the association between the absolute mortality risk and assessment of HER2 status.Results: There was a significant inverted U-shaped association between predicted mortality risk and HER2 status determination. Compared with patients with the lowest predicted mortality risk (quintile 1), patients with highest predicted mortality risk (last quintile) were significantly less likely to be tested for HER2 status, whereas those with intermediate predicted mortality risk (quintile 3) were more likely to be tested.Conclusion: Breast cancer prognostication using only patients with available biomarker status may lead to invalid results.

Adequate methods should be used to compare the features of effect measures - Corrected Proof

Ralf Bender — 2012-01-24

In a recent article on the considerable variation of the number needed to treat (NNT), Wisløff et al. used Monte Carlo simulations to compare the distribution of NNT with that of the relative risk (RR). The methods used have the following limitations and shortcomings.

Regarding adequate methods for comparison of effect measures - Corrected Proof

Torbjørn Wisløff, Odd O. Aalen, Ivar S. Kristiansen — 2012-01-24

We are pleased to observe that we agree with Prof Dr Bender that baseline risk should be considered when effect measurers are reported . When it comes to his critique, we are clearly aware that one can look at the number needed to treat (NNT) from a mathematical point of view, and that a main reason for the difficulties with this measure is that it is defined by inverting a difference. This creates a fundamental instability in the NNT. Our aim, however, is to illustrate the practical implications of this for medical studies in a simple way, and we believe we do this rather clearly.

Good interobserver agreement was attainable on outcome adjudication in patients with stable coronary heart disease - Corrected Proof

2012-01-19

Abstract: Objective: In clinical trials, agreement on outcomes is of utmost importance for valid estimation of intervention effects. As there is limited knowledge about adjudicator agreement in cardiology, we examined the level of agreement among three cardiology specialists adjudicating all possible events in a randomized controlled clinical trial of patients with stable coronary heart disease.Study Design and Setting: All information (hospital records, death certificates, etc.) was forwarded to two randomly selected blinded adjudicators. If they disagreed, the third arbiter had to choose the more likely of the two alternatives. Files of 5,475 nonfatal and 362 fatal events were evaluated.Results: For nonfatal outcomes, pairwise kappa values ranged from 0.75 to 0.80. The three adjudicators had 4.3%, 9.5%, and 6.1% of their nonfatal outcome classifications overruled by their arbiter. If stable angina pectoris, unstable angina pectoris, and acute myocardial infarction were treated as one, agreement increased minimally. For fatal outcomes, the pairwise kappa values ranged from 0.65 to 0.90. The three adjudicators had 12%, 9%, and 10% of their death classifications overruled.Conclusion: Specialists in cardiology can attain a reasonably high agreement on outcomes in patients with stable coronary heart disease.

Development and validation of clinical prediction models: Marginal differences between logistic regression, penalized maximum likelihood estimation, and genetic programming - Corrected Proof

2012-01-04

Abstract: Objective: Many prediction models are developed by multivariable logistic regression. However, there are several alternative methods to develop prediction models. We compared the accuracy of a model that predicts the presence of deep venous thrombosis (DVT) when developed by four different methods.Study Design and Setting: We used the data of 2,086 primary care patients suspected of DVT, which included 21 candidate predictors. The cohort was split into a derivation set (1,668 patients, 329 with DVT) and a validation set (418 patients, 86 with DVT). Also, 100 cross-validations were conducted in the full cohort. The models were developed by logistic regression, logistic regression with shrinkage by bootstrapping techniques, logistic regression with shrinkage by penalized maximum likelihood estimation, and genetic programming. The accuracy of the models was tested by assessing discrimination and calibration.Results: There were only marginal differences in the discrimination and calibration of the models in the validation set and cross-validations.Conclusion: The accuracy measures of the models developed by the four different methods were only slightly different, and the 95% confidence intervals were mostly overlapped. We have shown that models with good predictive accuracy are most likely developed by sensible modeling strategies rather than by complex development methods.

Lethal misconceptions: interpretation and bias in studies of traffic deaths - Corrected Proof

Donald A. Redelmeier, Christopher J. Yarnell — 2012-01-04

Abstract: Clinical epidemiology studies are vulnerable to subtle confounding, leading skeptics to claim that an odds ratio below three rarely indicates a clinically important finding. We argue that such a high threshold is inappropriate when interpreting traffic death studies in clinical epidemiology research. We review 10 concepts that emphasize the value of modest effect sizes by taking into account the baseline frequency, nonfatal disability, numbers needed to treat, shared responsibility, event diversity, behavioral offsets, measurement error, indirect reinforcement, delayed progression, and economic affordability. An awareness of these concepts may help when interpreting effect sizes in studies of traffic deaths.

Two-thirds of methodological research remained unpublished after presentation at Cochrane Colloquia: an empirical analysis - Corrected Proof

Sarah Chapman, Anne Eisinga, Sally Hopewell, Mike Clarke — 2012-01-04

Abstract: Objectives: To determine the extent to which abstracts of methodology research, initially presented at annual meetings of The Cochrane Collaboration, have been published as full reports and over what period of time. A secondary aim was to explore whether full publication varied in different methodological subject areas.Study Design and Setting: The Cochrane Methodology Register (CMR) was searched for all abstracts reporting methodology research, presented at the 11 Cochrane Colloquia from 1997 to 2007. EMBASE, PubMed, and CMR were searched for full publications of the same research.Results: We identified 908 eligible conference abstracts and found full publications for 312 (34.4%) of these, almost half of which (47.1%) had appeared by the end of the first year after the relevant Colloquium. The proportion of abstracts that had not been published by 3 years was 69.7%, falling to 66.2% at 5 years. Publication varied considerably between different methodological areas.Conclusion: Approximately two-thirds of methodological research studies presented at Cochrane Colloquia remain unpublished as full papers at least 5 years later. This highlights the importance of searching conference abstracts if one wishes to find as comprehensive and complete a sample of methodological research as possible.

Health awareness and associations with nonparticipation bias—a question of faith in life? - Corrected Proof

Olli Nummela — 2012-01-04

I would like to thank Ivy Shiue for her thoughtful and constructive comments on our article “Register-based data indicated nonparticipation bias in a health study among aging people” published in the Journal of Clinical Epidemiology, 2010 .

Increasing health awareness to decreasing nonparticipation bias in health studies - Corrected Proof

Ivy Shiue — 2012-01-04

It is with great interest to read “Register-based data indicated nonparticipation bias in a health study among aging people” by Nummela et al. , which addressed the differences in response rates across sociodemographic groups. The authors indicated that in the register-based data, they found participants with a tendency to have lower socioeconomic positions compared with nonparticipants. Poor health outcome is also more common seen in nonparticipants. Besides this observation, health beliefs and/or awareness may also be one of the reasons affecting response rates. Prior studies have shown that lack of health awareness, perception, or knowledge among nonparticipants was observed across public health programs and clinical trials , as well as nonparticipants, particularly among the elderly, often underestimate, ignore, or deny having high risk of certain diseases. Health beliefs, awareness, perception, or knowledge is actually related to socioeconomic status, including education and income. For primary prevention purpose, we may need to understand if unawareness could bring more influence than socioeconomic positions to avoid participating in health studies because health awareness is directly related to lifestyle behaviors, and lifestyle behaviors are indicators to health outcomes. Constant public education on importance of healthy living and research will then be required.

Within-person study designs had lower precision and greater susceptibility to bias because of trends in exposure than cohort and nested case–control designs - Corrected Proof

Jennifer M. Nicholas, Andrew P. Grieve, Martin C. Gulliford — 2011-12-26

Abstract: Objective: To compare precision and apparent bias between cohort, nested case–control, self-controlled case series, case–crossover, and case–time–control study designs.Study Design and Setting: Study designs were implemented to evaluate the association between thiazolidinediones (TZDs) and heart failure, TZDs and fracture, and liver enzyme–inducing anticonvulsants and fracture.Results: Effect estimates were similar for the cohort and case–control study; for the association between TZDs and fracture in women, the hazard ratio was 1.36 (1.18, 1.56) and odds ratio (OR) was 1.44 (1.21, 1.70). For this clinical example, the self-controlled case series gave upward bias when follow-up was censored at the outcome (incidence rate ratio [IRR], 7.08; 4.96, 10.09) but was otherwise unbiased (IRR, 1.41; 1.14, 1.75). The retrospective case–crossover OR was 3.24 (2.18, 4.80), which was reduced by either bidirectional sampling (OR, 1.20; 0.98, 1.46) or with the case–time–control design (OR, 1.40; 1.09, 1.81). Findings on apparent bias were similar for the other two clinical examples. In each clinical example, within-person designs had considerably lower precision than the cohort or case–control study designs.Conclusion: When long-term exposures are analyzed, within-person study designs may have lower precision and greater susceptibility to bias. Bias may be reduced by sampling follow-up both before and after the outcome or with the case–time–control study design.

Multicriteria benefit–risk assessment using network meta-analysis - Corrected Proof

2011-12-26

Abstract: Objective: To enable multicriteria benefit–risk (BR) assessment of any number of alternative treatments using all available evidence from a network of clinical trials.Study Design and Setting: We design a general method for multicriteria decision aiding with criteria measurements from Mixed Treatment Comparison (MTC) analyses. To evaluate the method, we apply it to BR assessment of four second-generation antidepressants and placebo in the setting of a published peer-reviewed systematic review.Results: The analysis without preference information shows that placebo is supported by a wide range of possible preferences. Preference information provided by a clinical expert showed that although treatment with antidepressants is warranted for severely depressed patients, for mildly depressed patients placebo is likely to be the best option. It is difficult to choose between the four antidepressants, and the results of the model indicate a high degree of uncertainty.Conclusions: The designed method enables quantitative BR analysis of alternative treatments using all available evidence from a network of clinical trials. The preference-free analysis can be useful in presenting the results of an MTC considering multiple outcomes.

The Global Rating Scale complements the AGREE II in advancing the quality of practice guidelines - Corrected Proof

2011-12-21

Abstract: Objective: To explore the role of a four-item Global Rating Scale (GRS) that could be used in place of the Appraisal of Guidelines, Research and Evaluation II (AGREE II).Study Design and Setting: A mixed four-factor design was used (User Type, Evaluation Type, Clinical Topic, Guideline). Participants were asked to read and evaluate a guideline using both the AGREE II draft and GRS or GRS only and to complete a series of questions regarding overall guideline quality, adoption, utility, and acceptability.Results: One GRS item varied as a function of User Type. Each item was a significant predictor of participants' outcome measures. All items were rated as useful by stakeholders. The GRS rating scores, outcome measures, and usefulness scores did not vary between the two Evaluation Type conditions. Correlations between the GRS and the outcome measures were stronger compared with those between the AGREE II draft and these measures.Conclusion: Although the GRS is less sensitive than the AGREE II in detecting differences in guideline quality as a function of User Type, its items did predict important outcome measures related to guideline adoption. The GRS may play a role in guideline evaluation, although further study is warranted.

The use of weekly text messaging over 6 months was a feasible method for monitoring the clinical course of low back pain in patients seeking chiropractic care - Corrected Proof

2011-12-12

Abstract: Objective: This study critically evaluates a new method of collecting frequent data using mobile phones and text messages. Fluctuating conditions such as low back pain (LBP) need frequent monitoring to describe the clinical course in detail and to account for individual and subgroup variations.Study Design and Setting: In this multicentre prospective observational study, 262 subjects with nonspecific LBP were followed with weekly text messages for 6 months, with the question “How many days this previous week has your low back pain been bothersome?” The text replies were instantly recorded in a data file to be merged with baseline and follow up data (age, gender, pain intensity, duration, and self- rated health) collected through ordinary questionnaires. The response rate, user-friendliness, and compliance of this method were evaluated.Results: The mean response rate for the text messages throughout the study was 82.5% and was unaffected by season. The method was found to be user friendly. Dropout was not affected by age and gender, but compliance was possibly somewhat affected by outcome.Conclusion: Weekly text messages are a useful method of data collection to examine the clinical course of LBP in the primary care sector.

Covariate adjustment increased power in randomized controlled trials: an example in traumatic brain injury - Corrected Proof

2011-12-12

Abstract: Objective: We aimed to determine to what extent covariate adjustment could affect power in a randomized controlled trial (RCT) of a heterogeneous population with traumatic brain injury (TBI).Study Design and Setting: We analyzed 14-day mortality in 9,497 participants in the Corticosteroid Randomization After Significant Head Injury (CRASH) RCT of corticosteroid vs. placebo. Adjustment was made using logistic regression for baseline covariates of two validated risk models derived from external data (International Mission on Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury [IMPACT]) and from the CRASH data. The relative sample size (RESS) measure, defined as the ratio of the sample size required by an adjusted analysis to attain the same power as the unadjusted reference analysis, was used to assess the impact of adjustment.Results: Corticosteroid was associated with higher mortality compared with placebo (odds ratio=1.25, 95% confidence interval=1.13–1.39). RESS of 0.79 and 0.73 were obtained by adjustment using the IMPACT and CRASH models, respectively, which, for example, implies an increase from 80% to 88% and 91% power, respectively.Conclusion: Moderate gains in power may be obtained using covariate adjustment from logistic regression in heterogeneous conditions such as TBI. Although analyses of RCTs might consider covariate adjustment to improve power, we caution against this approach in the planning of RCTs.

Reliability analysis for a proposed critical appraisal tool demonstrated value for diverse research designs - Corrected Proof

Michael Crowe, Lorraine Sheppard, Alistair Campbell — 2011-11-14

Abstract: Objective: To examine the reliability of scores obtained from a proposed critical appraisal tool (CAT).Study Design and Setting: Based on a random sample of 24 health-related research papers, the scores from the proposed CAT were examined using intraclass correlation coefficients (ICCs), generalizability theory, and participants’ feedback.Results: The ICC for all research papers was 0.83 (consistency) and 0.74 (absolute agreement) for four participants. For individual research designs, the highest ICC (consistency) was for qualitative research (0.91) and the lowest was for descriptive, exploratory and observational research (0.64). The G study showed a moderate research design effect (32%) for scores averaged across all papers. The research design effect was mainly in the Sampling, Results, and Discussion categories (44%, 36%, and 34%, respectively). The scores for research designs showed a majority paper effect for each (53–70%), with small to moderate rater or paper×rater interaction effects (0–27%).Conclusions: Possible reasons for the research design effect were that the participants were unfamiliar with some of the research designs and that papers were not matched to participants’ expertise. Even so, the proposed CAT showed great promise as a tool that can be used across a wide range of research designs.

Case study: A patient–clinician collaboration that identified and prioritized evidence gaps and stimulated research development - Corrected Proof

Brian S. Buckley, Adrian M. Grant, Cathryn M.A. Glazener — 2011-08-05

Abstract: Objective: To assess the effect of a research prioritization partnership that aimed to influence the research agenda relating to urinary incontinence (UI).Study Design and Setting: Research often neglects important gaps in existing evidence so that decisions must be made about treatments without reliable evidence of their effectiveness. In 2007–2009, a United Kingdom partnership of eight patient and 13 clinician organizations identified and prioritized gaps in the evidence that affect everyday decisions about treatment of UI. The top 10 prioritized research questions were published and reported to research funders in 2009. A year later, new research or funding applications relating to the prioritized topics were identified through reviews of research databases and consultation with funding organizations, elements of the research community, and organizations that participated in the partnership.Results: Since dissemination of the prioritized topics, five studies are known to have been funded, three in development; five new systematic reviews are under way, one is being updated; five questions are under consideration by a national research commissioning body.Conclusion: The partnership successfully developed and used a methodology for identification and prioritization of research needs through patient–clinician consensus. Prioritization through consensus can be effective in informing the development of clinically useful research.

Mathematical coupling does not account for the association between baseline severity and minimally important change values - Corrected Proof

Berend Terluin — 2011-08-02

Abstract: A number of studies have demonstrated a correlation between baseline severity and minimally important change (MIC) values. However, Browne et al. stated that these studies failed to account for “mathematical coupling,” and that, therefore, the correlation between baseline severity and MIC values may be spurious. The present article demonstrates that on the level of individual scores, mathematical coupling causes the observed baseline and change scores to correlate even in the absence of any true correlation between these variables. This phenomenon is because of the fact that change scores can only be estimated by subtracting the baseline score from the follow-up score, causing the baseline and change scores to share a common piece of error variance. However, MIC values are always determined on group level, and mathematical coupling does not affect group-level statistics or the correlation of these statistics across groups. Therefore, mathematical coupling does not account for the association between baseline severity and MIC values as suggested by Browne et al.