Journal of Clinical Epidemiology - Articles in Press

Formatting modifications in GRADE evidence profiles improved guideline panelists comprehension and accessibility to information. A randomized trial - Corrected Proof

2012-05-07

Abstract: Objective: To determine the effects of formatting alternatives in Grading of Recommendations Assessment, Development, and Evaluation (GRADE) evidence profiles on guideline panelists’ preferences, comprehension, and accessibility.Study Design and Setting: We randomized 116 antithrombotic therapy guideline panelists to review one of two table formats with four formatting alternatives. After answering relevant questions, panelists reviewed the other format and reported their preferences for specific formatting alternatives.Results: Panelists (88 of 116 invited [76%]) preferred presentation of study event rates over no study event rates (median 1 [interquartile range (IQR) 1] on 1–7 scale), absolute risk differences over absolute risks (median 2 [IQR 3]), and additional information in table cells over footnotes (median 1 [IQR 2]). Panelists presented with time frame information in the tables, and not only in footnotes, were more likely to correctly answer questions regarding time frame (58% vs. 11%, P<0.0001), and those presented with risk differences and not absolute risks were more likely to correctly interpret confidence intervals for absolute effects (95% vs. 54%, P<0.0001). Information was considered easy to find, easy to comprehend, and helpful in making recommendations regardless of table format (median 6, IQR 0–1).Conclusion: Panelists found information in GRADE evidence profiles accessible. Correct comprehension of some key information was improved by providing additional information in table and presenting risk differences.

A confidence interval approach to sample size estimation for interobserver agreement studies with multiple raters and outcomes - Corrected Proof

Michael A. Rotondi, Allan Donner — 2012-05-07

Abstract: Objective: Studies measuring interobserver agreement (reliability) are common in clinical practice, yet discussion of appropriate sample size estimation techniques is minimal as compared with clinical trials. The authors propose a sample size estimation technique to achieve a prespecified lower and upper limit for a confidence interval for the κ coefficient in studies of interobserver agreement.Study Design and Setting: The proposed technique can be used to design a study measuring interobserver agreement with any number of outcomes and any number of raters. Potential application areas include: pathology, psychiatry, dentistry, and physical therapy.Results: This technique is illustrated using two examples. The first considers a pilot study in oral radiology, whose authors studied the reliability of the mandibular cortical index as measured by three dental professionals. The second example examines the level of interobserver agreement among four nurses with respect to five triage levels used in the Canadian Triage and Acuity Scale.Conclusion: This method should be useful in the planning stages of an interobserver agreement study in which the investigator would like to obtain a prespecified level of precision in the estimation of κ. An R software package (R Foundation for Statistical Computing, Vienna, Austria), kappaSize is also provided that implements this method.

Using electronic medical records analysis to investigate the effectiveness of lifestyle programs in real-world primary care is challenging: a case study in diabetes mellitus - Corrected Proof

2012-05-07

Abstract: Objective: The increasing prevalence of diabetes suggests a gap between real world and controlled trial effectiveness of lifestyle interventions, but real-world investigations are rare. Electronic medical registration facilitates research on real-world effectiveness, although such investigations may require specific methodology and statistics. We investigated the effects of real-world primary care for patients with type 2 diabetes mellitus (T2DM).Study Design and Setting: We used medical records of patients (n=2,549) with T2DM from 10 primary health care centers. A mixed-effects regression model for repeated measurements was used to evaluate the changes in weight and Hemoglobin A1c (HbA1c) over time.Results: There was no statistically significant change in weight (+0.07kg, P=0.832) and HbA1c (+0.03%, P=0.657) during the observation period of 972 days. Most patients maintained their physical activity level (70%), and 54 % had an insufficient activity level. The variability in the course of weight and HbA1c was because of differences between patients and not between health care providers.Conclusion: Despite effective lifestyle interventions in controlled trial settings, we found that real-world primary care is only able to stabilize weight and HbA1c in patients with T2DM over time. Medical registration can be used to monitor the actual effectiveness of interventions in primary care.

Methodologists and context experts disagreed regarding managing conflicts of interest of clinical practice guidelines panels - Corrected Proof

Elie A. Akl, Renee Karl, Gordon H. Guyatt — 2012-05-07

Abstract: Objective: A new strategy to manage conflicts of interests (COIs) of a clinical guideline’s panelists gives primary responsibility to a methodologist, puts equal emphasis on intellectual and financial COIs, and excludes panelists with primary conflicts from drafting or voting on recommendations. We explored the views of the methodologists and content experts regarding the new strategy.Study Design and Setting: Before the guidelines chapter panels initiated their work, we conducted semi-structured personal interviews with the methodologists and the lead content experts. We analyzed the data qualitatively.Results: Twenty-four panelists participated. The methodologists thought that the new strategy increased their responsibility and authority. The lead content experts perceived their role label as unfair and reflecting a demotion. Whereas methodologists were concerned about potential conflicts with content experts, the lead content experts were uncomfortable with the “extra surveillance” by the methodologists. Whereas methodologists believed that the changes ensure more rigorous evidence-based guidelines, some lead content experts were worried that methodologists’ lack of content expertise and content expert attrition could hurt the quality of the guidelines.Conclusions: The methodologists and lead content experts were uneasy regarding their counterpart’s role. They disagreed about the potential effect of the new strategy on the quality of the guideline.

Self-reported diabetes is a valid outcome in pragmatic clinical trials and observational studies - Corrected Proof

2012-05-07

The Centers for Disease Control estimates that approximately 24 million adults, or about 8% of the US population, are considered to have diabetes (both diagnosed and undiagnosed) . The risk of developing diabetes increases with age, more so for women compared with men . The Women’s Health Initiative (WHI) is a data source that has tremendous untapped potential for exploring the risk factors for and the health consequences of diabetes in older women. From 1993 to 1998, the WHI enrolled 68,133 postmenopausal women, aged 50–79 years at baseline in its randomized controlled clinical trials . The cohort has been followed-up annually through 2010 and a follow-up period through 2015 was recently funded. Incident self-reported diabetes was obtained as one of the outcomes of the clinical trials, however was not independently confirmed by review of medical records. We sought to examine the agreement of self-reports of diabetes in the WHI clinical trials with medical record review, thereby obtaining estimates of both the positive and negative predictive value of self-report.

Incorrect calculation of numbers needed to treat affects confidence intervals more often than point estimates - Corrected Proof

Ralf Bender — 2012-05-07

In a recent article on the estimation of the number needed to treat (NNT) in studies with varying follow-up times, Suissa et al. concluded that 6 of 10 studies published in four major medical journals in 2009 did not compute the NNT correctly. This confirms and extends the previous result, that 17 of 34 articles published in the same journals between 2003 and 2005 applied an inappropriate calculation method to derive NNTs from time-to-event outcomes . There are two frequently used inadequate methods to calculate NNTs from time-to-event outcomes, namely the use of naive proportions neglecting censoring and the application of incidence rates rather than absolute risks . Suissa et al. presented published examples where the application of naive proportions or incidence rates leads to incorrect NNT point estimates. Besides incorrect point estimates, an additional problem is the incorrect calculation of confidence intervals (CIs) if naive proportions or incidence rates are used to estimate NNTs. This is important because in practice incorrectly calculated NNT point estimates are frequently close to the corresponding estimates calculated by an appropriate method . However, even if the use of an incorrect method has little effect on the point estimate, the corresponding incorrectly calculated CI may be misleading. For example, if naive proportions are used to estimate NNTs in the case of varying follow-up times and a large proportion of censored data, the corresponding CIs will be too narrow because censoring is not taken into account and the values used for the effective sample sizes are too large .

The living with chronic obstructive pulmonary disease scale was successfully adapted for use in Southern European (Italian and Spanish) and Eastern European (Russian) cultures - Corrected Proof

2012-05-07

Abstract: Objective: The living with chronic obstructive pulmonary disease (LCOPD) scale was developed in the United Kingdom and the United States to assess the quality of life (QoL) of people with chronic obstructive pulmonary disease. The scale was shown to be well accepted by patients, to be unidimensional, and to have very good psychometric properties. This article reports on the adaptation of the measure for use in representative Southern European (Italian and Spanish) and Eastern European (Russian) languages.Study Design and Setting: The same methodology was used in each country: translation panels were held to produce new language versions. These were then tested with patients to ensure face and content validity. Finally, test–retest postal surveys were conducted to establish internal consistency, reproducibility, and construct validity.Results: Few problems were found in translating the LCOPD, and patient interviews indicated that the measure was considered relevant and easy to complete. The validation surveys showed that the new versions of the LCOPD were consistent, reproducible, and had good construct validity. The psychometric properties of the new versions were similar to those found for the original U.K. and U.S. versions.Conclusion: The study was successful in adapting the LCOPD for use in Southern and Eastern European languages.

Rank minimization with a two-step analysis should not replace randomization in clinical trials - Corrected Proof

Tim Morris — 2012-05-07

Taves’ article promoting the wider use of minimization contains a number of concerning points, one or two of which need some comment. Ignoring my concerns about selection bias with rank minimization, this letter is primarily about the section titled A new statistical analysis convention.

GRADE guidelines 11—making an overall rating of confidence in effect estimates for a single outcome and for all outcomes - Corrected Proof

2012-04-30

Abstract: GRADE requires guideline developers to make an overall rating of confidence in estimates of effect (quality of evidence—high, moderate, low, or very low) for each important or critical outcome. GRADE suggests, for each outcome, the initial separate consideration of five domains of reasons for rating down the confidence in effect estimates, thereby allowing systematic review authors and guideline developers to arrive at an outcome-specific rating of confidence. Although this rating system represents discrete steps on an ordinal scale, it is helpful to view confidence in estimates as a continuum, and the final rating of confidence may differ from that suggested by separate consideration of each domain.An overall rating of confidence in estimates of effect is only relevant in settings when recommendations are being made. In general, it is based on the critical outcome that provides the lowest confidence.

Pluralism of viewpoints as the antidote to intellectual conflict of interest in guidelines - Corrected Proof

Allan D. Sniderman, Curt D. Furberg — 2012-04-30

Men try consciously to eliminate the atrocities of society but meanwhile they unheedingly build a civilization, which is more destructive of moral and personal values than anything intended in a primitive society. (Reinhold Niebuhr, 1923).

Nasopharyngeal swabs of school children, useful in rapid assessment of community antimicrobial resistance patterns in Streptococcus pneumoniae and Haemophilus influenzae - Corrected Proof

2012-04-30

Abstract: Objectives: The present study evaluates the feasibility of rapid surveillance of community antimicrobial resistance (AMR) patterns of Streptococcus pneumoniae and Haemophilus influenzae in India using nasopharyngeal swabs (NPSs) of school children. It compares the AMR data obtained with that of invasive and nasopharyngeal (NP) isolates studied previously. No one has done such surveillance since our study so we decided to publish and more clearly demonstrate the feasibility of the methodology we did.Study Design and Setting: This community-based, cross-sectional, cluster sample study had seven centers; each had two sites distant to them. Two hundred sixty school children per center were enrolled. NP swabbing was performed and isolates identified as S. pneumoniae and H. influenzae at each center were sent to reference laboratories.Results: From January to December 2004, 1,988 NP swabs were processed; 776 S. pneumoniae and 64 H. influenzae were isolated. The AMR patterns for S. pneumoniae to co-trimoxazole varied, with sensitivity as low as 6% in Mumbai, 29% in Chennai and Vellore, and 100% in Delhi and Lucknow. For H. influenzae, sensitivity rates to co-trimoxazole ranged from 22% to 62%. The AMR patterns for both bacteria in the present study with data from invasive and NP isolates studied earlier were similar.Conclusion: The study demonstrates that it is practical and feasible to rapidly assess the AMR patterns of both S. pneumoniae and H. influenzae in NPSs of school children in different geographic locations all over India.

Rebutal of anon’s critique: in fact randomization is the limiting probability distribution for imbalance with minimization - Corrected Proof

Donald R. Taves — 2012-04-30

The author takes issue with the proposed new statistical analysis convention that should be used with rank minimization and all other algorithms that fit the definition of minimization . He thinks balancing on all characteristics regardless of their importance is illogical for three reasons. First, he says that randomization, not minimization, guarantees that unknown characteristics will be balanced “in probability.” “In probability” means that if a study could be repeated an infinite number of times using randomization the effect of the unknown characteristic on the outcome will cancel out. This also is true with minimization if the unknown characteristic is not correlated with any minimized characteristic. In this case it will distribute independently, that is, randomly . However, if there is any correlation between the unknown characteristic and those used in minimization, the distribution of the outcome will tend to be more balanced than with randomization, that is, there will be fewer instances where the outcome will be statistically significantly different by chance alone. A demonstration of this synergism between correlated characteristics is shown in the original description of minimization . Using 15 characteristics it was shown that those that were correlated acted as though they were weighted more than those that were not. Also this interaction of correlated characteristics is the basis for doing analyses of covariance. Both considerations lead to an axiom: balancing any characteristic that is correlated with an unknown characteristic that is correlated with the outcome will decrease the probability of obtaining differences in the outcome by chance alone. In summary, minimization “in probability” cannot make the balance worse than randomization, it is the limiting probability distribution. Therefore, his first point is in error; minimization guarantees better balance of unknown characteristics than does randomization “in probability.”

Informed consent documents do not encourage good-quality decision making - Corrected Proof

2012-04-27

Abstract: Objective: Informed consent for research has emphasized information provision over support to people making a difficult decision. We assessed the extent to which existing informed consent documents (ICDs) conform to the International Patient Decision Aid Standards for supporting decision making.Study Design and Setting: One hundred thirty-nine ICDs for trials registered with ClinicalTrials.gov were obtained from study investigators. Using a four-point scale, two raters assessed each ICD on 32 items.Results: Overall agreement between raters was 95.1% (linear weighted kappa—0.745). For 12 items focused on providing enough information, conformity was above 50% for three, and 0% for another four. For all eight items focused on how to present outcome probabilities, conformity was below 20%. For two items focused on clarifying and expressing values, conformity was below 10%. For two items focused on improving structured guidance, conformity was below 5%. For four items focused on using evidence, one item showed conformity of 74%; all others showed conformity below 5%. For four items focused on transparency, conformity was high (above 60% for two, above 80% for the others).Conclusions: Existing ICDs do not meet most validated standards for encouraging good decision making. These standards make clear predictions about how one might improve ICDs ensure that research participants are fully informed.

Primary study authors of significant studies are more likely to believe that a strong association exists in a heterogeneous meta-analysis compared with methodologists - Corrected Proof

Orestis A. Panagiotou, John P.A. Ioannidis — 2012-04-27

Abstract: Objective: To assess the interpretation of a highly heterogeneous meta-analysis by authors of primary studies and by methodologists.Study Design and Setting: We surveyed the authors of studies on the association between insulin-like growth factor 1 (IGF-1) and prostate cancer, and 20 meta-analysis methodologists. Authors and methodologists presented with the respective meta-analysis results were queried about the effect size and potential causality of the association. We evaluated whether author responses correlated with the number of IGF-related articles they had published and their study results included in the meta-analysis. We also compared authors’ and methodologists’ responses.Results: Authors who had published more IGF-related papers offered more generous effect size estimates for the association (ρs=0.61, P=0.01) and higher likelihood that the odds ratio (OR) was greater than 1.20 (ρs=0.63, P=0.01). Authors who had published themselves studies with statistically significant effects for a positive association were more likely to believe that the true OR is greater than 1.20 compared with methodologists (median likelihood 50% versus 2.5%, P=0.01).Conclusion: Researchers are influenced by their own investment in the field, when interpreting a meta-analysis that includes their own study. Authors who published significant results are more likely to believe that a strong association exists compared with methodologists.

Rank minimization with a two-step analysis should not replace randomization in clinical trials - Corrected Proof

Brennan C. Kahan — 2012-04-27

Preventing selection bias is a key component of randomized trials. Thus, it was of great interest to read Taves commentary regarding new methods of preventing selection bias using rank minimization . Taves states that concerns of selection bias with rank minimization can be discounted based on four considerations. However, the reasoning behind a number of assumptions made in the commentary is unclear.

Rebutal of Kahan’s critique of rank minimization - Corrected Proof

Donald R. Taves — 2012-04-27

Kahan is correct that better referencing is needed to support an assumption that randomization with a block size of five, and therefore Rank Minimization also, is currently an acceptable way to avoid selection bias . McEntegart works for a group managing 337 clinical trials. He concluded that even the 40 trials that used center as one of the variables were better protected from selection bias with minimization than with randomization in blocks. He found that 85% of the latter used a block size of four or less , supporting the assumption that blocks of five are considered acceptable.

A framework for effective collaboration between specialist and broad-spectrum groups for delivering priority Cochrane reviews - Corrected Proof

2012-04-27

Abstract: Objectives: We aimed to develop and pilot a process for joint working between Cochrane Review Groups (specialist-area groups responsible for producing Cochrane reviews) and Cochrane Fields (broad-spectrum interest groups), for identifying high priority review topics and enhancing quality and dissemination of priority reviews.Study Design and Setting: We developed and piloted a framework for collaboration between a Cochrane Review Group (specializing in musculoskeletal injuries) and a Cochrane Field (focusing on health care of older people) for identifying, delivering, and disseminating priority Cochrane intervention reviews using hip fracture rehabilitation as an exemplar. The processes adopted included consultation of members of both the entities, mapping of trials from the Review Group’s Specialized Register, jointly establishing criteria for topic prioritization, identification of researchers, and facilitating provision of expert peer review from the field.Results: A framework for effective collaboration between a Cochrane Review Group and Cochrane Field for identifying and delivering priority Cochrane Reviews was devised and piloted. Additionally, two new Cochrane reviews, preceded by protocols, were published.Conclusion: The project demonstrated the feasibility and potential benefits of a structured collaboration between a Cochrane Review Group and a Cochrane Field for the identification and production of Cochrane reviews on priority topics.

An overused phrase: interpreted with caution - Corrected Proof

James M. Gaylor — 2012-04-25

A cursory scan of published biomedical literature reveals close variants of a recurrent and overused phrase: “These results, however, should be interpreted with caution.” Searching PubMed abstracts for the phrase “interpret∗ caution” retrieves 555 citations published between January 2010 and August 2011, and screening the first 100 abstracts reveals that 90 indeed contain this phrase in full and in 79 of them it appears in the conclusion sections of abstracts for primary studies and systematic reviews, potentially leaving in the mind of the reader the lingering and unsettling feeling of uncertainty in comprehending the results.

Risk factors for tuberculosis among health care workers in South India: a nested case–control study - Corrected Proof

2012-04-23

Abstract: Objective: The epidemiology of tuberculosis (TB) among health care workers (HCWs) in India remains under-researched. This study is a nested case–control design assessing the risk factors for acquiring TB among HCWs in India.Study Design and Settings: It is a nested case–control study conducted at a tertiary teaching hospital in India. Cases (n=101) were HCWs with active TB. Controls (n=101) were HCWs who did not have TB, randomly selected from the 6,003 subjects employed at the facility. Cases and controls were compared with respect to clinical and demographic variables.Results: The cases and controls were of similar age. Logistic regression analysis showed that body mass index (BMI) <19kg/m2 (odds ratio [OR]: 2.96, 95% confidence interval [CI]: 1.49–5.87), having frequent contact with patients (OR: 2.83, 95% CI: 1.47–5.45) and being employed in medical wards (OR: 12.37, 95% CI: 1.38–110.17) or microbiology laboratories (OR: 5.65, 95% CI: 1.74–18.36) were independently associated with increased risk of acquiring TB.Conclusion: HCWs with frequent patient contact and those with BMI <19kg/m2 were at high risk of acquiring active TB. Nosocomial transmission of TB was pronounced in locations, such as medical wards and microbiology laboratories. Surveillance of high-risk HCWs and appropriate infrastructure modifications may be important to prevent interpersonal TB transmission in health care facilities.

Ensuring relevance for Cochrane reviews: evaluating processes and methods for prioritizing topics for Cochrane reviews - Corrected Proof

2012-04-23

Abstract: Objective: The purpose of this study was to assess the presence and effectiveness of existing systems of prioritization for Cochrane review topics and to explore methods of improving those systems.Study Design and Setting: We surveyed groups of Cochrane review authors and recorded any evidence of their use of priority-setting processes or policies. To evaluate the effectiveness of the policies we encountered, we assessed them using two frameworks from the literature: “Accountability for Reasonableness” (1) and Sibbald’s 2009 framework (2) for successful priority setting. We then held two workshops with the subject groups to discuss our findings and their implications.Results: Of the 66 groups surveyed, 29 had a system in place to inform the selection or prioritization of topics for Cochrane reviews. Fifteen groups used a more comprehensive structured approach that eventually resulted in a list of ranked priority titles for authoring, updating, or disseminating Cochrane reviews. Most groups involved researchers, practitioners, and patients in their prioritization processes.Conclusion: Groups within The Cochrane Collaboration currently use a range of different priority-setting systems, some of which are more detailed than others. These differences often reflect the nature of The Cochrane Collaboration itself: given the topic breadth, history, and variety of international contexts present in the organization, a single unified system would not always be appropriate. All Cochrane entities, however, should have or develop strategic plans to improve the inclusiveness and transparency of their own prioritization processes, increase the number of finished prioritized reviews, and make more effective use of feedback from end users to increase the likelihood of producing reviews that have positive effects on health outcomes.

Directed acyclic graphs can help understand bias in indirect and mixed treatment comparisons - Corrected Proof

Jeroen P. Jansen, Christopher H. Schmid, Georgia Salanti — 2012-04-23

Abstract: Objective: To introduce and advocate directed acyclic graphs (DAGs) as a useful tool to understand when indirect and mixed treatment comparisons are invalid and guide strategies that limit bias.Study Design and Setting: By means of DAGs, it is heuristically explained when indirect and mixed treatment comparisons are biased, and whether statistical adjustment of imbalances in study and patient characteristics across different comparisons in the network of RCTs is appropriate.Results: A major threat to the validity of indirect and mixed treatment comparisons is a difference in modifiers of the relative treatment effect across comparisons, and statistically adjusting for these differences can improve comparability and remove bias. However, adjustment for differences in covariates across comparisons that are not effect modifiers is not necessary and can even introduce bias. As a special case, we outline that adjustment for the baseline risk might be useful to improve similarity and consistency, but may also bias findings.Conclusion: DAGs are useful to evaluate conceptually the assumptions underlying indirect and mixed treatment comparison, to identify sources of bias and guide the implementation of analytical methods used for network meta-analysis of RCTs.

Hearing, mobility, and pain predict mortality: a longitudinal population-based study - Corrected Proof

2012-04-23

Abstract: Objective: Measures of health-related quality of life (HRQL), including the Health Utilities Index Mark 3 (HUI3) are predictive of mortality. HUI3 includes eight attributes, vision, hearing, speech, ambulation, dexterity, cognition, emotion, and pain and discomfort, with five or six levels per attribute that vary from no to severe disability. This study examined associations between individual HUI3 attributes and mortality.Study Design and Setting: Baseline data and 12 years of follow-up data from a closed longitudinal cohort study, the 1994/95 Canadian National Population Health Survey, consisting of 12,375 women and men aged 18 and older. A priori hypotheses were that ambulation, cognition, emotion, and pain would predict mortality. Cox proportional hazards regression models were applied controlling for standard determinants of health and risk factors.Results: Single-attribute utility scores for ambulation (hazard ratio [HR]=0.10; 0.04–0.22), hearing (HR=0.18; 0.06–0.57), and pain (HR=0.53; 0.29–0.96) were statistically significantly associated with an increased risk of mortality; ambulation and hearing were predictive for the 60+ cohort.Conclusion: Few studies have identified hearing or pain as risk factors for mortality. This study is innovative because it identifies specific components of HRQL that predict mortality. Further research is needed to understand better the mechanisms through which deficits in hearing and pain affect mortality risks.

Classification and codification of rare diseases - Corrected Proof

Yllka Kodra, Bernardino Fantini, Domenica Taruscio — 2012-04-23

Rare diseases (RDs) are a group of disorders of very different etiology, whose common denominator is that they are low-prevalence diseases and for most of which there is no treatment available. Because of their low prevalence, they attract little interest from basic and clinical researchers and the scientific community in general, all of whom find it very difficult to secure public and/or private sources of financing for the study of the pathogenesis, diagnosis, and treatment of such diseases .

STrengthening the Reporting of OBservational studies in Epidemiology e Molecular Epidemiology STROBE-ME: an extension of the STROBE statement [J Clin Epidemiol 2011;64(12):1350-1363] - Corrected Proof

2012-04-16

An error in the text on page 1357 has been noted. In the section identified as “4.5. ME-4.1 – Report on the setting of the biological sample collection; amount of sample; nature of collecting procedures; participant conditions; time between sample collection and relevant clinical or physiological endpoints” the final sentence of the first paragraph states that “Proteins and cholesterol levels can be lowered by 5-15%.” This is incorrect, and the corrected text should read “Proteins and cholesterol levels can be increased by 5-15%.”

Author’s specialty and conflicts of interest contribute to conflicting guidelines for screening mammography - Corrected Proof

2012-04-13

Abstract: Objective: To examine the relationship between guideline panel members’ conflicts of interest and guideline recommendations on screening mammography in asymptomatic, average-risk women aged 40–49 years.Study Design and Setting: We searched the National Guideline Clearinghouse and MEDLINE for relevant guidelines published between January 2005 and June 2011. We examined the disclosures and specialties of the lead and secondary authors of these guidelines, as well as the publications of the lead authors.Results: Twelve guidelines were identified with a total of 178 physician authors from a broad range of specialties. Of the four guidelines not recommending routine screening, none had a radiologist member, whereas of the eight guidelines recommending routine screening, five had a radiologist member (comparison of the proportions, P=0.05). A guideline with radiologist authors was more likely to recommend routine screening (odds ratio=6.05, 95% confidence interval=0.57–∞, P=0.14). The proportion of primary care physicians on guideline panels recommending routine vs. nonroutine screening was significantly different (38% vs. 90% of authors; P=0.01). The odds of a recommendation in favor of routine screening were related to the number of recent publications on breast disease diagnosis and treatment by the lead guideline author (P=0.02).Conclusion: Recommendations regarding mammography screening in this target population may reflect the specialty and intellectual interests of the guideline authors.

Decisions about lumping vs. splitting of the scope of systematic reviews of complex interventions are not well justified: A case study in systematic reviews of health care professional reminders - Corrected Proof

Michelle C. Weir, Jeremy M. Grimshaw, Alain Mayhew, Dean Fergusson — 2012-04-13

Abstract: Objectives: Lumping and splitting refer to the scope of a systematic review question, where lumped reviews are broad and split are narrow. The objective was to determine the frequency of lumping and splitting in systematic reviews of reminder interventions, assess how review authors justified their decisions about the scope of their reviews, and explore how review authors cited other systematic reviews in the field.Study Design and Setting: A descriptive approach involving a content analysis and citation bibliometric study of an overview of 31 systematic reviews of reminder interventions.Results: Twenty-four of 31 reminder reviews were split, most frequently across one category (population, intervention, study design, outcome). Review authors poorly justified their decisions about the scope of their reviews and tended not to cite other similar reviews.Conclusion: This study demonstrates that for systematic reviews of reminder interventions, splitting is more common than lumping, with most reviews split by condition or targeted behavior. Review authors poorly justify the need for their review and do not cite relevant literature to put their reviews in the context of the available evidence. These factors may have contributed to a proliferation of systematic reviews of reminders and an overall disorganization of the literature.

The importance of measuring strength-of-preference scores for health care options in preference-sensitive care - Corrected Proof

R. Trafford Crump, Hilary A. Llewellyn-Thomas — 2012-04-11

Abstract: Objective: The objective was to determine whether a paired-comparison/Leaning Scale (LS) method: 1) could feasibly be used to elicit strength-of-preference scores for elective health care options in large community-based survey settings and 2) could reveal preferential subgroups that would have been overlooked if only a categorical-response format had been used.Study Design: Medicare beneficiaries in four different regions of the United States were interviewed in person. Participants considered eight clinical scenarios, each with two to three different health care options. For each scenario, participants categorically selected their favored option, then indicated how strongly they favored that option relative to the alternative on a paired-comparison bidirectional LS.Results: Two hundred two participants were interviewed. For seven of the eight scenarios, a clear majority (>50%) indicated that, overall, they categorically favored one option over the alternative(s). However, the bidirectional strength-of-preference LS scores revealed that, in four scenarios, for half of those participants, their preference for the favored option was actually “weak” or “neutral.”Conclusion: Investigators aiming to assess population-wide preferential attitudes toward different elective health care scenarios should consider gathering ordinal-level strength-of-preference scores and could feasibly use the paired-comparison/bidirectional LS to do so.

Optimization of literature searching - Corrected Proof

Josepha W.M. Plevier, Jan W. Schoones — 2012-04-11

Katchamart et al. advocate the importance of well-designed database-specific search strategies to support systematic reviews. As medical information specialists, we can endorse this conclusion.

Reporting of methods was better in the Clinical Trials Registry-India than in Indian journal publications - Corrected Proof

Prathap Tharyan, Aneesh Thomas George, Richard Kirubakaran, Jabez Paul Barnabas — 2012-03-29

Abstract: Objective: We sought to evaluate if editorial policies and the reporting quality of randomized controlled trials (RCTs) had improved since our 2004–05 survey of 151 RCTs in 65 Indian journals, and to compare reporting quality of protocols in the Clinical Trials Registry-India (CTRI).Study Design and Setting: An observational study of endorsement of Consolidated Standards for the Reporting of Trials (CONSORT) and International Committee of Medical Journal Editors (ICMJE) requirements in the instructions to authors in Indian journals, and compliance with selected requirements in all RCTs published during 2007–08 vs. our previous survey and between all RCT protocols in the CTRI on August 31, 2010 and published RCTs from both surveys.Results: Journal policies endorsing the CONSORT statement (22/67, 33%) and ICMJE requirements (35/67, 52%) remained suboptimal, and only 4 of 13 CONSORT items were reported in more than 50% of the 145 RCTs assessed. Reporting of ethical issues had improved significantly, and that of methods addressing internal validity had not improved. Adequate methods were reported significantly more frequently in 768 protocols in the CTRI, than in the 296 published trials.Conclusion: The CTRI template facilitates the reporting of valid methods in registered trial protocols. The suboptimal compliance with CONSORT and ICMJE requirements in RCTs published in Indian journals reduces credibility in the reliability of their results.

Introducing GRADE across the NICE clinical guideline program - Corrected Proof

2012-03-14

Abstract: Objectives: Grading of Recommendations Assessment, Development and Evaluation (GRADE) is a system for rating the confidence in estimates of effect and grading guideline recommendations. It promotes evaluation of the quality of the evidence for each outcome and an assessment of balance between desirable and undesirable outcomes leading to a judgment about the strength of the recommendation. In 2007, the National Institute for Health and Clinical Excellence began introducing GRADE across its clinical guideline program to enable separation of judgments about the evidence quality from judgments about the strength of the recommendation.Study Design and Setting: We describe the process of implementing GRADE across guidelines.Results: Use of GRADE has been positively received by both technical staff and guideline development group members.Conclusion: A shift in thinking about confidence in the evidence was required leading to a more structured and transparent approach to decision making. Practical problems were also encountered; these have largely been resolved, but some areas require further work, including the application of imprecision and presenting results from analyses considering more than two alternative interventions. The use of GRADE for nonrandomized and diagnostic accuracy studies needs to be refined.

Medical journal editor and scientific publication issue - Corrected Proof

Viroj Wiwanitkit — 2012-03-05

Sir, I read the recent report by Wong and Callaham with great interest. It is very interesting that a considerable portion of the medical editors have poor knowledge on publication ethics topics . The results in this study repeatedly confirm the finding in the previous study by Wager et al. . This finding is very important, and there is a need for improving the standards of medical editors.

E-mail invitations to general practitioners were as effective as postal invitations and were more efficient - Corrected Proof

2012-02-10

Abstract: Objective: To evaluate which of two invitation methods, e-mail or post, was most effective at recruiting general practitioners (GPs) to an online trial.Study Design and Setting: Randomized controlled trial. Participants were GPs in Scotland, United Kingdom.Results: Two hundred and seventy GPs were recruited. Using e-mail did not improve recruitment (risk difference=0.7% [95% confidence interval −2.7% to 4.1%]). E-mail was, however, simpler to use and cheaper, costing £3.20 per recruit compared with £15.69 for postal invitations. Reminders increased recruitment by around 4% for each reminder sent for both invitation methods.Conclusions: In the Scottish context, inviting GPs to take part in an online trial by e-mail does not adversely affect recruitment and is logistically easier and cheaper than using postal invitations.

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.