Sample size determination for bibliographic retrieval studies

BACKGROUND

Research for developing search strategies to retrieve high-quality clinical journal articles from MEDLINE is expensive and time-consuming. The objective of this study was to determine the minimal number of high-quality articles in a journal subset that would need to be hand-searched to update or create new MEDLINE search strategies for treatment, diagnosis, and prognosis studies.

METHODS

The desired width of the 95% confidence intervals (W) for the lowest sensitivity among existing search strategies was used to calculate the number of high-quality articles needed to reliably update search strategies. New search strategies were derived in journal subsets formed by 2 approaches: random sampling of journals and top journals (having the most high-quality articles). The new strategies were tested in both the original large journal database and in a low-yielding journal (having few high-quality articles) subset.

RESULTS

For treatment studies, if W was 10% or less for the lowest sensitivity among our existing search strategies, a subset of 15 randomly selected journals or 2 top journals were adequate for updating search strategies, based on each approach having at least 99 high-quality articles. The new strategies derived in 15 randomly selected journals or 2 top journals performed well in the original large journal database. Nevertheless, the new search strategies developed using the random sampling approach performed better than those developed using the top journal approach in a low-yielding journal subset. For studies of diagnosis and prognosis, no journal subset had enough high-quality articles to achieve the expected W (10%).

CONCLUSION

The approach of randomly sampling a small subset of journals that includes sufficient high-quality articles is an efficient way to update or create search strategies for high-quality articles on therapy in MEDLINE. The concentrations of diagnosis and prognosis articles are too low for this approach.

Radiographs of hip fractures were digitally altered to mask surgeons to the type of implant without compromising the reliability of quality ratings or making the rating process more difficult

OBJECTIVE

To devise and test techniques to blind outcome assessors in trials of hip fracture fixation.

STUDY DESIGN AND SETTING

We developed three techniques (Blackout, Subtraction, and Overlay) to mask radiographs of hip fractures fixated with cancellous screws or dynamic hip screws. Fifty orthopedic trauma surgeons each assessed 32 radiographs blinded with each technique.

RESULTS

All techniques achieved low rates of correct identification of screw type (14.9% for Blackout, 26.9% for Subtraction, 22.1% for Overlay) and high proportions of "don't know" responses (72.3%, 48.4%, 52.8%, respectively). The interrater reliability of reduction quality in the blinded images (intraclass correlation coefficient [ICC]=0.55-0.57) was similar to the reliability of the unblinded radiographs (ICC=0.60). Surgeons perceived 6.9% of the Overlay images as much more difficult to rate than unblinded radiographs, compared with 9.7% of Subtraction images (P=0.25) and 28.0% of Blackout images (P<0.001).

CONCLUSION

Three techniques of blinding radiographs of femoral neck fractures successfully mask surgeons to the type of implant fixated, do not compromise reliability of reduction ratings, and do not make rating most radiographs more difficult. Trialists should explore creative approaches to optimize blinding when designing trials, and should incorporate rigorous approaches to testing blinding success.

Use of latent class models to accommodate inter-laboratory variation in assessing genetic polymorphisms associated with disease risk

Background

Researchers wanting to study the association of genetic factors with disease may encounter variability in the laboratory methods used to establish genotypes or other traits. Such variability leads to uncertainty in determining the strength of a genotype as a risk factor. This problem is illustrated using data from a case-control study of cervical cancer in which some subjects were independently assessed by different laboratories for the presence of a genetic polymorphism. Inter-laboratory agreement was only moderate, which led to a very wide range of empirical odds ratios (ORs) with the disease, depending on how disagreements were treated.

This paper illustrates the use of latent class models (LCMs) and to estimate OR while taking laboratory accuracy into account. Possible LCMs are characterised in terms of the number of laboratory measurements available, and if their error rates are assumed to be differential or non-differential by disease status and/or laboratory.

Results

The LCM results give maximum likelihood estimates of laboratory accuracy rates and the OR of the genetic variable and disease, and avoid the ambiguities of the empirical results. Having allowed for possible measurement error in the expure, the LCM estimates of exposure – disease associations are typically stronger than their empirical equivalents. Also the LCM estimates exploit all the available data, and hence have relatively low standard errors.

Conclusion

Our approach provides a way to evaluate the association of a polymorphism with disease, while taking laboratory measurement error into account. Ambiguities in the empirical data arising from disagreements between laboratories are avoided, and the estimated polymorphism-disease association is typically enhanced.

Stop orders to reduce inappropriate urinary catheterization in hospitalized patients: a randomized controlled trial

BACKGROUND

Hospitalized patients frequently have urinary catheters inserted for inappropriate reasons. This can lead to urinary tract infections and other complications.

OBJECTIVE

To assess whether stop orders for indwelling urinary catheters reduces the duration of inappropriate urinary catheterization and the incidence of urinary tract infections.

DESIGN

A randomized controlled trial was conducted in three tertiary-care hospitals in Ontario, Canada. Patients with indwelling urinary catheters were randomized to prewritten orders for the removal of urinary catheters if specified criteria were not present or to usual care.

PARTICIPANTS

Six hundred ninety-two hospitalized patients admitted to hospital with indwelling urinary catheters inserted for < or = 48 h.

MEASUREMENTS

The main outcomes included days of inappropriate indwelling catheter use, total days of catheter use, frequency of urinary tract infection, and catheter reinsertions.

RESULTS

There were fewer days of inappropriate and total urinary catheter use in the stop-order group than in the usual care group (difference -1.69 [95% CI -1.23 to -2.15], P < 0.001 and -1.34 days, [95% CI, -0.64 to -2.05 days], P < 0.001, respectively). Urinary tract infections occurred in 19.0% of the stop-order group and 20.2% of the usual care group, relative risk 0.94 (95% CI, 0.66 to 1.33), P = 0.71. Catheter reinsertion occurred in 8.6% of the stop-order group and 7.0% in the usual care group, relative risk 1.23 (95% CI, 0.72 to 2.11), P = 0.45.

CONCLUSIONS

Stop orders for urinary catheterization safely reduced duration of inappropriate urinary catheterization in hospitalized patients but did not reduce urinary tract infections.

Blinding of outcomes in trials of orthopaedic trauma: an opportunity to enhance the validity of clinical trials

BACKGROUND

Blinding personnel in randomized controlled trials is an important strategy to minimize bias and increase the validity of the results. Trials of surgical interventions present blinding challenges not seen in drug trials. How often orthopaedic trauma investigators undertake blinding, and the frequency with which they could potentially utilize blinding, remains uncertain.

METHODS

We conducted a systematic review of all randomized controlled trials of orthopaedic trauma published from 1995 to 2004. Two reviewers assessed each trial for eligibility and extracted data regarding its characteristics, outcomes, reporting of blinding, and feasibility of blinding.

RESULTS

We included 171 unique randomized controlled trials spanning a variety of body regions and interventions. The most commonly reported outcomes were clinical (e.g., mortality or wound infection; 91% of trials), radiographic (83%), patient-reported (66%), and physiological results (e.g., range of motion; 56%). Less than 10% of the trials in each category reported the use of blinded outcome assessors. This contrasted with blinding that investigators could have accomplished: blinding was feasible with use of simple methods such as independent assessors, concealed incisions, and masked radiographs for 89% of clinical assessors, 89% of radiographic assessors, 96% of physiological assessors, and 35% of patient-reported assessors.

CONCLUSIONS

Trials in orthopaedic trauma typically measure many outcomes requiring judgment, but the individuals assessing those outcomes are seldom blinded. Investigators have the opportunity to enhance the validity of future clinical trials by incorporating simple blinding techniques.

Comparing estimates of influenza-associated hospitalization and death among adults with congestive heart failure based on how influenza season is defined

BACKGROUND

There is little consensus about how the influenza season should be defined in studies that assess influenza-attributable risk. The objective of this study was to compare estimates of influenza-associated risk in a defined clinical population using four different methods of defining the influenza season.

METHODS

Using the Studies of Left Ventricular Dysfunction (SOLVD) clinical database and national influenza surveillance data from 1986-87 to 1990-91, four definitions were used to assess influenza-associated risk: (a) three-week moving average of positive influenza isolates is at least 5%, (b) three-week moving average of positive influenza isolates is at least 10%, (c) first and last positive influenza isolate are identified, and (d) 5% of total number of positive isolates for the season are obtained. The clinical data were from adults aged 21 to 80 with physician-diagnosed congestive heart failure. All-cause hospitalization and all-cause mortality during the influenza seasons and non-influenza seasons were compared using four definitions of the influenza season. Incidence analyses and Cox regression were used to assess the effect of exposure to influenza season on all-cause hospitalization and death using all four definitions.

RESULTS

There was a higher risk of hospitalization associated with the influenza season, regardless of how the start and stop of the influenza season was defined. The adjusted risk of hospitalization was 8 to 10 percent higher during the influenza season compared to the non-influenza season when the different definitions were used. However, exposure to influenza was not consistently associated with higher risk of death when all definitions were used. When the 5% moving average and first/last positive isolate definitions were used, exposure to influenza was associated with a higher risk of death compared to non-exposure in this clinical population (adjusted hazard ratios [HR], 1.16; 95% confidence interval [CI], 1.04 to 1.29 and adjusted HR, 1.19; 95% CI, 1.06 to 1.33, respectively).

CONCLUSION

Estimates of influenza-attributable risk may vary depending on how influenza season is defined and the outcome being assessed.

An opportunity cost approach to sample size calculation in cost-effectiveness analysis

The inclusion of economic evaluations as part of clinical trials has led to concerns about the adequacy of trial sample size to support such analysis. The analytical tool of cost-effectiveness analysis is the incremental cost-effectiveness ratio (ICER), which is compared with a threshold value (lambda) as a method to determine the efficiency of a health-care intervention. Accordingly, many of the methods suggested to calculating the sample size requirements for the economic component of clinical trials are based on the properties of the ICER. However, use of the ICER and a threshold value as a basis for determining efficiency has been shown to be inconsistent with the economic concept of opportunity cost. As a result, the validity of the ICER-based approaches to sample size calculations can be challenged. Alternative methods for determining improvements in efficiency have been presented in the literature that does not depend upon ICER values. In this paper, we develop an opportunity cost approach to calculating sample size for economic evaluations alongside clinical trials, and illustrate the approach using a numerical example. We compare the sample size requirement of the opportunity cost method with the ICER threshold method. In general, either method may yield the larger required sample size. However, the opportunity cost approach, although simple to use, has additional data requirements. We believe that the additional data requirements represent a small price to pay for being able to perform an analysis consistent with both concept of opportunity cost and the problem faced by decision makers.

Developing common metrics of mechanical exposures across aetiological studies of low back pain in working populations for use in meta-analysis

OBJECTIVES

One of the challenges of conducting meta-analyses on the relationship between workplace mechanical exposures and low back pain is that mechanical exposures are reported in a wide variety of ways. We aimed to develop common metrics to apply in the translation of literature-based workplace mechanical exposures for use in meta-analyses, and to test the metrics' measurement properties.

METHODS

We developed a set of 7-point scales to capture the intensity of important aspects of mechanical exposures that may be related to the development of low back pain in workers. The scales represented three dimensions of mechanical exposures at work: (1) trunk posture, (2) weight lifted or force exerted and (3) spinal loading, and estimated both peak and cumulative loads. Measurement properties of the scales were tested through a survey of experts in biomechanics and ergonomics who were asked to rate literature-based workplace exposure definitions using the scales and provide estimates of their confidence in their ratings.

RESULTS

For each dimension the ratings for peak loads tended to be higher than the cumulative load ratings. The inter-rater reliability for the scales ranged from 0.3 to 0.5; we would need to average the ratings of at least four expert raters to have an acceptable level of reliability (>0.7). Inter-expert reliability was positively related to the experts' level of confidence in their ratings. In most cases the ranking of intensity ratings from the experts matched the ranking of exposure intensity from the original articles.

CONCLUSIONS

This study provides insight into estimating the intensity of literature-based mechanical exposure metrics using a common set of scales which can be applied across epidemiologic studies. These metrics may be useful to quantify the relationship between workplace mechanical exposure and low back pain in a systematic review and meta-analysis.

Human papillomavirus DNA versus Papanicolaou screening tests for cervical cancer

BACKGROUND

To determine whether testing for DNA of oncogenic human papillomaviruses (HPV) is superior to the Papanicolaou (Pap) test for cervical-cancer screening, we conducted a randomized trial comparing the two methods.

METHODS

We compared HPV testing, using an assay approved by the Food and Drug Administration, with conventional Pap testing as a screening method to identify high-grade cervical intraepithelial neoplasia in women ages 30 to 69 years in Montreal and St. John's, Canada. Women with abnormal Pap test results or a positive HPV test (at least 1 pg of high-risk HPV DNA per milliliter) underwent colposcopy and biopsy, as did a random sample of women with negative tests. Sensitivity and specificity estimates were corrected for verification bias.

RESULTS

A total of 10,154 women were randomly assigned to testing. Both tests were performed on all women in a randomly assigned sequence at the same session. The sensitivity of HPV testing for cervical intraepithelial neoplasia of grade 2 or 3 was 94.6% (95% confidence interval [CI], 84.2 to 100), whereas the sensitivity of Pap testing was 55.4% (95% CI, 33.6 to 77.2; P=0.01). The specificity was 94.1% (95% CI, 93.4 to 94.8) for HPV testing and 96.8% (95% CI, 96.3 to 97.3; P<0.001) for Pap testing. Performance was unaffected by the sequence of the tests. The sensitivity of both tests used together was 100%, and the specificity was 92.5%. Triage procedures for Pap or HPV testing resulted in fewer referrals for colposcopy than did either test alone but were less sensitive. No adverse events were reported.

CONCLUSIONS

As compared with Pap testing, HPV testing has greater sensitivity for the detection of cervical intraepithelial neoplasia. (Current Controlled Trials number, ISRCTN57612064 [controlled-trials.com].).

Focus on Function - a randomized controlled trial comparing two rehabilitation interventions for young children with cerebral palsy

BACKGROUND

Children with cerebral palsy receive a variety of long-term physical and occupational therapy interventions to facilitate development and to enhance functional independence in movement, self-care, play, school activities and leisure. Considerable human and financial resources are directed at the "intervention" of the problems of cerebral palsy, although the available evidence supporting current interventions is inconclusive. A considerable degree of uncertainty remains about the appropriate therapeutic approaches to manage the habilitation of children with cerebral palsy. The primary objective of this project is to conduct a multi-site randomized clinical trial to evaluate the efficacy of a task/context-focused approach compared to a child-focused remediation approach in improving performance of functional tasks and mobility, increasing participation in everyday activities, and improving quality of life in children 12 months to 5 years of age who have cerebral palsy.

METHOD/DESIGN

A multi-centred randomized controlled trial research design will be used. Children will be recruited from a representative sample of children attending publicly-funded regional children's rehabilitation centers serving children with disabilities in Ontario and Alberta in Canada. Target sample size is 220 children with cerebral palsy aged 12 months to 5 years at recruitment date. Therapists are randomly assigned to deliver either a context-focused approach or a child-focused approach. Children follow their therapist into their treatment arm. Outcomes will be evaluated at baseline, after 6 months of treatment and at a 3-month follow-up period. Outcomes represent the components of the International Classification of Functioning, Disability and Health, including body function and structure (range of motion), activities (performance of functional tasks, motor function), participation (involvement in formal and informal activities), and environment (parent perceptions of care, parental empowerment).

DISCUSSION

This paper presents the background information, design and protocol for a randomized controlled trial comparing a task/context-focused approach to a child-focused remediation approach in improving functional outcomes for young children with cerebral palsy.

TRIAL REGISTRATION

[clinical trial registration #: NCT00469872].

Effect sizes can be calculated for studies reporting ranges for outcome variables in systematic reviews

OBJECTIVE

To develop a method by which studies reporting the ranges (or maxima and minima) of observed outcomes can be included in systematic reviews, along with other studies reporting in the more usual way by using standard deviations (SDs).

STUDY DESIGN AND SETTING

An approach is proposed to allow a numerical conversion of a reported range from a continuous outcome into an equivalent SD.

RESULTS

The SD is estimated from the observed range times an appropriate conversion factor. Two examples (the first concerning a patient education program on adherence to drug treatment for rheumatoid arthritis, and the second investigating if cognitive behavior therapy could improve adherence to antiretroviral therapy and then lead to suppression of human immunodeficiency virus replication) demonstrate the calculations.

CONCLUSION

This note provides a simple method to allow studies that report outcome variability in terms of ranges to be included in systematic reviews by conversion to equivalent SDs. The method is entirely valid if the outcome variable is normally distributed, but for nonnormal data, some caution may be needed.

Methodologic discussions for using and interpreting composite endpoints are limited, but still identify major concerns

OBJECTIVE

To investigate the rationale, potential problems and solutions of using composite endpoints (CEPs) for the assessment of intervention effects.

STUDY DESIGN AND SETTING

This study is a systematic review. We searched MEDLINE, EMBASE, and the Science Citation Index, for publications appearing between 1980 and September 2005, and reviewed potentially informative textbooks. Eligible articles provided a commentary, analysis, or discussion of CEPs for any of the following areas: (1) rationale, (2) interpretation or meaning, (3) advantages, (4) limitations or conceptual problems, and (5) recommendations for use.

RESULTS

Seventeen articles and one textbook proved eligible. Decreases in sample size requirements and ability to assess the net effect of an intervention were the most commonly cited advantages. Authors noted the risk of misinterpretation when heterogeneity among components with respect to either patient importance or magnitude of treatment effects as the most salient disadvantage. There were discrepancies between authors concerning the usefulness of CEPs to avoid bias from competing risks and when the direction of the effect of therapy differs across components.

CONCLUSION

Methodologists have given limited attention to CEPs and their views are sometimes contradictory. Further work is needed to establish the role of CEPs in research and in guiding clinical practice.

Low-back pain definitions in occupational studies were categorized for a meta-analysis using Delphi consensus methods

OBJECTIVE

To determine which literature-based definitions of low back pain (LBP) could be combined to produce sufficiently similar sets for use in a meta-analysis.

STUDY DESIGN AND SETTING

A group of six international experts participated in an e-mail-administered Delphi process. Literature-based LBP definitions were preliminarily classified into 14 sets within four outcome types: pathology, symptoms and care-seeking, functional limitations, and participation. Experts independently rated their level of agreement that each outcome definition belonged in its assigned set using a seven-point Likert scale. After each round, results were synthesized and revised classifications were fed back to the experts who were asked to consider them before rerating the outcome definitions.

RESULTS

The experts completed three Delphi rounds and reached consensus on the categorization of 115/119 (97%) of the outcome definitions. There were 20 final sets of outcomes identified: three sets of pathology outcomes, two sets each of functional limitation and participation outcomes, and 13 sets of symptom and care-seeking outcomes.

CONCLUSIONS

In a research area that currently lacks uniformly accepted definitions of outcomes, we successfully used a Delphi consensus process to reach substantial agreement on combinable LBP outcomes that would be combinable for a meta-analysis.

Risk of hospitalization during influenza season among a cohort of patients with congestive heart failure

It is uncertain whether hospitalization among patients with congestive heart failure (CHF) increases during the influenza season. This retrospective cohort study used influenza surveillance data from the United States (1986-1987 to 1990-1991), clinical information from the Studies of Left Ventricular Dysfunction (SOLVD) database, and daily temperature data from the National Climatic Data Center to assess the effect of influenza season on hospitalizations in this cohort of patients. The overall hospitalization rate was higher during influenza seasons compared to non-influenza seasons [relative risk (RR) 1.08, 95% confidence interval (CI) 1.01-1.16]. Multivariable Cox modelling revealed an adjusted hazard ratio (HR) of 1.11 for hospitalization during the influenza season (95% CI 1.03-1.20, P=0.005). Overall death rates were also higher during influenza seasons than non-influenza seasons (RR 1.09, 95% CI 0.97-1.21), but the corresponding adjusted HR for death was not significant (HR 1.01, 95% CI 0.98-1.24, P=0.11). Patients with CHF have a greater risk of hospitalization during the influenza season than in the non-influenza season, supporting the current belief that patients with CHF should be regarded as a high-risk group.

Pneumonia and Lower Respiratory Infections in Nursing Home Residents: Predictors of Hospitalization and Mortality

OBJECTIVES

To compare predictors of hospitalization and death in nursing home residents with pneumonia and other lower respiratory infections (LRIs).

DESIGN

A nested cohort study.

SETTING

Nine nursing homes in southern Ontario.

PARTICIPANTS

Three hundred fifty-three nursing home residents with LRIs (enrolled in the control arm of a clinical trial).

MEASUREMENTS

Comorbidities, vaccination status, age, health-related quality of life, functional status, and vital statistics were evaluated as potential predictors of hospitalization and mortality at 30 days.

RESULTS

Moderate to high disease severity score on a practical severity scale was a strong independent predictor of hospitalization (odds ratio (OR)=7.12, P<.001) and mortality (OR=5.04, P=.003). Diagnosis of pneumonia, established using chest radiograph, was also associated with hospitalization (OR=2.43, P=.008) and mortality (OR=2.35, P=.02). Oxygen saturation (<90%) was a strong independent predictor of hospitalization (OR=3.02, P=.004) but was not a significant predictor of mortality in multivariable analyses. Diagnosis of congestive heart failure (OR=2.26, P=.02) was an independent predictor of hospitalization, whereas receipt of pneumococcal vaccine (OR=0.36, P=.01) and greater functional independence (OR=0.92, P=.02) were negatively associated with hospitalization.

CONCLUSION

In nursing home residents with LRI, severity of illness and radiographically confirmed pneumonia are predictive of death and hospitalization.

Estimation, power and sample size calculations for stochastic cost and effectiveness analysis

Various methods have been proposed to address uncertainty in economic evaluations of healthcare programmes. One approach suggested in the literature is to estimate separate confidence intervals for the incremental costs and effects of a new health programme in comparison with an existing programme. These intervals are then combined to generate a rectangular confidence region in the cost-effectiveness plane that implicitly defines a corresponding confidence interval for the incremental cost-effectiveness ratio (ICER). The same approach has been used to calculate sample sizes and study power. This application of the rectangle method is consistent with the adoption of ICERs and a threshold as a decision rule, this being the most commonly used approach in empirical applications of cost-effectiveness analysis, as well as the one recommended by agencies that assess medical technology around the world. In this paper, we first outline the rectangle method, and then propose a modification that recognises that separate inferences are being drawn on the cost and effectiveness domains, and that corrects for multiple statistical comparisons. The confidence rectangle is otherwise too small, the corresponding confidence interval for the ICER is too narrow and sample sizes are under-estimated. Our modification corrects these problems. A further difficulty is that the placement of the confidence rectangle around the null value is somewhat arbitrary, and does not correspond to a unique value of ICERs. As a result, different values of sample size and power for the estimation of ICERs can be obtained, depending on the null values of the cost and effectiveness. We conclude that it is important to clearly identify the analytic goal in terms of estimating differential costs, differential effects or a combination of the two using the ICER index. These ideas are illustrated using numerical examples.

Effect of exposure misclassification on the mean squared error of population attributable risk and prevented fraction estimates

Previous work has considered the effect of exposure misclassification on the bias of population attributable risk (AR) estimates, but little is known about the corresponding effects on their precision or mean squared error (MSE). This paper considers AR estimation in typical scenarios for case-control and cohort studies. The analogous index used when exposure reduces the risk--the prevented fraction (PF)--is also investigated. It is shown, through both theoretical and simulation results, that even with quite modest levels of exposure misclassification, the MSE can increase substantially, relative to the variance of AR estimated without measurement error. When exposure assessment is perfectly sensitive, there is no bias in AR but lack of measurement specificity can still cause a substantial loss of precision. In a few cases, the AR or PF with misclassified exposure can actually have smaller MSE; these exceptional cases arise when sensitivity is poor and the bias in AR or PF is relatively large. We conclude that while bias can be reduced by defining exposure on a highly sensitive basis, one must also consider the deleterious effect on precision by doing so. Loss of precision in the AR and PF estimates can be safely ignored only when the exposure measure is very accurate.

Studies reporting ROC curves of diagnostic and prediction data can be incorporated into meta-analyses using corresponding odds ratios

OBJECTIVE

To develop an approach by which studies describing the accuracy of diagnostic tests or clinical predictions can be combined in a meta-analysis, even though studies may report their results using different summary measures.

STUDY DESIGN

A method is proposed to allow algebraic and numerical conversion of values of the Receiver Operating Characteristic Area Under the Curve (AUC) summary statistic into corresponding odds ratios (OR). A similar conversion is demonstrated for the standard errors (SEs) of these summary statistics.

RESULTS

The conversion of the AUC values into OR values was achieved using a logit-threshold model. The delta method was used to convert the associated SEs. An example concerning predictions of mortality in the intensive care unit illustrates the calculations.

CONCLUSION

This paper provides an accessible method that permits the meta-analyst to overcome some of the difficulties implied by incomplete and inconsistent reporting of research studies in this area. It allows all studies to be included on the same metric, which in turn more easily permits exploration of issues such as heterogeneity. The method can readily be used for meta-analyses of diagnostic or screening tests, or for prediction data.

Structural equation and log-linear modeling: a comparison of methods in the analysis of a study on caregivers' health

BACKGROUND

In this paper we compare the results in an analysis of determinants of caregivers' health derived from two approaches, a structural equation model and a log-linear model, using the same data set.

METHODS

The data were collected from a cross-sectional population-based sample of 468 families in Ontario, Canada who had a child with cerebral palsy (CP). The self-completed questionnaires and the home-based interviews used in this study included scales reflecting socio-economic status, child and caregiver characteristics, and the physical and psychological well-being of the caregivers. Both analytic models were used to evaluate the relationships between child behaviour, caregiving demands, coping factors, and the well-being of primary caregivers of children with CP.

RESULTS

The results were compared, together with an assessment of the positive and negative aspects of each approach, including their practical and conceptual implications.

CONCLUSION

No important differences were found in the substantive conclusions of the two analyses. The broad confirmation of the Structural Equation Modeling (SEM) results by the Log-linear Modeling (LLM) provided some reassurance that the SEM had been adequately specified, and that it broadly fitted the data.

Patients undergoing knee surgery provided accurate ratings of preoperative quality of life and function 2 weeks after surgery

OBJECTIVE

To evaluate patients' ability to recall their preoperative self-reported quality of life, function, and general health 2 weeks postoperatively.

STUDY DESIGN AND SETTING

We randomized consecutive patients undergoing arthroscopic knee surgery to group I (assessments at 4 weeks preoperatively, on the day of surgery, and 2 weeks and 12 months postoperatively) or group II (assessments at 2 weeks and 12 months postoperatively). At each visit patients completed disease-specific, knee-specific, and generic health rating scales. At a median of 2 weeks postoperative (range, 0.6 to 14 weeks), patients completed questionnaires according to their recollection of their health 2 weeks before surgery.

RESULTS

Agreement between actual and recalled data was excellent for disease-specific (ICC(WOMET)=0.88 (95% CI 0.82-0.91), ICC(ACL-QOL)=0.86 (95% CI 0.75-0.91)), knee-specific (ICC(IKDC)=0.92 (95% CI 0.90-0.94), ICC(KOOS)=0.93 (95% CI 0.91 to 0.95), and general physical health (ICC(SF-36(PCS))=0.81 (95% CI 0.75-0.86)) instruments. Agreement for general mental health was moderate (ICC(SF-36(MCS))=0.67 (95% CI 0.58-0.75). Greater error associated with recalled ratings contributed to small increases in sample size requirements or small decreases in power to detect differences between groups.

CONCLUSION

Patients recalled their preoperative quality of life, function, and general health at 2 weeks postoperative with sufficiently high accuracy to warrant substituting prospectively collected baseline data with recalled ratings.

McMaster PLUS: a cluster randomized clinical trial of an intervention to accelerate clinical use of evidence-based information from digital libraries

BACKGROUND

Physicians have difficulty keeping up with new evidence from medical research.

METHODS

We developed the McMaster Premium LiteratUre Service (PLUS), an internet-based addition to an existing digital library, which delivered quality- and relevance-rated medical literature to physicians, matched to their clinical disciplines. We evaluated PLUS in a cluster-randomized trial of 203 participating physicians in Northern Ontario, comparing a Full-Serve version (that included alerts to new articles and a cumulative database of alerts) with a Self-Serve version (that included a passive guide to evidence-based literature). Utilization of the service was the primary trial end-point.

RESULTS

Mean logins to the library rose by 0.77 logins/month/user (95% CI 0.43, 1.11) in the Full-Serve group compared with the Self-Serve group. The proportion of Full-Serve participants who utilized the service during each month of the study period showed a sustained increase during the intervention period, with a relative increase of 57% (95% CI 12, 123) compared with the Self-Serve group. There were no differences in these proportions during the baseline period, and following the crossover of the Self-Serve group to Full-Serve, the Self-Serve group's usage became indistinguishable from that of the Full-Serve group (relative difference 4.4 (95% CI -23.7, 43.0). Also during the intervention and crossover periods, measures of self-reported usefulness did not show a difference between the 2 groups.

CONCLUSION

A quality- and relevance-rated online literature service increased the utilization of evidence-based information from a digital library by practicing physicians.

Randomized controlled trial of human papillomavirus testing versus Pap cytology in the primary screening for cervical cancer precursors: Design, methods and preliminary accrual results of the Canadian cervical cancer screening trial (CCCaST)

Since infection with oncogenic human papillomavirus (HPV) has been considered a necessary cause of cervical cancer, tests for oncogenic HPV types have been proposed as adjuncts or replacements to Pap cytology. We designed the Canadian Cervical Cancer Screening Trial (CCCaST) to compare the relative efficacy of HPV DNA testing and Pap cytology in primary screening for cervical cancer and its high-grade precursors. CCCaST randomized women aged 30-69 years in Montreal (Quebec) and in St. John's (Newfoundland) to 1 of 2 screening groups: focus on Pap (conventional) or focus on HPV testing (Hybrid Capture 2). Women in both arms received both tests, but were randomized as to their order, the first test being the index test. Women with an abnormal Pap test or a positive HPV test underwent colposcopy and biopsy, as did a random sample of women with a negative index test. CCCaST enrolled 9,667 women between October 2002 and October 2004. At enrolment, 2.8% had an abnormal Pap test, 6.1% had a positive HPV test and 1.1% were abnormal in both tests. ASC-US was the most frequent cytological abnormality, representing 64% of abnormal Pap results. The frequency of abnormal Pap and HPV results decreased with increasing age and the proportion of HPV-positive results increased with the severity of Pap abnormality. Efficacy analysis will determine if the extra referrals with HPV DNA testing will translate into a relevant increase in high-grade cervical cancer precursor detection. Because of its design, CCCaST will provide sound evidence for formulating cervical cancer screening strategies.

Vasopressin or Epinephrine for Out-of-Hospital Cardiac Arrest

STUDY OBJECTIVE

The use of vasopressin in patients with cardiac arrest presenting with specific rhythms is controversial. We performed an evidence-based emergency medicine review of evidence comparing vasopressin to epinephrine in structured cardiac arrest protocols.

METHODS

We searched MEDLINE, EMBASE, the Cochrane Library, and other databases for randomized trials or systematic reviews comparing vasopressin to epinephrine for adults with cardiac arrest and measuring survival to hospital discharge and neurologic function in survivors. We used standard criteria to appraise the quality of published trials and systematic reviews. We used the random effects model in supplementary analyses to summarize results and to test for significant differences across subgroups of patients presenting with different arrest rhythms.

RESULTS

We found 3 high-quality well-reported randomized trials and 1 rigorous meta-analysis. The evidence does not confirm a consistent benefit of vasopressin over epinephrine in increasing survival or improving neurologic outcome in survivors. Subgroup analysis reveals a large difference in effect of vasopressin over epinephrine in cardiac arrest patients with asystole, compared to other arrest rhythms, coming from within-trial comparisons. The difference is not consistent across otherwise similar trials, is not statistically significant, may reflect the application of multiple unplanned subgroup analyses, and is not supported by a plausible biological hypothesis.

CONCLUSION

Evidence from randomized trials does not establish a benefit of vasopressin over epinephrine in increasing survival to discharge or improving neurologic outcomes in adult patients with nontraumatic cardiac arrest.