Cluster randomized trials (CRTs) pose ethical challenges for investigators and ethics committees. This study describes the views and experiences of CRT researchers with respect to: (1) ethical challenges in CRTs; (2) the ethics review process for CRTs; and (3) the need for comprehensive ethics guidelines for CRTs.
Descriptive qualitative analysis of interviews conducted with a purposive sample of 20 experienced CRT researchers.
Informants expressed concern over the potential for bias that may result from requirements to obtain informed consent from research participants in CRTs. Informants suggested that the need for informed consent ought to be related to the type of intervention under study in a CRT. Informants rarely expressed concern regarding risks to research participants in CRTs, other than risks to privacy. Important issues identified in the research ethics literature, including fair subject selection and other justice issues, were not mentioned by informants. The ethics review process has had positive and negative impacts on CRT conduct. Informants stated that variability in ethics review between jurisdictions, and increasingly stringent ethics review in recent years, have hampered their ability to conduct CRTs. Many informants said that comprehensive ethics guidelines for CRTs would be helpful to researchers and research ethics committees.
Informants identified key ethical challenges in the conduct of CRTs, specifically relating to identifying subjects, seeking informed consent, and the use of gatekeepers. These data have since been used to identify topics for in-depth ethical analysis and to guide the development of comprehensive ethics guidelines for CRTs.
Cluster randomized trials; Research ethics; Informed consent; Clinical trials; Bioethics; Knowledge translation; Quality improvement; Implementation research
Food insecurity is a significant public health problem in North America and elsewhere. The prevalence of food insecurity varies by country of residence; within countries, it is strongly associated with household socioeconomic status, but the local environment may also play an important role. In this study, we analyzed secondary data from a population-based survey conducted in Québec, Canada, to determine if five local environmental factors: material and social deprivation, social cohesion, disorder, and living location were associated with changes in household food insecurity over a period of 6 years, while adjusting for household socioeconomic status (SES) and other factors.
Data from the Québec Longitudinal Study of Child Development, following same-aged children from 4–10 y of age, were analyzed using generalized estimating equations, to determine the longitudinal association between these environmental factors and food insecurity over a period of 6 years.
Of the 2120 children originally included in the cohort, 1746 (82%) were included in the present analysis. The prevalence of food insecurity was 9.2% when children were 4 y of age (95% CI: 7.8 – 10.6%) but no significant changes were observed over time. On average over the 6 year period, three environmental factors were positively related to food insecurity: high social deprivation (OR 1.62, 95%CI: 1.16 – 2.26), low social cohesion (OR 1.45 95%CI: 1.10 – 1.92), and high disorder (OR 1.76, 95%CI: 1.37 – 2.27), while living location and material deprivation were not related to food insecurity. These associations were independent of household SES and other social variables.
These results highlight the potential role of the local social environment in preventing and ameliorating food insecurity at the household level. Stakeholders providing food security interventions at the community level should consider interactions with local social characteristics and perhaps changing the social environment itself. Further intervention research also examining interactions with household-level factors could lead to the development of interventions that increase both household and community-level food security.
Food insecurity; Social capital; Social cohesion; Disorder; Deprivation; Neighbourhood; Longitudinal study; Environment; Context
The Ottawa Ethics of Cluster Trials Consensus Group sets out 15 recommendations for the ethical design and conduct of cluster randomized trials.
To determine the association between local environmental factors with child weight status in a longitudinal study, using a semi-parametric, group-based method, while also considering social and early life factors.
Standardized, directly measured BMI from 4–10 y of age, and group-based trajectory modeling (PROC TRAJ) were used to estimate developmental trajectories of weight change in a Québec birth cohort (n = 1,566). Associations between the weight trajectories and living location, social cohesion, disorder, and material and social deprivation were estimated after controlling for social and early life factors.
Four weight trajectory groups were estimated: low-increasing (9.7%); low-medium, accelerating (36.2%); medium-high, increasing (43.0%); and high-stable (11.1%). In the low-increasing and medium-high trajectory groups, living in a semi-urban area was inversely related to weight, while living in a rural area was positively related to weight in the high-stable group. Disorder was inversely related to weight in the low-increasing group only. Other important risk factors for high-stable weight included obesity status of the mother, smoking during pregnancy, and overeating behaviors.
In this study, associations between local environment factors and weight differed by trajectory group. Early life factors appear to play a more consistent role in weight status. Further work is needed to determine the influence of place on child weight.
Primary care plays a key role in the prevention and management of cardiovascular disease (CVD). We examined primary care practice adherence to recommended care guidelines associated with the prevention and management of CVD for high risk patients.
We conducted a secondary analysis of cross-sectional baseline data collected from 84 primary care practices participating in a large quality improvement initiative in Eastern Ontario from 2008 to 2010. We collected medical chart data from 4,931 patients who either had, or were at high risk of developing CVD to study adherence rates to recommended guidelines for CVD care and to examine the proportion of patients at target for clinical markers such as blood pressure, lipid levels and hemoglobin A1c.
Adherence to preventive care recommendations was poor. Less than 10% of high risk patients received a waistline measurement, half of the smokers received cessation advice, and 7.7% were referred to a smoking cessation program. Gaps in care exist for diabetes and kidney disease as 54.9% of patients with diabetes received recommended hemoglobin-A1c screenings, and only 55.8% received an albumin excretion test. Adherence rates to recommended guidelines for coronary artery disease, hypertension, and dyslipidemia were high (>75%); however <50% of patients were at target for blood pressure or LDL-cholesterol levels (37.1% and 49.7% respectively), and only 59.3% of patients with diabetes were at target for hemoglobin-A1c.
There remain significant opportunities for primary care providers to engage high risk patients in prevention activities such as weight management and smoking cessation. Despite high adherence rates for hypertension, dyslipidemia, and coronary artery disease, a significant proportion of patients failed to meet treatment targets, highlighting the complexity of caring for people with multiple chronic conditions.
Cardiovascular disease; Primary care; Diabetes; Evidence-based care; Preventive care; Quality of care
This article is part of a series of papers examining ethical issues in cluster randomized trials (CRTs) in health research. In the introductory paper in this series, we set out six areas of inquiry that must be addressed if the CRT is to be set on a firm ethical foundation. This paper addresses the sixth of the questions posed, namely, what is the role and authority of gatekeepers in CRTs in health research? ‘Gatekeepers’ are individuals or bodies that represent the interests of cluster members, clusters, or organizations. The need for gatekeepers arose in response to the difficulties in obtaining informed consent because of cluster randomization, cluster-level interventions, and cluster size. In this paper, we call for a more restrictive understanding of the role and authority of gatekeepers.
Previous papers in this series have provided solutions to the challenges posed by informed consent in CRTs without the need to invoke gatekeepers. We considered that consent to randomization is not required when cluster members are approached for consent at the earliest opportunity and before any study interventions or data-collection procedures have started. Further, when cluster-level interventions or cluster size means that obtaining informed consent is not possible, a waiver of consent may be appropriate. In this paper, we suggest that the role of gatekeepers in protecting individual interests in CRTs should be limited. Generally, gatekeepers do not have the authority to provide proxy consent for cluster members. When a municipality or other community has a legitimate political authority that is empowered to make such decisions, cluster permission may be appropriate; however, gatekeepers may usefully protect cluster interests in other ways. Cluster consultation may ensure that the CRT addresses local health needs, and is conducted in accord with local values and customs. Gatekeepers may also play an important role in protecting the interests of organizations, such as hospitals, nursing homes, general practices, and schools. In these settings, permission to access the organization relies on resource implications and adherence to institutional policies.
Despite evidence-based recommendations supporting long-term use of cardiac medications in patients post ST-elevation myocardial infarction, adherence is known to decline over time. Discontinuation of cardiac medications in such patients is associated with increased mortality.
This is a pragmatic, cluster-randomized controlled trial with blinded outcome assessment and embedded qualitative process evaluation. Patients from one health region in Ontario, Canada who undergo a coronary angiogram during their admission for ST-elevation myocardial infarction and who survive their initial hospitalization will be included. Allocation of eligible patients to intervention or usual care will take place within one week after the angiogram using a computer-generated random sequence. To avoid treatment contamination, patients treated by the same family physician will be allocated to the same study arm. The intervention consists of recurrent, personalized, paper-based educational messages and reminders sent via post on behalf of the interventional cardiologist to the patient, family physician, and pharmacist urging long-term adherence to secondary prevention medications. The primary outcome is the proportion of patients who report in a phone interview taking all relevant classes of cardiac medications at twelve months. Secondary outcomes to be measured at three and twelve months include proportions of patients who report: actively taking each cardiac medication class of interest (item-by-item); stopping medications due to side effects; taking one or two or three medication classes concurrently; a perfect Morisky Medication Adherence Score for cardiac medication compliance; and having a discussion with their family physician about long-term adherence to cardiac medications. Self-reported measures of adherence will be validated using administrative data for prescriptions filled.
This intervention is designed to be easily generalizable. If effective, it could be implemented broadly. If it does not change medication utilization, the process evaluation will offer insights regarding how such an intervention could be optimized in future.
Randomized trial; Medication adherence; Reminders
The effect of hospital-acquired infection with Clostridium difficile on length of stay in hospital is not yet fully understood. We determined the independent impact of hospital-acquired infection with C. difficile on length of stay in hospital.
We conducted a retrospective observational cohort study of admissions to hospital between July 1, 2002, and Mar. 31, 2009, at a single academic hospital. We measured the association between infection with hospital-acquired C. difficile and time to discharge from hospital using Kaplan–Meier methods and a Cox multivariable proportional hazards regression model. We controlled for baseline risk of death and accounted for C. difficile as a time-varying effect.
Hospital-acquired infection with C. difficile was identified in 1393 of 136 877 admissions to hospital (overall risk 1.02%, 95% confidence interval [CI] 0.97%–1.06%). The crude median length of stay in hospital was greater for patients with hospital-acquired C. difficile (34 d) than for those without C. difficile (8 d). Survival analysis showed that hospital-acquired infection with C. difficile increased the median length of stay in hospital by six days. In adjusted analyses, hospital-acquired C. difficile was significantly associated with time to discharge, modified by baseline risk of death and time to acquisition of C. difficile. The hazard ratio for discharge by day 7 among patients with hospital-acquired C. difficile was 0.55 (95% CI 0.39–0.70) for patients in the lowest decile of baseline risk of death and 0.45 (95% CI 0.32–0.58) for those in the highest decile; for discharge by day 28, the corresponding hazard ratios were 0.74 (95% CI 0.60–0.87) and 0.61 (95% CI 0.53–0.68).
Hospital-acquired infection with C. difficile significantly prolonged length of stay in hospital independent of baseline risk of death.
Primary care providers play an important role in preventing and managing cardiovascular disease. This study compared the quality of preventive cardiovascular care delivery amongst different primary care models.
This is a secondary analysis of a larger randomized control trial, known as the Improved Delivery of Cardiovascular Care (IDOCC) through Outreach Facilitation. Using baseline data collected through IDOCC, we conducted a cross-sectional study of 82 primary care practices from three delivery models in Eastern Ontario, Canada: 43 fee-for-service, 27 blended-capitation and 12 community health centres with salary-based physicians. Medical chart audits from 4,808 patients with or at high risk of developing cardiovascular disease were used to examine each practice's adherence to ten evidence-based processes of care for diabetes, chronic kidney disease, dyslipidemia, hypertension, weight management, and smoking cessation care. Generalized estimating equation models adjusting for age, sex, rurality, number of cardiovascular-related comorbidities, and year of data collection were used to compare guideline adherence amongst the three models.
The percentage of patients with diabetes that received two hemoglobin A1c tests during the study year was significantly higher in community health centres (69%) than in fee-for-service (45%) practices (Adjusted Odds Ratio (AOR) = 2.4 [95% CI 1.4-4.2], p = 0.001). Blended capitation practices had a significantly higher percentage of patients who had their waistlines monitored than in fee-for-service practices (19% vs. 5%, AOR = 3.7 [1.8-7.8], p = 0.0006), and who were recommended a smoking cessation drug when compared to community health centres (33% vs. 16%, AOR = 2.4 [1.3-4.6], p = 0.007). Overall, quality of diabetes care was higher in community health centres, while smoking cessation care and weight management was higher in the blended-capitation models. Fee-for-service practices had the greatest gaps in care, most noticeably in diabetes care and weight management.
This study adds to the evidence suggesting that primary care delivery model impacts quality of care. These findings support current Ontario reforms to move away from the traditional fee-for-service practice.
Urgent, unplanned hospital readmissions are increasingly being used to gauge the quality of care. We reviewed urgent readmissions to determine which were potentially avoidable and compared rates of all-cause and avoidable readmissions.
In a multicentre, prospective cohort study, we reviewed all urgent readmissions that occurred within six months among patients discharged to the community from 11 teaching and community hospitals between October 2002 and July 2006. Summaries of the readmissions were reviewed by at least four practising physicians using standardized methods to judge whether the readmission was an adverse event (poor clinical outcome due to medical care) and whether the adverse event could have been avoided. We used a latent class model to determine whether the probability that each readmission was truly avoidable exceeded 50%.
Of the 4812 patients included in the study, 649 (13.5%, 95% confidence interval [CI] 12.5%–14.5%) had an urgent readmission within six months after discharge. We considered 104 of them (16.0% of those readmitted, 95% CI 13.3%–19.1%; 2.2% of those discharged, 95% CI 1.8%–2.6%) to have had a potentially avoidable readmission. The proportion of patients who had an urgent readmission varied significantly by hospital (range 7.5%–22.5%; χ2 = 92.9, p < 0.001); the proportion of readmissions deemed avoidable did not show significant variation by hospital (range 1.2%–3.7%; χ2 = 12.5, p < 0.25). We found no association between the proportion of patients who had an urgent readmission and the proportion of patients who had an avoidable readmission (Pearson correlation 0.294; p = 0.38). In addition, we found no association between hospital rankings by proportion of patients readmitted and rankings by proportion of patients with an avoidable readmission (Spearman correlation coefficient 0.28, p = 0.41).
Urgent readmissions deemed potentially avoidable were relatively uncommon, comprising less than 20% of all urgent readmissions following hospital discharge. Hospital-specific proportions of patients who were readmitted were not related to proportions with a potentially avoidable readmission.
There is a need to find innovative approaches for translating best practices for chronic disease care into daily primary care practice routines. Primary care plays a crucial role in the prevention and management of cardiovascular disease. There is, however, a substantive care gap, and many challenges exist in implementing evidence-based care. The Improved Delivery of Cardiovascular Care (IDOCC) project is a pragmatic trial designed to improve the delivery of evidence-based care for the prevention and management of cardiovascular disease in primary care practices using practice outreach facilitation.
The IDOCC project is a stepped-wedge cluster randomized control trial in which Practice Outreach Facilitators work with primary care practices to improve cardiovascular disease prevention and management for patients at highest risk. Primary care practices in a large health region in Eastern Ontario, Canada, were eligible to participate. The intervention consists of regular monthly meetings with the Practice Outreach Facilitator over a one- to two-year period. Starting with audit and feedback, consensus building, and goal setting, the practices are supported in changing practice behavior by incorporating chronic care model elements. These elements include (a) evidence-based decision support for providers, (b) delivery system redesign for practices, (c) enhanced self-management support tools provided to practices to help them engage patients, and (d) increased community resource linkages for practices to enhance referral of patients. The primary outcome is a composite score measured at the level of the patient to represent each practice's adherence to evidence-based guidelines for cardiovascular care. Qualitative analysis of the Practice Outreach Facilitators' written narratives of their ongoing practice interactions will be done. These textual analyses will add further insight into understanding critical factors impacting project implementation.
This pragmatic, stepped-wedge randomized controlled trial with both quantitative and process evaluations demonstrates innovative methods of implementing large-scale quality improvement and evidence-based approaches to care delivery. This is the first Canadian study to examine the impact of a large-scale multifaceted cardiovascular quality-improvement program in primary care. It is anticipated that through the evaluation of IDOCC, we will demonstrate an effective, practical, and sustainable means of improving the cardiovascular health of patients across Canada.
This article is part of a series of papers examining ethical issues in cluster randomized trials (CRTs) in health research. In the introductory paper in this series, we set out six areas of inquiry that must be addressed if the cluster trial is to be set on a firm ethical foundation. This paper addresses the second of the questions posed, namely, from whom, when, and how must informed consent be obtained in CRTs in health research? The ethical principle of respect for persons implies that researchers are generally obligated to obtain the informed consent of research subjects. Aspects of CRT design, including cluster randomization, cluster level interventions, and cluster size, present challenges to obtaining informed consent. Here we address five questions related to consent and CRTs: How can a study proceed if informed consent is not possible? Is consent to randomization always required? What information must be disclosed to potential subjects if their cluster has already been randomized? Is passive consent a valid substitute for informed consent? Do health professionals have a moral obligation to participate as subjects in CRTs designed to improve professional practice?
We set out a framework based on the moral foundations of informed consent and international regulatory provisions to address each of these questions. First, when informed consent is not possible, a study may proceed if a research ethics committee is satisfied that conditions for a waiver of consent are satisfied. Second, informed consent to randomization may not be required if it is not possible to approach subjects at the time of randomization. Third, when potential subjects are approached after cluster randomization, they must be provided with a detailed description of the interventions in the trial arm to which their cluster has been randomized; detailed information on interventions in other trial arms need not be provided. Fourth, while passive consent may serve a variety of practical ends, it is not a substitute for valid informed consent. Fifth, while health professionals may have a moral obligation to participate as subjects in research, this does not diminish the necessity of informed consent to study participation.
This article is part of a series of papers examining ethical issues in cluster randomized trials (CRTs) in health research. In the introductory paper in this series, we set out six areas of inquiry that must be addressed if the CRT is to be set on a firm ethical foundation. This paper addresses the first of the questions posed, namely, who is the research subject in a CRT in health research? The identification of human research subjects is logically prior to the application of protections as set out in research ethics and regulation. Aspects of CRT design, including the fact that in a single study the units of randomization, experimentation, and observation may differ, complicate the identification of human research subjects. But the proper identification of human research subjects is important if they are to be protected from harm and exploitation, and if research ethics committees are to review CRTs efficiently.
We examine the research ethics literature and international regulations to identify the core features of human research subjects, and then unify these features under a single, comprehensive definition of human research subject. We define a human research subject as any person whose interests may be compromised as a result of interventions in a research study. Individuals are only human research subjects in CRTs if: (1) they are directly intervened upon by investigators; (2) they interact with investigators; (3) they are deliberately intervened upon via a manipulation of their environment that may compromise their interests; or (4) their identifiable private information is used to generate data. Individuals who are indirectly affected by CRT study interventions, including patients of healthcare providers participating in knowledge translation CRTs, are not human research subjects unless at least one of these conditions is met.
Clinicians informally assess changes in patients' status over time to prognosticate their outcomes. The incorporation of trends in patient status into regression models could improve their ability to predict outcomes. In this study, we used a unique approach to measure trends in patient hospital death risk and determined whether the incorporation of these trend measures into a survival model improved the accuracy of its risk predictions.
We included all adult inpatient hospitalizations between 1 April 2004 and 31 March 2009 at our institution. We used the daily mortality risk scores from an existing time-dependent survival model to create five trend indicators: absolute and relative percent change in the risk score from the previous day; absolute and relative percent change in the risk score from the start of the trend; and number of days with a trend in the risk score. In the derivation set, we determined which trend indicators were associated with time to death in hospital, independent of the existing covariates. In the validation set, we compared the predictive performance of the existing model with and without the trend indicators.
Three trend indicators were independently associated with time to hospital mortality: the absolute change in the risk score from the previous day; the absolute change in the risk score from the start of the trend; and the number of consecutive days with a trend in the risk score. However, adding these trend indicators to the existing model resulted in only small improvements in model discrimination and calibration.
We produced several indicators of trend in patient risk that were significantly associated with time to hospital death independent of the model used to create them. In other survival models, our approach of incorporating risk trends could be explored to improve their performance without the collection of additional data.
A high quality decision requires that patients who meet clinical criteria for surgery are informed about the options (including non-surgical alternatives) and receive treatments that match their goals. The aim of this study was to evaluate the psychometric properties and clinical sensibility of a patient self report instrument, to measure the quality of decisions about total joint replacement for knee or hip osteoarthritis.
The performance of the Hip/Knee Osteoarthritis Decision Quality Instrument (HK-DQI) was evaluated in two samples: (1) a cross-sectional mail survey with 489 patients and 77 providers (study 1); and (2) a randomized controlled trial of a patient decision aid with 138 osteoarthritis patients considering total joint replacement (study 2). The HK-DQI results in two scores. Knowledge items are summed to create a total knowledge score, and a set of goals and concerns are used in a logistic regression model to develop a concordance score. The concordance score measures the proportion of patients whose treatment matched their goals. Hypotheses related to acceptability, feasibility, reliability and validity of the knowledge and concordance scores were examined.
In study 1, the HK-DQI was completed by 382 patients (79%) and 45 providers (58%), and in study 2 by 127 patients (92%), with low rates of missing data. The DQI-knowledge score was reproducible (ICC = 0.81) and demonstrated discriminant validity (68% decision aid vs. 54% control, and 78% providers vs. 61% patients) and content validity. The concordance score demonstrated predictive validity, as patients whose treatments were concordant with their goals had more confidence and less regret with their decision compared to those who did not.
The HK-DQI is feasible and acceptable to patients. It can be used to assess whether patients with osteoarthritis are making informed decisions about surgery that are concordant with their goals.
shared decision making; patient centered care; quality measurement; osteoarthritis; total joint replacement; decision quality
Objectives To investigate the extent to which authors of cluster randomised trials adhered to two basic requirements of the World Medical Association’s Declaration of Helsinki and the International Committee of Medical Journal Editors’ uniform requirements for manuscripts (namely, reporting of research ethics review and informed consent), to determine whether the adequacy of reporting has improved over time, and to identify characteristics of cluster randomised trials associated with reporting of ethics practices.
Design Review of a random sample of published cluster randomised trials from an electronic search in Medline.
Setting Cluster randomised trials in health research published in English language journals from 2000 to 2008.
Study sample 300 cluster randomised trials published in 150 journals.
Results 77 (26%, 95% confidence interval 21% to 31%) trials failed to report ethics review. The proportion reporting ethics review increased significantly over time (P<0.001). Trials with data collection interventions at the individual level were more likely to report ethics review than were trials that used routine data sources only (79% (n=151) v 55% (23); P=0.008). Trials that accounted for clustering in the design and analysis were more likely to report ethics review. The median impact factor of the journal of publication was higher for trials that reported ethics review (3.4 v 2.3; P<0.001). 93 (31%, 26% to 36%) trials failed to report consent. Reporting of consent increased significantly over time (P<0.001). Trials with interventions targeting participants at the individual level were more likely to report consent than were trials with interventions targeting the cluster level (87% (90) v 48% (41); P<0.001). Trials with data collection interventions at the individual level were more likely to report consent than were those that used routine data sources only (78% (146) v 29% (11); P<0.001).
Conclusions Reporting of research ethics protections in cluster randomised trials is inadequate. In addition to research ethics approval, authors should report whether informed consent was sought, from whom consent was sought, and what consent was for.
This article is part of a series of papers examining ethical issues in cluster randomized trials (CRTs) in health research. In the introductory paper in this series, Weijer and colleagues set out six areas of inquiry that must be addressed if the cluster trial is to be set on a firm ethical foundation. This paper addresses the third of the questions posed, namely, does clinical equipoise apply to CRTs in health research? The ethical principle of beneficence is the moral obligation not to harm needlessly and, when possible, to promote the welfare of research subjects. Two related ethical problems have been discussed in the CRT literature. First, are control groups that receive only usual care unduly disadvantaged? Second, when accumulating data suggests the superiority of one intervention in a trial, is there an ethical obligation to act?
In individually randomized trials involving patients, similar questions are addressed by the concept of clinical equipoise, that is, the ethical requirement that, at the start of a trial, there be a state of honest, professional disagreement in the community of expert practitioners as to the preferred treatment. Since CRTs may not involve physician-researchers and patient-subjects, the applicability of clinical equipoise to CRTs is uncertain. Here we argue that clinical equipoise may be usefully grounded in a trust relationship between the state and research subjects, and, as a result, clinical equipoise is applicable to CRTs. Clinical equipoise is used to argue that control groups receiving only usual care are not disadvantaged so long as the evidence supporting the experimental and control interventions is such that experts would disagree as to which is preferred. Further, while data accumulating during the course of a CRT may favor one intervention over another, clinical equipoise supports continuing the trial until the results are likely to be broadly convincing, often coinciding with the planned completion of the trial. Finally, clinical equipoise provides research ethics committees with formal and procedural guidelines that form an important part of the assessment of the benefits and harms of CRTs in health research.
The cluster randomized trial (CRT) is used increasingly in knowledge translation research, quality improvement research, community based intervention studies, public health research, and research in developing countries. However, cluster trials raise difficult ethical issues that challenge researchers, research ethics committees, regulators, and sponsors as they seek to fulfill responsibly their respective roles. Our project will provide a systematic analysis of the ethics of cluster trials. Here we have outlined a series of six areas of inquiry that must be addressed if the cluster trial is to be set on a firm ethical foundation:
1. Who is a research subject?
2. From whom, how, and when must informed consent be obtained?
3. Does clinical equipoise apply to CRTs?
4. How do we determine if the benefits outweigh the risks of CRTs?
5. How ought vulnerable groups be protected in CRTs?
6. Who are gatekeepers and what are their responsibilities?
Subsequent papers in this series will address each of these areas, clarifying the ethical issues at stake and, where possible, arguing for a preferred solution. Our hope is that these papers will serve as the basis for the creation of international ethical guidelines for the design and conduct of cluster randomized trials.
Cluster randomized trials (CRTs) present unique methodological and ethical challenges. Researchers conducting systematic reviews of CRTs (e.g., addressing methodological or ethical issues) require efficient electronic search strategies (filters or hedges) to identify trials in electronic databases such as MEDLINE. According to the CONSORT statement extension to CRTs, the clustered design should be clearly identified in titles or abstracts; however, variability in terminology may make electronic identification challenging. Our objectives were to (a) evaluate sensitivity ("recall") and precision of a well-known electronic search strategy ("randomized controlled trial" as publication type) with respect to identifying CRTs, (b) evaluate the feasibility of new search strategies targeted specifically at CRTs, and (c) determine whether CRTs are appropriately identified in titles or abstracts of reports and whether there has been improvement over time.
We manually examined a wide range of health journals to identify a gold standard set of CRTs. Search strategies were evaluated against the gold standard set, as well as an independent set of CRTs included in previous systematic reviews.
The existing strategy (randomized controlled trial.pt) is sensitive (93.8%) for identifying CRTs, but has relatively low precision (9%, number needed to read 11); the number needed to read can be halved to 5 (precision 18.4%) by combining with cluster design-related terms using the Boolean operator AND; combining with the Boolean operator OR maximizes sensitivity (99.4%) but would require 28.6 citations read to identify one CRT. Only about 50% of CRTs are clearly identified as cluster randomized in titles or abstracts; approximately 25% can be identified based on the reported units of randomization but are not amenable to electronic searching; the remaining 25% cannot be identified except through manual inspection of the full-text article. The proportion of trials clearly identified has increased from 28% between the years 2000-2003, to 60% between 2004-2007 (absolute increase 32%, 95% CI 17 to 47%).
CRTs should include the phrase "cluster randomized trial" in titles or abstracts; this will facilitate more accurate indexing of the publication type by reviewers at the National Library of Medicine, and efficient textword retrieval of the subset employing cluster randomization.
Cluster randomized trials are an increasingly important methodological tool in health research. In cluster randomized trials, intact social units or groups of individuals, such as medical practices, schools, or entire communities – rather than individual themselves – are randomly allocated to intervention or control conditions, while outcomes are then observed on individual cluster members. The substantial methodological differences between cluster randomized trials and conventional randomized trials pose serious challenges to the current conceptual framework for research ethics. The ethical implications of randomizing groups rather than individuals are not addressed in current research ethics guidelines, nor have they even been thoroughly explored. The main objectives of this research are to: (1) identify ethical issues arising in cluster trials and learn how they are currently being addressed; (2) understand how ethics reviews of cluster trials are carried out in different countries (Canada, the USA and the UK); (3) elicit the views and experiences of trial participants and cluster representatives; (4) develop well-grounded guidelines for the ethical conduct and review of cluster trials by conducting an extensive ethical analysis and organizing a consensus process; (5) disseminate the guidelines to researchers, research ethics boards (REBs), journal editors, and research funders.
We will use a mixed-methods (qualitative and quantitative) approach incorporating both empirical and conceptual work. Empirical work will include a systematic review of a random sample of published trials, a survey and in-depth interviews with trialists, a survey of REBs, and in-depth interviews and focus group discussions with trial participants and gatekeepers. The empirical work will inform the concurrent ethical analysis which will lead to a guidance document laying out principles, policy options, and rationale for proposed guidelines. An Expert Panel of researchers, ethicists, health lawyers, consumer advocates, REB members, and representatives from low-middle income countries will be appointed. A consensus conference will be convened and draft guidelines will be generated by the Panel; an e-consultation phase will then be launched to invite comments from the broader community of researchers, policy-makers, and the public before a final set of guidelines is generated by the Panel and widely disseminated by the research team.
The exchange of information is an integral component of continuity of health care and may limit or prevent costly duplication of tests and treatments. This study determined the probability that patient information from previous visits with other physicians was available for a current physician visit.
We conducted a multicentre prospective cohort study including patients discharged from the medical or surgical services of 11 community and academic hospitals in Ontario. Patients included in the study saw at least 2 different physicians during the 6 months after discharge. The primary outcome was whether information from a previous visit with another physician was available at the current visit. We determined the availability of previous information using surveys of or interviews with the physicians seen during current visits.
A total of 3250 patients, with a total of 39 469 previous–current visit combinations, met the inclusion criteria. Overall, information about the previous visit was available 22.0% of the time. Information was more likely to be available if the current doctor was a family physician (odds ratio [OR] 1.75, 95% confidence interval [CI] 1.54–1.98) or a physician who had treated the patient before the hospital admission (OR 1.33, 95% CI 1.21–1.46). Conversely, information was less likely to be available if the previous doctor was a family physician (OR 0.38, 95% CI 0.32–0.44) or a physician who had treated the patient before the admission (OR 0.72, 95% CI 0.60–0.86). The strongest predictor of information exchange was the current physician having previously received information about the patient from the previous physician (OR 7.72, 95% CI 6.92–8.63).
Health care information is often not shared among multiple physicians treating the same patient. This situation would be improved if information from family physicians and patients' regular physicians was more systematically available to other physicians.
Stillbirth rate is an important indicator of access to and quality of antenatal and delivery care. Obtaining overall estimates across various regions of the world is not straightforward due to variation in definitions, data collection methods and reporting.
We conducted a systematic review of a range of pregnancy-related conditions including stillbirths and performed meta-analysis of the subset of studies reporting stillbirth rates. We examined variation across rates and used meta-regression techniques to explain observed variation.
We identified 389 articles on stillbirth prevalence among the 2580 included in the systematic review. We included 70 providing 80 data sets from 50 countries in the meta-analysis. Pooled prevalence rates show variation across various subgroup categories. Rates per 100 births are higher in studies conducted in less developed country settings as compared to more developed (1.17 versus 0.50), of inadequate quality as compared to adequate (1.12 versus 0.66), using sub-national sample as compared to national (1.38 versus 0.68), reporting all stillbirths as compared to late stillbirths (0.95 versus 0.63), published in non-English as compared to English (0.91 versus 0.59) and as journal articles as compared to non-journal (1.37 versus 0.67). The results of the meta-regression show the significance of two predictor variables – development status of the setting and study quality – on stillbirth prevalence.
Stillbirth prevalence at the community level is typically less than 1% in more developed parts of the world and could exceed 3% in less developed regions. Regular reviews of stillbirth rates in appropriately designed and reported studies are useful in monitoring the adequacy of care. Systematic reviews of prevalence studies are helpful in explaining sources of variation across rates. Exploring these methodological issues will lead to improved standards for assessing the burden of reproductive ill-health.