ABSTRACT
BACKGROUND: Previous studies concluded that there was an increased risk of non-fatal venous thromboembolism (VTE) with drospirenone. It is unknown whether the risk is differential by ethinyl-estradiol dosage. OBJECTIVES: To assess the risk of VTE with drospirenone and to determine whether drospirenone and ethinyl-estradiol 20 µg (DRSP/EE20) has a lower VTE risk than drospirenone and ethinyl-estradiol 30 µg (DRSP/EE30). METHODS: Our cohort included women aged 18-46 years taking drospirenone or levonorgestrel (LNG)-containing combined oral contraceptives (COCs) in the IMS claims database between 2001 and 2009. VTE was defined using ICD-9-CM coding and anticoagulation. The hazard ratio (HR) from Cox proportional hazards models was used to assess the VTE relative risk (RR) with drospirenone compared with levonorgestrel, adjusted by a propensity score used to control for baseline co-morbidity and stratified by EE dosage and user-type (new/current). RESULTS: The study included 238 683 drospirenone and 193,495 levonorgestrel users. Among new and current users, a 1.90-fold (95% CI, 1.51-2.39) increased VTE relative risk was observed for drospirenone (18.0 VTE/10,000 women-years) vs. levonorgestrel (8.9 VTE/10,000 women-years). In analysis of new users, DRSP/EE20 had a 2.35-fold (95% CI, 1.44-3.82) VTE RR versus LNG/EE20. New users of DRSP/EE30 observed an increased RR versus LNG/EE30 among women starting to use COCs between 2001 and 2006 (2.51, 95% CI, 1.12-5.64) but not between 2007 and 2009 (0.76, 95% CI, 0.42-1.39), attributable to an increased incidence rate with LNG/EE30 from 2007 to 2009. In direct comparison, DRSP/EE20 had an elevated risk of VTE compared with DRSP/EE30 (RR, 1.55; 95% CI, 0.99-2.41). CONCLUSIONS: We observed a modestly elevated risk of VTE with drospirenone, compared with levonorgestrel. The larger VTE incidence rate observed in DRSP/EE20 than in DRSP/EE30 and the increasing VTE incidence rate with levonorgestrel between 2007 and 2009 were unexpected.
Subject(s)
Androstenes/adverse effects , Ethinyl Estradiol/administration & dosage , Levonorgestrel/adverse effects , Venous Thromboembolism/drug therapy , Adolescent , Adult , Cohort Studies , Contraceptives, Oral/administration & dosage , Databases, Factual , Female , Humans , Middle Aged , Proportional Hazards Models , Risk , Time Factors , United States , Young AdultABSTRACT
BACKGROUND: Gastrointestinal injuries including gastric ulcers have been reported with oral bisphosphonate therapy. However, the risk of the more serious upper gastrointestinal bleeding (UGB) especially in the community setting with these drugs remains unknown. Similarly, the risk of UGB among users of both bisphosphonates and non steroidal anti-inflammatory drugs (NSAIDs) in the community is also unknown. AIM: To explore the risk of more serious UGB among users of bisphosphonates and the risk of UGB among users of both bisphosphonates and NSAIDs in the community. METHODS: We conducted a case-control study within a cohort of Quebec residents who had received a revascularization procedure from 1995 to 2004. Cohort members were followed up from the date of their first procedure until the earliest of: (1) study outcome, (2) date of death or (3) end of health care coverage. Cases were defined as those with the first diagnosis of a UGB. For each case, 20 controls were selected and matched to the cases by index date, age and cohort entry. Adjusted odds ratios for current use of bisphosphonates, NSAIDs and co-therapy of both drugs were computed. RESULTS: Within the initial cohort, 3253 incident cases of UGBs and corresponding 65 060 matched controls were identified. The adjusted odds ratio (OR) for UGB by current users of bisphosphonates was 1.01 (95% CI, 0.72-1.43). Current NSAID use was associated with an increased risk of UGB OR = 1.75; 95% CI, 1.53-1.99. The OR for use of bisphosphonates and NSAIDs was elevated OR = 2.00; 95% CI, 1.12-3.57. This risk was still elevated for users of bisphosphonates and COX-2 inhibitors [OR = 2.38 (95% CI, 1.26-4.50)]. CONCLUSION: We found no evidence of an increase in the risk of UGB among current users of bisphosphonates. The risk of combined NSAID and bisphosphonate therapy was increased, but this risk was not higher than the risk for NSAID users alone.
Subject(s)
Anti-Inflammatory Agents, Non-Steroidal/adverse effects , Cyclooxygenase 2 Inhibitors/adverse effects , Diphosphonates/adverse effects , Gastrointestinal Hemorrhage/chemically induced , Administration, Oral , Aged , Aged, 80 and over , Anti-Inflammatory Agents, Non-Steroidal/therapeutic use , Case-Control Studies , Cyclooxygenase 2 Inhibitors/therapeutic use , Diphosphonates/therapeutic use , Drug Interactions , Drug Therapy, Combination , Female , Gastrointestinal Agents/therapeutic use , Gastrointestinal Hemorrhage/epidemiology , Humans , Male , Odds Ratio , Quebec/epidemiology , Risk Factors , Treatment OutcomeABSTRACT
OBJECTIVES: Recent studies suggest that persons with congestive heart failure (CHF) may be at higher risk for short-term effects of air pollution. This daily diary panel study in Montreal, Quebec, was carried out to determine whether oxygen saturation and pulse rate were associated with selected personal factors, weather conditions and air pollution. METHODS: Thirty-one subjects with CHF participated in this study in 2002 and 2003. Over a 2-month period, the investigators measured their oxygen saturation, pulse rate, weight and temperature each morning and recorded these and other data in a daily diary. Air pollution and weather conditions were obtained from fixed-site monitoring stations. The study made use of mixed regression models, adjusting for within-subject serial correlation and temporal trends, to determine the association between oxygen saturation and pulse rate and personal and environmental variables. Depending on the model, we accounted for the effects of a variety of personal variables (eg, body temperature, salt consumption) as well as nitrogen dioxide (NO2), ozone, maximum temperature and change in barometric pressure at 8:00 from the previous day. RESULTS: In multivariable analyses, the study found that oxygen saturation was reduced when subjects reported that they were ill, consumed salt, or drank liquids on the previous day and had higher body temperatures on the concurrent day (only the latter was statistically significant). Relative humidity and decreased atmospheric pressure from the previous day were associated with oxygen saturation. In univariate analyses, there was negative associations with concentrations of fine particulates, ozone, and sulphur dioxide (SO2), but only SO2 was significant after adjustment for the effects of weather. For pulse rate, no associations were found for the personal variables and in univariate analyses the study found positive associations with NO(2), fine particulates (aerodynamic diameter of 2.5 microm or under, PM(2.5)), SO2, and maximum temperature, although only the latter two were significant after adjustment for environmental effects. CONCLUSIONS: The findings from the present investigation suggest that personal and environmental conditions affect intermediate physiological parameters that may affect the health of CHF patients.
Subject(s)
Air Pollution/adverse effects , Heart Failure/etiology , Heart Rate/physiology , Oxygen/blood , Weather , Aged , Aged, 80 and over , Atmospheric Pressure , Female , Health Status , Heart Failure/physiopathology , Humans , Male , Medical Records , Middle Aged , Multivariate Analysis , Particulate Matter/toxicity , Quebec , Regression Analysis , SeasonsABSTRACT
BACKGROUND: Cyclo-oxygenase-2 selective inhibitors have been associated with cardiovascular side effects, but previous studies have generally excluded people with previous myocardial infarction, thereby limiting our knowledge of their cardiotoxicity in this population. OBJECTIVES: To determine whether a history of myocardial infarction modified the risk of acute myocardial infarction associated with the use of various non-steroidal anti-inflammatory drugs (NSAIDs). METHODS: A population-based cohort of 122 079 elderly people with and without previous myocardial infarction newly treated with an NSAID between 1 January 1999 and 30 June 2002 were identified using the computerised health databases of Québec, Canada. A nested-case-control approach was used for the analysis, with controls matched by cohort entry and age. Current users of NSAIDs, those whose last prescription overlapped with the index date, were compared with those who were not exposed to NSAIDs in the year preceding the event. Rate ratios of acute myocardial infarction were estimated using conditional logistic regression and adjusted for potential confounders. RESULTS: Users of rofecoxib, both with and without previous myocardial infarction, were at increased risk of myocardial infarction, with a trend for greater risk among those with a previous event (rate ratio (RR) 1.59, 95% confidence interval (CI) 1.15 to 2.18 v RR 1.23, 95% CI 1.05 to 1.45; p = 0.14 for interaction). By contrast, celecoxib was only associated with an increased risk in people with previous myocardial infarction (RR 1.40, 95% CI 1.06 to 1.84 v RR 1.03, 95% CI 0.88 to 1.20; p = 0.04 for interaction). The available power was insufficient to reliably assess risks among patients with previous myocardial infarction treated with other NSAIDs, dose-response relationships or interaction with aspirin. CONCLUSIONS: Although only rofecoxib use was associated with an increased risk of myocardial infarction in those without a previous event, both rofecoxib and celecoxib were associated with an excess risk of acute myocardial infarction for current users with a history of myocardial infarction. A large randomised trial is required to more completely and reliably assess the cardiovascular safety of celecoxib and traditional NSAIDs in this population of high-risk patients.
Subject(s)
Cyclooxygenase 2 Inhibitors/adverse effects , Lactones/adverse effects , Myocardial Infarction/chemically induced , Pyrazoles/adverse effects , Sulfonamides/adverse effects , Sulfones/adverse effects , Aged , Case-Control Studies , Celecoxib , Confounding Factors, Epidemiologic , Cyclooxygenase 2 Inhibitors/therapeutic use , Female , Humans , Lactones/therapeutic use , Logistic Models , Male , Myocardial Infarction/drug therapy , Pyrazoles/therapeutic use , Recurrence , Risk , Sulfonamides/therapeutic use , Sulfones/therapeutic useABSTRACT
Heart failure affects over 500,000 Canadians, and 50,000 new patients are diagnosed each year. The mortality remains staggering, with a five-year age-adjusted rate of 45%. Disease management programs for heart failure patients have been associated with improved outcomes, the use of evidence-based therapies, improved quality of care, and reduced costs, mortality and hospitalizations. Currently, national benchmarks and targets for access to care for cardiovascular procedures or office consultations do not exist. The present paper summarizes the currently available data, particularly focusing on the risk of adverse events as a function of waiting time, as well as on the identification of gaps in existing data on heart failure. Using best evidence and expert consensus, the present article also focuses on timely access to care for acute and chronic heart failure, including timely access to heart failure disease management programs and physician care (heart failure specialists, cardiologists, internists and general practitioners).
Subject(s)
Health Services Accessibility , Heart Failure/therapy , Patient Selection , Follow-Up Studies , Humans , Randomized Controlled Trials as Topic , Risk Factors , Time FactorsABSTRACT
The incidence of end-stage renal disease (ESRD) owing to diabetes has continued to increase despite the extensive use of angiotensin-converting enzyme (ACE) inhibitors to prevent diabetic nephropathy, primarily from evidence of short-term effectiveness. We assessed the long-term effect of ACE inhibitors on the risk of ESRD. We formed a population-based cohort of all diabetic patients treated with antihypertensive drugs in the Province of Saskatchewan, Canada, between 1982 and 1986. The patients were followed up to the end of 1997 to identify cases of end-stage renal failure. A nested case-control analysis was used with the controls matched to each case on age, diabetes type, and duration of follow-up. The cohort comprised 6102 subjects, of which the 102 cases who developed end-stage renal failure were matched to 4129 controls. Relative to thiazide diuretic use, the adjusted rate ratio of end-stage renal failure associated with the use of ACE inhibitors was 2.5 (95% confidence interval 1.3-4.7), whereas it was 0.8 (95% confidence interval 0.5-1.4) for beta-blockers and 0.7 (95% confidence interval 0.4-1.3) for calcium antagonists. The rate ratio of end-stage renal failure with the use of ACE inhibitors was 0.8 (95% confidence interval 0.3-2.5) during the first 3 years of follow-up, but increased to 4.2 (95% confidence interval 2.0-9.0) after 3 years. ACE-inhibitor use does not appear to decrease the long-term risk of end-stage renal failure in diabetes. Our data suggest instead that ACE inhibitors might actually increase this risk, which may possibly contribute to the continued increasing incidence of ESRD owing to diabetes.
Subject(s)
Angiotensin-Converting Enzyme Inhibitors/adverse effects , Diabetes Mellitus/drug therapy , Diabetic Nephropathies/etiology , Kidney Failure, Chronic/etiology , Adrenergic beta-Antagonists/therapeutic use , Adult , Aged , Angiotensin-Converting Enzyme Inhibitors/therapeutic use , Antihypertensive Agents/adverse effects , Antihypertensive Agents/therapeutic use , Calcium Channel Blockers/therapeutic use , Case-Control Studies , Cohort Studies , Diabetic Nephropathies/prevention & control , Female , Humans , Kidney Failure, Chronic/prevention & control , Male , Middle Aged , Sodium Chloride Symporter Inhibitors/therapeutic useABSTRACT
In 2004, the Canadian Cardiovascular Society formed an Access to Care Working Group with a mandate to use the best science and information available to establish reasonable triage categories and safe wait times for common cardiovascular services and procedures through a series of commentaries. The present commentary is the first in the series and lays out issues regarding timely access to care that are common to all cardiovascular services and procedures. The commentary briefly describes the 'right' to timely access, wait lists as a health care system management tool, and the role of the physician as patient advocate and gatekeeper. It also provides advice to funders, administrators and providers who must monitor and manage wait times to improve access to cardiovascular care in Canada and restore the confidence of Canadians in their publicly funded health care system.
Subject(s)
Cardiovascular Diseases/therapy , Health Services Accessibility , National Health Programs , Patient Rights , Referral and Consultation , Canada , Gatekeeping , Health Care Rationing , Health Priorities , Health Services Accessibility/economics , Health Services Accessibility/legislation & jurisprudence , Humans , National Health Programs/legislation & jurisprudence , National Health Programs/organization & administration , Patient Rights/legislation & jurisprudence , Social Responsibility , Time Factors , Triage , Universal Health Insurance , Waiting ListsABSTRACT
In 2004, the Canadian Cardiovascular Society formed an Access to Care Working Group with a mandate to use the best science and information available to establish reasonable triage categories and safe wait times for common cardiovascular services and procedures through a series of commentaries. The present commentary discusses the rationale for access benchmarks for urgent cardiac catheterization and revascularization, including hospital transfer in the setting of non-ST elevation acute coronary syndromes. The literature on standards of care, wait times, wait list management and clinical trials was reviewed. A survey of all cardiac catheterization directors in Canada was performed to develop an inventory of current practices in identifying and triaging patients. The Working Group recommended the following medically acceptable wait times for access to diagnostic catheterization and revascularization in patients presenting with acute coronary syndromes: for diagnostic catheterization and percutaneous coronary intervention, the target should be 24 h to 48 h for high-risk, three to five days for intermediate-risk and five to seven days for low-risk patients; for coronary artery bypass graft surgery, the target should be three to five days for high-risk, two to three weeks for intermediate-risk and six weeks for low-risk patients. All stakeholders must affirm the appropriateness of these standards and work continuously to achieve them. However, some questions remain around what are the best clinical risk markers to delineate the triage categories and the utility of clinical risk scores to assist clinicians in triaging patients for invasive therapies.
Subject(s)
Angina, Unstable/therapy , Health Services Accessibility/standards , Myocardial Infarction/therapy , Triage/standards , Angioplasty, Balloon, Coronary , Benchmarking , Canada , Cardiac Catheterization , Coronary Artery Bypass , Health Services Accessibility/statistics & numerical data , Humans , Patient Transfer , Risk Assessment , Syndrome , Time Factors , Waiting ListsABSTRACT
The Canadian Cardiovascular Society is the national professional society for cardiovascular specialists and researchers in Canada. In the spring of 2004, the Canadian Cardiovascular Society Council formed an Access to Care Working Group in an effort to use the best science and information to establish reasonable triage categories and safe wait times for access to common cardiovascular services and procedures. The Working Group has elected to publish a series of commentaries to initiate a structured national discussion on this very important issue. Access to treatment with implantable cardioverter defibrillators is the subject of the present commentary. The prevalence pool of potentially eligible patients is discussed, along with access barriers, regional disparities and waiting times. A maximum recommended waiting time is proposed and the framework for a solution-oriented approach is presented.
Subject(s)
Death, Sudden, Cardiac/prevention & control , Defibrillators, Implantable , Health Services Accessibility/statistics & numerical data , Canada , Humans , Time Factors , Waiting ListsSubject(s)
Angiotensin-Converting Enzyme Inhibitors/therapeutic use , Coronary Disease/drug therapy , Perindopril/therapeutic use , Angiotensin-Converting Enzyme Inhibitors/economics , Drug Costs , Humans , Myocardial Infarction/prevention & control , Perindopril/economics , Randomized Controlled Trials as Topic , Treatment OutcomeABSTRACT
PURPOSE: Congestive heart failure is an important cause of patient morbidity and mortality. Although several randomized clinical trials have compared beta-blockers with placebo for treatment of congestive heart failure, a meta-analysis quantifying the effect on mortality and morbidity has not been performed recently. DATA SOURCES: The MEDLINE, Cochrane, and Web of Science electronic databases were searched from 1966 to July 2000. References were also identified from bibliographies of pertinent articles. STUDY SELECTION: All randomized clinical trials of beta-blockers versus placebo in chronic stable congestive heart failure were included. DATA EXTRACTION: A specified protocol was followed to extract data on patient characteristics, beta-blocker used, overall mortality, hospitalizations for congestive heart failure, and study quality. DATA SYNTHESIS: A hierarchical random-effects model was used to synthesize the results. A total of 22 trials involving 10 135 patients were identified. There were 624 deaths among 4862 patients randomly assigned to placebo and 444 deaths among 5273 patients assigned to beta-blocker therapy. In these groups, 754 and 540 patients, respectively, required hospitalization for congestive heart failure. The probability that beta-blocker therapy reduced total mortality and hospitalizations for congestive heart failure was almost 100%. The best estimates of these advantages are 3.8 lives saved and 4 fewer hospitalizations per 100 patients treated in the first year after therapy. The probability that these benefits are clinically significant (>2 lives saved or >2 fewer hospitalizations per 100 patients treated) is 99%. Both selective and nonselective agents produced these salutary effects. The results are robust to any reasonable publication bias. CONCLUSIONS: beta-Blocker therapy is associated with clinically meaningful reductions in mortality and morbidity in patients with stable congestive heart failure and should be routinely offered to all patients similar to those included in trials.
Subject(s)
Adrenergic beta-Antagonists/therapeutic use , Heart Failure/drug therapy , Heart Failure/mortality , Bayes Theorem , Hospitalization/statistics & numerical data , Humans , Research Design/standards , Survival AnalysisABSTRACT
Little is known about how physicians make decisions when the evidence is incomplete or controversial. While thrombolysis improves survival following acute myocardial infarction (AMI), conflicting evidence exists as to any specific agent's superiority, particularly if cost-effectiveness is considered. Using a Bayesian hierarchical model, the authors examined the patient, physician, and hospital characteristics that are related to the decision-making process concerning the choice of thrombolytic agent in a prospective registry of 1,165 AMI patients receiving thrombolysis. Tissue plasminogen activator (t-PA) was administered to 432 patients (31.8%) and streptokinase (SK) to the remainder. The presence of an anterior infarction, a previous myocardial infarction, low blood pressure, a cardiologist decision maker, younger age, and receiving treatment within six hours after the start of symptoms were independent predictors of receiving t-PA. The levels of importance that physicians accorded to these patient characteristics differed according to their practicing institutions. Generally, they followed evidence-based medicine and reasonably targeted high-risk patients to receive the more expensive t-PA. However, they also preferentially treated younger patients, where only a small absolute advantage appears to exist.
Subject(s)
Decision Making , Fibrinolytic Agents/therapeutic use , Myocardial Infarction/drug therapy , Streptokinase/therapeutic use , Thrombolytic Therapy , Tissue Plasminogen Activator/therapeutic use , Aged , Bayes Theorem , Evidence-Based Medicine , Female , Fibrinolytic Agents/economics , Humans , Logistic Models , Male , Middle Aged , Prospective Studies , Quebec , Registries , Streptokinase/economics , Tissue Plasminogen Activator/economicsABSTRACT
OBJECTIVE: To estimate the magnitude of the clinical benefits that may result from use of abciximab at the time of angioplasty and the cost of achieving them. DATA SOURCES: Four published randomized control trials. DATA SYNTHESIS: Meta-analysis of outcomes at six months. RESULTS: Use of abciximab in comparable high risk populations, in the manner described in these trials, is estimated to have the following effects: It does nto influence mortality within the first six months. It reduces the rate of myocardial infarction (MI) by 3.3/100 treatments with a 95% CI of 1.6 to 5.2. It may reduce the need for revascularization (angioplasty or coronary artery bypass graft) by 2.1/100 treatments (95% CI -1.0 to 5.0). It does not cause any significant increase in major hemorrhagic events. There is no evidence that it influences restenosis rates. The net cost per MI prevented would be approximately $44,000, ranging from approximately $29,000 to $71,000 on sensitivity analysis. The net cost per adverse event prevented (MI plus revascularization procedure) would be approximately $27,000 (sensitivity analysis $16,000 to $57,000). Use of abciximab for all of the approximately 17,487 angioplasties carried out in Canada each year may prevent 395 myocardial infarcts and 186 revascularization procedures, at an overall cost of approximately $29 million and a cost effectiveness of approximately $50,000 per adverse event prevented. (This assumes the same proportional reduction in events as in these four studies, and that 35% of procedures are high risk). SIGNIFICANCE: Possible eventual prolongation of life due to fewer periprocedural MIs with abciximab use cannot be quantified. Thus, these estimates of cost effectiveness cannot be used to compare this intervention directly with others in terms of dollars per life year saved. The field is evolving rapidly and these conclusions may soon have to be modified. Increasing use of stents will probably slightly reduce, but not abolish, the health benefits of abciximab use. These estimates are based on only four trials. However, until more trials are completed they provide the best available evidence on which to base policy decisions.
Subject(s)
Angioplasty, Balloon, Coronary , Angioplasty, Balloon , Antibodies, Monoclonal/therapeutic use , Coronary Disease/surgery , Immunoglobulin Fab Fragments/therapeutic use , Platelet Aggregation Inhibitors/therapeutic use , Abciximab , Angioplasty, Balloon/economics , Angioplasty, Balloon, Coronary/economics , Antibodies, Monoclonal/pharmacology , Clinical Trials as Topic , Coronary Disease/drug therapy , Costs and Cost Analysis , Female , Humans , Immunoglobulin Fab Fragments/pharmacology , Male , Platelet Aggregation Inhibitors/pharmacology , Randomized Controlled Trials as TopicABSTRACT
OBJECTIVE: To describe the various components of the delay to thrombolytic treatment for patients with acute myocardial infarction (MI) and to identify the hospital and patient characteristics related to these delays. DESIGN: Cohort analysis from a hospital registry of patients receiving thrombolytic treatment. SETTING: Forty acute care hospitals in Quebec. SUBJECTS: All 1357 patients who received thrombolysis between January 1995 and May 1996. MAIN OUTCOME MEASURES: Time from onset of symptoms to arrival at hospital and the various components of the in-hospital delay. RESULTS: The median delay before presentation to hospital was 98 (interquartile range [IR] 56 to 180) minutes and was longer for women (p < 0.001), patients over 65 years of age (p < 0.001) and patients with diabetes mellitus (p < 0.01). The median time from arrival at hospital to thrombolysis was 59 (IR 41 to 89) minutes, the medical decision-making component taking a median of 12 (IR 4 to 27) minutes. Women (p < 0.005), older patients (p < 0.001) and patients with a past history of MI (p < 0.001) had increased in-hospital delays to thrombolysis. Delays were longer in community hospitals (p < 0.05) and low-volume centres (p < 0.01) and when a cardiologist made the decision to administer thrombolysis (p < 0.001). Multivariate analysis showed that increased age (odds ratio 1.5, 95% confidence interval 1.3 to 1.7, p < 0.001) and having the medical decision made by a cardiologist (odds ratio 1.8, 95% confidence interval 1.6 to 2.0, p < 0.001) were independently associated with an increased risk of being in the upper median of in-hospital delays. CONCLUSIONS: Despite certain improvements, there remain substantial delays between symptom onset and the administration of thrombolysis for patients with acute MI. A large part of the delay is due to the hesitation of patients (particularly women, older patients and patients with diabetes) to seek medical attention. Although the median time for medical decision-making appears reasonable, care must be taken to ensure that all patient groups receive timely evaluation and therapy. The delay associated with having the treatment decision made by a cardiologist probably represents a marker for more difficult, complex cases. Methods should be developed to permit specialty consultation, if needed, while minimizing treatment delays. Community and low-volume hospitals may require special attention.