Comparison of "accelerated" tissue plasminogen activator with streptokinase for treatment of suspected myocardial infarction

Prophylactic anticoagulation to prevent venous thromboembolism in traumatic intracranial hemorrhage: a decision analysis

Introduction

Patients with intracranial hemorrhage due to traumatic brain injury are at high risk of developing venous thromboembolism including deep vein thrombosis (DVT) and pulmonary embolism (PE). Thus, there is a trade-off between the risks of progression of intracranial hemorrhage (ICH) versus reduction of DVT/PE with the use of prophylactic anticoagulation. Using decision analysis modeling techniques, we developed a model for examining this trade-off for trauma patients with documented ICH.

Methods

The decision node involved the choice to administer or to withhold low molecular weight heparin (LMWH) anticoagulation prophylaxis at 24 hours. Advantages of withholding therapy were decreased risk of ICH progression (death, disabling neurologic deficit, non-disabling neurologic deficit), and decreased risk of systemic bleeding complications (death, massive bleed). The associated disadvantage was greater risk of developing DVT/PE or death. Probabilities for each outcome were derived from natural history studies and randomized controlled trials when available. Utilities were obtained from accepted databases and previous studies.

Results

The expected value associated with withholding anticoagulation prophylaxis was similar (0.90) to that associated with the LMWH strategy (0.89). Only two threshold values were encountered in one-way sensitivity analyses. If the effectiveness of LMWH at preventing DVT exceeded 80% (range from literature 33% to 82%) our model favoured this therapy. Similarly, our model favoured use of LMWH if this therapy increased the risk of ICH progression by no more than 5% above the baseline risk.

Conclusions

Our model showed no clear advantage to providing or withholding anticoagulant prophylaxis for DVT/PE prevention at 24 hours after traumatic brain injury associated with ICH. Therefore randomized controlled trials are justifiable and needed to guide clinicians.

Thrombolytic therapy guided by a decision analysis model: are there potential benefits for patient management?

BACKGROUND

Although thrombolytic therapy improves the outcome of myocardial infarction, it is associated with increased risks of stroke and bleeding; these risks may outweigh the benefits of therapy. The risks and benefits of thrombolysis, for any individual clinical situation, can be explicitly estimated by means of decision analysis.

HYPOTHESIS

The aim of this study was to compare the actual use of thrombolytic agents for suspected acute myocardial infarction (AMI) with the management preferred by a decision analysis model.

METHODS

Admission data prospectively obtained in 262 consecutive patients admitted to a rural community hospital's coronary care unit with suspected AMI, as well as clinical decisions and outcomes, were reviewed and analyzed.

RESULTS

Seventeen deaths from AMI and no major strokes were observed, compared with 18.30 deaths and 0.85 major strokes predicted by a decision analysis model. Forty-seven of 84 patients with confirmed AMI and 3 of 178 without AMI were given a thrombolytic agent, compared with 65 patients with and 7 without AMI who had decision analysis-guided therapy. Decision analysis-guided therapy could have saved 3.7 additional lives and gained 29.6 life years, but produced 0.4 extra strokes. Changing the quality adjustment for stroke or heart failure would not have altered the treatment preferred by decision analysis in any of the 262 cases studied. Some patients were predicted to benefit considerably from thrombolysis with little extra risk of stroke and vice versa: all cases must, therefore, be assessed individually.

CONCLUSIONS

A decision analysis model can guide thrombolytic therapy by promptly defining its risks and benefits.

The cost-effectiveness of thrombolysis administered by paramedics

OBJECTIVE

The objective of this study is to estimate the expected health outcomes, costs and cost-effectiveness of changing from current practice, where thrombolytic therapy is given in hospital, to paramedic practice where thrombolytic therapy is administered by appropriately trained paramedics (pre-hospital) for STEMI patients.

METHODS

A decision-analysis microsimulation model was constructed with a 30-day component and a long-term health state transition component. A brief review of the literature was undertaken to obtain data on time-to-needle to populate the model. The primary health outcome was quality-adjusted life years (QALYs); secondary outcomes included cardiac events, procedures and survival. Costs to the Australian healthcare system for the rest of life were taken as the analytical perspective.

RESULTS

On average, STEMI patients gain 0.13 QALYs at an additional life-time cost of $343. The incremental cost-effectiveness ratios were $3428 per life-year gained and $2601 per QALY gained. These estimates were robust to changes in a range of assumptions and parameter values. The most important factor was the time-to-needle - the greater the difference between current practice times and paramedic practice times, the greater the health benefits and lower the cost per QALY (and life-year) gained. A key factor in the model was the substantially lower incidence of heart failure from earlier time-to-needle. Importantly, there was little change in the cost per QALY gained for a wide range of ages; thus, there is no argument to limit thrombolysis by paramedics to above or below an age threshold.

CONCLUSIONS

Paramedics administering thrombolysis can avert some STEMI deaths and the pre-hospital administration of thrombolysis is good value for money.

A review of health care models for coronary heart disease interventions

This article reviews models for the treatment of coronary heart disease (CHD). Whereas most of the models described were developed to assess the cost effectiveness of different treatment strategies, other models have also been used to extrapolate clinical trials, for capacity and resource planning, or to predict the future population with heart disease. In this paper we investigate the use of modelling techniques in relation to different types of health intervention, and we discuss the assumptions and limitations of these approaches. Many of the models reviewed in this paper use decision tree models for acute or short term interventions, and Markov or state transition models for chronic or long term interventions. Discrete event simulation has, however, been used for more complex whole system models, and for modelling resource-constrained interventions and operational planning. Nearly all of the studies in our review used cohort-based models rather than population based models, and therefore few models could estimate the likely total costs and benefits for a population group. Most studies used de novo purpose built models consisting of only a small number of health states. Models of the whole disease system were less common. The model descriptions were often incomplete. We recommend that the reporting of model structure, assumptions and input parameters is more explicit, to reduce the risk of biased reporting and ensure greater confidence in the model results.

An unadjusted NNT was a moderately good predictor of health benefit

BACKGROUND AND OBJECTIVE

Whether the number needed to treat (NNT) is sufficiently precise to use in clinical practice remains unclear. We compared unadjusted NNTs to quality-adjusted life years (QALYs) gained, a more comprehensive measures of health benefit.

STUDY DESIGN AND SETTING

From a subset (n = 65) of a dataset of 228 cost-effectiveness analyses, we compared how well NNTs predicted clinically important QALY gains using correlation analysis, multivariable models and receiver-operator curve (ROC) analysis.

RESULTS

NNT was inversely correlated with QALY gains (P < .001); this relationship was affected by quality of life and life-expectancy gains of treatment (P <or= .04). The NNT is a moderately accurate predictor of treatments that provide large health benefits (area under ROC 0.74-0.81). For ruling out therapies with low QALY gains (threshold <or=0.125 to <or=0.5 QALYs), an NNT >15 had a sensitivity of 82% to 100%. For ruling in therapies with high QALY gains (threshold >or=0.125 to >or=0.5 QALYs), an NNT <or=5 had a specificity of 77%.

CONCLUSION

Using NNT thresholds of <or=5 and >15 to rule in and out therapies with large QALY gains may provide general guidance regarding the magnitude of health benefit.

Model of Evidence-Based Dental Decision Making

Shared decision making empowers patients as informed consumers, helping them to make clinical decisions that optimize their personal oral health. Evidence-based dentistry supports this process by providing best evidence that, when presented in visual and interactive formats, focuses consultation time on treatment or therapy options and their trade-offs. Currently, an explosion of evidence and technological advancements have necessitated a cooperation that translates into an interdisciplinary approach to care delivery. For evidence-based dentistry, this interdisciplinary approach includes the interaction of researchers, clinicians, and patients in promoting individual patient care. To facilitate this interaction, a computerized model of evidence-based dental decision making is presented to manage knowledge in its application to clinical practice. This model includes the use of decision aids and a decision tree composed of multiple clinical practice guidelines. These aids allow for quantifying treatment options in terms of estimates using probability, utility, and cost data. With these estimates, decision analysis and the flexibility to manipulate data provide patients with increased control and acceptance of the decisions that they make about their personal oral health.

Determining an appropriate threshold for referral to surgery for gastroesophageal reflux disease

BACKGROUND

Persistent symptomatic gastroesophageal reflux disease (GERD) can be treated with medication or surgery. The purposes of this study were (1) to determine how poor the quality of life on medication would need to be to justify assuming the risks of surgery, and (2) to estimate the proportion of patients currently on medication whose quality of life is below this value.

METHODS

We developed a Markov decision analysis model to simulate health outcomes (measured in quality adjusted life years [QALY]) over 10 years for medication and surgery in patients with typical GERD symptoms. We included probabilities of events obtained from a systematic literature review. Quality of life adjustments, expressed as utilities, were drawn from a survey of 131 patients 1 to 5 years after antireflux surgery. By using this model, we calculated what quality of life on medications would change the optimal strategy from medication to surgery (threshold). To determine the proportion of patients below this value, we prospectively surveyed 40 medically treated GERD patients at our hospital.

RESULTS

Surgery resulted in more QALYs than medical therapy when the utility with medication use was below 0.90. Sensitivity analysis showed this value to be relatively insensitive to reasonable variations in surgical risks (mortality, failures, reoperation) and quality of life after surgery. Among those surveyed on medications, 48% fell below this threshold and would be predicted to benefit from surgery.

CONCLUSION

Our model suggests that surgery would likely benefit a high proportion of medically treated GERD patients. Individual assessment of quality of life with GERD should be considered to aid clinical decision making.

An off-the-shelf help list: a comprehensive catalog of preference scores from published cost-utility analyses

PURPOSE

The Panel on Cost-Effectiveness in Health and Medicine recommends an organized collection of preference measure values for health states that can be used in costutility analyses (CUAs). The authors sought to construct a catalog of preference scores from published CUAs, organize the catalog by clinical categories, and identify methods of preference score assessment.

METHOD

The authors systematically searched Medline and other databases to identify original CUAs published through 1997. Information was abstracted on the health state descriptions, corresponding preference scores, method of preference score elicitation, and the source of the estimate.

RESULTS

Two hundred twenty-eight CUAs were appraised. The authors found 949 health states and corresponding preference scores. Most frequently, health states pertained to the circulatory system (21.7%), health states were valued by experts (35.8%), and values were derived through community-based preference scores (23.5%).

CONCLUSION

A catalog of preference scores for health states can be constructed. The catalog (http://www.hsph.harvard.edu/organizations/hcra/cuadatabase/ intro.html) may provide a useful reference tool for producers and consumers of CUAs but also underscores the methodologic variation and inconsistencies present in the field.

Health-related quality of life after stroke a comprehensive review

BACKGROUND AND PURPOSE

We performed a comprehensive review of all quality-of-life (QOL) estimates for stroke appearing in the peer-reviewed literature between 1985 and 2000. We examine variation in QOL weights and the rigor of methods used to assess QOL and discuss the implications for cost-utility assessment and resource allocation decisions.

METHODS

Through a systematic search, we identified 67 articles that met our inclusion criteria. A team of trained researchers read each article and followed detailed guidelines to extract QOL weights and other parameters. This effort yielded 161 QOL estimates for stroke-related health states. All estimates were measured on a 0 to 1 scale, with 0 representing the worst outcome and 1 representing the best.

RESULTS

QOL estimates range from -0.02 to 0.71 (n=67) for major stroke, from 0.12 to 0.81 (n=14) for moderate stroke, from 0.45 to 0.92 (n=38) for minor stroke, and from 0.29 to 0.903 (n=42) for general stroke. Although QOL should decrease with severity, there were many instances in which the QOL for major stroke as reported by one study exceeded the QOL for moderate stroke as reported by another. The same reversal was found for moderate and minor stroke, and it occurred even when both authors used similar assessment methods and subject populations. Authors of cost-utility and decision analyses rarely base their choice of QOL weights on their own primary data (19%). When obtaining weights from secondary sources, some authors (23%) chose QOL weights for a severity of stroke that did not match the severity for which they sought data.

CONCLUSIONS

QOL estimates for stroke vary greatly and are not always estimated in sound fashion. This impedes the comparability and quality of the cost-effectiveness studies that use these QOL weights and hampers good resource allocation decisions.

Decision analysis in the selection, design and application of clinical and health services research

Research evidence will never be sufficient to tell us how treatments compare in every group or class of patients. It will always be necessary to particularize from the general to the specific. The ad hoc way this is normally done contrasts with the rigour of the research process. However, extrapolating from the general to the particular can be made equally rigorous by Decision Analytic modelling. Furthermore, research evidence is not enough when decisions turn on multiple objectives, which must be traded off against each other. Research evidence and patients' preferences can be reconciled, again rigorously, by using Decision Analysis to show how probabilities and values interact. In addition, research should be designed around likely clinical impact, and this in turn can be made explicit by Decision Analytic modelling. In particular, sample size calculations should be based on such explicit modelling, rather than current procedures, which seldom seek to defend the size of clinical effects sought.

Patient preferences for thrombolytic therapy in acute myocardial infarction

BACKGROUND

Despite extensive professional debate regarding the optimal thrombolytic therapy strategy in acute myocardial infarction (AMI), patient preferences have not been explored.

METHODS

Preferences among patients with known or suspected coronary artery disease for treatment with tissue plasminogen activator (tPA) or streptokinase (SK) for AMI were determined using a questionnaire presenting GUSTO-1 trial and drug cost data. Preferences were based on consideration of 30-day mortality (M) alone, hemorrhagic stroke rate (SR) alone, overall preference (M + SR), drug acquisition costs, and the estimated annual costs of using a single agent to treat all AMIs. Cost-related responses were provided under payer designations of self, third-party insurance, and federal government.

RESULTS

The response rate was 81% (101/125 patients). tPA was preferred by 84%, and SK by 66%, for M alone and SR alone, respectively (chi 2, p < 0.01). Overall preference (M + SR) favored tPA (78%, p < 0.01). tPA preference decreased to 43% considering drug acquisition costs under the self-pay option (p < 0.01 vs M + SR). Similar trends of lesser magnitude were also observed for the third-party and government-payer options.

CONCLUSIONS

Under conditions of zero cost and consideration of mortality plus stroke-risk data, tPA were preferred overall due to its lower mortality. Introduction of drug-cost data significantly shifted the preference toward SK, particularly under the self-payer designation. Patient preferences for thrombolytic therapy in AMI indicate tradeoffs between clinical attributes and costs, and should assist in framing medical debate and decision making.

Cost-effectiveness of carotid endarterectomy in asymptomatic patients

PURPOSE

The purpose of this study was to determine the cost-effectiveness of carotid endarterectomy for treating asymptomatic patients with > or = 60% internal carotid stenosis, based on outcomes reported in the Asymptomatic Carotid Atherosclerosis Study (ACAS).

METHODS

A cost-effectiveness analysis was performed using a Markov decision model in which the probabilities for base-case analysis (average age, 67 years; 66% male; perioperative stroke plus death rate, 2.3%; ipsilateral stroke rate during medical management, 2.3% per year) were based on ACAS. The model assumed that patients who had TIAs or minor strokes during medical management crossed over to surgical treatment, and used the NASCET data to model the outcome of these now-symptomatic patients. Average cost of surgery ($8500), major stroke ($34,000 plus $18,000 per year), and other costs were based on local cost determinations plus a review of the published literature. Cost-effectiveness was calculated as the incremental cost of surgery per quality-adjusted life year (QALY) saved when compared with medical treatment, discounting at 5% per year. Sensitivity analysis was performed to determine the impact of key variables on cost-effectiveness.

RESULTS

In the base-case analysis, surgical treatment improved quality-adjusted life expectancy from 7.87 to 8.12 QALYs, at an incremental lifetime cost of $2041. This yielded an incremental cost-effectiveness ratio of $8,000 per QALY saved by surgical compared with medical treatment. The high cost of care after major stroke during medical management largely offset the initial cost of endarterectomy in the surgical group. Furthermore, 26% of medically managed patients eventually underwent endarterectomy because of symptom development, which also decreased the cost differential. Sensitivity analysis demonstrated that the relative cost of surgical treatment increased substantially with increasing age, increasing perioperative stroke rate, and decreasing stroke rate during medical management.

CONCLUSION

For the typical asymptomatic patient in ACAS with > or = 60% carotid stenosis, our results indicate that carotid endarterectomy is cost-effective when compared with other commonly accepted health care practices. Surgery does not appear cost-effective in very elderly patients, in settings where the operative stroke risk is high, or in patients with very low stroke risk without surgery.

'Evidence-based', 'cost-effective' and 'preference-driven' medicine: decision analysis based medical decision making is the pre-requisite

Three broad movements are seeking to change the world of medicine. The proponents of 'evidence-based medicine' are mainly concerned with ensuring that strategies of proven clinical effectiveness are adopted. Health economists are mainly concerned to establish that 'cost-effectiveness' and not 'clinical effectiveness' is the criterion used in determining option selection. A variety of patient support and public interest groups, including many health economists, are mainly concerned with ensuring that patient and public preferences drive clinical and policy decisions. This paper argues that decision analysis based medical decision making (DABMDM) constitutes the pre-requisite for the widespread introduction of the main principles embodied in evidence-based medicine, cost-effective medicine and preference-driven medicine; that, in the light of current modes of practice, seeking to promote these principles without a prior or simultaneous move to DABMDM is equivalent to asking the cart to move without the horse; and that in fact DABMDM subsumes and enjoins the valuable aspects of all three. Particular attention is paid to differentiating between DABMDM and EBM, by way of analysis of various expositions of EBM and examination of two recent empirical studies. EBM, as so far expounded, reflects a problem-solving attitude that results in a heavy concentration on RCTs and meta-analyses, rather than a broad decision making focus that concentrates on meeting all the requirements of a good clinical decision. The latter include: ensuring that inferences from RCTs and meta-analyses to individual patients (or patient groups) are made explicitly; paying equally serious attention to evidence on values and costs as to clinical evidence; and accepting the inadequacy of 'taking into account and bearing in mind' as a way of integrating the multiple and distinct elements of a decision.