Objective: There is increasing empirical evidence for the existence of bias in the publication of primary clinical research, with statistically significant results being published more readily, more quickly, and in higher impact journals. Meta-analysis of individual patient data (IPD) may represent a gold standard of “secondary” clinical research, giving the best possible summary of current evidence for a particular question, but publication of these may also be subject to bias. This study aimed to explore which factors might be associated with publication of IPD meta-analyses and to identify potential sources of bias.

Methods: For all known IPD meta-analysis projects in cancer, the responsible investigator was surveyed by means of a questionnaire to determine descriptive characteristics of the meta-analysis, the nature of the results, and details of the publication history.

Results: There is no good evidence that overall publication status of meta-analyses in cancer is dependent on the statistical or clinical significance of the results. However, those meta-analyses with nonsignificant results did seem to take longer to publish and were published in lower impact journals compared with those with more striking results.

Conclusions: Based on the current data, there seems to be no strong association between the results of IPD meta-analyses in cancer and publication.

The literature search for Bloom et al.'s review (Bloom BS, Retbi A, Dahan S, Jonsson E. Evaluation of randomized controlled trials on complementary and alternative medicine. IJTAHC, 2000;16:13–21) adopted and inadequate search strategy (an appropriate search of the Cochrane library finds nearly four times as many trials), excluded foreign languages, in which many CAM trials are published, and examined only blinded studies, when many CAM techniques (chiropractic, yoga) and associated outcomes (back pain, anxiety) are difficult to blind. The authors then complain about a lack of data. Is there such a term as “Procrustean” review?

Objective: Our aim was to assess the impact of six recommendations regarding drug prescription on the clinical practices of French psychiatrists. The recommendations were part of the conclusions of a consensus conference entitled “Long-term therapy of schizophrenia” (Paris, January 1994).

Methods: The impact of the conference was assessed on the basis of awareness of the existence of the conference, knowledge of its conclusions, and actual changes in clinical practice. We performed: a) a survey of a representative sample of 396 psychiatrists 2 years after the conference; and b) an analysis of changes in drug prescriptions in a cohort of 2,407 patients with schizophrenia under treatment at the time of the conference.

Results: Overall, 78% of interviewed psychiatrists were aware of the existence of the conference and 70% of its conclusions. Declared prescription practices conformed with conference conclusions about 60% (10%–95%) of the time. No difference in practices was noted between psychiatrists who were aware of the recommendations and those who were not. Single neuroleptic prescriptions increased in the cohort study in line with the main conference recommendation. The increase was small, but significant from 51.1% to 56.4%, and mainly concerned patients recently put on treatment. Contrary to recommendations, prescriptions of anticholinergics plus neuroleptics inexplicably rose from 48.2% to 54.3%.

Conclusion: Small changes in prescription habits occurred in the wake of the consensus conference, but we cannot really ascribe them to a direct impact of the conference. Despite the great pains we took in disseminating the conclusions of the conference as widely as possible, it is clear that a more forceful action plan (e.g., including continuous medical education) is required.

The use of economic evaluation in decision making appears to have increased over the past few years and economic evaluation is looked upon as another measure to help contain costs and improve efficiency in an evidence-based decision-making environment. Following the examples of Australia and the Canadian Province of Ontario, four European Union (EU) countries (Finland, the Netherlands, Portugal, and the United Kingdom) have recently introduced economic evaluation guidelines. In addition to the Australian and Canadian guidelines, which constitute a hurdle to reimbursement, the paradigm that seems to be evolving in the four EU countries follows a similar route. Finland and the Netherlands seem to be moving toward the notion of a fourth hurdle to reimbursement, whereas the National Institute for Clinical Excellence in England and Wales was in principle meant to influence practice, although in reality this essentially acts as a hurdle to reimbursement, requiring a different data set to that used by regulatory authorities. Whereas the Portuguese guidelines were developed to assist in preparing economic submissions to support reimbursement decisions, they are unclear about when such evidence will be required and also discuss the dissemination of economic evidence to broader audiences. The introduction of these guidelines poses a number of challenges to policy makers, the implications of which are analyzed in the paper: a) to ensure that economic evaluations are carried out scientifically without industrial or political bias; b) to define an acceptable methodology that would increase their credibility; and c) to address certain practical issues ranging from deciding how to use economic evaluations in policy making to setting up new institutions or improving the coordination and dissemination of evidence. The variation in the use of economic evaluation guidelines in the four EU countries highlights the differences in national pharmaceutical policies and is in line with policy makers' continuous attempts to contain costs. While the paper critically discusses the guidelines, it also points out that a series of methodologic issues need to be addressed if economic criteria are to be introduced in policy making with the aim to improve resource allocation. The paper concludes that economic evaluation as a discipline is beginning to impact on policy, whereas the consistent use of economic evaluation results is, in principle, being adopted by policy makers but needs to go a step further to reach practitioners.

Objectives: To determine if published figures on the proportion of articles included in systematic reviews and identified in electronic databases are applicable to an example from medical imaging.

Methods: A systematic review was performed. Additionally, sensitivity and precision of a MEDLINE search were compared with values from three published searches, each customized for a specific field.

Results: All articles included in the systematic review were in electronic databases. The MEDLINE search had low precision compared with searches in other fields.

Conclusions: In a specific area of medical imaging, electronic databases, including MEDLINE, are reliable sources of articles.

Objectives: To study the socioeconomic impact of rheumatic illness in Sweden and to discuss the consequences for technology assessment studies.

Methods: A cost-of-illness study based on data from official statistics and treatment studies.

Results: The total socioeconomic cost was 52 billion Swedish kronor (SEK) in 1994. The imbalance between direct (10% of total) and indirect costs (90effectiveness of the healthcare sector, the need for new treatment methods, appropriate information systems, and technology assessment studies as well as the institutional arrangements for rehabilitation and basic medical research.

Conclusions: A discussion of solutions for financial cooperation between county councils and regional social insurance offices should be considered. The new biotechnological pharmaceuticals will increase the cost for drugs in health care about 20 times, but the total socioeconomic cost for society may remain at the same level due to a decrease of inpatient costs and indirect costs for loss of production as well as a decrease of transfer payments from social insurance. It is unavoidable that the new pharmaceuticals require priority discussions and active resource allocation in health care and in other sectors of society.

Objectives: Single-vessel bypass can often be accomplished through less invasive techniques than conventional coronary artery bypass graft (CABG) at substantially lower cost. We undertook a study to empirically determine the cost savings associated with one such technique, left anterior small thoracotomy (LAST).

Methods: Reviewing medical and billing records, we measured the difference in hospitalization costs between two methods of coronary bypass surgery. The study groups consisted of 50 patients who underwent LAST and 28 who underwent single-vessel conventional CABG during 1995 and 1996. A subsequent validation sample of 50 patients who underwent LAST was also analyzed. Hospitalization costs were estimated using a relative value unit methodology and were risk-adjusted for both perioperative risk factors and changes in operating room technology.

Results: Risk-adjusted hospitalization costs for those undergoing LAST were $9,510 and $12,546 for the CABG control subjects (p < .01), with differences in surgical costs reflecting over 62% of this overall difference. Differences in average length of stay were under a half-day (10.0 for LAST vs. 10.46 for CABG). Only one inpatient fatality was reported; therefore, no inference regarding mortality differences could be made.

Conclusions: LAST is substantially less costly than conventional surgery, and the savings are potentially greater if hospital length of stay is reduced to a clinically recommended time of 2 days.

There is a time delay between the final recruitment of patients to a randomized controlled trial and the publication of results. The practical options available to decision makers during this gap can be listed according to whether all treatments are already widely available or whether at least one has been restricted to the trial. When the treatments are already in widespread use, the options are simply either to stop randomizing or to continue. When one trial treatment is restricted, there are further options: a) withdraw the restricted treatment altogether, pending the final analysis; b) continue to offer randomization, with a view to providing further data should these be needed; or c) make the intervention widely available to patients who would have previously been eligible for the trial. In this paper, we discuss the relative advantages and disadvantages of each option and discuss their attendant ethical implications. In particular, we suggest that continuing randomization is an option worthy of serious consideration. Randomizing patients acts as a “hedge” against the need for more data, given that sample size calculation is an inexact science. However, patients must be made aware of the basis on which randomization is offered.

Objectives: To collect and summarize information published by INAHTA agencies on the indications, habitual attitudes and practices, and economic and legal implications of preoperative evaluation in elective surgeries.

Method: The authors appraised the information contained in six papers published between 1989 and 1999 in Sweden, France, Basque Country, the Netherlands, United Kingdom, and Catalonia. The section on indications in preoperative evaluation does not present global conclusions. The sections addressing habitual attitudes and practices among physicians and those addressing economic and legal considerations cover only the similarities among the reports and the main ideas relating to these issues.

Results: The conclusions found in the reports about indications in preoperative evaluation are similar or differ slightly, e.g., as regards age limits in patients for whom the tests are recommended. However, more important differences are shown in other areas, especially in reports where consensus methods were used. In some instances, the opinions, attitudes, and customary practices of professionals during the preoperative stage do not concur with the recommendations extracted from the assessment reports and the customary practice of doctors. In relation to economic considerations, a substantial quantity of resources could be liberated if the recommended general clinical practices were followed. From the point of view of civil law, the evidence-based recommendations could be considered as a kind of coded lex artis.

Objective: A basic issue in randomized controlled trials (RCTs) is whether we can safely assume comparability between groups at baseline with respect to all potentially important prognostic factors. In other words, did randomization work sufficiently well? In small trials balanced allocation procedures are employed, whereas in large-scale trials simple randomization will do. The question is: When should balancing be considered?

Methods: We performed a simulation study in which we varied the number of categories in the prognostic factors and the number of patients.

Results: Simulation showed that, in all instances, a balancing procedure almost always led to perfect or almost perfect balance, while the imbalance with simple randomization was larger. To study the effect of balanced and random allocation on subgroup analyses in our OME trial, we compared the quotient of the width of the confidence intervals (CI). The widest CI in random allocation over the 13 hospitals was on average 13% wider than in balanced allocation.

Conclusion: Investigators should always consider balanced allocation, especially in categories with a low number of patients and when subgroup analysis over many categories is requested.

Objective: Our study aims to provide a practical contribution to the field of medical technology assessment within a new paradigm. This paradigm indicates the need for more comprehensive technology assessments in the development stage of a new technology.

Method: We introduce a method, based on Saaty's Analytic Hierarchy Process, which quantitatively supports discussions between the various actors that shape the technology's development and diffusion. These discussions focus on technical, medical, social, and economical requirements relevant to the design and diffusion of the new technology.

Discussion: In contrast to more traditional technology assessments, our method encompasses the perspectives of the diverse actors in the social context of technology development and diffusion. It influences their decision making on technology design and diffusion in order to improve this technology's later clinical as well as social effectiveness.

There is a growing interest in the potential use of conjoint analysis (CA) as a tool for estimating patient preferences within the economic evaluation of healthcare technologies. It has been suggested that where cost is included as one of the attributes within the exercise, CA can be used to estimate willingness to pay (WTP) for healthcare interventions indirectly within the framework of cost-benefit analysis. While recognizing the potential advantages of CA as a technique for the assessment of benefits in health care, this paper recommends several avenues of investigation that need to be carefully explored before an informed judgment can be made as to the validity of eliciting WTP values using the CA methodology. It is argued that much can be learned from the use of qualitative research techniques in examining individuals' understanding and interpretation of CA questions where cost is included as an attribute in the exercise. Additionally, further research is required in defining techniques for establishing the most appropriate levels for the cost attribute in a CA exercise and in determining the sensitivity of WTP estimates to the levels that are chosen for the cost attribute.

Objectives: This study compared the relative cost-effectiveness of stroke care provided in London and Copenhagen.

Methods: Hospitalized stroke patients at centers in London (1995–96) and Copenhagen (1994–95) were included. Each patient's use of hospital and community health services was recorded for 1 year after stroke. Center-specific unit costs were collected and converted into dollars using the Purchasing Power Parity Index. An incremental cost-effectiveness ratio (ICER) was calculated comparing a Copenhagen model of stroke care to a London model, using regression analysis to adjust for case-mix differences.

Results: A total of 625 patients (297 in Copenhagen, 328 in London) were included in the analysis. Most patients in London (85%) wereadmitted to general medical wards, with 26% subsequently transferred to a stroke unit. In Copenhagen, 57% of patients were directly admitted to a stroke or neurology unit, with 23% then transferred to a separate rehabilitation hospital. The average length of total hospital stay was 11 days longer in Copenhagen. Patients in Copenhagen were less likely to die than those in London; for patients with cerebral infarction the hazard ratio after case-mix adjustment was 0.53 (95% CI from 0.35 to 0.80). However, a lower proportion of patients with hemorrhagic stroke died in London. The ICER of using the Copenhagen compared with the London model of care ranged from $21,579 to $37,444 per life-year gained for patients with cerebral infarctions.

Conclusions: The ICERs of the Copenhagen compared with the London model of care were within a range generally regarded as cost-effective.

Objectives: The EUR-ASSESS project was undertaken by a large number of members of European health technology assessment (HTA) agencies and programs with the aim of improving coordination through developments in several areas of HTA. At the conclusion of the project, an independent survey was undertaken to reassess the importance of the original objectives of EUR-ASSESS and determine the impact of the project on the way HTA activities are organized and carried out in Europe.

Methods: A postal questionnaire survey and a series of more in-depth personal interviews.

Results: EUR-ASSESS was generally considered successful in achieving its aims, particularly in creating an informal network of people and organizations, improving the understanding of the work of others, improving the ability to focus the objectives of HTA, facilitating the sharing of experience and mutual learning opportunities, exchanging ideas on research agendas, and developing a common language. However, significant differences emerged in the expectations of different actors in European HTA, and in their perceptions of the impact of the project. These reflect the wide diversity of approaches existing in Europe.

Conclusions: Health technology assessment in Europe is changing its nature from a set of national or local initiatives with relatively limited impact to a truly European movement with strong links and coordination between its individual components. Important lessons can be learned from the experience of EUR-ASSESS, particularly in the light of an increasing commitment of the European Commission to play a role in the development of technology assessment.

This paper considers the question of what ought to be valued in the context of measuring the outcomes of healthcare interventions. The answer is discrete health states in the case of the quality-adjusted life-year (QALY) model and an entire health profile in the case of the healthy-years equivalent (HYE) approach. How well the weighted average of values attached to the former approximates the overall value attached to the latter depends on the validity of the assumptions of the QALY model. The paper considers some of the empirical literature relating to them. One of the most important assumptions, which from the limited evidence available appears not to hold, is additive separability. However, it is argued that violation of this assumption does not in itself invalidate the QALY approach, since in some circumstances it might be more appropriate to elicit the value of a health state independently of the states that succeed it. Investigation into this issue is identified as one of the key areas where future research efforts should be directed.

Objectives: Several instruments for measuring health-related quality of life (HRQOL) have been developed, and others are under construction. The problem is whether the different HRQOL measures show comparable results. We first compared the functional relationship of the Nottingham Health Profile (NHP) and the 15-dimensional measure of HRQOL (15D) in hip and knee replacement patients. The hypothesis was that condition or intervention does not affect the functional relationship between NHP and 15D changes.

Methods: We assessed the agreement of the instruments by comparing observed changes in the 15D and its fitted values derived by regressing the 15D by the NHP dimensions. Patients (n = 452) were recruited consecutively from seven Finnish orthopedic departments during April 1991–May 1992. HRQOL was measured prior to surgery and 6, 12, and 24 months postoperatively.

Results: There was a different functional relationship between the HRQOL instruments in hip and knee patients; they agreed upon the direction of changes in HRQOL in 84% and showed opposite signs in 12%. The NHP showed significantly more improvement in quality of life than the 15D.

Conclusions: The two instruments were in slight but significant disagreement. Estimates of effectiveness can vary according to the instrument used. Comparisons of effectiveness in healthcare programs measured by differing instruments need information on the functional discrepancies between the instruments in the conditions and interventions in which they are applied.

Objective: The purpose of this study is to analyze the perception of the content of health technology assessment (HTA) among health professionals applying for a state grant of DKK 10 million.

Methods: A total of 113 applications were received and analyzed. When conducting the analysis, it was assumed that the applicants' maximum five-page project description would reflect: a) the applicants' perception of what an HTA is; b) how the assessment was to be conducted; and c) what the results were going to be used for.

Results: More than 40% of all applications focused on treatment; in 51% only one or two professional groups were to be involved (thus interdisciplinarity was questionable); only 22% of the HTA cases were intended to form the basis for political/administrative decisions; in general, the HTAs were planned far less comprehensively than was relevant; 76% of the projects did not include a formal synthesis phase; 41% intended to use diffusion as the only method for publication of the HTA result.

Conclusions: The analysis reveals several areas where DIHTA has to make an effort in order to secure that HTA in fact constitutes a comprehensive and well-documented basis for decision making. These areas concern the following topics: multidisciplinarity, the objective of HTA, comprehensiveness, the synthesis phase, and publication and utilization of the HTA result.

This paper presents the results of a contingent valuation study measuring willingness to pay (WTP) for treatment of patients with von Willebrand's disease. Median WTP for treatment of this disorder was $1,500 or $3,500, depending on how the initial bid was structured. Regression analysis shows that income, education, and a category rating scale for health status were significant in predicting WTP. The adjusted annual WTP was $2,178. WTP surveys may increasingly be useful for health technology assessment. Starting point bias in how the bids are structured must be recognized.

Objectives: To determine whether unpublished dissertations are an important source of trials for systematic reviews.

Methods: In a review of infant massage, we identified 17 dissertations. We recorded whether each dissertation was included in the review and, if so, whether it contributed data to any analyses.

Results: Thirteen dissertations were excluded, 11 because of poor reporting of methodology. Three dissertations were also published as journal papers. The data from the remaining dissertation were not included in any analysis. Reviewing data on the Cochrane Library, only one of 878 reviews included data from a dissertation that might have changed a review's conclusions.

Conclusions: Searching for and retrieving unpublished dissertations involves considerable time and effort and appears to influence the conclusions of a review only rarely.

This document focuses on the scientific evidence presently available on the patient–doctor relationship. As the patient seeks care, the physician represents only one of many relationships that the patient forms, and relationships with physicians and other professional caregivers have a mutual influence on each other. This critical, systematic review is based on approximately 800 studies that were selected according to predetermined quality criteria. Most of the studies were performed in the United States, but several are from Canada, the Netherlands, Great Britain, and Sweden. Most studies involve patients treated in ambulatory care settings.