Health Care Technology Assessment (HTA): Developing & Using the Evidence in Health Care Decisions

(The following were presented during the Second Plenary Session at the ISPOR 1st Latin America Conference, 9, 10, 11 September 2007, Cartagena, Colombia)

What Is Health Technology Assessment - And How Is European Network For HTA (EUnetHTA) Developing Transnational Collaboration?

Finn Børlum Kristensen MD, PhD, Director, Danish Centre for Health Technology Assessment (DACEHTA), Monitoring and HTA, National Board of Health, Copenhagen, Denmark, Adjunct professor, University of Southern Denmark - for the European Network for HTA, EUnetHTA

This article will address: 1) Health Technology Assessment (HTA); 2) the relationship between HTA and policy processes; 3) the need for transnational collaboration in HTA in Europe; and 4) the European collaborative project - EUnetHTA.

Health Technology Assessment
Health technology refers to the practical application of scientific knowledge in health care: All the kinds of interventions that we can do in relation to patients and to health of the population. Health technology is much more than devices and machines, it is more than pharmaceuticals, it is also e.g. cognitive interventions in the area of psychiatry, and it is diagnostic procedures, not necessarily using a lot of technical equipment. Technology assessment in health care is a multidisciplinary field of policy analysis. It studies the medical, social, ethical, and economic implications of development, diffusion, and use of health technology [1].

HTA is multidisciplinary; you cannot really do a full HTA without having clinical researchers from health care practice involved together with researchers from other areas, such as health economics, epidemiology, organization research, and perhaps social scientists to cover patient and citizen aspects.

HTA is a kind of policy analysis. Policy analysis means analysis of different ways of implementing alternatives, say, a new technology, and seeing the consequences of different options -and HTA falls into this broad area. HTA does not just address the medical implications but also the broader societal implications, the implication for the patient, as well as the economic implications.

HTA may very well inform regulatory processes. In fact, increasingly HTA has a defined role in the development of the basis for regulatory decisions - but it is not regulation itself. It provides information that helps pointing to different options and consequences, but does not give one direct answer (“yes / no”). In other words, it is important for understanding the concept and practice of HTA that HTA is not mandating policies: The resulting report is not the decision on a policy.

It is fundamentally important that HTA is firmly rooted in research and research methods. However, this aspect will not be developed further in the article.

In the Danish HTA model we say that we will address the technology, the patient, the economy, and the organization [2]. “Organization” is organization of health care, use of human resources, use of doctor time / nurse time and other issues related to the application of the technology. “Patient” does not refer to the direct clinical outcomes etc. of the technology or the side-effects (which would fall under the headline “technology”) but the patient as a whole individual, the patient as a member of society. Questions here might be: 'will it improve the possibility for the patient to get back to work quicker?' “Economy” covers health economic issues as applied in e.g. pharmacoeconomics. The work is multidisciplinary using rigorous methods. It should be transparent, and it should be synthesized into a report. It is very important to ensure that this process is as transparent as possible for the user to know how the HTA came from these questions to the synthesis to the report.

The Relationship between HTA and Policy Processes
Ideally HTA is linking - and in practice HTA is increasingly linked - into policymaking and decision-making. The role of HTA has been compared with that of a bridge between research and decision-making [3]. The Figure illustrates the close relation between HTA and policy-making and depicts the interdependence, yet division of work, between research-based assessment and decisionmaking. A successful process from policy question(s) to an HTA report that informs policy spans across paradigms in a conscious and transparent way.

When preparing a decision you have one or more policy questions on ways of implementing policy options, say, in the area of diabetes or rheumatology, or a new vaccine. A “translation process” produces a number of HTA questions - these are more specific questions that are formulated in a way that is actually answerable. One question might be: 'what is actually the size of this health problem?' leading to: “how big is the population that is going to be targeted with the potential intervention?” and “which negative health consequences can be avoided?' etc.

Figure 1. From policy questions to HTA reports that inform policy

Figure 1 illustrates that the process of doing HTA moves across domains. To the left is a decision-making domain, and regulation is included in that. In this domain people are developing the policies for decision-makers politicians to decide; it can also be clinical policies at the clinical level. To the right is the research domain. People in academia work in the research domain and do not see them as having anything directly to do with the decision-making domain. They're doing research. The paradigm of policy-making is the theoretical basis of those who are supporting the decision-makers. Policy-makers are increasingly people who have an academic background from university in political science, economics, public health etc. They are developing policies for the decision- makers, and learned about research in their studies. Most of the HTA parSeptember/ October 2008 ISPOR CONNECTIONS 15 adigm is actually in the research domain but it extends into the decision-making domain: HTA should be practical in the sense that it should be useful for planning and decision-making.

Figure 1 depicts from left to right the process from policy questions leading into more specific HTA questions which are handled in the research domain, reported - and then fed back to the decision-making domain. Currently the ways of reporting and feeding back into the system are developed in various countries. It is important to ensure clarity on who is doing what from HTA to decision. HTA and decision-making is not the same.

In England and Wales the National Institute for Health and Clinical Excellence (NICE) do appraisals using the evidence coming from HTA in a process that leads to guidance [4]. This is policy-making beyond HTA, and even involves a decision. HTA is building on robust methods coming from research. Decisions involve other important criteria as well - this is where the appraisal comes in. Currently there is discussion in Europe about: 'where does the assessment (policy analysis) stop' and 'where does “appraisal” (policy-making) start'? There is not one answer; it varies from country to country setting.

The Need for Transnational Collaboration in HTA in Europe
Researchers from academic groups in UK did a study of HTA that took a fresh look at the use of HTA and the potentials of HTA in Europe [5]. These are some of the conclusions from that report: HTA can play a valuable role but it has to be transparent, timely, relevant, in-depth and usable. This is very much asking for practical implementation of scientific methods. This assumes that HTA is built on sound methods. It has to be available when the decision-makers need this kind of information, and not after the decision was made. It has to use robust methods, and be supplemented by other important criteria. The authors conclude that if you maximize the potential of HTA the decision-makers will actually be helped.

So what kind of challenges does this raise? One challenge is about avoiding unnecessary duplication when several health systems express a need for HTA. Take an example like PET scanning. A lot of parallel reports have come out and differed in terms of their conclusions. It was raised as a problem for international HTA that reports were rather different - also in terms of the kind of advice one could derive from them.

In the area of HPV vaccination against cervical cancer we tried to coordinate in Europe, about 8 countries did HTAs on the same topic - in most cases using the same evidence. We need to use best practice, share what can be shared, and find out how to produce information in a shared way, that can then be adapted into national or regional versions, taking into account that the regions differ, even within the same country, or at least across Europe.

International collaboration in the area of HTA has been there for more than 20 years. There is an international organization called INAHTA which is an association for agencies. It has more than 40 members from the Americas, Europe and Austral-Asia. It's been very useful for agencies to participate in this organization, especially during establishment. There's also an international society for HTA, HTAi, which has an international journal and annual meetings.

The European Collaborative Project - EUnetHTA
The European Network for HTA is a response to an express need that came from the member states in the European Union and the European Commission. They asked for a sustainable network for HTA Europe. EUnetHTA is a 3-year project from 2006-08, and it is big - it has 34 Associated Partners who contribute financially and receive a grant from the EU (BOX 1). There are also 29 Collaborating Partners including international organizations. Partners are from nearly all of the EU countries. There are also a few institutions from other countries like the AHRQ in the USA and CADTH in Canada.

The objective of the project is to establish an effective and sustainable network. It shall help informing policy decisions. That said, research and academic work is absolutely necessary for progress, again taking into account that HTA is practically oriented. It should be useful for decision-makers. The project intends to bring agencies together with research institutions, and have links to health ministries that do the policy-making. This will promote effective support to policy decisions in the member states.

Health care policy is a national issue in the EU. Coordination of methods is an EU issue with the member states. EUnetHTA partners want effective use of their national resources, and there are many examples, where we could do better in terms of reducing duplication of efforts. This does not mean that there should be one HTA for the whole of Europe on e.g. HPV vaccination or the treatment of Alzheimer's disease, but there might be components of the work that can be shared. Through transnational work partners will be able to increase the output from the often relatively small agencies that are doing HTA in the countries. The vision is that we can share some work between counties thus avoiding parallel work from scratch on the same topic at the same time and be able to have more output.

It is also a EUnetHTA objective to strengthen the link between HTA and health care policy-making. In the HTA environment there are still some people who really prefer not to have anything to do with policy- and decision-making at all. They want to stay in their academic setting and the research domain only. This can work well provided that such contributors to HTA work closely together with e.g. HTA agency people who ensure the policy relevance of the reporting.

It has proven helpful that the work in the EUnetHTA project was organized in several work packages with objectives, milestones and deliverables described in a protocol (See www.eunethta.net for more information).

Examples of EUnetHTA Work
It has already been mentioned that HTA is implemented differently, and it is difficult to extract data from existing reports because the structure of the reports and the methods are not sufficiently clearly described in the reports coming out of HTA agencies. In parallel two work packages either de novo address a technology with HTA, or adapt existing HTA reporting, leading into reporting and informing policy.

One Work Package is about trying to define common core information that can be shared when doing HTA. The project is testing the possibility of producing core information on a certain technology that can be shared across national reporting. Not least data on efficacy data and clinical effectiveness data may be shareable. The project develops two models - one model for therapeutic interventions, and one model for diagnostic interventions. Two examples of “HTA core information” will be produced in the project; the first core HTA, which is drug-eluting stents, the second on multislice CT of coronary arteries.

EUnetHTA use 9 domains that were defined in the previous EU project' EURASSESS):

• Current use of the technology (implementation level);
• Description and technical characteristics of technology;
• Safety;
• Effectiveness;
• Costs, economic evaluation;
• Ethical aspects;
• Organizational aspects; and
• Social aspects

The domains define the landscapes of HTA that you should consider when you plan a concrete HTA. You don't necessarily have to address them all. In certain technologies it may be useful to look at legal aspects while in most cases it may be completely irrelevant. One would expect domains like effectiveness and safety, issues of cost and economic evaluation, etc. to be covered in an HTA.

The work is centered round defining a common core of information. In the domain of ethics, is there a big or a small part of the whole of ethical issues > 16 September/October 2008 ISPOR CONNECTIONS that can be defined as a common core? Probably not very much will be shareable because in Europe we have people with e.g. different religious and moral approaches to the individual and society. So probably in ethics it is quite difficult to share a lot of the ethical discourse.

In the domains of efficacy and effectiveness the common core may extend quite a bit into national reporting with shareable information from published reports and systematic reviews like Cochrane Systematic Reviews. Under each domain there are a number of topics - topics in the domain of clinical effectiveness could be e.g. influence on mortality, survival. Those outcomes could be defined into more specific issues. It's analytic work; you divide things into the smallest bit that is necessary to have a genuine element of information that may be shared.

Another Work Package is about adapting existing HTA reports into new settings.. This work is trying to dig into existing reports and trying to take out the things that might be transferable. Currently, agencies (especially in smaller countries) sometimes take existing reports from another country, if practicable for language and other reasons, and make a national version, to be able to deliver HTA reporting sufficiently quickly. There are problems with getting information out of existing reports in the right way. There is now an “Adaptation Toolkit” which is being tested for usefulness.

These examples illustrate what should be in place if we want to have real transnational collaboration. We need to have a common taxonomy and a common framework - and we need to find out to what extent there is a common core of information that can be shared irrespective of health system.

Conclusion
In summary, transnational HTA collaboration does not interfere with maintaining national governance in healthcare. EUnetHTA is doing practical, methodological work to share information. We are not taking the decision power out of the hands those who make decisions in the area of the use of health technology. There is a lot of enthusiasm in EUnetHTA and time will show how far we will get.

References

1. International Network of Agencies for Health Technology Assessment. About HTA. http://www.inahta.org/HTA/ (accessed 25 February 2008)
2. Kristensen FB, Sigmund H (eds.). Health Technology Assessment Handbook. 2nd Ed. Copenhagen: Danish Centre for Health Technology Assessment, National Board of Health, 2007.
3. Battista RN, Hodge NJ. The development of health care technology assessment. An international perspective. Int J Technol Assess Health Care 1995;11:287-300.
4. Walley T. Health technology assessment in England: assessment and appraisal. MJA 2007;187:283-5.
5. Sorenson S, Kanavos P, Drummond M. Ensuring Value for Money in Healthcare - the Role of HTA in the European Union. In: Financing Sustainable Healthcare in Europe (Cox Report), 2007 (http://www.sustainhealthcare.org/cox.php).
6. Banta HD et al. (eds.). Introduction to the EUR-ASSESS Project. Int J Technol Assess Health Care 1997;13:133-43.

Results of The NEVALAT ProjectResults of The NEVALAT Project

Joan Rovira, Universidad de Barcelona, Spain and Guillermina Albarracín, Barcelona, Spain

GLOBAL OVERVIEW
This article presents the results of the NEVALAT (Network for the Economic Evaluation of Health in Latin America) Project. This project was a follow-up of a previous initiative, called the REDEFS, Red de Economía y Financiamiento de la Salud [Financing and Health Economy Network] that was promoted in the early 90's by the World Bank, PAHO and the IDB. By the mid 90's, these institutions discontinued the sponsorship of the REDEFS and so those who had participated - researchers in Latin America and some researchers in Europe - saw the need to get organized in order to maintain the collaboration. They prepared a project and applied for a grant to the DG XII (Directorate for Research) of the European Commission (EC) and began working on this project from 2000 to 2004. There were teams from the following countries: Argentina, Brazil, Colombia, Cuba, Mexico, Nicaragua, Peru and Uruguay in Latin America and from Spain and the UK in Europe. The overall outcome of the NEVALAT project is an informal open network of contacts between professionals and a set of virtual information assets and analytical tools that are publicly available but quickly becoming out-dated.

The NEVALAT activities were organized into several modules or work packages. One of the most important parts of the project was the Work Package aimed at assessing how far economic evaluation had already been implemented for decision making purposes, as well as the identification of the relevant decision-makers and of future opportunities for using economic assessment in the region. The current situation was analyzed by means of a survey aimed at eliciting 1) the mechanisms actually used to allocate health resources in the countries of the region (there were 10 countries surveyed at that point in time); 2) how resources were being allocated; and 3) whether it was any evidence that economic assessment was being used for decision-making regarding priority setting, resource allocation, pricing and financing health services and technologies, etc. The results of the survey are available at Iglesias, Drummond and Rovira [1]. At the time the survey was done (2002) the scenario was quite discouraging and disappointing, as it became evident that, although a certain number of good quality studies had been produced, economic evaluation was practically not used for any policy and management purposes in the region. As has been evidenced recently by the presentations given at the ISPOR 1st Latin America Conference, the situation seems to be improving in many countries, especially in Brazil, where some regulations requiring the presentation of economic evaluations have been introduced, but also in Mexico and Argentina. The situation has moved forward and there is an explicit and growing interest on the part of decision-makers on this approach.

A second Work Package was aimed at eliciting the knowledge, actual use and attitudes towards economic evaluation in health care among decision makers in each country. The main purpose was to assess whether the limited use of economic evaluation was out of ignorance, or because of other reasons: e.g. they believed that it is not ethical to base clinical and health policy decisions on money and cost considerations, or they did not trust the results of the studies. Information on the attitudes of several groups of health professionals and decision- makers was collected using a combination of techniques - focus groups, personal interviews and postal surveys -in some countries of the project, especially in Argentina and Cuba. The answer to the question “Why is economic evaluation not much used in Latin America?” is an important one for professionals and organizations committed to the diffusion of this approach in the region. In Europe and in other countries with a longer tradition in economic evaluation, those who sponsor and do the studies are often surprised to know that most decision-makers are not making a regular use of economic assessment tools, and the reaction is often one of blaming those decision-makers and assuming that they either do not want to do it for some hidden reasons, or because they do not know or understand the techniques. Critics seldom realize that decision-makers might have very rational and legitimate reasons for not using economic evaluations, such as, the lack of transparency, of the method- • 59 partners • 34 Associated Partners (national/regional HTA agencies, MoH, research institutions, international organisations (Cochrane Collaboration) • 25 Collaborating Partners (e.g. WHO, OECD, CoE) • 24 EU countries ( ÷ Slovakia) • 2 EEA (Norway, Iceland) • Switzerland •4 institutions in countries outside Europe (Israel, Australia, Canada, USA) September/October 2008 ISPOR CONNECTIONS 17 ology, the validity of the studies or the relevance of the results for the actual situations. The results of that part of the project have clear implications for sponsors and analysts of economic evaluation studies in the region: before involving themselves in financing and producing economic assessment it would be advisable for them to learn more about the users' needs and perceptions and to ensure that the studies they produce are reliable and relevant to the target audiences.

The Work Package on Standardization of the Methodology tried to promote standardization as it is already done in some countries. But no progress was attained along these lines except for Cuba. A possible explanation for this is that at the time the project was conducted, no agencies in the region were using economic evaluation for financing, pricing or regulatory purposes and they consequently felt no need for, nor were they receptive to proposals of standardization. As part of the process of standardization NEVALAT promoted, among other tools, the development of national unit costs databases, and PREVE, a software program for carrying out cost-effectiveness analyses, so as to facilitate the implementation and transference of the results of studies across countries. It also produced a web page (www.nevalat.org) with an inventory of the economic assessments carried out in the region in Spanish, Portuguese and English and a directory of researchers and centers involved in economic evaluation in health.

LITERATURE REVIEW OF ECONOMIC EVALUATION STUDIES IN LATIN AMERICA
The review of economic assessment studies done in the region was done by means of a common form for summarizing the studies, which was as simplified version of the one used by the Office of Health Economics, Health Economics Evaluations Database (OHE-HEED). The OHE had developed a database that included more than 25,000 references of economic assessment papers and studies. The use of a compatible form was aimed at allowing the comparison of what was being done in Latin America vis-à-vis the rest of the world. The review started with a series of electronic searches on the internet and at the OHE-HEED, but it was complemented with manual searches of grey literature at national level by the researchers of the participating countries.

The first compilation was carried out by Jerry Crosby, of the Health Economic Office, which conducted a study of 115 papers identified in the OHE-HEED as full economic evaluation studies performed in all Latin American countries, published in the period 1992-2002. He did a descriptive analysis according to several variables. Most of the studies - 50 out of the 115 - were drug assessments. The data showed a large proportion of studies that involved infectious diseases when compared to the aggregate results for all countries. In relation to the studies' sponsorship, the results show that the pharmaceutical industry did not have a majority role in financing these studies, although this information was only available in roughly half of the studies. Most studies had been financed by government or by public agencies. These results were presented at the ISPOR 9th Annual International Meeting [4].

A second analysis was published by Rovira and Sanabria [2]. It comprised about 300 economic studies including economic assessments of health technologies, but also costs and cost of disease studies. It covered the period 1960 and 2003, and included both published and unpublished studies, but only from the countries that were members of the NEVALAT project. The review did not make a quality assessment of the papers, but is only a descriptive bibliometric analysis aimed at finding some general trends in the literature.

Iglesias, Drummond and Rovira [1] also included a list of references of full economic evaluation studies found in the OHE-HEED database and in the NHS Economic Evaluation Database at the University of York.

It is worth mentioning, as well, the review published in 2006 by Machado, Iskedjian and Einarson [3]. They carried out a search in MEDLINE, EMBASE, LILACS and SIETES using the following key words: cost-effectiveness, costutility, cost reduction, cost-benefit, and all the names of all South American countries. They selected studies that had considered both costs and outcomes of a technology. The papers were rated for quality according to a list of 12 criteria. The quality of most studies was evaluated as “poor” or “acceptable” and the overall level was considered relatively low in comparison to the results of similar exercises done in other countries.

The last updating of the literature that has been done for the present paper took into account the previous reviews quoted above. It is worth noting the limited level of overlapping across the groups of papers selected in each review. For instance, the NEVALAT review [1] (93 papers) and the Machado et al (2005) review (24 papers) have only 12 papers in common in the period 1984-2002. Similarly, Machado and Iglesias have only 5 common papers in the same period. Out of the 93 papers selected in the NEVALAT review, only 33 were in the York database. These findings highlight how much searches on a given topic can yield widely different results according to the methodology, keywords, databases and selection criteria applied.

The updating of the inventory of economic evaluation studies in Latin America tried to be consistent with the previous studies on the topic. Table 1 summarizes the search strategy. Table 2 displays the number and time distribution of the studies found in the various reviews mentioned in the paper. Our last search in Pubmed (last column) identified 143 out of the 174 studies found by all searches combined (first column). This updating exercise added 18 studies from the period covered by the NEVALAT review [1] that had been missed for some reason in the previous search. For the period posterior to the NEVALAT search, 2003-2007, 65 papers were added from the York University database, 5 from Machado et al. [4] and 25 from the additional PubMed search done by the authors of this paper.

Table 1. Summary description and results of the search by source

Table 2. Distribution of studies by source and year of publication

To sum up, the search results of the NEVALAT project, up to 2003 (Iglesias et al) included 93 references, but the present review raises this number to 174 as of August 2007. The distribution of studies by year of publication suggests a real leap between 1995 and 1998 and an apparent stabilization in the number of publications thereafter (Graph 1). In other words the number of published studies rose from 2 or 4 studies per year to about 14 or 15 per year on average. The distribution by country shows a high concentration in a few countries: the six countries with the higher number of studies Brazil, Mexico, Argentina, Colombia, Chile and Peru account for over 90% of the total number of studies (Graph 2). It is also worth mentioning that most of the studies are single country studies: only 12 (14%) of the studies are multi-country. Multi-country studies involve usually (10 studies) Brazil and Argentina. NEVALAT tried to promote cross-national studies assuming this would improve the quality and efficiency and the possibility of comparing results across countries.

Graph 1. Evolution of the number of published studies

Graph 2. Distribution of studies by country (only countries with three or more studies)

CONCLUSIONS
The NEVALAT project was aimed at supporting the production and use of economic evaluation studies by building a network of researchers, decision-makers and institutions interested in the topic and facilitating some key elements, such as general information, databases on researchers, studies, standard tools such as software programs and unit cost databases and other analytical tools.

The use of economic evaluation for decision-making purposes is still limited in Latin America, but its role is rapidly growing.

According to the literature searches done in the context of the NEVALAT project and by other researchers, there are a considerable number of studies on economic assessment in health. The quality may not be optimal, according to the sample analyzed by Machado et al, but the skills for improvement seem to be available in the region.

The NEVALAT initiative needs new inputs or continuity somewhere else. ISPOR might provide the appropriate environment for providing a new momentum to the evolution of economic evaluation in Latin America.

The full report and tools of the NEVALAT Project is available upon request from Joan Rovira, University of Barcelona, joanrovira@ub.edu.

REFERENCES

[1] Iglesias CP, Drummond MF, Rovira J. NEVALAT Project Group. Health-care decision-making processes in Latin America: problems and prospects for the use of economic evaluation. International J Technol Assess Health Care 2005;21:1-14.

[2] Rovira J, Sanabria C. La evaluación económica en salud y la toma de decisiones en Iberoamérica. Revista de la Facultad de Ciencias Económicas de la UNMSM 2005;10:27-46.

[3] Machado M, Iskedjian M, Einarson TR. Quality assessment of published health economic analyses from South America. Ann Pharmacother 2006;40:943-9.

[4] Rovira J. Developing strategy for promoting the use of economic evaluation in health care decision making in Latin America [oral presentation]. ISPOR 9th Annual International Meeting, Arlington, VA, USA, May 16-19, 2004.

HTA Experiences from the University HealthSystem Consortium

Karl Matuszewski MS, PharmD, Senior Director, Clinical Knowledge Service, University HealthSystem Consortium, Clinical Practice Advancement Center, Oak Brook IL USA

Many from the pharmaceutical and device industry believe that individuals who assess medical technologies are evil technology suppressors who don't like innovation. That is not true. Assessors love medical technology. The only problem is that people who assess medical technology for a living are the types who when their wife asks them “does this dress make me look fat?” will think about it deeply for five seconds and boldly answer “not really.”

There are a number of examples of very good medical technologies that many health technology assessment (HTA) agencies have looked at - lung transplants, liposomal amphotericin, implantable cardioverters and defibrillators - all great technologies, not likely to be misused. Clinicians are knowledgeable and do good patient screening when using these technologies. Then there are categories of medical technologies that are good, but often misused, such as antibiotics, bariatric surgery, pulmonary artery catheters, and red blood cell transfusions. When a clinician is faced with a parent whose child has an ear infection, it is (at least in the United States) very easy to prescribe an antibiotic while knowing in all reality that the infection is probably of viral origin and the antibiotic is doing no good. Finally, to compile a list of technologies that are truly bad, you have to go into the distant past for blood-letting and lobotomies. There have been some recent examples, such as intraoptic nerve decompression and aprotinin, although in many cases over-aggressive marketing and inappropriate clinical use may be the culprit.

HTA is of great interest to government agencies and insurance companies that pay for a substantial portion of health care. But, it should also be of interest to academic medical centers (AMCs) because of the tremendous amount of new medical technology introduced into the market every year. About 20-30 new molecular entities are approved annually in the US by the FDA, as well as thousands of medical devices via the 510(k) process demonstrating substantial equivalence to existing devices, or via pre-market authorization (PMA) requiring some level of clinical evidence. AMCs purchase these products and often hope that some level of reimbursement will be forthcoming.

The US health care system is complex and confusing, with numerous payers, such as Medicare, Medicaid, and thousands of commercial insurers administering policies for services at over 4,000 hospitals. These institutions often function as independent entities, with little or no collaboration or coordination of services. Dilemmas exist with one payer paying for a specific medical techSeptember/ October 2008 ISPOR CONNECTIONS 19 nology and another not paying. This 'schizophrenic environment' would seem to be a reasonable call for sophisticated assessment by hospital for incorporating any new technologies into routine practice.

Unfortunately, the state of formal HTA in US hospitals is dismal. Probably less than 25% have any sort of technology assessment infrastructure and process, excluding the P&T committee and formularies that evaluate a new drug's addition to the clinical armamentarium. New devices and medical/surgical procedures are rarely evaluated. There is usually limited discussion about clinical evidence, safety, and all the issues that formal technology assessment reviews. Many promote quality of care as the number one goal for American hospitals - it is often in our mission statements - but financial viability will be equally vital in the coming years. AMCs are beginning to recognize that they can no longer be “the best at everything.” Training health care professionals and conducting biomedical research while caring for patients is an enormous task. Perhaps the era of a single regional institution being a center of excellence in oncology, neurology, cardiology and all the “-ologies” in medicine has passed. Greater coordination and communication between all health service providers in an area will be required.

There are several sources of technology assessment information available to hospitals. Some payers, such as the Blue Cross and Blue Shield Association, have sophisticated in-house HTA capabilities. They make their assessments available to the public on their website. Other payers are not as transparent, and often the first time a hospital finds out about a policy decision is when a clinician uses the technology and the claim is denied. HTA in this case is rarely a prospective process - it is often a retrospective and mysterious process, which US health providers find very frustrating. Another source of HTA information is the Agency for Healthcare Research and Quality (AHRQ), a US governmental agency that contracts with designated evidence practice centers to create public technology assessments. Several US organizations (e.g., ECRI, Hayes, Inc., Health Technology Center, Health Care Advisory Board) provide HTA services for a fee.

The University HealthSystem Consortium (UHC), formed in 1984, is an alliance of 102 academic medical centers and 184 of their affiliated hospitals representing approximately 90% of the nation's non-profit academic medical centers. UHC offers its members specific programs and services to improve clinical, operational and patient safety performance. UHC's mission is to advance knowledge, foster collaboration, and promote change to help members succeed in their respective markets. UHC's vision is to be a catalyst for change, accelerating the achievement of clinical and operational excellence.

UHC's Clinical Knowledge Service has tried to fill the gap for AMC HTA resources using a variety of formats. It has produced assessments, reports, bulletins, case studies, and monographs. These resources are available online to UHC members in “health care topics,” under clinical specialty headings. Topics of 2005 and 2006 reports included:

• Dual-modality positron emission tomography/computed tomography imaging in oncology;
• Embol-X aortic embolus protection device;
• Patient-controlled fentanyl transdermal delivery device;
• Transcatheter closure of secundum atrial septal defects;
• Kyphoplasty for vertebral compression fractures;
• Bone morhogenetic proteins in spinal fusion;
• Lumbar total disc replacement;
• Bar-coded medication administration;
• ICU Telemedicine; and
• Ultrafiltration in heart failure

These reports are usually between 30-50 pages, with 50-100 references. Another UHC offering called TechFlash has been available to UHC members since 2005. TechFlashes are based on an individual request from a UHC member, with a response turnaround in 3-4 weeks. They provide a summary review of a specific medical technology, with over 90 completed since June 2005.

This format is preferred by many health care organizations to aid in decisionmaking. Topics that have been tech flashes in 2007 are listed below, and the majority has been on devices:

• Intravenous immunoglobulin off-label use;
• Hemoglobin-based oxygen carriers;
• Perineural pain pumps in ambulatory patients;
• Embolic protection systems;
• CardioMEMS Endosure intraortic pressure sensor;
• ProDisc lumbar disc arthroplasty;
• Barrx HALO endoscopic RF ablation for Barrett's esophagus;
• Arista AH absorbable hemostat;
• Total ankle replacement systems;
• InSpectra StO2 tissue oxygen monitor;
• BrainLab intraoperative MRI;
• US Com ultrasonic cardiac output monitor;
• Orthofix Physio-Stim bone growth stimulator;
• Cardiva Boomerang wire;
• CR Bard Simpulse Varicare wound management system;
• INSORB absorbable subcuticular skin stapler;
• Prestige cervical disc;
• Audiocor for optimizing heart failure treatment; and
• EsophyX transoral fundoplication

AMCs need to build or refine their infrastructure to make greater use of HTA information. There should be something analogous to the P&T process - a multidisciplinary committee charged with technology decision-making. Someone has to be accountable for either right decisions or wrong decisions. Of key importance is picking the right projects - not doing TA on everything - only the ones that are going to have a major impact on the organization. Success can be achieved by holding regular TA meetings, having policies and procedures in place for this committee, and with ongoing financial and administrative support for this process.

The era of evidence-based medicine demands that we know which technology is good, which technology is mediocre and has the potential for misuse, and which technology is truly no better than the alternative. Rarely is there a single individual you can go to in an AMC or any hospital and say 'What technologies do we have?' The pharmacy department could print a list of the formulary, usually about 2000 drugs, and you could look through that list and know what is available. In terms of devices and medical/surgical procedures, it would be rare that a single individual or source would have knowledge of that institution's capabilities. You would have to go to 20 - 30 different departments and talk to numerous clinicians before a true picture of the institution's technology inventory would emerge.

So how do US hospitals evaluate new technology today? Randomly, financially on a pure return of investment, or is it a political process? Often in AMCs a new medical technology is used as a recruitment tool, and when the new clinician decides to leave, the institution is left with a technology that no other clinician has an interest in using. Surgical robotic systems, specialty transplant programs, and advanced imaging modalities are examples with high investment costs with dubious clinical and financial returns.

HTA in the US is a sporadic and uncoordinated activity performed by various public/private and non-profit/for profit organizations for varying motives. The complex - and most would say inefficient - nature of the US health system is only recently embracing HTA concepts of evidence based medicine, comparative effectiveness, and systematic reviews as a basis for efficient and rational decision-making. Who will lead and how will the system change is very much open to debate. Western European countries have a considerable lead on dealing with the issues of access, quality, and cost of population health, and many Latin American countries seem to be in a position to reap the benefits of formal HTA in providing the health services to their populations in an equitable manner.

  Issues Index | 2008 Issues Index