Pressures

The worldwide experience of cost pressure on health services has affected the NHS since its creation in 1948. The NHS, being a service industry, has experienced a higher rate of price inflation than the economy as a whole. The ageing population has posed additional cost pressures; the percentage of people over 65 grew from 9 percent in 1961 to 13 percent in 2001 (35). Rising expectations of service levels reflect the growth in personal wealth. In the United Kingdom, gross domestic product per head in 2000 was US$25,000—approximately double the 1961 level in real terms (15). And health technology itself has been a major pressure, as elsewhere. The numbers (in new drugs alone—there are some fifty new chemical entities per year where only a handful of drugs were available in the 1950s) and nature (the increasing complexity of diagnostics, pharmaceuticals, and care settings) are both accelerating cost drivers. New technologies have a tendency not only to be more expensive than their predecessors (if they displace them at all), but also to widen the treatable patient group (46).

The United Kingdom, possibly more than elsewhere, has been through a long period of exposure to doubts about the assumed universal effectiveness of health-care technology. The thalidomide scandal of the 1960s led directly to the current drug licensing system: the establishment of the Medicines Controls Agency, with its remit to license pharmaceuticals on the basis of “safety, efficacy, and quality,” has been an important step in the history of health technology assessment. UK research on variations in health care dates back to Glover's analysis of tonsillectomy in the 1930s (22); and the international HTA movement over the past 10 years has furnished numerous examples of current practice that may be less effective than current alternatives, including notably, radical mastectomy in breast cancer (14), anti-arrhythmics in myocardial infarction (2), and albumin for critically ill patients (1). The corollary of the overenthusiastic adoption of ineffective technologies has been the delayed adoption of important effective new technologies, for instance, streptokinase in myocardial infarction (2) and tamoxifen in early breast cancer (13).

These effects have been near universal in the developed world. However, particular features of UK health care have affected the timing and trajectory of the health technology assessment response. The United Kingdom has a significant indigenous pharmaceutical industry, which is very resistant to assessment and control. It has a prestigious medical press, notably the British Medical Journal and the Lancet. Arguably, winning the evidence-based medicine argument with the editors of these journals was a factor which accelerated HTA in the United Kingdom.

NHS Environment

The NHS itself has been critically important, with the crucial features being a centralized financial structure and considerable inertia. These have allowed the UK Treasury to keep tighter control of overall health service expenditure than in other countries. Equally, the Government has always been involved in the management of the NHS and so is seen to be responsible for sorting out its problems. It has, therefore, readily engaged in policy initiatives designed to tackle cost pressure on the one hand and quality problems on the other.

Furthermore, the overall objectives of the NHS in 1948, that is, to meet all need for advice, treatment, and care (31), set the service an arguably impossible aim. The 1979 Royal Commission on the National Health Service recognized the impossibility of meeting all need and demand and slightly modified the rhetoric accordingly with the objectives of providing a broad range of services of a high standard, responsive to local needs, satisfying “reasonable expectations” of its users (49).

But it was the 1991 internal market reforms that played a crucial role in the development of HTA policy. These reforms created purchasers and providers of health care, linked by contractual financial arrangements. This critical change created, in the Health Authorities (now Primary Care Trusts), a formal set of authorities with ultimate responsibility for the overall value for money (cost-effectiveness) of the care that was provided (30).

Although there had been a considerable health services research agenda both in the universities and in health districts, it was the creation of purchasing organizations that triggered a general focus on cost-effectiveness in treatment policy.

Historical Perspective

The pivotal developments in the United Kingdom over the past 25 years have been facilitated by a rich history of empirical health services research. The academic environment had its origins in the 18th century English and Scottish enlightenment (57), with UK effectiveness research dating at least from James Lynd's controlled trial of six different treatments for scurvy in the 18th century (17). The rekindlers of this tradition in the latter part of the 20th century included Archie Cochrane and Thomas McKeown. Cochrane's “Effectiveness and Efficiency, Random Reflections on Health Services” (6) identified both the paucity of evidence on which much of health care was based and strongly advocated the randomized controlled trial (RCT) as the solution. McKeown's “The Role of Medicine” (27) challenged the idea that major improvements in health were principally a consequence of acute care.

Meanwhile, health economics as a distinct academic specialty stemming from welfare economics as a whole, has its origins in the 1960s and 1970s. An early focus was on economic evaluation and an academic literature on the cost-effectiveness of elements of health care was soon built up. But this early work on economic evaluation was relatively poorly connected with either the moves toward evidence-based medicine (18) (starting at McMaster University in Canada) or with the health technology assessment initiative (which started in the United States with the development of the Office of Technology Assessment in the 1970s) (61).

Integration in the United Kingdom of these three elements—economic evaluation, evidence-based health care, and health technology assessment—started with the birth of official “HTA” early in the 1990s (52). It was triggered by a parliamentary report Priorities in Medical Research (25), which noted the gross disproportion of funding between basic research on the one hand and applied research on the NHS and its health-care strategies on the other. The creation of a new post of Director of Research and Development followed in 1991, along with the preparation of an official applied research strategy for the NHS (12). This was further informed by a report to the Department of Health calling for HTA to deal with the multiple uncertainties in the effectiveness of health care experienced by the NHS (13); and by a wider discussion in the NHS and the research community on the need for HTA (for example, in the Caversham Conference [24]). By the mid-1990s, there was a substantial structure of officially sponsored HTA activity. This was supported by many preexisting health services research and development activities within the NHS and the universities, which were similar in their objectives but now became directly aligned to the formal HTA structures.

Current Organization

The different facets of the organization of HTA in the United Kingdom may be best understood by using the distinction made earlier between assessment and appraisal. The distinction was enshrined with the establishment of NICE in 1999 but existed before, at least informally.

Assessment

Assessment is led by universities and produced by universities and in-house research teams within industry and the NHS. It is funded by industry (particularly the pharmaceutical industry), health charities, government funded “research councils,” and since 1993, by the NHS Research and Development (R&D) HTA program. (The NHS R&D HTA program is referred to hereafter as “the R&D program.” Within the NHS, it is usually referred to as “the HTA program,” but that title is avoided here because of possible confusion with other parts of the overall HTA system.) All are important to the HTA process. The R&D program is entirely oriented to HTA projects but has a smaller budget (currently approximately £10 million per year) than the Research Councils, charities, and industry (which spend far more on health-care research, much of it, of course, not HTA related). The latter fund more HTA primary research, but in the United Kingdom, primary research (the gathering of new data in RCTs or from clinical or routine databases) to evaluate health technologies is recognized as part of the HTA process. Indeed, the R&D program spends 90 percent of its budget on new RCTs. Secondary research (systematic review, with or without economic modeling) was given a tremendous boost by the establishment of the R&D program. Some secondary research is now also funded by the other sources. Pharmaceutical companies, in particular, have been investing much more in the provision of systematic review and health economic assessments since the establishment of NICE. Tertiary research (we use the term to include guidelines and other forms of disease-wide syntheses of evidence) is necessary because of the narrowness and increasing complexity of systematic reviews. They are principally government funded and are targeted at policy makers (particularly health-care commissioners). Examples include the Needs Assessment series (51) and the various guidelines produced by professional groups and lately sponsored by NICE. Tertiary research is informed by individual HTAs and so has become closely allied to the HTA process.

The various producers of HTA generate three types of HTA according to the customer and funder of the work. The three types are (i) researcher-curiosity reports–wide (unspecified) customer; (ii) the R&D program—NHS customer (20); and (iii) the HTA-NICE process—NICE (and ultimately the NHS) as the customer.

The characteristics of these processes are set out in Table 1. Universities hosting researcher-curiosity research tend to have well-established primary research-focused departments. Those responding to the R&D program include a wide range of university departments often in collaboration with clinical experts. The HTA-NICE reviews are specifically contracted from seven specialist review centers at Aberdeen, Birmingham, Exeter, Liverpool, Sheffield, Southampton, and York.

With the expansion of HTA work, several University and other centers have established specialist and coordinating roles, in particular:

• The UK Cochrane Centre in Oxford, established in 1992, helped to found the Cochrane Collaboration and to establish the Cochrane Library and the Cochrane Database of Systematic Reviews.
• The NHS Centre for Reviews and Dissemination at the University of York, established in 1993 as one of the first specialist centers producing reviews, maintains three HTA-relevant databases (Database of Abstracts of Reviews of Effectiveness, the NHS Economic Evaluations Database, and the HTA database).
• The National Coordinating Centre for HTA at the University of Southampton, established in 1996, is responsible for managing the R&D program on behalf of the Department of Health.
• The National Coordinating Centre for Service Delivery and Organization (SDO) at the London School of Hygiene and Tropical Medicine, established in 2000, is responsible for managing the SDO program on behalf of the Department of Health.
• The Department of Health's National Research Register (56)—a database of ongoing and recently completed research projects funded by, or of interest to, the NHS—and Research Findings Electronic Register (9)—a database providing “prompt sight” of the findings of completed research funded by the NHS/Department of Health.
• Multiple centers supporting training and education, such as the Critical Appraisal Skills Program (45).

Researcher-Curiosity Health Technology Assessment

Nearly all of this work is funded “responsively,” that is, researchers have an idea for an HTA question to be researched and submit proposals to a funder, which then decides whether to fund it. This strategy applies whether the funder is spending public money (such as the Medical Research Council [MRC]) or private money (such as a research-funding charity or industry). This approach does not mean that no priority setting occurs. The MRC, for instance, has certain areas within which it encourages researchers to apply. Charities are focused around specific health problem areas. Commercial companies will only fund HTA in areas that are commercially important to them. However, priority setting is shaped by the interests of funders and not by health or health-care needs; it is limited and implicit rather than needs-based and explicit.

R&D Program

The R&D program has a formal and explicit process of agenda setting that has evolved over the 10 years of the program's existence. It identifies technologies of potential importance by occasional widespread consultations (postal and electronic) with clinicians, NHS managers, patient groups, NHS organizations, and policy makers; by horizon scanning; and by considering HTA recommendations in systematic reviews.

Each year, this process generates some 900 suggestions, which have to be turned into a short list of approximately thirty to forty HTA priorities for which funding can be made available (5;41). This priority setting is done through three panels, currently covering pharmaceuticals, therapeutic procedures, and diagnostics/screening. Panel members are clinicians, researchers, managers, and health-care consumers. The panels consider the long list of suggestions and, using the criteria in Box 1, choose a subset for more detailed investigation. Short briefing papers called vignettes are produced on each of these topics to help inform the panels' discussions. Each vignette takes a few days to produce and has the headings shown in Box 2. The panels then make their recommendations and the final prioritization decisions are taken by an overarching Prioritization Steering Group.

HTA-NICE process

NICE was established in 1999; so it is a young organization, and its processes are still evolving. At the time of writing, the prioritization process has four stages, although as this structure has been the subject of some controversy, it may change in the future. Note that the process is not directly under NICE's control: technologies for NICE's work program are referred by the Department of Health and the National Assembly for Wales. The process occurs largely within the Department of Health.

1. The first stage is the generation of two plausible lists of candidate technologies: (a) Existing and new technologies are identified by consultation with the NHS, professional associations, and national policy makers. This list is shortened by a committee in the Department of Health. (b) The National Horizon Scanning Centre produces a list of emerging and new technologies—those that have not yet made an impact on the NHS but may be expected to do so within the next few years.
2. The two lists are then merged and considered by two Department of Health committees: the Technologies Advisory Group, and the Joint Planning Group. They assess the technologies against set criteria (Box 3), consider their likely importance, and also the balance of NICE's total work program.
3. The third stage is a political decision made by the government. Health ministers consider the recommendations emerging from the second stage; announce their provisional decision and consult on it; and then make a final decision on NICE's work program.
4. Finally, there is a postscript conducted by NICE itself. NICE is not free to add to or remove from the list of topics, but it does conduct a public exercise to determine the scope of its priorities—that is, what technologies/comparators, patient groups, and outcomes it should consider.

Dissemination of Researcher-Curiosity Health Technology Assessment

For work in this category, the routes from HTA to decision-making are circuitous. Because such HTA usually takes the form of primary research, it will need to be incorporated into systematic reviews or other HTA reports before it is accessible to most decision makers. Nonrandomized HTA evidence is not readily accessible in this way, unlike randomized controlled trials that are all available through the Cochrane Library. Moreover, many RCTs will have been incorporated by the Cochrane Collaboration into Cochrane systematic reviews, increasing the likelihood that they will contribute to the tertiary research that decision makers typically look for. If it is, then, to have an impact on health care, this will be on an “as-needed” basis, contributing to evidence-based patient care, clinical guidelines, and national clinical policy.

Dissemination of R&D Program HTA

Decision makers' use of HTAs coming from the R&D program will again be on an as-needed basis, contributing to evidence-based patient care, clinical guidelines, and national clinical policy. (It is likely that, compared with the previous category, the relative balance will be shifted more to the latter two groups.) Assessments produced by the HTA program, in general, can be expected to have a bigger and quicker impact on health care than those in the previous category, for two reasons. First, the needs-led approach to the HTA program priority-setting process ensures that the assessments are on topics of high importance to NHS decision makers. Second, because many of the assessments are produced as reviews, they are already in the synthesized form that decision makers need.

Dissemination HTA-NICE Reports

Health technology assessment reports supporting NICE appraisal decisions are produced to order, on topics produced to order; they, therefore, have a much simpler dissemination route, that is, they are an important source of evidence for the NICE Appraisal Committee. Although they may contribute to evidence-based patient care decisions, it is unlikely that they will contribute to guidelines outside national clinical policy—simply because the existence of national clinical policy will discourage the production of independent local guidelines.

Where HTA Has and Has Not Changed Practice

Few studies have systematically examined the influence of HTA on changes in health-care policy and practice over time. We, therefore, need to fall back on the general observations made above, illuminated by occasional specific examples and other general principles. Researcher-curiosity HTA contributes to changes in health-care practice by feeding into the sea of knowledge. Its effectiveness will be patchy and slow, due to delays at every stage of the process. One prominent example of changes in practice and health outcome that can plausibly be attributed to HTA is the recent “revolution in coronary care” and associated increased survival (58). The R&D program provides a much more direct route into practice. An example of its effectiveness can be seen in the case of prostate cancer screening. After the 1997 publication of two HTA reports (4;50), the Department of Health in 1997 issued an instruction to the NHS that “population screening for prostate cancer … should not be provided by the NHS or offered to the public until there is new evidence of an effective screening technology for prostate cancer” (64).

NICE appraisals guidance is new to the NHS, the first examples dating from late 1999. It was not until late 2001 that the Department of Health made NICE guidance “compulsory,” when it announced that local NHS commissioners in the future would have to make funding available within three months to support NICE decisions (11). Not surprisingly, there is not yet any research-based evidence on NICE's effectiveness, although it is plausible that, given the high managerial and political profile of NICE, its guidance will be widely implemented (7). The Department of Health has commissioned an evaluation of the effectiveness of early NICE guidance from the University of York and the results of this are expected later in 2004.

Monitoring and Compliance

Discussion of monitoring and compliance represents a subtle but important switch, from the language of assessment to that of appraisal. There is no specific monitoring to ensure that decisions about individual patient care are evidence-based. All NHS organizations have a general responsibility, vested in the chief executive, to provide care of high quality. This is operationalized for clinical staff in terms of clinical governance, “a framework through which NHS organizations are accountable for continually improving the quality of their services and safeguarding high standards of care by creating an environment in which excellence in clinical care will flourish” (23). Individual doctors, who by law must all be registered with the General Medical Council (GMC) to work as doctors, have a responsibility, amongst other things, to pay attention to HTA. For instance, the GMC's Maintaining Good Medical Practice says, “to help maintain quality, clinical teams will normally use … the standards set by professional organizations and … recommended clinical guidelines” (21).

The status of clinical guidelines is variable. Practitioners may be monitored for compliance, chiefly through local management systems. Sanctions for noncompliance will depend on the seriousness of the breach. Although they could in principle include dismissal, this severity is unlikely. Conversely, there are no special incentives for compliance except in a few special cases for general practitioners (GPs). GPs are independent contractors rather than employees of the NHS and their remuneration includes elements related to compliance with specific clinical guidelines—for instance, reaching certain immunization targets.

The implementation of national clinical policy is constrained by English law. Any clinical guidance from any body is only advisory, and doctors must exercise their professional judgment in deciding whether to follow it. NICE guidance, however, appears to be an exception to this. Although NICE stresses that its “guidance is guidance only, and is not intended to override the clinical judgment of clinicians dealing with individual cases,” the implementation of NICE guidance was strengthened in 2001 by a change in government policy. This change insisted that those who commission NHS services must not deny funds for treatments approved by NICE when more than 3 months have passed since the NICE guidance was issued.