Bridging the gap between health economic analysis and the real world

Health economic analysis on pharmaceuticals has focused on obtaining approval for use from governments. After which, the analysis has not had much impact on physicians in determining their use or the conditions for which products are prescribed. Consequently, the industry has focused it’s attention on marketing their products on the basis of their effectiveness, and not their cost effectiveness. It has been directed, not to payers, but primarily to individual physicians and sometimes patients, in countries where direct-to-consumer advertising is allowed. The power of the payers has tended to be limited to increasing copayments to brands where generics can be substituted, partly because it was difficult to monitor the conditions under which the product was actually prescribed, and partly because governments and payers do not want to be in a position of explicitly rationing or denying the use of specific pharmaceuticals. They have preferred to leave such decisions to physicians who are implicitly forced to do so in order to stay within the constraints imposed by global budgets or through reimbursement mechanisms.

However, the pressure to bridge the gap between health economic analysis and the real world has increased. For those making the analysis, there is a greater need for accountability, more scrutiny of the assumptions made and more need to incorporate negative results from clinical trials as they are all likely to be made public. Policy makers have come under similar pressure to justify their decisions by showing the evidence and, if any value judgments have been made, to explicitly state this and on what grounds. Moreover, researchers and policy makers must communicate in a way that is readily comprehensible to the public. Transparency is being demanded by the public who are not willing to defer to professional or bureaucratic authority.

Health economist researchers must acknowledge the gap between the assumptions and ideal conditions under which cost-effectiveness analysis is conducted, and how the product is going to be actually used in the real world. Clinical trials seldom include elders or those with complications since the patient’s idiosyncratic characteristics tend to confound the results or would not be ethically allowed. Yet, most of the actual prescribing would be done precisely for these patients. The messy conditions which physicians must deal with could be one excuse for not following what is recommended by evidence-based medicine, and not merely their ingrained professional inertia. Perhaps more important is the technology creep, or the expansion of use from the restricted conditions under which a product was originally approved. Even if its use is expanded to conditions that remain off-label, it is difficult to monitor such infringements since physicians have discretion in making their diagnosis, especially in gray areas in which it is difficult to set explicit criteria. Thus, while a product may have been cost effective under the strict conditions as defined in the model, it would no longer be so in the real world. A classic example would be the use of cimetidine, when restricted to treating peptic ulcers it would be cost effective compared with performing surgery, but when expanded to common gastritis, it may not be so.

Future analysis

Where should future analysis focus? One option is to continue as before. That is, to conduct health economic analysis for the approval process and, when appropriate and necessary, to selectively promote the results to physicians after launch. However, this approach could be counter-productive in the long run as it may bring into question the validity and relevance of the analysis. In Japan, the initial interest in health economic analysis has lapsed. It is not required at all by the government and prices continue to be set by comparing the innovativeness of the new product with its comparator. Other reasons exist why analysis is not conducted, such as the difficulty of obtaining data and the lack of those capable of performing the analysis [1,2], but the main reason is that the government was not convinced of its value.

Alternative approach

The alternative approach is to conduct the analysis in the real world where clinical decisions are made. Despite efforts by physicians to maintain their autonomy, there is increasing pressure to adhere to standardized protocols. In inpatient care, clinical pathways set the treatment schedule for every day that the patient is in hospital. In ambulatory care, disease management programs have algorithms that precisely set the specific conditions for prescribing each medication and its dosage. Physicians can of course choose not to follow the protocol. However, should they opt to do so, they must provide convincing evidence of the patient’s unique clinical conditions that justifies their falling out of line. This trend would become more pronounced as electronic medical records (EMRs) become more prevalent. As they will be linked to order-entry systems, the standard prescription that appears on the screen would be the one chosen by the physician in almost all cases. Prescribing other pharmaceuticals would become a time-consuming job, in addition to which they must later justify their aberrations to their peers and payers.

Thus, health economic analysis should focus on evaluating the justification for changing to a new product in prototypical clinical pathways and disease management programs, and not in isolation to the clinical context. Evaluation would be directed less on outcomes, but more on process and efficiency savings by decreasing the lengths of stay, ordering of diagnostic tests and prescribing of other medications. This shift in emphasis may not be welcomed by the pure-minded, but since risk adjusting for outcomes is a complex task that requires enormous databases, it would be more realistic to use a combination of literature reviews and expert opinion when designing protocols. Once the prototypical model has been developed, tailor-made modifications can be made to meet the specifications of each healthcare organization. This task may not be so daunting as it may appear since the specific conditions are likely to remain within the range of the sensitivity analysis made in the prototypical model However, it does mean that the health economic analysis must be made with more input and feedback from clinicians and administrators.

Feasibility & impact

A formidable barrier to such changes is the absolute shortage of health economists. Few, in comparison with the need, are trained and most tend to work for pharmaceutical companies. However, as in other health professions, there could be both practitioners and academics. A standard set of skills, if supported by software, could perhaps be disseminated among the administrative staff of healthcare organizations. The data available for analysis would probably be even more limited but assumptions can be made on firmer grounds due to more contact with the field. Health economists must also place greater emphasis in making their methodology and results more comprehensible and relevant to clinicians and administrators. The use of more sophisticated models would perhaps be better targeted to allocating resources for investment purposes at the macro, national or regional level.

From the industry’s perspective, the greater reliance on protocols means that it may not be so productive to have their marketing targeted on individual physicians. Whereas formerly the penetration of a new product would have been a gradual process as more physicians become aware of its effectiveness, in the future, its adoption will increasingly become an all-or-nothing decision for each healthcare organization. If incorporated into the protocol, virtually all physicians would use it, if not, few if any would use it. Thus, the marketing strategy must be redirected into establishing a working relationship with the healthcare organization for the development and revisions of protocols.

Another major change in corporate strategy that is needed is taking full advantage of EMRs. EMRs allow the conducting of phase four mega-studies at considerably less cost and time. Unforeseen side effects after launch can also be more easily detected through the use of EMRs. Thus, instead of the current situation in which most health economic analysis is conducted by the industry in a sometimes adverse relationship to the government and payers, as the process becomes more clinically oriented, the relationship could become more consultative and constructive. This would be the way forward for more effective and efficient use of health economic analyses.