Copyright © 2017. Interamerican Development Bank. If you wish to republish this article, please request authorization at

By Ursula Giedion.

Lee este artículo en español aquí.

As health technologies evolve, the ways in which these technologies are evaluated and applied to the public sphere to ensure effectiveness also change. The Health Technology Assessment (HTA), in particular, is a systematic and technically robust process that examines a health technology (often new) from different perspectives. Using the available evidence, and then balancing the different positive and negative results, it produces recommendations for public policy. Its main purpose is to inform health care policies at the national, regional or local level. Such decisions may relate to procurement, funding or adequate use of health technologies and also to disinvestment in obsolete or ineffective technologies.

Given the importance of HTA and its scientific approach, one might expect that the recommendations of different HTA agencies serving similar populations in reasonably comparable contexts produce similar decisions. But this is not the case, as Dr. Panos Kanavos from the London School of Economics (LSE) illustrated in a recent webinar by CRITERIA*, an Inter-American Development Bank-supported network on priority setting. One of his slides (see below) actually shows significant differences in medicine-related recommendations given by HTA agencies across different jurisdictions.

HTA Recommendations on Medicines, by Jurisdiction

*National Institute for Health and Care Excellence: **Scottish Medicines Consortium; ***Dental and Pharmaceutical Benefits Agency; ****Haute Autorité de Santé.

Source: LSE 2016 presented in webinar by Dr. Kanavos found at Breve 18 of Red CRITERIA.

The drugs marked in orange are the ones where divergent recommendations were received across countries, even among neighboring countries (e.g. England, Scotland, Sweden and France). Take the example of Gefitinib. It was listed with conditions in England, listed without conditions in Scotland, in Sweden it was not accepted, and the benefit was thought to be very minor in France. These differences can be observed for all type of drugs (oncologic drugs, orphan drugs as well as some neurologic drugs).

Why does This Happen?

The difference seems to be related mainly to different interpretations of the same evidence. This shows that evidence-based decision making has as much to do with evidence itself as with its judgement and appraisal.  Social values intervene to i) judge the evidence due to the limits of scientific evidence itself (there is uncertainty, studies may have flaws, there are transferability issues, information is often incomplete and of varying quality etc.); and ii) to incorporate preferences and values that are not easily measured such as fairness, patient autonomy, severity of disease, among others. This, in turn, pinpoints the importance of explicit, deliberative processes through which the available evidence is systematically appraised to come up with a decision. It is during this process of deliberation that social value judgements are made and need to be taken into consideration when making a coverage decision.

In other words, as a result, a decision maker in Costa Rica, Colombia or Brazil may well take advantage of the evidence produced elsewhere after having sorted out the generalizability issues, and still have to go through their own, explicit and transparent deliberative processes to come up with a decision. Consequently, their decisions may well end up being different. Ergo? Thinking about how to implement fair, transparent and participatory deliberative processes is just as important as the evidence itself. There is much to be said in favor of these processes on grounds of principle. But, be aware: we only have limited evidence on the effectiveness of such deliberative processes and how to implement them. Ultimately, this will be the definitive question.

How is this process approached in your country? Who is responsible and how transparent is it? Tell us about it in the comments section below or by mentioning @BIDgente on Twitter.

Download Breve 18: Procurement Policies for Pharmaceuticals

Ursula Giedion es la coordinadora de la Red CRITERIA en el Banco Interamericano de Desarrollo.

*CRITERIA is an initiative by the Inter-American Development Bank (IDB) that aims to support countries in the Latin American and Caribbean region so they have the strong evidence and high quality institutions needed to improve their prioritization and public spending assignment of health. Through stronger processes, collaborative work, knowledge and experience exchange in subjects related to health prioritization, CRITERIA shares ideas and alternative ways to advance towards universal coverage in health in contexts of limited resources and growing costs. Sign up HERE to get the latest information.

Recommended Posts
  • Dr andres freiberg

    excelente artículo de Ursula! la prioritización en salud es un proceso político que cada jurisdicción deberá organizar de acuerdo a lo que quiera obtener. un par de cosas hay que mencionar: la primera es que el tema candente de hoy no es tanto la incertudimbre en resultados en medicina (ya bien descrita por Arrow en 1963) sino los precios. cuando se inventó no hubo necesidad de evaluar penicilina porque su efectividad en todos los casos estaba fuera de duda. si los últimos biológicos costaran poco quizas solo se evaluaría su perfil de efectividad vs seguridad.. pero en los fármacos del artículo son muy caros y no tan efectivos, y por lo tanto, es de esperarse que distintas agencias de ETS formulen distintas recomendaciones. no es una falla del proceso de ETS en sí sino todo lo contrario. El riesgo de no advertir este simple concepto es que la tendencia de cobertura se sesgara al país que más tecnologías cubra, y en salud, más no es necesariamente mejor. y la razón es que a la larga los sistemas se quedan sin dinero para cubrir las necesidades más esenciales con los medicamentos más asequibles.

Dejar un comentario

Start typing and press Enter to search