The significant excuse of statistics

28 Apr
Cristian Tebé

HG Wells never said that “statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write”. If he had said it, he would have been 100% right. Politicians, administrators, scientists, everyone has an indicator, an average or a p-value at the ready to back up their arguments. The source of this information is not always clear and occasionally, the interpretation, or the results themselves, are incorrect.

One notable example of this is the controversy which arose in the UK in February and which saw a group of doctors presenting the British Secretary of State for Health, Jeremy Hunt, with a three-metre-high edition of the book “How to read a paper”. Hunt, in defense of his “seven-day NHS” plan, stated that in the UK, stroke patients admitted to hospital on weekends were more likely to die. In a letter to the Sunday Times, 59 top neurologists accused Hunt of misrepresenting statistical results and using outdated data to justify his policies.

How to read a paper - Twitter

I do not know if we should be giving books away, or what size they should be, but it might be interesting to take advantage of this space to reflect on the use and abuse of statistics and the almost religious fascination with significant p-values. One particular jingle, that of statistical significance, reminds me of that whole “scientifically proven” claim sported by many products advertised on television when I as a child. A statistically significant result is the seal of approval we all seek relentlessly, but we would do well to remember the tale of Pahom and ask ourselves, how many p-values does a researcher need? Statistically significant or statistics seen as a significant excuse.

To be continued.

Post written by Cristian Tebé Cordomí (@Cristiantb), Statistical Advisory Service at Bellvitge Biomedical Research Institute and Associate Professor at Universitat Rovira i Virgili.

Never before have we been so healthy and at the same time, never before have we been so sick

21 Apr
Lluís Bohígas

When I was born in 1950, life expectancy at birth was 65 years. I have now reached the 65-year mark and as such, according to the data, I should be starting to lose expectations. Fortunately, in recent years, medicine, health and lifestyles have improved and my reviewed life expectancy has increased, according to data from Idescat, by 20 years. This means that I have gained one year of life for every 3 years I have lived. That’s not bad! I am delighted with the figures, but there is one thing that worries me. That thing is called healthy life expectancy. This is calculated by combining mortality data which gives us life expectancy, together with morbidity data which can tell us how many years we are expected to be unhealthy. My healthy life expectancy is currently 12 years, in other words, I can spend 12 of the 20 years I have remaining healthy, and 8 unhealthy. Statistically speaking, that is.

And what diseases face me on the road ahead? Well the most serious life-threatening diseases are cardiovascular disease and cancer. Suffice to say that many of the gains we have achieved in terms of extending life expectancy have come from reducing cardiovascular mortality. Cancer is also in decline, thanks to improvements and research in health. Certain types of cancer are less common due to the decrease in the number of people who smoke. For example, lung cancer is declining in men but has increased in women. For other forms of cancer, there is now very effective treatment and medication that can relegate the disease to chronic status. However, despite the fact that these diseases are serious, they do not have us visiting the doctor every day. We visit the doctor complaining from high blood pressure, cholesterol, diabetes, heart failure, etc. Some of these diseases are linked to the largest public health problem today: obesity. Tobacco or alcohol are no longer the major public health problems, but instead it is obesity. The number of cases of are growing day by day and it is considered to be the root cause of certain diseases such as diabetes.

Obesity is caused by two factors: overeating and taking less exercise than we should. Most modern developments result in us doing less exercise: elevators, escalators, teleconferencing, etc., and a lot of the media input encourage us to eat more or to consume sugary drinks. The combination of these factors mean that today around 15% of the Catalan population is obese.

The most common disease in people over 65 years old today is called polypharmacy, in other words, the use of more than three types of medications on a daily basis. In many cases this can be more than 10 different drugs and there are some people who have a daily consumption of 20 drugs. Each drug is prescribed to address a particular health problem, and health problems have multiplied. Nowadays, we don’t have only one disease, we have several, and each disease has its own therapeutic arsenal. Doctors must monitor patients, not only to control the disease, but also to ensure they prescribe drugs that do not have adverse effects on the patient’s other conditions. When visiting our GP, they have to take into account a number of parameters, and a visit to the hospital can mean appointments with a number of different medical services. One of the most serious health issues today is derived from interactions between different drugs or therapeutic activities that can act as obstacles to each other.

Medicine has come a long way since I was born. We have discovered so much about diseases and we now have much more effective medicines and surgery to cure many conditions. Nevertheless, medicine is poorly prepared for the new patient: elderly, with several diseases, some mental (Alzheimer’s, dementia, depression, etc.), without family, and prescribed a lot of different medication. Because the system cannot cope with the complexity, they end up sending the patient to an elderly care nursing home.

For several years, voices have been raised in healthcare circles demanding the reorganization of services to meet the demands of these new patients. These voices have difficulty making themselves heard and it is very difficult to change the way health services operate. We have reduced mortality thanks to specialization and superspecilisation with diseases being defined with greater precision, but the patient has been reduced to a sack of diseases. The new medical perspective asks to view the patient as a whole and not only each one of the diseases they suffer from. There is still much work to do.

The current paradox is that we have never been so healthy and at the same time, we have never been so sick. Never before have we been able to live for so long in good health and never before have we lived so long with so many diseases to trouble us, to restrict us and make us dependent on health services. Faced with this situation, health services must adapt and patients must take a more active role in caring for their diseases.

Post written by Bohígas (@bohigasl), economist.

The Observatory, gateway to health open data

14 Apr

Núvol Open dataThe information generated by the interaction of citizens and the healthcare system keeps increasing in size. Just to get an idea, last year, only in Catalonia there were nearly 45 million consultations in primary care centres, more than 700.000 patients were hospitalised, and over 150 million prescriptions were made. However, there is much more stored in administrative records (diagnostic tests, medical imaging, hospital prescriptions, expenditures, etc.) which are kept and managed in large databases. Government officials are responsible for the safe keeping of this information, and it may use it to improve the quality of healthcare and for healthcare planning purposes.

Furthermore, advances in the development, interoperability and crosslinking of the different information systems are making it easier to gather a large amount of data that will contribute to better characterise both the general and the patient population, and they are essential to assess the results of healthcare policies.

There is a wealth of opportunities with the increasing amount of data, all of them available in electronic formats and with more quality, and with the better linking between administrative databases. Thus, the information gathered leads to new ways of generating knowledge, especially when multiple data sources are combined (genetic, environmental, socio-economic, etc.) and made available to the citizens.

This turns information into a valuable asset for planning and assessment, but also for third parties, especially in research and in initiatives aimed at enhancing the use of open data.

Open data are a actually a philosophy, as they represent a practice that encourages the free access of data for everyone, without technical limitations. This means that the original files containing the data are available to the public in the most structured way as possible. This enables any computer system to read them, and even to easily develop software based on them.

This trend towards freeing the access to data is parallel to the need of the Catalan healthcare system of managing the whole life cycle of information, from the generation of information to the knowledge dissemination.

Information and communication technologies, and information systems become key strategic allies to achieve the above objectives, and to succeed in the integration, transparency, assessment and accountability by the healthcare system and its different actors.

In the case of Catalonia, the Autonomous Government is committed to a progressive disclosure of the available public data while respecting the privacy, safety and property limitations applicable in each case through the Open Data portal, where all data are indexed and characterised. This is done following the international trends regarding the disclosure of public data, and it counts with the advice from the W3C experts (World Wide Web Consortium).

The Department of Health is thus also joining the initiative of supporting free access to data and public information. This will enable to further advance towards an open government system, based on the values of transparency, service and efficiency, on promoting the generation of value through reusing public information, on easing the internal organisation of the information systems, and on fostering interoperability among the components of the healthcare system.

The Catalan Health System Observatory collaborates in this project by favouring the knowledge about the healthcare sector in Catalonia, and by supplying the citizens with health information to assess the healthcare system itself, to support decision-making and to favour transparency and accountability. With this objective, the Observatory is strongly determined to unveil to the general public all the information regarding their healthcare system. Along with other products, the Observatory publishes on its website a set of health and healthcare activity indicators, consisting of texts, charts and open data files.

Catalan Health System Observatory

Additionally, all the data published in the Observatory reports (Results Centre, Crisis & health, etc.) are also available to the public as open data formats and infographics.  Finally, the Observatory website provides a link to the open data portal and a collection of health open data available up to date.

Check the open data available at the Observatory website!

Open data gencat - open health data


Post written by Montse Mias (@mmias70) and Anna García-Altés (@annagaal).

Ioannidis and the industry: a persistent distortion

7 Apr

John PA IoannidisJoanMVPons is a scientist and professor originally from Greece, currently working at Stanford (Meta-research Innovation Center – METRICS) who is, undoubtedly, among the most prolific authors of medical scientific literature.

Some of his papers, alone or in collaboration, have had a great impact. Who does not remember the one entitled “Why most published research findings are false?”. Nowhere in his large output will you find trivialities, and he recently came to Barcelona to speak about defective research and even about the waste of resources this implies. But we will leave the latter topic for another occasion.

The paper by this author I want to comment on is the one written in collaboration entitled “Undue industry influences that distort healthcare research, strategy, expenditure and practice: a review” published in 2013 in European Journal of Clinical Investigation.

One might think that all has been already said about the (bad) influence of drug and health care products industries. There is even a literary body or genre in biomedical scientific publications exclusively devoted to this topic. And books abound, too. All that could be said has been said. Well, actually it hasn’t. Undue influence, such as biases, is far more subtle than we think. It is often hard to tell how, similarly to interest conflicts in biomedical research or in prescription practices, the one who does it refuses any influence, since science could not admit it, as its own deontology doesn’t. Fools!

What is interesting about this paper is its review nature, not only for the number of papers gathered, but because it provides a more integrated (re)view of the different elements upon which industry acts, or is allowed to act. It should be noted that the interests and profit of the drug and health care products industry are quite legitimate, but it clearly shows some specifics that put it aside from other manufacturing industries, and not just because its important investment in R+D+i. It is believed to be one of the most profitable industries, possibly due to its large margins, but also because human diseases and ailments are here to stay, even though their end –which both the poor and the rich want to delay– is ultimately inescapable.

Ever since I learned it, I am very fond of a quotation by George W Merck (1894-1957) who for 25 years chaired the drug company that bears his family’s name (1925-1950). As this visionary man said: “We try to remember that medicine is for the patient. We try never to forget that medicine is for the people. It is not for the profits. The profits follow, and if we have remembered that, they have never failed to appear. The better we have remembered it, the larger they have been.“ I wonder what he would think of it now.

Coming back to Ioannidis and his paper, there he follows the outline of how this (bad) influence acts, and the main elements –which changed with time– upon which it exerts its distorting effect, although not as an exclusive factor. Governments, as with other industries also regulated by them, play an essential role.

Evidence based medicine - Clinical practice guidelines - Medical practice

Post written by Joan MV Pons.