The pharmaceutical industry has had a fair amount of bad press recently. We have seen accusations of price gouging across several companies, large companies have been accused of monopolization, and there is the constant arguments over a lack of research for less profitable drugs, regardless of how many lives they may save. It has made it a difficult market in which to operate, which is one of the reasons we have seen five of the largest nine pharmaceutical mergers occur in the last decade and each of the top nine taking place in the last 17.
This difficulty in the markets has led them to turn to data in order to help both make them more profitable and also predict market conditions ahead of time. Forecasting market conditions is arguably the most important, not only for the companies themselves but also for the patients they are helping.
Unfortunately, the better the business for pharmaceutical companies, the worse the world's health is, after all you would only ever sell more pills if people needed to take them. Despite this unfortunate fact, this has some big advantages in terms of forecasting and the data available to these companies, which, on the flip side, can then also help healthcare providers.
If you were to look at it from a simple perspective of the more a certain drug is used in a certain area, the higher occurrence of the illness that drug treats, it becomes clear how useful this can be.
Illness and disease are not created in isolation and are therefore trackable and often relatively predictable. For instance, if there is a wet summer forecast for a hot and humid area, there is likely to be an increase in the mosquito population and therefore a higher likelihood of diseases including West Nile, Eastern Equine Encephalitis, LaCross virus, and both Western and St. Louis Encephalitis. With this in mind and weather data often being relatively reliable, pharmaceutical companies can increase production of specific drugs where and when there is likely to be an increase in demand.
In addition to this they can track the spread of a disease in order to supply drugs not only to areas where the disease is now, but where it will be in the future. For instance, during the Ebola outbreak in West Africa people migration and death rates were tracked through the use of mobile phone data and the National Geospatial-Intelligence Agency's geospatial intelligence products. It helped to take preventative measures in areas that were likely to be impacted next and drugs to treat or prevent the disease could be sent to these areas.
However, it is not always in predicting future diseases spreading where pharma companies can look to the future. Simply matching conditions with the needs for similar years and looking at population changes within the interim can help to predict future demand. For instance, you may know that you sold 5% more cold remedies in China when the temperature was low and precipitation was high, so if this year is the same you can predict a 5% increase in sales and factor in population growth to increase these even further.
On the other side of the coin, the geographical sales data from pharma companies can help to identify the prevalence of illness across the world. For instance, if there is an increase in the sales of a specific drug in a certain area, it would suggest a coinciding increase in a specific illness too. This can then be used by governments and health agencies to identify areas that may be at risk. This can even be used for conditions, so if they saw an increase in sales of a drug in one area where the temperature range and precipitation was the same as another, it would be sensible to expect an increase in the illness the drug was treating in an area with similar conditions.
One of the major advantages of the huge amount of data that pharma companies could mine from is also a disadvantage. Almost everything can have an impact on people's health - increased temperatures means more hay fever, decreased temperatures means more colds, living in the countryside decreases the chances of respiratory infection, unless it's during harvest season. The huge number of variables gives them the opportunity to forecast to an almost unbelievable degree of accuracy, but they need to make sure they are focussing on the key data points and just working out what those are is complex in itself.