With population figures at a global all time high, life expectancy longer than it has ever been in history and an increasing demand for healthcare across the board, healthcare budgets have never been so stretched. The USA saw the introduction of the Deficit Reduction Act (DRA) in 2006/07 in attempt by the CMS to contain healthcare spend and across Europe regulatory, pricing & reimbursement environments are getting tougher in an attempt to restrict/claw back healthcare spend wherever possible.
Healthcare organisations are facing the inevitable – radical change in the approach & delivery of patient treatment will have to happen if the spiralling cost burden is to be contained.
In the shorter term, better allocation of resources through the implementation of IT systems (“tele-health” & Electronic Medical Records) were thought to be (at least in part) a way forward, enabling chronically ill patients to monitor their progress & care at home thus reducing burden on healthcare professionals.
Prevention is better than Cure:
A significant emerging trend is the move toward preventative healthcare – through increased screening to detect chronic conditions in their nascent stages and preventative vaccinations.
In the UK the Prime Minister, Gordon Brown, recently announced his intention that by 2011, the NHS is to provide a national screening programme for “key” diseases (diabetes, stroke/TIA, heart & kidney), via expansion of diagnostic procedure usage.
Healthcare of the future, it seems will be firmly routed in prevention rather than cure.
Impact on the Pharma Industry, in particular Forecasters:
The majority of pharmaceutical forecasts have traditionally included some kind of patient estimation as a cornerstone. Historically, patient-based forecasting has been based firmly in the arena of “prevalence”, sometimes “incident” patients were considered too (especially for acute disease conditions, infections & diagnostics). Perhaps only epidemiologists, vaccine manufacturers & those working in the cardiovascular markets had any real experience of estimating populations ‘at risk’.
A new focus on the approach to healthcare delivery will require a subtle yet significant re-work by the pharma analyst (and indeed the “City” investor analyst) from basing product evaluations on diseased patients to basing them on those at risk.
Epidemiologists already study disease transmission, a significant part of their workload being to determine & identify those populations more “at risk” than the general population.
But, finding the data & putting it together will require new skills, if the data exist at all. Most traditional patient database sources (including commercial databases as well as disease registries) provide prevalent patient population estimates – making it difficult enough to make the step transition to incident patients, let alone the further step in the disease development to elevated risk status. Finding at risk populations requires different consideration. It will become harder & harder to find off-the-shelf ready-to-go population estimates as the market shifts to requiring these patient risk estimations.
Key Factors to Consider:
When trying to make general estimates about the size of risk populations, they need to be defined in relation to the product in question. Some populations will be much easier to estimate than other, e.g. if the recommendation is for the use of a product in all those turning 60, it is a relatively simple matter to find population statistics on the number of people aged 59 coupled with mortality rates for the same age cohort. Others will be much harder to estimate. In either case, the following questions will provide some idea of how the populations could start to be investigated and how simple/complex it may be to find the data:
1. Before a patient develops the disease/condition, what marks them out?
2. Are there any known risk factors and are they easily detectable?
3. How many people in a given population manifest those risk factors?
Further Pitfalls:
Since the data can be harder to find for risk populations, uncertainty around the data increases eroding the predictive confidence of the final product valuation.
To add an extra layer of difficulty, for most diseases there are multiple risk factors and many diverse populations that could have one or more risk factors. So, finding a patient number of all those at risk (and stratifying the risk to find those at particularly high risk), becomes ever more complex. In this situation, it is easy to start double counting patients and over-estimating the market.
And Finally…….
It is inevitable that in many areas of healthcare this new focus will become apparent, and require a change in the approach to building forecasts, perhaps also leading to a change in management’s perspective & use of forecast models. The face of healthcare is changing. As forecasters, we should at least be prepared for it!
