Do We Know What’s Best For Our Health?

The lessons learned from the household goods sector offers a salutary lesson for healthcare providers.  There used to be a clear divide between customers and the goods they wanted to buy.  This divide was the shop assistant who would take your requirements, offer some advice and then provide you with the goods that met your needs.

We are all fully aware of the way this has fundamentally changed as technology has slowly but surely eroded the role of the gatekeeper so we now directly serve ourselves not only in store but of course from the comfort of our own homes.  In the era of online ratings and expert reviews freely available, the idea that we look to someone for advice for our grocery shopping now seems very quaint and anachronistic.

Of course, the same process is now happening in healthcare. People are increasingly presenting themselves to their physician with the benefit of having Googles their symptoms so they often have a reasonable idea of what is wrong with them.  There is a vast number of ways in which technology, such as fitness devices, is available for facilitating more health-conscious lifestyles without requiring the support from health provisioning services.

But this is scratching the surface, as there is a  proliferation of new healthcare services that are breaking down the barriers between patients and healthcare provision:

  • Diagnostics and testing: A range of labs that process samples for hospitals are making these available for home usage.  For example, Swedish company Werlabs offers comprehensive blood testing via home visits.  Thriva offers finger-prick blood tests you take at home.
  • Medical treatments: A range of companies restricting the way we access prescriptions by codifying the consultation process into an online questionnaire.  This is then approved by a physician before the drugs are dispatched.   Dr. Ed is a good example, with the offer of a ‘discreet, affordable online consultations and prescription treatments without needing to see a doctor face-to-face’.
  • Mental treatments:  Services such as like Ieso digital health offer one-to-one cognitive behavioural therapy (CBT) with a therapist via secure instant messaging via digital devices.
  • Medical data:  This is increasingly being made available to patient access with the potential for identifying patterns that have long-term health implications.  Whilst data is currently held across disparate patient records, platforms such PatientsKnowBest are aggregating it and making it available le through easy to use portals.
  • Managing diseases:  Behaviour change programs are increasingly productised, which means they can be scaled across a large number of patients.

However, with these developments comes a new set of challenges.  Whilst we may not necessarily need the help of a shop assistant to guide us in terms of which washing powder to purchase, we can reasonably ask how well equipped we are to evaluate the results of medical tests.

Biomarker testing is a good case in point here and perhaps at the core of personalized medicine. Biomarker testing looks for these molecular signs of health and picks up abnormalities that can be a sign of risk of cancer.  One of the most widely used forms of this is BRCA mutation testing which has been recommended by the U.S. Preventive Services Task Force (USPSTF) since 2005 for women whose family history demonstrates an increased risk for BRCA-related cancers.

As the cost of testing has been reduced and awareness has increased there has been a large increase in take-up.  With this, recent studies have found that BRCA tests performed in unaffected women increased significantly from 24.3% in 2004 to 61.5% in 2014. On the other side among 220,000 BRCA mutation carriers in the U.S., it is estimated that more than 90% have not been identified.

A large number of people are self-referring who do not fall within the USPSTF guidelines.  In fact, approximately 60%–80% of patients referred for genetic counselling and testing do not meet the referral requirement based on family history.  This clearly has downsides not only because of increased costs but also because there is always a possibility of a false positive which can have serious outcomes for the person involved.

So we can immediately start to see some issues which are fundamentally psychological in nature.  People are overestimating their risk but at the same time, those who are in high-risk categories are potentially under-estimating and not getting the necessary tests.  It is likely that healthcare practitioners are under-estimating risk in some cases and failing to prescribe testing.

The issues do not finish here.  There are a whole set of challenges associated with interpreting test results.  So, for example, 71% of women who undergo BRCA testing receive ambiguous or uninformative negative test results.  In these instances, patients then need to make decisions with very uncertain information, a classic set of conditions for psychological factors to take a more important role.

And finally, there is a range of treatment options available which could include surveillance and risk avoidance with the way through to chemoprevention and prophylactic surgery. There is no simple way to move from test results to treatment decisions.  Again, psychological factors come into play.

The areas that are likely to offer the most value to inform this issue includes:

  • Trust in information source: Trust remains a huge issue in healthcare; digital providers will need a detailed understanding of the psychology of trust if they are to maintain a strong relationship with patients
  • Cassandra’s Regret: This is the psychology of not wanting to know.  It is all too easy in a self-service environment to turn away from tests that may reveal unpalatable insights about your health.  How can providers ensure that they are helping patients to make the right choices in these instances?
  • Understanding risk: Humans are famously hugely fallible when it comes to interpreting risk.  The way that information is presented therefore becomes critical.  For example, many people struggle with probabilities (e.g. the probability that a woman has breast cancer is 0.8%) but are much more comfortable with natural frequencies e.g. eight out of every 1000 women have breast cancer).

As we see ever more personalisation of healthcare there are new decision architectures – as the expert has much less of an intervening role in patient decision making. As such it is hugely important for healthcare brands to understand and respond appropriately to this if they are to maintain their trusted role and authority with patients.

By Colin Strong

My thanks to Dr Tamara Ansons for her input to this article


[1] Tim Harford (2014) ‘Where next for behavioural economics’ Financial Times (

[1] Simonson, Itamar (2007) Will I Like A “Medium” Pillow? Another Look At Constructed And Inherent Preferences

[1] George Lowenstein & Peter Ubel (July 14 2010) Economics Behaving Badly. The New York Times