How much would you pay to extend your life by a year?

Your answer today, when you are healthy, will likely be something you won't agree with when you are old and sick. Research shows we do not plan well for our future selves.

If you have recently been to the movies or watched TV, you will have likely seen the NTUC Income commercial where somebody's future older self scolds his current younger self for not properly planning for the future. The commercial raises an interesting question about our ability to predict what our future self is likely to want.

A fantastic study published in the prestigious journal Science suggests that our predictions are likely to be wrong. It presents something the authors call the "end-of-history illusion". The authors surveyed more than 19,000 adults and asked them to report how much they had changed in the past and to predict how much they would likely change in the future.

Results showed that regardless of how old they were, people generally responded that they felt they had changed a lot in the past but were unlikely to change much in the future. In other words, in the moment, we all believe we know our true selves, assuming such a true self exists, but we are almost surely wrong. As the illusion makes clear, when asked again at any point in the future, we are likely to make the same erroneous claim.

The end-of-history illusion suggests that our current and future selves are likely to disagree on many issues, but it is the concept of present bias that provides insight into which side the two parties will take.

Present bias implies an irrational preference for current over future consumption, and therefore too little investment in the future. From the future self's perspective, the current self will exercise too little, eat too much, not save for a rainy day and, as the commercial suggests, buy too little insurance. By the time the current self becomes the future self, it is too late and all that is left to do is to regret the decisions made by their former selves.

People, when asked about the future, believe they are unlikely to change much in the future. A study's results suggest that patients with life-limiting illnesses are unlikely to receive care consistent with their preferences.
People, when asked about the future, believe they are unlikely to change much in the future. A study's results suggest that patients with life-limiting illnesses are unlikely to receive care consistent with their preferences. ST PHOTO: SEAH KWANG PENG

There are many practical implications of these biases. Our work at the Lien Centre for Palliative Care aims to explore the extent to which these and other biases influence treatment choices for life-limiting illnesses.

Caregivers were far more aggressive in their willingness to pursue treatments with only moderate survival benefits... As a result, patients who do not have a say in their treatment are likely to be overtreated compared to what they would receive if actively involved in the treatment decisions.

In a word, the results are alarming and suggest that patients with life-limiting illnesses are unlikely to receive care consistent with their preferences.

As an example, we surveyed over 500 healthy older Singaporeans and 320 cancer patients to explore how much each group would be willing to pay for moderately life-extending treatments and other end-of-life services.

Healthy older adults stated, on average, that they would pay less than $3,000 to extend their life by one year if diagnosed with a life-limiting illness such as advanced cancer.

Cancer patients, on the other hand, were willing to pay roughly $18,000, six times what healthy adults thought they would pay if in the same situation. Clearly, the current and future selves are seeing things differently.

In Singapore, as with many other Asian countries, patients often defer to family members to make decisions as to which end-of-life treatments to receive.

Anecdotally, we are told that up to one-third of patients at the National Cancer Centre Singapore either do not know or pretend not to know that they have cancer. For them, all treatment decisions are made by the family, with input from the doctor.

This would not be problematic if patients and their family caregivers had similar views on end-of-life treatments. However, a second study conducted by our team reveals that is unlikely to be the case, presumably because caregivers want to retain hope and avoid any regret for not doing everything within their power to extend the life of their loved one.

In this study, we compared cancer patients' willingness to pay for end-of-life treatments with that of their family caregivers. We found that caregivers were far more aggressive in their willingness to pursue treatments with only moderate survival benefits.

For a treatment that would extend the patient's life by one year, in contrast to the $3,000 stated by healthy adults for extending their own life and the $18,000 stated by patients, caregivers would pay over $61,000 - more than three times what patients would pay for themselves. As a result, patients who do not have a say in their treatment are likely to be overtreated compared with what they would receive if actively involved in the treatment decisions.

One might hope that doctors would intervene to ensure overtreatment does not occur. However, a third study suggests this is unlikely to be the case. In this study, we surveyed 285 local doctors and gave them hypothetical scenarios describing patients with life-limiting illnesses but with characteristics that varied by age, expected survival, cognitive status and treatment costs. For each scenario, we asked the physicians whether or not they would recommend life-extending treatments.

Results showed a lack of consistency in physician recommendations. For example, for a 75-year-old patient who is not cognitively impaired whose life could be extended by one year at a cost of $55,000, roughly 45 per cent of physicians stated they would recommend the life-extending treatment and the remainder said they would not. This is close to a coin toss and suggests that if a patient were to get a second or third opinion on the recommended course of treatment, it would almost surely differ from the first. This would clearly cause great anxiety on the part of the patient and family.

In every scenario we considered, for patients with no cognitive impairment, results were similarly mixed. This is not surprising and should not be taken as an indictment of the healthcare system.

However, it suggests that physicians should educate patients and their families on the clinical benefits of various treatment options but because clinical benefit is only one of many factors that influence treatment choices for patients with life-limiting illnesses, physicians are not in a good position to make treatment decisions on behalf of the patient. This is best left to the patient, with input from the family.

So what are the implications of the above?

In short, the results suggest that we cannot count on our current selves to properly forecast what our future selves would want if diagnosed with a life-limiting illness. Most likely, we will underestimate our future demand and therefore not plan appropriately.

Contrarily, our loved ones, not wanting to give up hope and wanting to avoid future regret, are likely to push us towards treatments that we will feel are not worth the expense, thus exposing the family to significant financial risk.

Physicians, often spurred by a healthcare system that pursues aggressive treatments even with limited survival benefits, are unlikely to go against the wishes of the family if the care has any potential to extend life.

There are no easy fixes to these problems. MediShield Life provides a partial solution to the problem by forcing individuals to have insurance who otherwise may not have purchased a plan for themselves. However, this is only a partial fix because MediShield Life does not provide complete coverage for many costly treatments.

Advance care planning and open discussions about treatment choices if diagnosed with an advanced illness are other recommended solutions. These discussions should take place early and often, and should include considerations about costs but also discussions about place of death, and trade-offs between quality of life and care that modestly extends life but potentially at a low quality.

One finding our research makes clear is that patients value dying at their place of choice, avoiding severe pain and receiving well-coordinated healthcare where they are treated with dignity and respect far more than they value moderate increases in life expectancy.

Caregivers, providers and policymakers need to understand what matters most to patients as they approach the end of life and work to ensure those priorities are met.

• Professor Eric Finkelstein is the director of the Lien Centre for Palliative Care at Duke-NUS Medical School.

A version of this article appeared in the print edition of The Straits Times on March 31, 2016, with the headline 'How much would you pay to extend your life by a year?'. Print Edition | Subscribe