This article was also published on Smart Blogs

Eric Siegel, Ph.D.

Retirement kills more people than hard work ever did.
—Malcolm Forbes

I'm not afraid of death; I just don't want to be there when it happens.
—Woody Allen

Who benefits by predicting your behavior? Organizations do—companies, government agencies, and political campaigns. They employ predictive analytics, technology that learns from data to render per-person predictions, one individual at a time.

The payoff for predicting extends beyond boosting sales and winning elections: everyone benefits when this technology strengthens the fight against risk, crime, and even spam.

In these efforts, each important thing a person does can be valuable to predict, namely click, buy, steal, drop out of school, quit your job, donate, crash your car, or vote.

So how about the final thing each of us do, die? In fact, there are five reasons organization predict your death. Sometimes they do it with altruistic intent, for healthcare-related purposes. In other cases, there's a financial incentive—they predict death for the money.

To begin with, there are two fairly well-known reasons to predict when an individual's death will come:

1. Healthcare: predicts death to help prevent it. For example, Riskprediction.org.uk predicts your risk of death in surgery, based on aspects of you and your condition, in order to help inform medical decisions. In other work, psychiatric research predicts which patients are at the greatest risk of suicide.

2. Life insurance: prices policies according to predicted life expectancy. A growing number of life insurance companies go beyond conventional actuarial tables and employ predictive analytics to establish mortality risk. It's not called death insurance, but their core analytical competency is to calculate when you are going to die.

Beyond life insurance, it turns out health insurance also predicts death—of policyholders. Until recently, death prediction has not been within the usual domain for health insurance. On the surface, given that the ulterior motives of health insurance are at times under scrutiny, one may imagine dubious implications. For what purpose do they predict dying?

We will return to this question—for now, here's a bit more about how death prediction works.

Standard actuarial methods assess mortality risk from a handful of factors such as age, gender, whether the individual smokes and drinks, Body Mass Index, and psychological outlook (e.g., "optimistic"). These are the attributes you may enter—right now, if you like—into http://www.death-clock.org to calculate the Grim Reaper's ETA. This website bases its predictions on data from the World Health Organization.

Predictive analytics extends beyond the limits of standard actuarial methods to incorporate a greater range of factors, and to combine them—for each individual being predicted—by way of more sophisticated mathematical models. In healthcare, for example, a patient's diagnostic codes and lab results provide further predictive oomph. Moving to a wider range of domains, here are a few more colorful examples of risk factors:

Solo rockers die younger than those in bands. Although all rock stars face higher risk, solo rock stars suffer twice the risk of early death as rock band members. This may be due to the fact that band members benefit from peer support and solo artists exhibit even riskier behavior (factoid courtesy of public health offices in the UK).

Men on the Titanic faced much greater risk than women. A woman on the Titanic was almost four times as likely to survive as a man. Most men died and most women lived. This may be due to the fact that priority for access to life boats was given to women.

Retirement is bad for your health. For a certain working category of males in Austria, each additional year of early retirement was shown to decrease life expectancy by 1.8 months. This may be due to the fact that unhealthy habits such as smoking and drinking follow retirement (factoid courtesy of the University of Zurich).

Some organizations predict when death will arise not by natural causes, by instead by accident, or even intentionally, in the cases of wartime battles and murder.

3. Law enforcement and military: predict kill victims in order to protect. U.S. Armed Forces conduct research to analytically predict terrorist attacks. Researchers also assess the risk to individual soldiers, e.g., when parachuting. Law enforcement in Maryland applies predictive models to detect inmates more at risk to be perpetrators or victims of murder. Further, university and law enforcement researchers have developed predictive models that foretell murder among those previously convicted for homicide.

4. Safety institutes: predict system failure casualties. For example, researchers have identified aviation incidents that are five times more likely than average to be fatal, using data from the National Transportation Safety Board.

We come now to the final item: Why would a health insurance company predict death? Fear not, it's actually done for benevolent purposes.

5. A top-five U.S. health insurance company: predicts the likelihood an elderly insurance policy holder will pass away within 18 months in order to trigger end-of-life counseling, e.g., regarding living wills and palliative care. The predictions are based on clinical markers in the insured's recent medical claims.

While the more fortunate elderly are surrounded by caring family fretting about comfort care, many aren't as lucky. In lieu of the doting supervision of family, many nearing the end of life will greatly benefit from pertinent screenings and service offerings, often available only by way of accurate, timely targeting.

Despite the benefits of this work, predicting death is so sensitive that the health insurance company in question must keep its humanitarian activity a secret. An employee of this company told me the predictive performance is strong, and the project is providing clear value for the patients. Despite this, those at the company quake in their boots that the project could go public, agreeing only to speak to me anonymously. "It's a very sensitive issue, easily misconstrued," the employee said.

Given the sensitivity of a predicted passing, some organizations feel it's better not to know. Industry leader John Elder (Elder Research, Inc.) tells of the adverse reaction from one company's human resources department when the idea of predicting employee death was put on the table. Since death is one way to lose an employee, it's in the data mix. In a meeting with a large organization about predicting employee attrition, one of John's staff witnessed a shutdown when someone mentioned the idea. The project stakeholder balked immediately: "Don't show us!" Unlike health care organizations, this human resources group was not meant to handle and safeguard such prognostications.

Nevertheless, whether by accident, murder, or natural causes, organizations have made a science of predicting when we each will die.

But is there prediction after death? It turns out that death is not the final event to be predicted for a life. The Chicago Police Department predicts whether a murder can be solved. The department found that characteristics of a homicide and its victim help predict whether the crime will be solvable.

Eric Siegel, Ph.D., is the founder of Predictive Analytics World (www.pawcon.com)—coming in 2013 to Toronto, San Francisco, Chicago, Washington D.C., Boston, Berlin, and London—and the author of Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie... (February 2013, published by Wiley). For more information about predictive analytics, see the Predictive Analytics Guide (www.pawcon.com/guide).