New life expectancy calculator guides you to changes that could extend your life. Here’s how it works.

Dr. Doug Manuel can tell you when you are likely to die.

No, it’s not a premise for Hollywood’s latest thriller. Dr. Manuel leads the research team that developed the life expectancy calculator embedded in the newly revamped Heart&Stroke Risk Assessment.

Dr. Manuel, a senior scientist at the Ottawa Hospital Research Institute at the University of Ottawa, walked us through the research and shared his ideas on how the calculator can help Canadians.

How did you come up with the life expectancy calculator?

We originally did the work for a 2012 report we created for Public Health Ontario and the Institute for Clinical Evaluative Sciences called Seven More Years. We estimated the burden of life expectancy in Ontario related to certain health behaviours – smoking, alcohol, food and exercise. We estimated that on average, life expectancy was reduced by about seven years.

Then it occurred to us that this model actually works quite well for individual people. So we created a life expectancy calculator.

How did you figure out the impact of each behaviour on how long someone is likely to live?

We took the Ontario sample of the Canadian Community Health Survey, which is done by Statistics Canada. That includes about 78,000 people who were followed for up to 10 years. We basically looked at who died. Then we created predictive models based on whether they were a current, former or “never” smoker, plus the amount they smoked, leisure time physical activity, alcohol (number of drinks per week) and diet.

We included age, sex, body mass index and ethnicity, and neighbourhood for socioeconomic status – all things shown to be associated with increased risk. We also used our knowledge of epidemiology, of what’s been shown in past studies to cause increased mortality.

Were you surprised by any of the results?

I’m always surprised by the impact of healthy living. We saw a 20-year difference in life expectancy between the healthiest and unhealthiest person. That’s massive.

By healthiest person we mean someone who is meeting all the general recommendations for healthy living, compared to someone who is sedentary and smokes and doesn’t eat fruit and vegetables.

Is there one behaviour that has the biggest impact on someone’s life expectancy?

The most common risk is diet. Most people don’t eat seven fruits and vegetables in a day. So if you look at the behaviour that the greatest proportion of people could change, it would be diet.

But individually, for people who smoke, that is by far the biggest risk.

The questionnaire asks how often someone eats carrots. Will carrots add years to your life?

We probably get more comments on carrots than on anything!

We do know that for heart disease, carrots have been shown to be protective. But in the context of life expectancy, eating carrots is more of an indication that you’re getting a varied healthy diet with a range of fruits and vegetables.

We were a little bit limited in what diet data we could use, so that’s one reason why carrots are there.

How reliable is the calculator?

I would use it as benchmark; I wouldn’t rely heavily on it for each individual person. But that said, its statistical properties are very good. For instance, it’s much better than, say, the Framingham Risk Scorefor heart disease, in terms of its predictive properties.

Will you add more risk factors to the calculator in future?

We will be adding more of what we call intermediate and distal risk factors, such as blood pressure and cholesterol. And we will be adding more questions and calculators for specific chronic diseases.

We started out trying to have a good estimate of life expectancy that applies to most people. Over time, as our statistical power increases – as we follow people for longer and we have larger sample sizes – then we can get to smaller risks as well.

We’re also very interested about public health risks such as air pollution and more elements of diet too; for example, we are working to add sodium.

We know that people don’t always give reliable answers to questions about their health behaviours. How does the calculator handle that?

As researchers we look at that problem quite a bit – how to make self-reported information more reliable.

For example, the Canadian Health Measures Survey is based on the same questions (as the Canadian Community Health Survey) but adds some physical testing such as blood and urine, basically to check responses.

For smoking we know that the answers are very good; for physical activity and diet not so good. For alcohol, we know that people under report their drinking by a considerable amount. So the reliability varies.

One thing to remember is that we’ve built the calculator based on self-reported information, and when you take the questionnaire you are using your own self-reported information. In general, the way the questions are asked in the Heart&Stroke Risk Assessment is the same way they were asked in the survey.

Why do you think the life expectancy calculator is important?

It’s useful as an information tool, to engage people so they can be informed about their choices.

The Heart&Stroke Risk Assessment does a great job in further engaging people, asking, “Do you want to change? Here are some resources to help you.”

In fact I think the Foundation’s Risk Assessment is the best one available in the world right now.

For us, the calculator is about informing public policy as well. We can’t have informed health policy without knowing how it affects your family, your community, yourself.

As a physician, how do you use the calculator in your own practice?

I find it really helpful with my patients. When a doctor asks, “How’s your diet?” the answer is usually, “Pretty good.” But when you actually sit down and ask how many fruits and vegetables the person eats, you get a much more meaningful answer.

So I’ll get my patients to fill out the questionnaire in the waiting room. Then they bring it in and it saves me time asking questions. The answers are better and the life expectancy part immediately engages them in discussion.

I’m always surprised how people answer it. I’m surprised how many people eat no fruits and vegetables at all – a lot of my patients.

• Dr. Manuel’s research team maintains their calculator at projectbiglife.ca. The Heart and Stroke Foundation worked with his team to adapt the calculator into our Risk Assessment at heartandstroke.ca/risk

Source: Heart and Stroke Foundation