Statistical affairs of the heart
27/03/2013 § 1 Comment
I’m getting to the age when doctors want me to worry about my heart. The Heart Foundation, with their mildly neurotic ‘Fulfil a lifetime’ slogan, wants to help. I can use their on-line calculator to ‘know my numbers’ — to calculate my risk of a heart attack. Well, not a bad idea, eh? A little more information, a warning of things to come?
I’m reasonably healthy, if you can’t tell by my typing. The calculator asks about age, gender, ethnicity, smoking, blood pressure, etc., etc. It’s gathering those known risk factors to calculate a heart attack risk score tailored to me.
But really, the calculator is about striking fear into the, uh, hearts of users. How do I know this? Two ways.
First, the calculator is skewed towards bad news. I didn’t have my cholesterol numbers handy, so it calculated two results. The first result was based on average cholesterol numbers. Then the calculator told me that ‘one in four people’ has an elevated risk, and that’s the number it used to calculate my ‘heart age’. I went back and changed all the inputs to be as positive and healthy as possible, and the calculator still estimated that my heart age was two years older than my chronological age. No matter what I did, the calculator always assumed the worst, and never allowed me to have a healthy heart — a heart younger than my age.
Secondly, the risk numbers don’t add up. The lowest risk possible is a 2% risk of heart attack or stroke; mine came out at 7%. Both numbers are below the ‘mild’ level of 10%. Now, bear in mind that surviving a heart attack outside a hospital is uncommon: there is ‘an overall survival following cardiac arrest [out of hospital] of 6.8%.’ So, the risk numbers are very close to the mortality numbers — a 1.9%, 6.5%, or 9.3% chance of dying from a heart attack in the next 5 years.
Statistics NZ helpfully provides mortality statistics on-line. At age 45, males on average have a 0.98529 5-year survival rate. That is, they have about a 1.5% chance of dying from all causes combined.
The ‘and stroke’ bit complicates the calculations, but let’s just stick with the heart attacks. The Heart Foundation calculator seems to suggest that my best-case odds are 1.9% chance of dying in the next 5 years, and the mild risk category is anything up to 9.3% chance. These are much worse odds than Stats NZ is giving me, at less than 1.5%.
How can I reconcile these numbers? Well, the stroke part was left out, so I could go back and include them (but I won’t for now). The calculator picked up many risk factors, so it isn’t about age or health status. And even the best-case numbers don’t add up, never mind the other ones.
I think the healthy heart calculator overestimates the risk of heart attack, presumably to raise your awareness of the downside risk of high blood pressure, diabetes, and other conditions. That suggests it isn’t about information, but rather manipulation. All for a good cause, of course — a healthy, fulfilled population. And what better tool for manipulation than fear and anxiety?
Statistics shall be my tin-foil hat.