Self-reported data on health risk appraisals has its pros and cons, despite the failure of many wellness and benefits professionals to acknowledge the pros. What about data obtained at worksite biometric health screenings? How valid is it?

[For this post, "biometric" refers to the measurement of biological test results, as opposed to its increasingly popular usage describing the identification of individuals based on their biology. "Valid" is used in a more casual, rather than statistical, context.]

No health test is perfect. For an example, let’s consider blood pressure, which most people would consider one of the more straightforward biometric health measurements. There are at least two major branches of potential error in measuring blood pressure.

1. An individual’s blood pressure is much more variable than most people think. If your participants are not questioning why their onsite screening blood pressure is different from their blood pressure measured in their doctor’s office, they either are unusually well informed about blood pressure variability, or you have not gotten them to pay close enough attention to their numbers.

What affects an individual’s actual blood pressure?

• For some people, blood pressure spikes at the doctor’s office (“white coat hypertension“)
• Blood pressure tends to be higher at work.
• Blood pressure changes throughout the day. It tends to be higher from 6:00 am to 12 noon, just when most worksite health screenings are taking place.
• Blood pressure is affected by caffeine, physical activity, and smoking.

2. Professionals charged with measuring blood pressure — even in a clinician’s office, but especially at worksite health screenings — often fail to adhere to best practices in blood pressure measurement. How much of a difference can measurement error make? Here are some examples, based on a review of research by medical device manufacturer Welch Allyn, of how much difference measurement error can make to systolic blood pressure measurements:

• Using a cuff that’s too small can overstate the measurement by 10 to 40 mmHg
• Placing the cuff over clothing can increase or decrease the measurement 10 to 40 mmHg
• Having the participant’s back unsupported or feet dangling can increase the measurement 5 to 15 mmHg
• The participant crossing their legs is associated with overmeasurements of 5 to 8 mmHg
• Failure to rest, sitting quietly for 3 to 5 minutes before the pressure is measured, can add 10 to 20 mmHg
• If the participant is talking during the measurement, their reading is likely to be 10 to 15 mmHg higher than their true pressure.
• A full bladder is associated with overmeasurement of 10 to 15 mmHg

According to the European Society of Hypertension

No matter which measurement device is used, blood pressure will always be a variable phenomenon influenced by many factors, not least being the circumstances of measurement itself. Considerable variability may occur in blood pressure from moment to moment with respiration, emotion, exercise, meals, tobacco, alcohol, temperature, bladder distension and pain, and blood pressure is also influenced by diurnal variation…

Inarguably, blood pressure is an essential and widely accepted biometric test, especially when multiple readings are taken by well trained clinicians. But a biometric test like blood pressure measurement, just like self-reported health status, is imperfect. Neither should be taken as engraved in stone, nor be summarily dismissed. When one of your participants races into your onsite screening, cup of coffee in hand, on the way to their 8 a.m. meeting, you cannot safely assume that their blood pressure measured at the screening is more accurate than their self-report of their blood pressure measured at their doctor’s visit two weeks prior. It may or it may not be.

In the next post, we’ll take a quick look at another common worksite biometric health screening test: Cholesterol.

[This is one in a series of posts about health risk and health risk appraisals.]