Canadian researchers say that home blood pressure monitors are wrong most of the time, and could cause problems for people who rely on them to make informed health decisions.

"High blood pressure is the number one cause of death and disability in the world," said lead researcher Jennifer Ringrose of the University of Alberta. "Monitoring for and treating hypertension can decrease the consequences of this disease. We need to make sure that home blood pressure readings are accurate."

Ringrose and her team tested dozens of home monitors and found they weren't accurate within five mmHg about 70 percent of the time. The devices were wrong by 10 mmHg about 30 percent of the time.

The study results are extremely relevant since many doctors ask their hypertensive patients to keep track of their blood pressure readings and report their results.

There are steps, however, that patients can take to minimize inaccurate readings.

"Compare the blood pressure machine measurement with a blood pressure measurement in clinic before exclusively relying upon home blood pressure readings," advised Ringrose.

"What's really important is to do several blood pressure measurements and base treatment decisions on multiple readings," she said. "Taking home readings empowers patients and is helpful for clinicians to have a bigger picture rather than just one snapshot in time."

The researchers compared the results of 85 volunteers' home monitors with the gold standard — two observers taking several blood pressure measurements simultaneously, blinded to one another, with a third person ensuring agreement between both observers' readings.

While the average difference between the home monitors and the gold standard measurements was acceptable, the majority of individual devices demonstrated clinically-relevant inaccuracy. The team also found that readings were more inaccurate in men than in women. They believe there are many factors that could account for their findings.

"Arm shape, arm size, the stiffness and age of blood vessels, and the type of blood pressure cuff are not always taken into account when a blood pressure machine is designed and validated," said study co-author Raj Padwal Padwal. "Individual differences, such as the size, age and medical background of the person using the blood pressure monitor are also contributing factors."

The study was published in the American Journal of Hypertension.

Blood pressure testing has been called into question by other studies. A Harvard study also found that home blood pressure monitors weren't always accurate in up to 15 percent of patients. Researchers from Great Britain's University of Exeter found that checking pressure in one arm — the usual method  — could let some people with problems fall through the cracks, and that testing both arms could spot people at risk of heart disease, even in those who appear healthy.

Researchers analyzed the medical records of more than 3,000 people between the ages of 50 and 70, who had a greater risk of cardiovascular disease, but were otherwise healthy.  They found that a difference of only 5 points (5mm Hg) in systolic blood pressure measurements between the two arms doubled the risk of dying from heart-related disease during the eight-year follow-up.