Researchers at the university's Institute of Cardiovascular and Medical Sciences calculated the variability of blood pressure, over long periods of time, one, four, five and nine years, for 14,522 hypertensive patients attending the Glasgow blood pressure clinic.
Their results showed that the magnitude of visit-to-visit blood pressure variation was a strong predictor of mortality, independent of long-term average blood pressure.
Even people who would be considered well-controlled in terms of blood pressure values at each visit showed a higher risk if they had wide swings in their blood pressure readings between clinic visits.
High blood pressure is considered a silent killer, as it has no symptoms and, if untreated, can result in early stroke, heart attack and death.
The current treatment of high blood pressure revolves around regular blood pressure checks and adjusting treatment to get the blood pressure down to safe levels.
The study was led by Dr Sandosh Padmanabhan, who said yesterday: "Blood pressure is inherently variable and will fluctuate due to a complex interaction of various factors.
"Factors such as stress, seasonal effects and people not taking medication regularly can cause increased blood pressure fluctuations.
"The research has implications for how we best manage hypertension in patients. For example, physicians will need to give more consideration to blood pressure variability when monitoring and treating high blood pressure. The results of our study also highlight the importance of not only taking blood pressure medicines to reduce blood pressure but also taking them regularly."
The study is published in the journal Hypertension.