Beta-blockers are often prescribed for patients who are asymptomatic and for whom the benefits of therapy are likely to be small. Side-effects are therefore of great importance. Symptoms such as muscle fatigue and peripheral coldness are commonly associated with treatment with beta-blockers but these subjective phenomena are difficult to detect and quantify so that their prevalence and severity are uncertain. Conventional symptom questionnaires may be relatively insensitive and visual analogue scales (VAS), which permit interval quantification of subjective phenomena, may have some advantages. They appear to be able to detect the presence of tiredness of the legs and peripheral coldness among hypertensive patients taking beta-blockers and are more sensitive than descriptive or 'category' questionnaires. However, some patients find VAS hard to understand and a minority of patients score erratically for all symptoms. A most detailed explanation is of paramount importance. Scores derived from VAS surveys are not normally distributed and should be examined using non-parametric statistical methods; normalising transformations do not improve the sensitivity of the method.