One form of exercise that is becoming increasingly popular with the older generation is strength training. Recent research and training studies have shown that high-resistance exercises have provided substantial strength gains in older adults. However, this means that proper evaluation of 1RM (the amount of weight that can be lifted for one repetition) is becoming increasingly important when attempting to establish training loads. Unless the individual is confident or medically fit to lift maximal loads, a prediction equation for determining 1RM must be used. A number of these currently exist, and recently a team of researchers at Western Washington University set out to determine the validity of 1RM equations for older adults (K.M. Kutzen, L.R. Brilla & D. Caine, 'Validity of 1RM prediction equations for older adults'. The Journal of Strength and Conditioning Research, Vol. 13, No. 3, August 1999).

Fifty-one subjects (30 female and 21 male), all aged over 60, took part in two experimental conditions. The subjects were tested on 11 machines ranging from triceps press and bench press to supine leg press. During session one, the repetitions to fatigue data for predicted 1RM equations were collected. Five to eight days later, the subjects completed an actual 1RM on each exercise machine. The researchers found that correlations between actual and predicted 1RM scores demonstrated a moderate to strong relationship for all exercises, with upper-extremity exercises producing the higher coefficients (0.77-0.90) as compared to the lower-extremity exercises (0.61-0.81). In all of the exercises, the average predicted 1RM was lower than the actual 1RM. All of the six prediction equations used underestimated the actual 1 RM by a range of 1-10 kg, depending on the type of lift.

The researchers concluded that the use of prediction equations to determine 1RM was a useful tool for professionals working with clients in this age group. They felt that a 7-10 repetition measurement for predicting 1RM could easily be included in a training programme and could be used to quickly establish training loads. Of the six prediction equations tested, no one equation consistently outperformed the others in terms of accuracy or strength of prediction. The study team also recommended that future research should attempt to establish population-specific prediction equations.