In a report to the 1984 IRCOBI conference, the author and two colleagues demonstrated the application of the Maximum Likelihood method to biomechanical research data. The method was used to interpret field accident data or laboratory research data into a cumulative risk function for injuries. By using that method the problem with so called censored data - so common in biomechanical research - was easily overcome. The present report is a follow-up of the 1984 report, giving answers to received comments, and also including recent developments in connection with the method. A procedure to complement the risk function with a confidence band has been developed. It is shown that the method can comprise both censored as well as non-censored data in the same analysis. A heuristic goodness of fit measure is suggested. In conclusion, the described method seems to be an efficient tool for biomechanical research work.