Measurement of the protective effect of attenuated poliovirus vaccine.

Abstract

The assessment of the protective effect of attenuated poliovirus vaccines is particularly difficult. The technique which has been outstandingly successful with other vaccines, such as inactivated poliovirus vaccine, whooping-cough vaccine, and B.C.G., is that of the controlled prophylactic trial, but there are reasons why the method is not readily applicable to this particular problem. In the first place, many of the countries which experienced high morbidity from poliomyelitis in the past have used inactivated virus vaccines on a wide scale, and it would not be easy to allocate individuals in a population randomly to receive either inactivated or live virus vaccine ; allocation at random to either attenuated virus or no vaccine at all is clearly unethical. Secondly, because of the widespread use of inactivated vaccines or for other reasons, the morbidity in most of these countries is now rather low and very large populations would have to be studied to give any chance of seeing sufficient paralytic poliomyelitis to provide an assessment. Thirdly, because there is the possibility of the vaccine virus spreading from individuals who had been given vaccine to their contacts the group given inactivated vaccine might include an unknown propor tion of infected persons ; as a consequence the polio myelitis experience of the groups as allocated might not provide a true measure of the difference in performance of the prophylactics given to them. Also, because of the risk of spread to contacts and the possibility of increase in virulence of the vaccine virus on repeated human passage, live virus vaccine has often been fed so far as possible to whole populations simultaneously and the deliberate withholding of live vaccine from part of the population, as required in a controlled trial, has been specifically avoided. As an alternative to the controlled trial as a method of assessment, reliance has been placed either on the comparison between areas in which attenuated virus vaccines have or have not been used, or on the trend of morbidity before and after the use of such vaccines. The past experience of this disease has, however, been so unpredictable in both time and place that such comparisons cannot immediately indicate whether a specific prophylactic is responsible for any change in pattern. Repeated demonstrations of the association of vaccination with reduced morbidity would carry more conviction, but such evidence would not provide a quantitative assessment of protective effect ; there would be no possibility, as existed, for example, in the Medical Research Council's trials of whooping-cough vaccines (Medical Research Council, 1956), of measuring the relative performance in the field of different vaccines, or of relating such potency to the results of laboratory tests. In this respect the ability to produce antibodies may not be a valid method of comparing inactivated and attenuated vaccines because the former do not aim to produce alimentary immunity.