National Participation Trends, 1955-61 in the Poliomyelitis Vaccination Program

THE largest vaccination program in the his¬ tory of the United States began with the licensing of Salk poliomyelitis vaccine for mass production in April 1955.By August 1957, about 150 million doses of Salk vaccine had been administered.Approximately one-half the population under 40 years of age had re¬ ceived at least one inoculation, and almost one- third had received three inoculations.There were sizable annual gains in the number of persons inoculated during the period August 1957-September 1961 (chart).The number of doses of Salk vaccine administered had more than doubled by the end of this 4-year period.By September 1961, 77 percent of the popula¬ tion under 40 years of age had received at least one inoculation, about 67 percent had received three or more inoculations, and 40 percent had received the booster inoculation.Trend statistics in this report are based on five national household surveys conducted an¬ nually since August 1957.These surveys were sponsored by the Public Health Service and conducted by the Bureau of the Census (1) as supplements to its monthly Current Population Survey (CPS).To ascertain the Salk vaccination status of the surveyed population, the following ques¬ tions were added to the regular CPS schedule: Has ?Yes ?had any polio injections or shots?? NoHow many shots has ? 1 ? 3 ? 2 ? 4 had already? Interpretation of DataIn this paper, the statistics are expressed principally as percentages of the population that received a stated number of inoculations of Salk poliomyelitis vaccine.Since these statis¬ tics are baSed on a sample, they are subject to sampling errors (table 1).These errors per- tain to the estimates derived from the 1957, 1960, and 1961 surveys, each of which utilized the entire monthly CPS sample of 35,000 house¬ holds.The 1958 and 1959 estimates were based on information from a subsample of one-half the households in the CPS sample.The sampling errors for these estimates are roughly 40 to 50 percent larger than the errors shown in table 1.Survey estimates are also subject to non- sampling errors due to nonresponse and to re¬ sponse bias.Nonresponse bias was trivial since more than 99 percent of the respondents re¬ ported the number of inoculations received.It is more difficult, however, to evaluate response bias.According to recheck interviews of a subsam¬ ple of persons interviewed in August 1957, the reporting errors were small.About 97 percent of the original and recheck interviews agreed on whether or not the person had received inoc¬ ulations, and the agreement on the number of inoculations received was almost as high.The numbers of doses of Salk vaccine released