In reference to the latter, these allocations do not include the cost of control work in connection with carrying out the provisions of the biologics law. It should also be borne in mind that the National Microbiological Institute is not a categorical institute dealing with only one field of disease. Its activities cover a wide range of infection and parasitic disease.

No research on poliomyelitis as such is supported by the National Microbiological Institute under its research-grants program. The institute does, however, support a number of studies in universities and other research institutions in the broad field of virus diseases. Fundamental knowledge derived from these studies will be applicable to many problems in infectious diseases, including poliomyelitis.

FUTURE NEEDS

There are three areas of research which from the standpoint of the control and the prevention of poliomyelitis may be considered to be of preeminent importance. The first of these is the development of a specific vaccine and the appraisal of its prophylactic potentialities against this disease. In an infection which appears to be transmitted to susceptible individuals primarily by contact with infected persons, and in which a solid substantial immunity results from infection, all considerations point to the desirability of protection by creating immunity with a vaccine. Pending the development of such biologic preventive, the passive transfer of protective antibodies by means of gamma globulin will serve to prevent or modify the severity in many cases. It is evident, however, that a good vaccine which will stimulate its recipient to generate his own antibodies the rest of his life would be more effective and cheaper. Much progress is being made in the development of such a material, but it must be remembered that the first such experimental product is not necessarily the most effective, the safest, or the least costly. Research to discover better vaccines should go on as an important element of a planned program of research on poliomyelitis.

The second great need in the control of this disease is a chemotherapeutic agent which will cure the infection promptly, alleviate its acute symptoms, and prevent the crippling or lethal paralysis which might otherwise ensue. Much work has been done in this area with no significant success thus far. Unlike the search for the best vaccine, which should be the goal of an orderly, planned progression of studies and experiments, it seems as likely that the specific medication against poliomyelitis may be found as the result of research entirely unrelated to infantile paralysis. That has been the situation with respect to certain other diseases. A decade ago, for example, a new and highly successful drug, chloroquine, was developed to combat malaria. Today this drug is finding clinical application in studies of rheumatoid arthritis and certain skin diseases. The same is true of another antimalarial preparation, developed since the Korean conflict, which has proved promising in studies of leukemia and toxoplasmosis in experimental animals.

The third area of essential knowledge in which more work should be in progress is the specific pathogenesis of paralytic poliomyelitis. It is now known that, in the early stages of poliomyelitis, the virus circulates in the blood. What is the incident or accident or circumstance within or external to the body of such an individual which, one time in a hundred or more, permits the virus to enter and damage the central nervous system causing paralysis? Ancillary to this consideration is the determination of whether or not this rare circumstance is controllable or preventable. If so, another means of protection against paralysis might become available; if not, we would realize that we must put our reliance entirely on vaccines.

With respect to the vaccines now undergoing development and evaluation by various nongovernmental groups sponsored largely by voluntary foundations, it would probably be safe to say that the final outcome of this effort would not be influenced substantially by Federal appropriations to augment work under way. The same is probably true of chemotherapeutic research, for it is more than likely that every new antibiotic or new drug which is significantly successful in combating any infectious disease is promptly tested against poliomyelitis. Thus it would seem that congressional support might wisely be directed to broad noncategorical research of a fundamental nature-the acquisition of knowledge which has profound implications in many disease areas.

One such area, for example, is upper-respiratory infections, of which the common cold is a conspicuous example. Here is a disease problem which has long and stubbornly resisted solution. The economic burden that the common cold imposes on our people is great enough in itself to justify a much stronger research effort than is now being supported. The same is true of a number of

other diseases caused by viruses, rickettsiae, and fungi. For investigators, these problems all have a common denominator: a paucity of fundamental knowledge. Before we can develop preventive measures against such ailments, or better methods of treatment, we need to know more about the life processes of disease organisms-their metabolism, their genetic constitution, and many other factors concerning their structure and behavior, and host-parasite relations.

The specific benefits to be derived from a broad approach to research problems are impossible to predict. From past experience, however, we know that new knowledge frequently finds unexpected application to problems far removed from the original area of investigation. The point which is worth emphasizing here is that the basic knowledge which we seek in critical research is essentially noncategorical. In our efforts to find answers to pressing problems, we must remember that imaginative science seldom reaches its full potentialities when support is inflexible and work is restricted to rigidly specified problems. The record of history shows that tomorrow's practical results are almost always the progeny of today's free curiosity.

Appropriated funds to Public Health Service used for poliomyelitis-Research and diagnosis

By Robert E. Serfling, Ph. D., and Ida L. Sherman, M. S.

Dr. Serfling and Mrs. Sherman are Chief and Assistant Chief, respectively, of the Statistics Section, Epidemiology Branch, Communicable Disease Center, Public Health Service, Atlanta, Ga.

In the early summer of 1894 inhabitants of the Otter Creek Valley in western Vermont became aware of a strange paralytic disease in their community. It seemed to select younger children as particular victims although a few adults were also affected. This first notable epidemic of poliomyelitis in the United States was carefully investigated by Dr. C. S. Caverly, president, Vermont State Board of Health. His field studies, constituting a classic model of "shoe-leather" epidemiology, mark the beginning of our modern knowledge of poliomyelitis. In the following years other investigations conducted in the Caverly pattern demonstrated the extensive distribution of the disease and indicated the need for systematic morbidity reporting.

Massachusetts, in 1907, was the first State (1) to require notification of all cases of poliomyelitis. In 1910, the Surgeon General of the Public Health Service requested all States to submit reports on poliomyelitis for 1909 and 1910. This initiated national reporting of poliomyelitis, although regular inclusion of reports from all States was not achieved until about 1922. Since that time a large body of data has accumulated, forming an increasingly comprehensive base for continuing analyses of elementary epidemiological characteristics, such as secular trends, geographic distribution, and seasonal variations. A number of excellent analyses have been published (2, 3, 4, 5), but the most recent include only the years through 1946 in the United States (6, 7). Sabin (8) reviewed epidemiological characteristics of a number of poliomyelitis outbreaks throughout the world.

Certain epidemiological characteristics of poliomyelitis have changed with time. In the United States, annual rates both of reported cases and of deaths have shown an upward trend, particularly during the past decade. In earlier

decades a lower incidence was observed in Southern States that in Northern States. This difference is less apparent in recent year.

Continuing studies and analysis of these elementary epidemiological characteristics and their changes are indicated for any infectious disease that remains as a serious problem in the country. Furthermore, discovery of the value of gamma globulin in the prophylaxis of poliomyelitis (9, 10) poses difficult problems as to the best way to utilize available supplies.

The fullest possible knowledge of the current epidemiological pattern of poliomyelitis as revealed by morbidity and mortality reports may be useful in guiding administrative decisions. For these reasons the present paper has been prepared. The past history of poliomyelitis as recorded in published literature and official reports was reviewed and special attention was given the period 1932 to 1952 with particular emphasis on changing patterns of the past 5 to 10 years.

NATIONAL INCIDENCE

The trend of the national case and death rates in the United States during the period 1910-52, is shown in figure 1. Since both cases and deaths were not reported from the same group of States during earlier years, the rates in figure 1 were based on reports from those States which reported both cases and deaths. The record of national incidence of poliomyelitis in the United States falls naturally into four periods:

1. Prior to 1909, when information on incidence, except for a few States, depended on descriptive accounts in the epidemiological literature.

2. From 1909-16, when published reports were available from some States. These were supplemented by special studies of Lavinder, Freeman, and Frost, and their summaries give more complete information on the period.

3. From 1917 through the epidemic year 1931, during which time reporting gradually became more complete. By the middle 1920's most States were reporting annually.

4. From 1932 until the present, during which time national reporting of both cases and deaths has been essentially complete. Because of the obvious differences which characterize national reporting in the different periods, they are presented separately in this discussion.

Early years

During the 1894 outbreak in Vermont, Caverly collected information on 132 cases. Of these, 119 had shown paralysis, 7 had died before paralysis was noted, and 6, although exhibiting symptoms characteristic of early stages of the illness, had not developed paralysis. In Rutland, the largest community in the affected area, 55 of the 12,000 inhabitants had been stricken-an attack rate of 460 cases per 100,000 population. In nearby Proctor, a town of 2,000 persons, 27 cases had occurred. The remaining 50 cases were scattered through a dozen small communities in the area.

Two years later, in a final report of his investigations (11), Caverly concluded that the disease was epidemic poliomyelitis, possibly a variant form of the infantile paralysis which in the United States had been known principally as an endemic disease of relatively infrequent occurrence.

In the succeeding years similar outbreaks occurred with increasing fre quency in many parts of the country. In 1908, 2 papers (12, 13) summarized accounts of 17 poliomyelitis epidemics in the United States and others throughout the world. By this time, outbreaks had been described in Alabama, California, Florida, Illinois, Maine, Massachusetts, Michigan, Missouri, New York, Pennsylvania, and Wisconsin. These seem generally to have been of

smaller scale than the Rutland episode.

In New York City, however, the 1907 outbreak was the largest then recorded in any place. The impact led to an extensive retrospective investigation initiated in October of 1907. The study (14) was conducted by mail and produced detained information on 752 cases. It was estimated that in all, about 2,500 cases had occurred. Although cases were somewhat concentrated on the east side of Manhattan, the epidemic had extended northward to Poughkeepsie and throughout the western end of Long Island. The case fatality was estimated to be 5 to 7 percent, about half that of the Rutland outbreak.

In Massachusetts, where smaller outbreaks had been noticed since 1893, incidence was also high in 1907 and the State made poliomyelitis a reportable disease. In 1908, 130 cases were reported in Massachusetts (1). In midwestern Minnesota, 150 cases were recorded, and in Wisconsin, 408. In the following year, 1909, a great outbreak struck in Nebraska. Description of this epidemic (15) was also based on a retrospective study. In answers to letters,

58 physicians reported 999 cases in 18 counties. The greatest number of cases, 384, was reported from Polk County, with a population of 10,000. Douglas County (1910 population, 168,546) reported 79 cases, a rate slightly less than 50 per 100,000 population, and comparable to that of the New York City epidemic of 1907. In 1910 Massachusetts reported 845 cases, Pennsylvania 1,112, and in the Midwest, Minnesota and Iowa each recorded more than 600 cases. In the Far West, nearly 400 cases occurred in the State of Washington. On August 9 of that year (1910) the Surgeon General of the Public Health Service initiated the request that started national reporting of poliomyelitis. In the same year provision was also made for separate classification of poliomyelitis deaths in the national vital statistics summary.

Figure 1. Annual poliomyelitis case and death rates in States reporting both cases and deaths, United States, 1910-52.

SOURCES: Cases, 1910-50-Public Health Service: The Notifiable Diseases (Pub. Health Rep. Supp.). 1951-National Office of Vital Statistics: Reported Incidence of Notifiable Diseases in the United States, 1951, Annual Supplement to Weekly Morbidity Report, vol. 2, No. 53, 1953. 1952-National Office of Vital Statistics: Morbidity and Mortality Weekly Report, vol. 1, Nos. 1-53 inclusive. Deaths, 1910-49, Vital Statistics of the United States. U. S. Bureau of the Census, 1910-44; U. S. Public Health Service 1945-49. 1950-51 data from advance releases, 1952 data from 10 percent mortality sample, Jan.-Nov., National Office of Vital Statistics.

Years 1909-16

For 1909 only 3 States submitted morbidity reports to the Public Health Service and in the following 7 years the numbers ranged from 11 to 29. After the 1916 epidemic, Lavinder, Freeman, and Frost published a summary of morbidity and mortality for the period 1909-16 which incorporated data from a number of States not included in the earlier national summaries. For the years 1909-15 they obtained additional State morbidity reports and also mortality reports for some States in which no morbidity data were available. For the latter, estimates of cases were made from reported poliomyelitis deaths, assuming a case fatality rate of 20 percent. For States in which only certain cities were in the registration area deaths for the entire State were estimated from those in the registration areas. Since Lavinder, Freeman, and Frost had noted that in States for which registration was complete the urban death rate was lower than the rural rate and that a case fatality as high as 20 percent occurred only rarely, they believed that error in their estimates had been in the direction of underestimation. Because of the care which went into this study, their figures for annual incidence of poliomyelitis in the Nation for 1909-16 are quoted below:

During the 5 years following the 1910 outbreaks, the estimated national rates declined. In these years the largest outbreaks took place in 1912, when New York reported 1,108 cases and California 531. These were the only instances in which a State reported more than 500 cases, although Massachusetts, New York, Pennsylvania, and Virginia reported from 100 to 500 cases annually, and Illinois and Ohio reported from 100 to 500 cases in 4 of the 5 years.

The great epidemic of 1916, although leading to highest rates in the Northeastern States, also struck severely in the north central area and in Montana. States with rates of approximately 20 per 100,000 population or larger are listed in table 1.

TABLE 1.-States reporting 20 or more poliomyelitis cases per 100,000 population

For the year 1916, Lavinder, Freeman, and Frost obtained morbidity reports from all but 4 States, amounting to a total of 29,061 cases. From these they estimated the national rate to have been 28.5 cases per 100,000 population. The total number of cases in the District of Columbia and the 27 States which reported to the Public Health Service was 27,363. Among these States, the average rate was 41 cases per 100,000 population. Years 1917-31

After 1916, no additional efforts were made to obtain complete information on reported cases of poliomyelitis. Annual case rates for the Nation, computed on the basis of the populations of those States submitting reports, and death rates for the death-registration States are shown in table 2. The period