I would like to address my remarks to the general problem of accident research as we in the Department of Labor see it, although not necessarily from the point of view of our departmental responsibilities.

One of the leading research scientists in this field, Dr. William Haddon, Jr., the Director of the epidemiology residency program of the New York State Department of Health, defines accidents simply as the "unexpected occurrence of injury."

He identifies two general classes of accidents:

First, those caused by interference with normal whole body or local body energy exchange. An example would be suffocation on the former level, frostbite on the latter.

Secondly, "the delivery to the body of amounts of energy in excess of the corresponding local or whole body injury thresholds."

There are at least five forms of this second class of accidents: Those resulting from mechanical energy, as for instance, motor vehicle accidents; thermal energy, as in second or third degree burns; electrical energy, as in electrocution; ionizing radiation, as an overdose of X-rays; and finally, chemical energy produced by the wide variety of chemicals developed by modern science or by the more classifical plant and animal toxins. (Poison ivy would be an example of that.)

When your committee first began to inquire into the subject it would have been necessary for a witness such as myself to say that while we know accidents will happen, just how often and to whom was a matter for considerable speculation.

This is no longer the case, thanks to the National Health Survey conducted by the Department of Health, Education, and Welfare.

In the period of July 1959 to June 1961, for example, some 44,995,000 members of the civilian noninstitutional population incurred injuries of one kind of another. About 18.8 million of the injuries occurred in the home, 8.1 million took place on the job, 4.7 million occurred in motor vehicle accidents.

It would appear that perhaps one person in five is injured in the course of the year. On the subject of the cost of injury, we are not nearly so well informed, even at the level of macroeconomics.

The National Safety Council estimates that the total cost of accidents and accidental injuries and deaths in the United States during 1961 was $14.5 billion.

This is a conservative figure, as it ought to be. The Safety Council is careful not to appear to exaggerate the dimensions of the problem in order to increase its importance.

The amount of $14.5 billion represented 2.8 percent of the gross national product of the United States in 1961. Our inclination in the Department of Labor would be to raise that amount to at least 3 percent of the GNP.

I don't have to point out to you that for the period 1957 to 1962, the annual increase in GNP was running at only 3 percent a year: accidents were in a sense depriving us of our growth, or much of it. I would also add that $14.5 billion, the low figure I mentioned earlier, was in excess of the total expenditure for research and development by_Government, industry, and nonprofit institutions in that year. It was only slightly less than the expenditure for all new buildings, and about equaled the purchase price of all new and used cars.

Accidents are very much a household problem for the Federal Government. We have some 110,000 injuries a year reported under

the Federal Employees Compensation Act, which is administered by the Department of Labor.

Last year we expended $46 million from the compensation fund for injuries to Federal-civilian employees. We have reason to believe that the number of injuries that occur is greater than those that are reported.

For example, in the Post Office in fiscal 1962, their Safety Division reported 84,000 injuries for its employees, although only 49,000 were reported under the Compensation Act.

This would mean that one postal employee out of seven was injured on the job during this period.

Studies by the Department of the Interior indicate that indirect costs of injuries are at least equal to the direct costs. In the Department of Defense, accidents are a problem of considerable proportion, as you would expect.

It would appear that in 1960, the overall cost of accidental injury and death to active-duty military personnel and their dependents was in the neighborhood of $285 million.

Property damage cost would double this figure, so that the total would amount to something like 1 percent of the Defense budget. Injuries and deaths to active-duty personnel only in motor vehicle accidents came to $83 million, in direct costs, with property damage probably an equivalent amount.

In general, something like half the total cost of accidents is accounted for by motor-vehicle accidents. Our information about the cost of motor-vehicle accidents is by no means complete or adequate, but the problem is so large that a fair idea of its dimensions can be had. In the course of the next 5 years, which is 1963 to 1967, inclusive, we would be willing to predict that the total cost of motor-vehicle accidents will come to some $42 billion.

Something like 190,000 to 195,000 persons will be killed and considerably more will be disabled.

The question of concern to this committee is what to do. We know about the cause of accidents and what are we likely to learn about the cause of accidents and what are we likely to learn about their prevention?

I reply that quite a bit is known about accident prevention as an applied technique, and the record of industrial safety quite substantiates this fact.

However, very little is known about the etiology of accidents and it would be my judgment that the absence of basic research data on this subject is now beginning to retard our progress.

We have gone about as far as you can go in the "hard hat and goggles" approach to the problem. The next breakthrough will require a far more sophisticated understanding of the nature of the problem.

I would judge that at least two factors can be identified which have held us back in this field. The first is that while we define accidents as unexpected events, they do in fact, seem to have an explanation. Most of these explanations are tautological: You say, "I fell because I slipped," but somehow they satisfy curiosity and they allay fear. The nagging sense of mystery and of danger which led men thousands of years ago to begin the study of infectious diseases, has simply not grown up around the subject of accidents, even though the morbidity

and the mortality resulting from accidents is just as real as that resulting from infectious diseases.

The second factor that I would identify is probably related to the first. The study of accidents has so far attracted only a very few persons from the learned professions. The result is that so far very little has been learned about the subject.

Here and there the field of accident prevention has touched upon the data of a recognized profession and almost always with excellent results.

As Congressman Nelsen would agree, the highway engineers have performed miracles by the simple application of the rigorous standards of their calling.

These standards provide for elaborate methods of analysis and testing and, most importantly, for strict accountability for failure.

Should a major bridge on the interstate highway system collapse one night, you can be sure that the engineering profession will in short order determine why it happened and who was responsible, but if the traffic safety campaign should fail utterly to achieve its announced objectives, the likelihood is not a word will be said about it.

Indeed, the authors are likely to continue to take credit for their efforts as if the campaign had been a vast success. There is simply no tradition of self-criticism in this field. Without that criticism there is unlikely to be much progress.

Traffic safety happens to be the area in which most accident prevention activity has occurred, therefore, it provides most of the bad examples. Let me cite an important one here. Highway accidents began to be a significant source of morbidity and mortality in the United States about one half century ago. About a generation ago, they reached epidemic proportions and have continued at that level

since.

In an effort to keep with the problem, for little more than a generation agencies of local, State, and Federal Governments have been compiling statistics about accidents, injuries, and deaths with a diligence and industry seems to grow as the years go by.

But that has been an almost wholly uncritical effort. As a result, it has been almost wholly useless. It is my impression, and it is the firm opinion of research workers for whom I have the greatest regard, that with perhaps one or two exceptions all the vast accumulation of data about automobile accidents over the past half century has contributed almost nothing to our understanding of the cause and prevention of accidents.

It is worth calling to the attention of the committee that it was only 3 years ago that the first scientific information about the nature of pedestrian fatalities was published in this country, and this information was obtained by the deceptively simple process of stopping some 200 persons who happened to be walking the streets of New York City at certain times and places.

There are not more than a handful of research reports in this field of the quality customary in other scientific fields, and almost all of these have been developed by non-Government groups, generally using non-Government data.

(Here, I would exclude the Cornell Crash Injury Research Studies, which are perhaps a special case.) The general rule may be laid down that in no field that I know of, is the disproportion so great between

expenditure on data collection and similar efforts and the production of scientific acceptable results.

Just to intercede with one point, it is not just that I feel that we neven't learned anything from this data collection process. This is a more serious matter than simply the failure to obtain new information. I fear that we have been mislead by such data, and we find ourselves in the position described by the old saying that, "It's not ignorance that hurts so much as knowing all those things that ain't so."

Accident statistics have, for example, shown that some drivers have more accidents than others, which has led most Government agencies in this field to assume there is something special about the multiaccident drivers, although in fact, most of them may be nothing more than innocent victims of the Poisson Distribution.

The simple fact is that data collection is not research. The collection of undependable data is not much of anything.

However, the great bulk of our expenditures in this field for half a century have been confined to this area. I believe the committee would be alarmed to learn how meager are the efforts we make in other directions.

The best information obtainable by the Bureau of Labor Statistics. is at the present time in the United States there are only about 50 competent researchers engaged in basic research in the accident prevention field.

The National Safety Council, for example, has on its staff only two persons engaged full-time in accident research and only one of these persons is presently a Ph. D. The other will be.

These are very scarce people to come by. This brings me to the subject of a National Accident Prevention Center. As you know, Mr. Chairman, within the administration, there are people who are anxious to see progress in this field who are as yet uncertain as to what precisely would be the best institutional arrangements for making such progress.

I would prefer to leave that to persons better qualified than I to judge such matters. However, I do most emphatically wish to support your concern that something be done.

The Nation needs a center of some kind where persons can be trained and gain experience at the professional level in the field of accident prevention research.

If we are to get past the 50 mark, as it were, such an effort must be made. I assume further that such an effort must be directly related to the medical profession, as much as this may pain our friends in the behavorial sciences, although they, of course, will be much involved.

It is my understanding this view is held by a significant number of medical doctors, although I cannot attest to that statement.

I would, however, quote Dr. Haddon once more who wrote recently: The time has come to stop regarding injury causation and research as somehow mystically difficult and different from the sequences with which we have long successfully dealt in the infectious disease and other areas, since there is no convincing evidence that this is the case other than its frequent assertion.

It would be my hope that agencies such as the Department of Labor would become much involved in assisting in such research and also that we would have even a larger role in applying the results.

I would ask the forbearance of the committee to expand for just one last moment on these possibilities.

Two items. First, an example from the field of traffic safety. One of the most pressing concerns in this field is that of the design of passenger cars. As the committee has abundantly demonstrated, very little is known about the relation of design to accidents and injuries. We do know that in 1960, for example, I believe the number is nearly one licensed motor vehicle out of seven was involved in a traffic accident.

Last year, your committee heard testimony that between onequarter and two-thirds of all automobiles manufactured in the United States are sooner or later involved in a personal injury accident.

We have also recently had an excellent study that indicates compulsory motor vehicle inspection may significantly reduce accident rates. All these factors point to the probable importance of automobile maintenance.

Now, there were in 1960, some three-quarters of a million automobile mechanics employed in the United States. About a third worked in the service department of new and used car dealers. Another third worked in repair shops. The remaining third worked in gasoline service stations, for manufacturers, and others.

Over the next decade, we will have to train between 350,000 and 400,000 new mechanics to make up for natural attritions and to provide for the increase in the total number of motor vehicles at a ratio of 1 mechanic for about every 90 vehicles.

It is an interesting fact that, while we have very high standards of apprenticeship and licensing in service trades, as plumbing and electrical work, in general, these requirements have not been adopted in the field of automobile maintenance.

But this surely should not mean that we have no established standards whatever, although I fear this is generally the case.

It seems to me, therefore, that it would be an excellent thing if we were to learn more about what is now the level of training of automobile mechanics and what, if any, public standards of training ought to be set.

I would think we ought to also learn a great deal more about the relationship of automobile design to efficient maintenance in terms of the skills and training of the mechanics' work force.

Is it possible, for instance, to mass produce automobiles that require more maintenance skill than the current work force possesses? If so, ought we to change the design, retrain the work force, or both?

These are questions where basic and applied research meet. I am sure the Department of Labor would be most interested to join in a study of this kind insofar as it concerns the Manpower Development and Training Act, the occupational safety movement, and, of course, Federal Safety Standards.

The Department of Labor would also, for example, be most interested to learn more about the problem of injuries to older workers. In his recent message on aid to our senior citizens, President Kennedy expressed the concern of the administration that older workers be permitted to continue in employment as long as they needed and wished to do.

He said, "Denial of employment opportunities to older persons is a personal tragedy. It is also a national extravagance, wasteful of human resources.