NERVOUS EXHAUSTION.

DURATION OF ILLNESS AT LEAST ONE MONTH-DATE OF ATTACK
PRIOR TO APPLICATION.

The death rate from suicide and heart disease was appreciably higher than the standard.

OTORRHOEA.

I At time of examination.

II One attack within two years.

III One attack between two and five years.

IV Two or more attacks, the last within two years of application.

The influence of medical selection in pulling the mortality down considerably below the expected, is well shown in this group. No doubt the cases were mostly mild ones to have given such a favorable result; otherwise, one might imagine otorrhoea to be a desirable asset.

Another point that appears to be shown is that a history of recurrent attacks is followed, in spite of a rigid medical selection, by a mortality above the expected.

This chart shows the material arranged in three classes:

Class One, with the diagnosis certain; thoroughly treated,— meaning two years of continuous treatment, and one year's freedom from symptoms thereafter.

Class Two, with the diagnosis certain; not thoroughly treated, or no details of treatment given.

Class Three, with the diagnosis doubtful; probably little or no treatment.

These cases of syphilis were insured long before the spirochaeta pallidae were discovered and before the modern ideas of treatment were introduced.

It seems reasonable to suppose that the attitude of the Companies toward those cases that had not undergone so-called "thorough treatment" was much more critical and severe than it was toward those who had been thoroughly treated. The latter were probably more freely accepted for insurance.

This illustrates again the influence of selection in shaping the results.

In the third class the extra mortality was most surely due to the presence of a considerable number of cases of syphilis, as there was no other known factor that could have so influenced the mortality.

In this chart the material is arranged in two groups; namely, those cases in which the diagnosis was certain, and those in which it was doubtful (without the aid of laboratory tests).

The material in each of these groups is arranged so as to show the mortality in different decades of ages.

In the second group-diagnosis doubtful-it is pretty certain

that the mortality can be accounted for only by a considerable number of cases without clinical symptoms that would be recognized as positive cases of syphilis with present-day diagnostic aids.

In the first group the greatest mortality occurred in the age period from 40 to 50; in the second group in the age period from 30 to 40.

The fact that the greatest mortality in the first group was deferred to a decade later may be due to the influence of treatment, for it is certain that in the first group the cases had much more treatment than those in the second group. In the second group-diagnosis doubtful-it can safely be assumed that the cases had little or no treatment.

"In the syphilis cases the death rate from pneumonia, diabetes, and suicide was much higher than normal. From locomotor ataxia, softening of the brain, and paralysis combined, the death rate was five times the normal.”

DISCUSSION.

DR. ROOT (Hartford): Gentlemen, the facts brought out by Dr. Rowley in his paper and illustrated by his charts are hardly discussible because he has figures covering experience of life insurance underwriters for a period of twenty-five years. All that can be discussed in this meeting is the significance of those charts. Of course, my experience, as a life underwriter, coincides exactly with his and the experience of the Aetna Life is practically identical with the experience derived from the study of the medicoactuarial table illustrated by Dr. Rowley.

There is one point, however, which I think is discussible and opens up a very interesting field for speculation, viz., the influence of immunity in tuberculosis. Now years ago, consumption or tuberculosis was regarded as a family inheritance and the majority of life insurance companies turned down applicants without discussion. I remember well in my early days in insurance work elaborate classifications were made of a father and one brother, and a mother or a sister or both, and more than one death within a certain period led to the declination of the risk. Experience led us to believe that was hardly a fair way to treat them, and the discovery of Koch, the discovery that it was an infection, led us to change our views. Then, too, the study of mortality from tuberculosis among insured lives showed a decrease with which we are all familiar, viz., the last twenty years now there has been a steady decrease in

mortality from tuberculosis. What is the reason? Study Dr. Rowley's charts again. Five years after the appearance of pleurisy shows a mortality of thirty per cent above the normal average. Five years after blood spitting shows a high average mortality. But if you compare those mortality readings, as we have done in the Aetna, with other groups, the figures show a material modification. That is to say, cases taken between twenty-five and thirty-five show a mortality as Dr. Rowley has indicated. Cases between forty-five and fifty-five showing their previous history of pleurisy at five or ten years intervals show a very much better mortality. In other words, the history of pleurisy at twenty-five and forty-five years of age shows a marked difference.

Then, too, I began to notice that in many of our deaths claims of risks insured in the late sixty's, seventy's and eighty's, we found many instances. I collected a large number of extremely bad family histories from tuberculosis. Why they were taken, I don't know, for I was fully aware of my predecessor, Dr. Russell's opinions. But it was perfectly obvious that many men taken with a history of tuberculosis after forty or forty-five years of age turned out to be pretty good risks. Summing up all these considerations, it seemed to me fairly obvious that the older an individual grew the less liable he was to tuberculosis, and the older he grew the more you could count on his immunity. And acting on that principle, for the last ten years we have been very much more lenient in taking those cases than we were prior to that. I have gone a step further in that I regret that I haven't figures with me that I could show you, but I haven't had time to prepare the chart-but the last five years we have been accepting cases forty-five years of age and over who showed a distinct history of pulmonary tuberculosis at least ten years prior. We have accepted cases as standard lives that ten years prior had blood spitting, fever, and a diagnosis of tuberculosis was made, and some of them had been to a sánitarium, but had lived ten years since with normal health; and so far our mortality has no increase over the average of that class. In other words, I think it is a fair inference that mankind civilized and living under normal average conditions is exposed to tuberculosis from the cradle to the grave; that as he lives and as he survives, and because he survives, he develops a gradually increasing immunity to the disease and by the time he is forty or forty-five years of age from a life insurance statistical point of view life underwriters are warranted in assuming that, other things being equal if his general habits and weight are normal and if his digestion is up to the normal average, that they are justified in assuming that he, at that age, possesses an immunity that will protect him as a class against tuberculosis as a special factor in mortality.

DR. BLUMER (New Haven): I was particularly interested in the chart regarding syphilis and I would like to ask Dr. Rowley to tell us some