and from 0.4 to 0.9 percent for chest X-ray, serologic test for syphilis, blood and urine sugar, and urine albumin tests. The hemoglobin test led to a positive diagnosis in only one case (fig. 3d).

In evaluating these results, it must be remembered that almost 28 percent of those screened out were not followed up medically. These figures, thus, seriously underestimate the prevalence of morbidity among the group tested.

TABLE 5.-Results of diagnostic followup by test, June 18-Nov. 30, 1951

Percent with newly discovered positive diagnosis in total tested.

2.3

4.6

Experience with specific tests

Further indications of the value of specific tests in this multiphasic battery may be cited from the data to be presented in the final report. The 70 millimeter X-ray, as interpreted in this experience, resulted chiefly in the detection of pulmonary disorders. All cardiac cases detected by X-ray were also screened out by blood pressure or electrocardiogram, and many cardiac cases were not indicated at all by the X-ray.

Diastolic blood-pressure readings proved twice as indicative of confirmed disease as did the systolic. Blood-pressure measurement led to the diagnosis of 228 cases of confirmed cardiovascular disease in which both the electrocardiogram and X-ray were negative.

The value of electrocardiograms in this particular age and occupational group was demonstrated. This procedure led to the diagnosis of 156 cases of cardiovascular disease in which both blood pressure and X-ray findings were negative. Both blood and urine sugar tests were positive in less than half of the confirmed cases of diabetes, with an approximately equal number of cases screened out by either test alone.

Specific diagnostic results

Of the 3,994 men tested, 2,521 had 1 or more positive screening tests. Table 6 shows the total diagnostic findings among 1,413 of these men who had received 1 or more positive diagnoses. It reveals that 1,087 new diagnoses were established, 1,107 previously known conditions were noted, and 124 diagnoses were not stated as new or previously known. Particularly significant are the instances of newly discovered disease detected as a result of the specific battery of tests: 6 active or possibly active pulmonary tuberculosis, 1 lung and 2 esophageal cancer, 28 syphilis, 46 diabetes, 138 obesity, 210 refractive errors, 85 impairment of hearing, 339 cardiovascular disease, and 6 nephritis.

In addition, many other important disease conditions not related to the screening tests were first discovered during the examination which followed referral for some positive test. These included 4 malignant neoplasms, 50 nervous-system disorders, 47 respiratory conditions, and numerous other important diseases.

A further byproduct of this multiphasic screening venture was the establishment of medical supervision for many of the conditions which, although previously known, had been neglected.

TABLE 6.-Newly discovered and previously known diagnoses, by selected

Factors influencing the pattern of diagnoses

Obvious factors affecting the pattern of disease revealed by the multiphasic screening program for the group as a whole are the older ages of these men, their occupation, and the specific array of tests upon which medical referrals were based. Within the group, occupational differences in detected morbidity were not striking. As expected, however, disease rates showed a marked correlation with age. Seventy-seven percent of those with one or more positive diagnoses were over 45 years of age, while this age group comprised only 64 percent of the total population under study.

A striking disease factor was that of overweight. Previous knowledge of the extent of obesity in this group of workers had prompted the designers of the project to establish the screening level as 40 percent above the ideal weight. Yet, almost a 10th of the total group exceeded this level, and over two-fifths of the men were 20 percent or more overweight.

The degree of overweight did not increase directly with age. While the proportion of men 20 percent or more overweight was significantly higher among those 45 years of age and over, this age group did not contain a significantly higher proportion of men 40 percent or more overweight. The reason may be that grossly overweight men in the older age groups tend to leave the industry through death or disability.

Table 7 and figure 4a indicate the influence of weight among men positive to

• In all references to significance, chi square test at .01 level was used.

selected screening tests. Most pronounced is the relation of hypertension and positive urine albumin to weight level. The men 40 percent or more overweight also showed significantly higher rates of positive findings for electrocardiogram and blood sugar tests.

Table 8 and figure 4b reflect similar relationships for persons with positive clinical diagnoses. The proportion of men with positive diagnoses (excluding, of course, the diagnosis of obesity) increases with the degree of overweight. Hypertensive disease shows the most striking correlation with weight level. Among those 40 percent or more overweight, the diagnosis of diabetes was significantly more frequent than in the total group. Other cardiovascular-renal disorders were not significantly related to weight.

Further investigation is definitely needed to clarify the relationship of the socio-occupational conditions affecting longshoremen, the high obesity rates among this group, and the prevalence of metabolic and cardiovascular disorders. TABLE 7.-Weight classification of men tested, men with positive results in selected tests, and percentage positive in each weight classification

1

1 Weight classified according to the deviation from the mid value of the ideal weight range for men of medium frame, ages 25 and over. Statistical Bull. Metropolitan Life Insurance Co. 24:6-8 (June), 1943. Unsatisfactory tests were considered negative.

The base for each row of percentages is the number of men tests in that weight classification.