TABLE 52-Estimated annual number of disabling illnesses of 8 days or more in 1943-44 for Negro men with railroad earnings of $150 or more in 1942

1 Excludes those of unknown age; distribution based on a 2-percent sample of workers with railroad earnings.

1 The number of workers in each age group was estimated by applying the percentages in column (a) to the known total in column (b).

The number with earnings of $150 or more in each age group was estimated by obtaining the ratio of the known total in column (d) to the total in the sample appearing in column (c) and applying that ratio to the number in the sample for the respective age groups.

TABLE 53.-Estimated morbidity rates for Negro railroad men, by age group

Annual number of compensable days of illness per man

Estimat- Compen- Average days in sample 2 Estimat

for white men !

ed rates sable for Negro days per

ed com. pensable days per Negro

man

1 Frequency of Sickness and Nonindustrial Accidents Causing Disability Lasting Eight Calendar Days or Longer Among 60,000 White Male Railroad Employees, 1930-34, Inclusive, by William M. Gafafer (reprint No. 1921), p. 8, Public Health Reports, Apr. 15, 1938.

2 Frequency and severity rates for illness of 8 days or more from a tabulation sent with letter from U. S. Public Health Service written by Hugh P. Brinton, Jan. 1, 1943.

The railroad study gave data only for illnesses lasting 8 days or longer. Sickness benefits cannot be paid until there are 8 days or more of illness in a 2-week period. But once there is such an initial period of illness, benefits will be paid for days of sickness in excess of 4 days in 2 weeks. We have therefore brought together an estimate with respect to illnesses of 5 days or more (many of of the 5-to-7-day illnesses, of course, not being compensable) and, as a matter of general information, an estimate of all illnesses lasting one day or more.

The total number of illnesses of one or more days and illnesses of five or more days for men were estimated on the basis of experience over a period of years in a large eastern public utility reported by the United States Public Health Service, Division of Industrial Hygiene, Absenteeism, by Dr. William M. Gafafer, table 2. These data are reported in table 54. The experience here indicates that the number of illnesses of one or more days is 6.35 times as large as the number of illnesses of 8 days or more. The number of illnesses of 5 days or more is more

than 2.23 times as large as the number of illnesses of 8 days or more. Multiplying the number of illnesses of 8 days or more for men by 2.23 gives a total of 565,000 of 5 days or more. Similarly a total of 1,610,000 for all illnesses for men is reached by multiplying 253,000 by 6.35.

Illnesses of women.-The railroad study which formed the basis for the estimate as to the number of illnesses of 8 days or more among male railroad workers contains no data on women. Women have, at least until recent years, been represented only to a minor degree in the railroad industry. Even among qualified workers in the benefit year 1943-44, women constituted slightly less than 4 percent.

The number of cases of illnesses of 8 days or more was estimated on the basis of the number of cases per 1,000 women in relation to the number of cases per 1,000 men for cases of 8 days or more experienced by employees of the public utility (table 54). The number of cases per 1,000 women workers was found to be 1.55 times as high as for each 1,000 men workers (219.1 as compared with 141.2). This differential was applied to the number of illnesses of like duration per 1,000 men multiplied by the number of thousands of women qualified for benefits. It was assumed that the qualified women were white (only about 4 percent of the women were Negro, one-fifth or one-sixth of 1 percent of all qualified workers). The rate of illnesses of 8 days or more for white males (table 51) was 122.1. The rate for women would be, therefore, 55 percent more, or 189.3.

TABLE 54.-Number of nonindustrial disabilities and of days of disability per 1,000 per year, by sex

Based on study on Absenteeism, by William M. Gafafer, from Manual of Industrial Hygiene and Medical Services in War Industries, table 2.

2 The maximum number of days of sickness in compensable registration periods.

On the basis of 79,000 qualified women in the benefit year 1943-44, the sickness rate indicates 15,000 cases of illness lasting 8 days or more. The number of cases of one or more days as shown on table 54 is about 8.4 times those of 8 days or more. The number of 1-day or more cases amony all qualified women in 1943-44 would therefore be 126,000. The number of 5 day or more cases is 2.2 times the number of 8-day or more cases. The number of such for the same group is estimated at 33,000.

We estimated the 8-day or more cases for men at 253,000 and for women at 15,000. We need to add about 0.7 percent for qualified employees whose sex and age were not reported. The number of cases of illness lasting 8 days or more therefore works out to be 270,000.

Number of days of illness.—We are interested in knowing the number of cases of illness to get an idea of how many people are affected, and particularly how many persons will have an illuess long enough to become eligible for benefits under the plan for sickness benefits proposed in S. 293. In making an estimate

of the amount of sickness benefits, not only the number of illnesses of 8 days or more, but also the duration of such illnesses (and the number of 5-, 6-, and 7-day illnesses among persons who already have been sick 8 days or more within the year and have recovered) must be calculated.

As in the determination of the number of illnesses, the number of days of illness of 8 days or more for white men was made from the basic data for the 60,000 railroad employees. Table 55 shows the distribution of such illnesses for white males by duration and by age group. From this was derived the average duration per case for each age group. In table 57 this average was applied to the frequency rates per 1,000 for each age group to obtain an average number of days of illness for each such employee, which when multiplied by the total number of employees in each age group gives the total days of illness. Adding the totals for the age groups gives a total of 8,899,000 days in illnesses of 8 days or more for white men of known age for the benefit year 1943-44.

The number of days of illnesses of Negro men was estimated on the basis of the differential rates given in the right-hand part of table 53. These differentials were applied to the respective average number of days of illness per white man to yield an average number of days per Negro. The sum of the rates for Negroes related to the sum of the rates for white men provided a differential, which when applied to the average number of days of illness in sicknesses of 8 days or more for white men resulted in an average number of days of illness per Negro. The total number of Negro men multiplied by the average number of days produced the total number of days of illness-1,734,000 for Negro men of known age in illnesses of eight or more days.