WAR CASUALTIES
By
Albert G. Love,

II. LOSS OF MAN POWER IN WAR.

J. LOSSES IN THE THEATER OF OPERATIONS.

31. Cases disposed of in the Zone of the Interior.--The loss of man power in the Theater of Operations forces of which the medical department has knowledge is due to: (a) noneffectives in hospital, i.e., patients receiving treatment in hospital both in the Theater of Operations and Zone of the Interior; (b) deaths from various causes within or without the hospital; (c) cases disposed of in the Zone of the Interior after being sent there for treatment. This latter group includes the men who are permanently incapacitated physically for further military service. In the following discussion, the cases sent from the Theater of Operations to the Zone of the Interior for further treatment are divided into two groups and are so treated throughout: (a) cases in hospital; (b) cases disposed of after reaching the Zone of the Interior by return to duty, death, or discharge for disability. The cases in hospital are losses in the Theater of Operations strength no matter where the hospitals are located, but cases disposed of in the Zone of the Interior may represent an unnecessary loss in the Theater of Operations strength if too many men who are ultimately returned to duty are sent to the Zone of the Interior.

As stated above, if sufficient care is exercised, the number of cases sent to the Zone of the Interior can be limited practically to potential disability cases, and the additional loss of man power will be small. Thus, if as suggested, 3% of disease and nonbattle injury cases, 6% of the gassed ones, and 20% of the gunshot cases are sent to the Zone of the Interior, the excess over the physically unfit will be only 1% of the cases of diseases and nonbattle injury cases, 2% of the gassed cases, and 5% of the gunshot ones, and even this excess would consist of cases requiring prolonged treatment.

The cases remaining in hospital in the Zone of the Interior of those sent there for treatment are shown by Figs. 41-43, but those disposed of after arrival there by return to duty, discharge for disability, or death are shown by Figs. 47-49.

a. Disease and nonbattle injury patients.---The number of cases so disposed of among those admitted for diseases And nonbattle injuries with a daily admission rate of 1.00 per 1000 Theater of Operations strength is shown by Fig. 47. This graph illustrates that the loss of man power in the Theater of Operations from the disposition of such cases in the Zone of the Interior increases in proportion to the percentage of hospital cases returned there for treatment. Thus the difference in such losses when 3% or 9% are so sent to the Zone of the Interior, accumulated to the end of one year would be 28.95--9.65, or 19.30, per 1000 Theater of Operations strength. To visualize this, let us assume that a force of 2,000,000 men is operating in the Theater of Operations in a well sanitated area in a temperate climate, and that the daily admission rate to hospital is 1.40 per 1000 per day. Then the number of men leaving hospital in the Zone of the Interior during one year if 3% are sent there would be 27,020, and if 9% are sent 81,060, including in each instance only 18,014 disability cases.

If the military authorities decide to send to the Zone of the Interior all Theater of Operations disease and nonbattle injury cases requiring hospital treatment for longer than 60 days, which will be 12.39% of all admissions (Fig. 38), the accumulated losses from this cause at the end of the year with a daily admission rate of 1.00 per 1000, would be 39.85 (9.65 X 12.39 ÷ 3) per 1000 men. In a force of 2,000,000 with a daily admission rate of 1.40 per 1000, the accumulated losses from cases disposed of in the Zone of the Interior at the end of a year would be 111,580 (39.85 X 1.49 X 2000) ; of this number 93,566 (111,580-18,014) are duty cases. Obviously these men may be sent again to the Theater of Operations, time and conditions permitting.

It will be observed from Fig. 47 that losses of this character continue to occur indefinitely as long as the basic conditions remain the same.


Fig. 47.*---Theater of Operations cases of diseases and nonbattle injuries disposed of in the Zone of the Interior by return to duty, death, or disability discharge when sent to the Zone of the Interior for further treatment.

*NOTE:

(a) Total cases disposed of in one year equals total admissions less all patients in hospital at the end of that period.
(b) Cases disposed of in the Zone of the Interior during one year, equals those sent there less the ones in the hospital.
(c) The relationship of (b) to (a) gives the percentage of the total dispositions' which occur in the Zone of the Interior.
(d) It is estimated that this same percentage relationship applies to cases disposed of during any period, as one month, etc.


b. Gas and gunshot patients.---Similar information in regard to cases wounded by poisonous gases and by gunshot missiles are shown by Figs. 48 and 49. The same remarks apply to those two graphs as to Fig. 47.


Fig. 48.---Theater of Operations war gas cases disposed of in the Zone of the Interior by return to duty, death, or disability discharge when sent to the Zone of the Interior for further treatment.

NOTE: For explanation of the method used in calculating the above data see Note on Fig. 47.


c. Total of Theater of Operations patients.---

(1) With a constant rate and a constant strength.---As an illustration of the difference in the number of cases disposed of in the Zone of the Interior in one year when different percentages of cases are sent there, let us make the following assumptions: (a) Two expeditionary forces, each of 2,000,000, both in well sanitated areas in the temperate zone and both engaged in continued and severe military combat: (b) a daily admission rate to hospital of 1.40 per 1000 strength from diseases and nonbattle injuries, .31 from wounds by poisonous gases, and .69 from wounds by gunshot missiles; (c) that one of the expeditionary forces sends to the Zone of Interior 3% of admissions for diseases and nonbattle injuries, 670 of cases wounded by poisonous gases, and 20% of those wounded by gunshot missiles, while the other sends 9%, of the first and second named and 30% of the third.


Fig. 49.---Theater of Operations gunshot cases disposed of in the Zone of the Interior by return to duty, death, or disability discharge when sent to the Zone of the Interior for further treatment.

NOTE: For explanation of the method used In calculating the above data see note on Fig. 47.

Table 6.---Number of cases disposed of during one year in the Zone of Interior, from the two expeditionary forces.


The difference (210,116-113,058 = 97,058) is the excess over the number of men physically disabled or requiring prolonged treatment, and consists of men who will be able eventually to return to duty. They must either be sent again to the Theater of Operations from the Zone of the Interior or be replaced, and even if the former is done unnecessary time is lost.

(2) With a constant rate but an increasing strength.----Fig. 50 shows a method of estimating the Theater of Operations patients to be disposed of in the Zone of the Interior when the admission rates from the various causes, as specified, are constant but when there is an increasing strength. The method of computation is the same as that outlined (see p. 36) for Fig. 25, the principal difference being that in Fig. 50 there are three causes of admissions considered, whereas in Fig. 25 there is only one.

d. Number of Theater of Operations patients to be handled in the Zone of the Interior. The bottom line of Fig. 50 shows the total Theater of Operations patients to be handled in the Zone of the Interior under the conditions outlined. Such data may be of value when estimating requirements for hospital ships and hospital trains.

32. Deaths.---In addition to the noneffectives in hospital and the cases disposed of in the Zone of the Interior from among those sent there, losses occur as the result of deaths.

A larger part of the fatalities which result from disease and nonbattle injuries is among the cases admitted to hospital, but some occur among men not in hospital. Both classes of deaths are counted here as hospital admissions and are included in the total of such cases.


Fig. 50.---A. Method of computing the Theater of Operations hospital patients to be disposed of in the Zone of the Interior when certain percentages of patients are sent there from a command in the Theater of Operations where there is an increasing strength and where there are admissions from both battle and nonbattle causes at a constant rate.

B. The total Theater of Operations patients to be hospitalized in the Zone of the Interior. (See Fig. 46).

C. The grand total Theater of Operations patients to be handled in the Zone of the Interior.


On the other hand, the larger part of the deaths which result from military combat occurs on the field of action. The last named is not included in the number of admissions for wounds, but in this study they are proportioned to the number of men wounded so that the expectancy of such losses can be computed from the admission rate for the wounded.

a. Fatality rates.---The fatality rate for hospital cases in the American Expeditionary Forces in 1918 from diseases and nonbattle injuries, including deaths from such causes not in hospital, was 3.70% (2); while that from poisonous gases, including only such as occurred in hospital, was 1.73% (2), and similarly the one from gunshot cases was 8.12%(2). The deaths from diseases and nonbattle injuries included, however, a large number from the unusually widespread and fatal epidemic of influenza with accompanying pneumonias. After excluding the unusual number of deaths from these causes, it is estimated that 1.44% of hospital disease and nonbattle injury cases in the Theater of Operations may be expected to die. This fatality rate is one and two-thirds times as great as the usual peace time one among troops in the United States. In the following estimation of losses among troops in the Theater of Operations, the fatality rates used are 1.44%, 1.73%, and 8.12% for hospital cases of disease and nonbattle injuries, gas, and gunshot wounds respectively.

b. Duration of treatment of fatal cases.---Men who ultimately die as the result of gunshot wounds do so after a shorter interval of time than do those from either of the other two causes. Thus during the first 15 days of treatment 85.36% of the deaths from gunshot wounds occurred, as compared with 76.37% of those from gas wounds, and 64.56% of those from diseases and nonbattle injuries.


Table 7.---Percentage of the total deaths in hospital from the three classes of causes in the American Expeditionary Forces, which occurred in any group at different time intervals.

Note: For basic formulae see Figs. 96, 97 and 98.


c. Killed in action.----In addition to 13,691 deaths from battle wounds in hospital in the American Expeditionary Forces, 36,694 men were killed on the field of battle. [Note: The A. G. O. figures are: Killed 37,541; died of wounds 12,934; total 50,475.(11) Used here: Killed 36,694; died of wounds 13,691; total, 50,385.]

There is no information available upon which to base an estimate of how many of those who were killed outright died from poisonous war gases and how many from gunshot missiles. Hearsay evidence indicates. that comparatively few men died as the result of poisonous gases before they reached hospital.

d. Killed in relation to number wounded.---An examination of the casualties of infantry regiments by combat days in the American Expeditionary Forces shows that the proportion of the killed to the number gassed decreased as the number of reported gassed cases increased; on the other hand, the proportion of the killed to the number of men wounded by gunshot missiles remained practically constant. Thus the proportion of the killed to the number wounded by gas was almost twice as great when there were 10 gassed as when there were 100, while the proportion of killed to wounded by gunshot missiles was the same when there were 10 wounded as when there were 100.[Note: For basic formula see Fig. 100.]

We have assumed here simply for ease in calculation that 1000 men were killed in action by poisonous gases, and the remainder (35,694), by gunshot missiles. Under this assumption the percentage of the killed to the wounded were: Gas 1.42% ; gunshot 23.25% ; or as expressed in the usual way, it is assumed there was 1 killed in action by poisonous gases to 70.4 so wounded, and 1 killed by gunshot missiles to 4.3 wounded.

e. Total deaths.---These relationships of killed in action to the wounded, and fatality rates for cases in hospital to the number of cases of diseases and nonbattle injuries, of gas wounds, and of gunshot wounds were used in computing the data shown for deaths by Figs. 51, 52, and 53. These Figures show graphically the accumulated number of deaths as time advances computed on the basis of 1.00 admission per day to hospital per 1000 strength in the Theater of Operations.

33. Total losses in the Theater of Operations.---

a. Disease and nonbattle injury patients.---The total accumulated losses from month to month in the Theater of Operations as the result of diseases and nonbattle injuries from (a) noneffectives in hospital (patients in hospital), (b) deaths, and (c) cases disposed of after being sent to the Zone of the Interior (including disability cases), are shown graphically by Fig. 51. As stated in the legend of the graph, all of the data are based upon 1.00 admission per day per 1000 Theater of Operations strength.

The noneffectives, or number of patients in hospital, stabilize after one year and consequently there is no additional loss from this factor thereafter; but the accumulated number of deaths and of cases disposed of in the Zone of the Interior, from cases sent there, continues to increase as long as the basic conditions remain the same.


Fig. 51.*---Loss of manpower in the Theater of Operations from diseases and nonbattle injuries due to deaths, noneffectives in hospital (patients), and cases disposed of in the Zone of the Interior when sent there for further treatment.

*NOTE: For basic formula see:

(a) Deaths, Fig. 96, (2).
(b) Patients in hospital, Fig. 84, (1).
Total losses equal sum of the two above (a) and (b), Plus cases disposed of in the Zone of the Interior as per Fig. 47.


The losses from noneffectives in hospital and from deaths are shown separately and combined. To the combination of the two there must be added losses from the cases disposed of in the Zone of the Interior. As has been shown on page 69 and by Fig. 47, this latter factor in the total loss rate increases in proportion to the percentage of cases sent to the Zone of the Interior. The following table shows the movement of the cases in and out of the hospital, with the summation of the losses up to the end of 360 days, when 3%, 6%, or 9% of all admissions are sent to the Zone of the Interior.


Table 8.---Disposition of disease and nonbattle injury patients admitted to hospital during the first 360 days of operations in the Theater of Operations, when the daily admission rate from such admissions is 1.00 per 1000 Theater of Operations strength.

To convert these data into terms of any admission rate with any strength, multiply any item shown above by the product of the admission rate by the strength in thousands. Thus if the expected daily admission rate to hospital is 1.40 per 1000 and the strength 2,000,000 multiply the items in question by 1.40 x (2,000,000 1000), or 2,800.

As shown on page 55 if the military authorities consider it advisable to send to the Zone of the Interior all disease and nonbattle injury Theater of Operations hospital cases averaging more than a certain time in hospital, the percentage of cases to be so transferred can be found by referring to Fig. 38. Then by referring to Fig. 47, and multiplying the data shown there on the 3% line by the number of times that 3 is contained in the predetermined percentage of cases to be returned to the Zone of the Interior, the cases disposed of in the Zone of the Interior under such conditions can be obtained (see Table 3). The results so obtained, when added to the accumulation of patients in hospital and of deaths at any time, will show the total losses.

b. Gas patients.---Fig. 52 shows in a similar way the losses which occur as the result of poisonous gases, both from the various reasons and also the total. As stated on p. 76, there are no data available upon which to base an estimate of the number of cases killed in action by poisonous gases. It is assumed for ease of calculation that 1000 of the men killed in action died from gas. The detail in regard to the deaths is shown in the table on the graph, and if it is desired to increase or decrease the 1000, the items can be changed correspondingly and the total losses so altered.


Fig. 52.*---Loss of manpower in the Theater of Operations from war (poisonous) gases due to deaths, noneffectives in hospital (patients), and cases disposed of In the Zone of the Interior when sent there for further treatment.

*NOTE: For basic formulae see:

(a) Deaths in hospital, Fig. 97, (2).
(b) Killed in action, Fig. 97, (4).
(c) Patients in hospital, Fig. 87, (a), (1).
Total losses equal sum of the above (a), (b), and (c), plus cases disposed of the Z. of I., as per Fig. 48.


The following table shows the movement of the gassed cases in and out of hospital with the summation of the losses up to the end of 360 days, when 3%, 6%, or 9% of all gas admissions are sent to the Zone of the Interior.


Table 9.---Disposition of gas casualties during the first 360 days of battle, when the daily admission rate from poisonous gases is 1.00 per 1000 Theater of Operations strength.

To convert any of the items above into terms of any admission rate with any strength, multiply the ones in question by the product of the admission rate by the strength in thousands. Thus if the expected daily admission rate is .31 per 1000 and the strength 2,000,000, multiply by .31 X (2,000,000 ÷ 1000), or 620.

c. Gunshot patients.---Fig. 53 shows in a similar way the losses from gunshot casualties in detail and in totals.


Fig. 53.*---Loss of manpower in the Theater of Operations from gunshot missiles due to deaths, noneffectives in hospital (patients), and cases disposed of in the Zone of the Interior when sent there for further treatment.

*NOTE: For basic formulae see:

(a) Deaths in hospital, Fig. 98, (2).
(b) Killed in action, Fig. 98, (4),.
(c) Patients in hospital, Fig. 88, (a), (1).
Total losses equal sum of the three above (a), (b) and (c) plus cases disposed of in the Z. of I., as per Fig. 49.


The following table shows the movement of such cases in and out of hospital, with the summation of the losses up to the end of 360 days, when 10%, 20%, or 30% of such cases are sent to the Zone of the Interior.


Table 10.---Disposition of gunshot casualties during the first 360 days of battle. when the daily admission rate from gunshot missiles is 1.00 per 1000 Theater of Operations strength.

d. Total patients.---The following table shows the movement of the total cases in and out of hospital, with the summation of the losses up to the end of 360 days.


Table 11.---After 360 days of battle: (1) Total cases to be accounted for, including the killed in action; (2) cases returned to duty in the Theater of Operations; (3) total losses.

34. Total losses by separate months.---

a. Disease and nonbattle injury patients.---Fig. 51 shows the summation of losses from month to month from disease and nonbattle injuries, when the daily admission rate is 1.00 per 1000 Theater of Operations strength. Each month's losses are added to the accumulated losses of the preceding months, and the curve is consequently a constantly ascending one, although there is a progressive decline in the loss rate for each separate month, with a descending curve (Fig. 54) until it stabilizes at the end of the 12th month. The rapid fall of this latter curve (Fig. 54) during the first few months is caused by the rapid increase in the number of patients returning to duty from the hospital (Fig. 55). After the first six months there is but little change in the loss rate from cases disposed of in the Zone of the Interior, among those sent there from the Theater of Operations, and also from the deaths; but the total for each month continues to decline until the end of one year, or until the outflow of patients from hospital equals the inflow. After the end of the year, the loss rate from the deaths in the Theater of Operations and the disposition of cases in the Zone of the Interior (including disability cases) will continue as long as the basic conditions do.


Fig. 54.---Loss of manpower from diseases and nonbattle injuries in the Theater of Operations during successive months.

NOTE: Calculated from data from the three upper curves, Fig. 51.

Fig. 55.---Cases of diseases and nonbattle injuries returned to duty in the Theater of Operations during successive months.

NOTE: Number returned to duty by months equals number admitted during that month less the month's losses as per Fig. 54.


The decline in the loss rates during the first 12 months is therefore due chiefly to the increasing number of patients returning to duty from hospitals. Consequently as stated in the last sentence of the legend of Fig. 54 these data apply only to each individual group of men. This group may be a field army, corps, division, or a regiment, or lesser body, but the essential point is that each one must establish its own increasing flow of men out of hospital to replace those entering hospital before its loss rate can decline.


Fig. 56.---Loss of manpower from war (poisonous) gases in the Theater of Operations, during successive months.

NOTE: Calculated from data from the three upper curves, Fig. 52.

Fig. 57.--Gas cases returned to duty in the Theater of Operations during successive months.

NOTE: Number returned to duty be months equals Dumber admitted during that mouth plus the number killed in action, less the month's losses as per Fig. 56.


The 3%, 6%, and 9% curves again show the importance as a loss factor of the disposition of cases in the Zone of the Interior from among those sent there.


Fig. 58.---Loss of manpower from gunshot injuries in the Theater of Operations during successive months.

NOTE: Calculated from data from the three upper curves, Fig. 53.

Fig. 59.---Gunshot cases returned to duty in the Theater of Operations during successive months.

NOTE: Number returned to duty by months equals number admitted during that month plus the number killed in action, less the month's losses as per Fig. 58.


b. Gas patients.---Fig. 56 shows the same data for casualties from poisonous gas cases as Fig. 54 does for those from diseases and nonbattle injuries. Losses from gas casualties also stabilize at the end of the 12th month, as does the return of patients to duty from hospital (Fig. 57). The loss rate from gas casualties after the 12th month is a little greater than the one from diseases and nonbattle injuries, due to the assumed number killed in action by gas, and to the assumed larger fatality rate for gas cases.

c. Gunshot patients.---The very great and continued importance of gunshot casualties as a loss factor is again shown by Fig. 58. As a result of the long duration of treatment of cases of this character, the loss rate continues to decline even after the 16th month (480th day), and until the 920th day. During the last several months shown on the graph, the cases disposed of in the Zone of the Interior, the deaths in hospital, and the assumed number killed in action account for a stabilized loss rate of 18.37, 15.39, and 12.41, according to the percentage of admissions sent to the Zone of the Interior, and the gradual increase of the noneffectives in hospital for the excess, as shown by Fig. 58, above them. With the larger less rates, fewer men are returned to duty from hospital (Fig. 59).

d. With a constant strength and a constant admission rate.---Here as elsewhere the possibilities of sanitation and the climate of the Theater of Operations area, the amount of seasoning of the troops, and military resistance of the enemy must be estimated before the total loss rate can be determined.

Table 12 again shows the influence of the percentage of cases which are sent to the Zone of the Interior upon the loss rate.


Table 12.---Losses per 1000 men in any group of them by months of service in the Theater of Operations, when the daily hospital admission rate per 1000 total strength is: (a) From diseases and nonbattle injuries, 1.40; (b) from gases .31; (c) from gunshot wounds, .69, or approximately the same rates as occurred in the American Expeditionary Forces, from July 1 to November 11, 1918 (excluding the influenza epidemic).

Note: Basic data from Fig. 54, 56 and 58 multiplied by 1.40, .31, and .69 respectively.


Under both sets of conditions, the loss rate falls rapidly after the first month, in the first declining from 56.82 per 1000 men during the first month to 13.75 in the 12th, and in the second from 58.34 in the first to 18.50 in the 12th month.

e. With an increasing strength but constant admission rate.---

(1) From one cause.-The problem of determining the probable loss rate for an expeditionary force is seldom so simple as the one above, for it will rarely happen that an entire force will be sent into an expeditionary area at one time. Then again, while in the interest of simplicity an average daily admission rate from diseases and nonbattle injuries may be used throughout, the military resistance encountered and the amount of military combat engaged in will probably vary so greatly that it will be necessary to use more than one rate.

Let us consider first the simpler of the two situations; that is, where we have a constantly, although not uniformly, increasing force with an average daily admission rate from diseases and nonbattle injuries. Fig. 60 shows the method of computing the data for such a problem. The strength of the military force with monthly increases is approximately the same as existed in the American Expeditionary Forces to the middle of each month; and the daily admission rate, the average one for the year (including the influenza epidemic).


Fig. 60.---Losses from diseases and nonbattle injuries by months under approximately the same conditions as occurred in the American Expeditionary Forces where there was a material increase in the strength each month, and where the constant average daily admission rate to hospital only from these causes was 1.65 per 1000 American Expeditionary Forces strength. It is here assumed that 3% of these admissions were sent to the Zone of the Interior. The symbols used throughout the body of the table represents the product of a monthly loss at a daily admission rate of 1.00 per 1000 strength multiplied by a monthly increase in strength in thousands. Thus 1 B (17.44 X 38.) =663.

NOTE: Data for losses each month (second line from top), see Fig. 54.


As we have found the greatest loss rate for any one force is during the first month because the patients, or noneffectives, in hospital are rapidly increasing and comparatively few men are returning to duty. As each group enters the Theater of Operations and establishes its own flow of patients out of hospital, its loss rate declines. The strength of each monthly increment in the Theater of Operations (Fig. 60) is multiplied by the initial month loss rate at the basic daily admission rate of 1.00 per 1000 strength, and during the following month by the loss rate for the second month, etc. The items as computed for the men who were in the Theater of Operations on the middle of January, and for each monthly increment during the next 10 months, are summated from side to side, and from above downward. The several sums so obtained are then multiplied by the daily admission rate of 1.65 per 1000; i.e., the one that occurred in the American Expeditionary Forces.


Fig. 61.---Method of computing the losses to be expected in the Theater of Operations in a command in which there is an increasing strength; where there are admissions from both battle and nonbattle causes; and also when certain percentages of the patients are sent to the Zone of the Interior.

The results at the bottom of the graph (Fig. 60) shows the losses which occurred from diseases and nonbattle injuries during each month in the strength of the total American Expeditionary Forces, and those on the extreme right hand margin, the total losses in each month's group of men from the time it entered the Theater of Operations until November 11, 1918.

(2) From three causes.---Fig. 61 shows a method of computing the losses in the Theater of Operations from the three causes with rates averaged for the entire American Expeditionary Forces during the year 1918. In Fig. 60, the losses are shown for each individual month, whereas in Fig. 61 the accumulated losses are shown, the February figure including those for January, etc.

f. With an increasing strength and also an increasing admission rate.---.Figs. 60 and 61 illustrate the method of calculating losses in any expeditionary force when there is an increase each month in the strength, but with a constant average daily admission rate. Figs. 62 and 63 show, however, how to compute the data when there is not only an increasing strength, but also an increasing admission rate. The same basic principle applies to the latter factor as to the former. Thus a group of men with an average daily admission rate establishes a return flow of men from the hospital at that rate, but when the rate increases materially and consequently more men are admitted to hospital, the return flow from that increased number must be established before there can be a decline in the specific loss rate.


Fig. 62.---Losses front war (poisonous) gases by months under approximately the same conditions as occurred in the American Expeditionary Forces where there was a material increase in: (a) The strength each month; (b) also in the admission rate at different periods. The average daily admission rates from war gases per 1000 American Expeditionary Forces strength were: Jan. to June 0.10; July to Sept. 0.22; Oct. to Nov. 0.41. It is here assumed that 6% of the admissions from gas were sent to the Zone of the Interior. The symbols used throughout the body of this table represent the product of a monthly loss by a monthly increase in strength in thousands and by a daily admission rate. Thus 1BS (23.53 x 38 x .10) = 89.

NOTE: Data for losses each month (second line from top), see Fig. 56.


Fig. 20 shows the daily admission rates per 1000 total American Expeditionary Forces strength from gas and from gunshot missiles during each month, and also during several grouped periods. From this graph, it is apparent that during the early weeks in 1918 the American Expeditionary Force troops engaged in but little military combat, but as time advanced the amount and intensity of the fighting increased and finally culminated in the prolonged and severe fighting during September, October, and early November.


Fig. 63.---Losses from gunshot injuries by months under approximately the same conditions as occurred in the American Expeditionary Forces where there was a material increase in: (a) The strength each month; (b) also in the admission rate at different periods. The average daily admission rates per 1000 American Expeditionary Forces strength were: Jan. to June 0.11; July to Sept. 0.53; Oct. to Nov. 0.96. It is here assumed that 20% of the gunshot admissions were sent to the Zone of the Interior. The symbols used throughout the body of this table represent the product of a monthly loss by an increase in strength in thousands and by a daily admission rate. Thus 1BS (36.39 x 38 x .11) =152.

NOTE: Data for losses each month (second line from top), see Fig. 58.


For convenience in calculating the data on Figs. 62 and 63, the time from January 1 to November 11, 1918 is divided into three periods, although more exact results could be obtained by dividing the entire time more nearly in accordance with the intensity of fighting.

As an illustration of the method of determining losses from gunshot missiles (Fig. 63), the January strength is multiplied by each loss rate from No. 1 to No. 11 at the basic daily admission rate of 1.00 per 1000; and also for each month of its service in the American Expeditionary Forces, (11 months) by the admission rate of .11. The increase in the admission rate of .42 (.53-.11) in July is then multiplied by the January strength and the initial five loss rates. These additional losses are entered in the table on the line with the January strength under the months from July through November. The further increase in the admission rate of .43 [.96-(.42 + .11)] in October is then multiplied by the January strength and the two initial loss rates. These additional losses as computed are entered on the January line under October and November as they occurred.

In computing the losses for the increase in strength during February, that increment is multiplied by the loss rates from No. 1 to No. 10 inclusive, then throughout by .11, by .42 for the last five groups, and by .43 for the last two. For the March increase in strength, the loss rates from No. 1 to No. 9 are used, for the April one, No. 1 to No. 8, etc. For the July strength, the only loss rates used are from No. 1 to No. 5 inclusive, and consequently the second multiplication is by .53 (.11 + .42). Similarly the October increase is first multiplied by the first two loss rates and then by .96 (.53 + .43).

The following table shows a summary of the monthly losses as computed on Figs. 60, 62, and 63.


Table 13. Total losses in the American Expeditionary Forces by months.

As nearly as can be estimated, the total losses in the American Expeditionary Forces to November 11, including deaths, cases in hospital and cases returned to the United States, was about 267,000, or about 9,000 more than the total 258,155 as shown above. The results then of these calculations approximate the actual losses as closely as could be expected from a computation in which grouped average rates and strengths are used.

To repeat, the rapid decline in loss rate as shown by Figs. 54, 56, and 59 occurs only when an expeditionary force has a constant strength throughout with constant admission rates from all causes. The loss rates per 1000 total Theater of Operations strength in the last column of Table 13 show how much such rates depend upon the increases in strength and upon the intensity of fighting.

K. LOSSES IN THE MOBILIZATION AREA.

The losses in the mobilization area consist of:

a. Noneffectives in hospital.
b. Deaths and discharges for disability among patients in hospital.
c. Deaths and discharges for disability among men not in hospital.

After the mobilization camps are filled, there is but little increase in the strength of the command. As the troops are trained and equipped they are sent to the expeditionary area and are replaced by incoming recruits. Consequently the number of men mobilized is greater than the average daily strength for one year. Thus in 1918, the former was 3,310,246, and the latter 1,310,246.


Fig. 64.*---Loss of manpower in the mobilization area (U.S.) among those actually enrolled in the military service from diseases and nonbattle injuries, due to deaths, discharges for disability, and noneffectives in hospital (patients).

*NOTE: For basic formula see: (a) Patients in hospital, Fig. 82, (2). (b) Deaths and discharges in hospital, Fig. 95 (2). (c) Deaths and discharges not on sick report based on experience in U. S., 1918.


The troops while in training establish a return flow of patients from hospital, and when they depart their sick in hospital remain behind, thus acting as a replacement flow for the incoming recruits. The ultimate number of noneffectives will then be divided among the total number of men passing through the mobilization area rather than among the average daily number present for the year. The men in camp during the first few weeks, when the noneffective rate rises quite rapidly, lose a larger percentage than those during the later periods when it is more nearly stabilized; but for all practical purposes we may disregard this and consider that the noneffectives are distributed evenly among the men mobilized. Then as in Fig. 64, with an ultimate noneffective rate of 20.36 to the average annual strength, the same rate to each 1000 men mobilized was

1,310,246

3,310,246

20.36

X

-----------

÷

-------------

=

8.06.

1000

1000

The time element is an important one in connection with deaths and discharges among patients in hospital, for the number of cases with their associated deaths and discharges for disability increase in proportion to the length of the time period, like the noneffective rate. Consequently this class of losses must be proportioned to the total number of men mobilized rather than to the daily strength averaged for one year. With a loss rate from these causes at the end of one year of 14.67 per 1000 in an average daily strength of 1,310,246, the corresponding one for 3,310,246 men mobilized is

1,310,246

3,310,246

14.67

X

-----------

X

-------------

=

5.81.

1000

1000

The third class of losses are from deaths and from discharges for disability (of which the latter comprised 95% of the two during 1918), among men not on sick report. This group consisted chiefly of the physically unfit men whose deficiencies escaped the scrutiny of the local and camp examining boards. The physical imperfections of such men soon became apparent to officers of organization and to the medical officers assigned to them, and they were discharged as physically unfit. Probably all such cases can be culled out within a month or six weeks, and there should be no material increase in the percentage of them thereafter, no matter how long the men remain in the mobilization area. Since the time element need not be considered in connection with them, as it is assumed that the training period will be long enough for their detection and elimination, the rates for them can be based upon the actual number of men mobilized. In 1918 it was 22.24 per 1000 men (Fig. 64).

Since the total losses in, man power in the mobilization area is 52.05 per 1000 men mobilized (See Fig. 64), or 5.20%, the number of men available for duty with any expedition will be 100%---5.20%---94.80%. For each 1000 men required for a large expeditionary force, when the troops are to be equipped and trained in the mobilization area, the number that will be required, after passing the camp examining board will be,

94.80% = 1000
100% = 1054.85

Since 6.6%,(12) of all men sent to the military camps by the Local Boards were found to be physically unfit, the percentage of the men actually called for service who will eventually be available for duty with an expeditionary force will be:

100%-(5.2% + 6.6%) = 88.2%.

Then for each 1000 men required for an expeditionary force, the number to be called after passing the rather careful scrutiny of the Local Examining Boards will be:

88.25 = 1000
100% = 1133.79

Of this number of men called there will be:

1. Found physically unfit by Camp Examining Boards 74.83
2. Noneffectives, deaths, and discharges as physically unfit in training camps 58.96
3 Available for duty with an expeditionary force 1000.00

Total to be called .

1133.79

Since 14.6% of all Class I men were found physically disqualified for full military service by local boards; and an additional 8.0% as available only for limited or domestic service by the local and camp examining boards, the percentage of Class I men between the ages of 21 and 30 who were eventually available for combat service with the expeditionary force was 100%-(14.6% +8.0% +6.6% +5.2% ) = 65.6%. Then for each 1000 men for combat duty required for an expeditionary force, the number of Class I men, ages 21 to 30, who must be available will be:

65.6% = 1000
100% = 1524.39

Of this number of men called there will be:

1. Found physically unfit by local examining board 222.56
2. Found physically fit for only limited service by local and camp examining boards 121.95
3. Found physically unfit by camp examining board . . . . . 100.61
4. Non-effectives, deaths, and discharges as physically unfit in training camps 79.27
5. Available for combat duty with an expeditionary force. 1000.00

Total Class I men

1524.39

.

III. MEDICAL SERVICE IN THE COMBAT ZONE.

The care of the sick and injured of the units operating in the Theater of Operation and the evacuation to general hospitals are important functions of the Medical Department. The efficient and systematic performance of this duty assists materially in insuring the mobility of the troops in the front area, and it is also a very important factor in maintaining the morale of the troops.

The character of cases to be treated and evacuated are:

a. Diseases and nonbattle injuries.

b. Battle wounds from:

(1) Gunshot missiles.
(2) War gases.

L. DISEASES AND NONBATTLE INJURIES

35. Total in the Theater of Operations.---Our experience shows (Fig. 18) that the daily admission rate from diseases and nonbattle injuries to hospital and quarters combined for seasoned troops operating under favorable conditions in a temperate climate in an area such as the American Expeditionary Forces, with a large number of men in the line of communications area, was 2.50 per 1000. It further shows (Fig. 1) that 56% of such cases were admitted to hospital and 44% were treated in quarters, infirmaries, improvised hospitals, etc. We know also (p. 3) that approximately one half as many cases are treated as dispensary cases as in hospital and quarters combined.

Under such conditions as outlined above the average daily sick rate per 1000 men among such troops would be:

Hospital sick 1.40
Quarters sick .... 1.10
Dispensary sick 1.25

If this rate is increased 20% to allow for normal variations, the daily sick rate per 1000 men in the Theater of Operations is 4.50, of which approximately one third would be sent to station or general hospitals, one third would be cared for in quarters, dispensaries, etc., and one third would be returned to duty with their organization. In the further discussion of this subject no mention will be-made of the care of the dispensary sick, but there will always be a large number of such cases to be treated, and supplies and personnel must be available for this purpose.

36. In the Combat area.---When troops enter the combat zone and then advance to the battle line, the incidental hardships and unfavorable conditions probably would cause an increase in the sick rate.

To obtain data in regard to the actual increase to be expected, a study was made of the medical records of the infantry regiments of the 1st, 3rd, 26th and 42nd Divisions, all active combat divisions, for the period June 1 to October 31, 1918, inclusive.(1)


Table 14.---Daily admission rates from diseases by systems by months, among each 1000 white enlisted men in combat divisions and in the total American Expeditionary Forces, June to October, 1918.

*Admission rates to hospital.
**Estimated total admission rates, including cases in quarters,

NOTE: The Combat Divisions include the Infantry Regiments of the 1st 3rd, 26th and 42nd Divisions. The rates for the Combat Division,, apparently include as hospital cases those ordinarily treated in quarters; whereas those computed for the total American Expeditionary Forces include only hospital cases. The estimated ones which include also quarters cases can be more properly compared with those for the Combat Divisions (Front Area).


Since the Divisions selected were composed of white men, their rates are compared with those for the total white troops in the American Expeditionary Forces.(3) The computed data as presented by Tables 14 and 15, and Figs. 65 and 66, shows that there was more sickness among the front area white troops than in the total group. It is apparent also that the excess among the front line troops occurred each month and from each class of disease, the most important increase being in the number of cases of respiratory and digestive diseases.


Fig. 65.---Daily admission rates by classes of diseases per 1000 strength of white troops in the front area, and also in the total American Expeditionary Forces, for the period from June 1 to October 31, 1918. (1, 3)

Table 15 and Fig. 66 also indicate that there was more sickness among the troops on the battle line than among the other front area troops. The data, as computed, shows that the average daily hospital admission rate from diseases only for the white troops in the front area was 3.76 per 1000, as compared with 1.26 for the total American Expeditionary Forces,(3) or practically 3 to 1. If the admissions from nonbattle injuries are included the rate of 1.26 is raised to 1.40; and if the ratio of 3 to 1 is continued, the 3.76 is raised to 4.20 (hospital only).


Fig. 66.---Daily admission rates from diseases per 1000 strength of white troops in the front area, and in the total American Expeditionary Forces, for the period June 1st to-October 31st, 1918, and also for each month of that period. (1, 3)

Before any real comparison can be made, however, of the rates for the combat divisions with those for the total American Expeditionary Forces, the character of the cases included in the two sets of rates must be considered. As stated above, many of the cases of sickness among the men in the training area were treated in quarters, infirmaries, etc., and were not made of record. Among the combat divisions, however, and especially so when in the actual combat area, a large part of the sick had to be evacuated, and consequently were admitted to either temporary or permanent hospitals where records were made. Apparently then, the comparison should be between the total admission rate in the American Expeditionary Forces; that is, the one including both hospital and quarters cases, and the one available for the combat division. For this purpose, the estimated rates to include both hospital and quarter cases for the total American Expeditionary Forces are shown both in the tables and on the graphs. The comparison is then between 2.50 and 4.20. Consequently,' we would expect the sick rate to be 1.70 (1.68) as great among the personnel of the front line combat divisions as in a total area, such as the American Expeditionary Forces. Furthermore, when the troops are on the actual battle line this excess may be 2.00 to 1.00 instead of 1.70 to 1.00.


Table 15.---Daily Admission Rates from Disease by Systems among each 1000 White Enlisted Men, on the battle line, among other troops in combat divisions, and in the total American Expeditionary Forces, June to October, 1918.

Even with the estimated increase added, the total American Expeditionary Forces rates are still much lower than those for the Combat Divisions (Front Area) for each month, and for each group of diseases with the two exceptions, one for infectious not specified (which is lower), and the other for tuberculosis (which is the same) (see Table 14 and Fig. 65).


Table 16.---Relative number of cases of sickness among the front area troops as compared with the estimated total in the American Expeditionary Forces set as a standard at 1.00.

M. BATTLE CASUALTIES---COLLECTION OF DATA.

The following data show the average and maximal casualty rates, and also how often rates of various magnitude occurred among the United States troops in the American Expeditionary Forces during 1918. The information as assembled is for the American part of the First American Army and its component units, and also for other selected Divisions and regiments. The sources of the information and the method of assembling the data are as follows.

37. Composition of Organizations.---

a. First American Army.---The Report of the First Army (13) was used to determine its component organizations from day to day. This information was supplemented by information obtained from "Field Operations, Volume VIII, The Medical Department of the United States Army in the World War".(14) The Corps of the First Army were I, III, and V, from September 26 to November 11; the IV from September 26 to October 12; and the Divisions serving with the French, for this study, grouped and included as a Corps, from Sept. 26 to Nov. 11.

b. Corps.---The report of the First Army(13) was also used to determine the Divisions in each Corps from day to day, and also those in the Army Reserve. "Field Operations"(14) was again used to supplement the information obtained. In addition, "Battle Participations of Organizations of the American Expeditionary Forces in France, Belgium, and Italy"(10) was used quite freely to verify the location of Divisions, and especially to determine when a Division left the line for reserve or vice versa.


Fig. 67.---Average daily number of American divisions in the First American Army, in each of its Corps, both in line (L) and in reserve (R), and also in the Army Reserve for the period of 47 days, from September 26 to November 11, 1918. (3)

NOTE: The IV Corps was a part of the First Army only until October 12, a period of 11 days. During that time the average daily number of divisions in line was 3.18 and in reserve 1.53, whereas the figures for the entire 47 days was L-1.15 and R-.55 as shown above.


Fig. 67 shows the daily average number of American Divisions in each Corps and also in the First American Army during the Meuse-Argonne offensive.

c. Divisions.---The information in regard to the operations of Divisions, was obtained from the "Battle Participations of Organizations of the American Expeditionary Forces" etc.(10). This information was supplemented by a mimeographed copy of "Brief Histories of Divisions, U. S. Army, 1917---1918"(15), June, 1921, from the Historical Branch, War Plans Division, General Staff.

38. Strength of the Organizations.---

a. First Army.---The strength of the First Army is given on page 113 of the "Report of the First Army"(13). Examination of the casualties in organizations by days, as will be referred to later, showed that only 91% of the total casualties were distributed by divisional organizations by days.[Note: The original medical cards, upon which the casualties were reported, showed the regiments or staff corps to which each patient was attached, but there was no reference to Division. The 9% of undistributed casualties occurred among staff or corps troops, not in regiments.] Apparently then only 91% were included in the total casualties assembled for the First Army. Therefore, the daily strength of the First Army as shown in the official report, was reduced in each instance to 91% to compensate for the difference in the total of casualties as they occurred and as assembled.

b. Corps.---The daily strength of the individual corps was found by multiplying the assumed strength of a division; that is, 24,000, by the number of divisions, including those in reserve, in the Corps, and then adding 33% [Note: The increase of 33% for Corps troops is based upon the strength of a type corps composed of three divisions and the equivalent of the strength of one division in corps troops. Since the daily number of divisions in the various corps was often greater than three while the number of corps troops probably remained more nearly constant, the increase of 33% for the latter probably results in too great a strength for the corps, and consequently in too low a rate. Hence an increase of from 5% to 10% in such rates as are given may be justifiable to cover corps troops not included in divisional organizations. The total figures thus obtained were reduced then in each instance to 91% to compensate for the unassembled casualties, as referred to above.

c. Divisions.---An examination made by the Historical Branch of the War College of the returns of 14 Combat Divisions on or about September 26, 1918 showed that the average strength of each one was 24,128 men. It was assumed then that 24,000 was a fair figure to use in determining the strength of the corps, as referred to above. But for the purpose of calculating rates for Divisions, the 24,000 was reduced to approximately 91%, or to 22,000, to compensate for the unassembled divisional casualties.

d. Regiment.---An examination made by the World War Division of The Adjutant General's Office of reports of the Infantry Regiments in the 1st, 42nd, and 80th Divisions on October 31, 1918 showed that the average strength of each regiment was 2,732. To compensate for the lag in reporting casualties, the round number of 2,500 was used in calculating regimental rates.

39. Sources of Casualty data.---

a. Immediate source of information. The immediate source of the casualty data was the tables in the latter part of "Medical and Casualty Statistics, Part 2, Volume 15, The Medical Department of the United States Army in the World War"(3) In these tables the number of gas wounded, gunshot wounded, and killed in action is given for each day for each Infantry, Artillery, and Engineer Regiment, and each Machine Gun Battalion.

b. Original source of information.---The original source of the information was: (a) For the number wounded by war gasses and by gunshot missiles, the Sick and Wounded Report cards forwarded from the American Expeditionary Forces to the Office of the Surgeon General; (b) for the killed in action, a nominal list of the deaths in the American Expeditionary Forces arranged by organization and by day, prepared in the Office of The Adjutant General and loaned to the Surgeon General's Office in 1921. It will be noted that in some instances during severe engagements, such as when the 27th and 30th Divisions participated in the attack on the Hindenburg line (Somme offensive) on the 29th and 30th of September, 1918, apparently all of the killed on the two days were reported as of the first day of the engagement. This results in a too high casualty rate for the first of the two days and in one too low for the second.

40. Engagements Studied.---

a. First Army and Corps.---The Meuse-Argonne offensive from September 26 to November 11.

b. Divisions.---All of the Divisions comprising the Corps of the First Army, including those with the French. In addition, the following were included: [Note: The Second Division could not be included because the data by days for its Marine organization were not available.]

1. First Division, Aisne-Marne, July 18-23.
2. Third Division, Aisne-Marne, July 18-31.
3. Fourth Division, Aisne-Marne, August 2-7.
4. Twenty-Sixth Division, Aisne-Marne, July 18-25.
5. Twenty-Seventh Division, Somme offensive, Sept. 26-Oct. 20.
6. Thirtieth Division, Somme offensive, Sept. 26-Oct. 20.
7. Thirty-Second Division, Aisne-Marne, July 30-August 6.
8. Forty-Second Division, Aisne-Marne, July 25-August 3.

c. Infantry regiments.---All of the Infantry regiments of the Divisions referred to; and, in addition, the 9th and 23rd Infantry regiments of the Second Division in the Aisne-Marne, July 18 and 19.

41. Method of Assembling Data.---

a. Infantry regiments.---In the study of casualties by regiments, only Infantry regiments were included, because the Artillery and Engineer regiments had relatively few casualties as compared with the Infantry. The tabulation of the casualties for the Infantry regiments included only those assigned to Divisions operating on the front line, and none of those assigned to Divisions in the Corps or Army Reserve. It was, however, impossible to select out the Infantry Regiments when in the reserve of the selected Divisions.

b. Divisions.---The casualties as assembled for Divisions include those for the four Infantry regiments; three Machine Gun Battalions; the Engineer regiment; and the three Artillery regiments, when operating with the Divisions.[Note: No allowance could be made for the absence of Artillery regiments on certain days. Consequently the casualty rates as computed for such days are slightly lower than they should be.] The daily casualties of these organizations were added to find those for the Divisions. As stated above, the Division strength was reduced to 91% of the total to compensate for the casualties which occurred in the organizations of the Divisions but for whom such data could not be assembled. In the study of the Divisions per se, the casualties of a Division was included only for the days when the Division was actually on the front line.

c. Corps.---In assembling the data for the Corps by day, the casualty data for all of the Divisions with the Corps on that date, both on the line and in reserve, were included. These were added to find the total for the Corps. In some instances, when more than one day was necessary to change a Division from the line to reserve or vice versa, the casualty rate for the Division shown as in reserve may be seemingly quite high.

d. Army.---The daily casualties of the Divisions in the Army Reserve were added to the total of those in the Corps to find the aggregate for the Army.

42. Percentage of gas wounded, gunshot wounded, and killed.---The term battle casualties as used here includes wounds by war gases, wounds by gunshot missiles, and the killed in action. In planning the evacuation of the wounded some approximate estimate must be made of the percentage of the gas and of the gunshot wounded in the total casualties.

As shown by Fig. 99, the daily proportion of gas wounded to gunshot wounded decreased as the number of the latter increased; and also as shown by Fig. 100, the daily proportion of gas wounded to the number killed in action decreased as the latter increased, while the proportion of gunshot wounded to the number killed remains practically constant. In other words the relative number of men wounded each day by war gasses decreased as the intensity of the military combat increased.

Fig. 68 shows how the proportion of the gas wounded, gunshot wounded and killed in the total casualties varied according to the resistance. It suggested that the approximate average percentage distribution of casualties in Infantry regiments in engagements with casualties varying from 100 to 400 (rates varying from 47.9 to 16%) be used in estimating the percentage of killed in action, gas wounded, and gunshot wounded in the total casualties. We would then assume that in severe engagements 16% of the casualties would be the "killed in action", 64% gunshot wounded, and 20% gas wounded. There would then be 4 gunshot wounded to 1 killed in action, and 3.2 gunshot wounded to 1 gas wounded.

The percentage of killed to wounded would then be:

(a) 16 To when there are both gas and gunshot wounded;
(b) 20% when there are only gunshot wounded.

Other American Expeditionary Forces experience indicates that in open warfare the killed would be 16-2/3% (1 to 5) when war gases are not used.


Fig. 68.---Variation in the percentage that the gassed, gunshot wounded and killed in action were of the total casualties according to the severity of the engagement. (3)

NOTE: It is suggested that the approximate average percentages be used in estimating the distribution of total casualties. The data on the left hand margin show the actual number of casualties.


N. BATTLE CASUALTIES---PRESENTATION OF DATA.

In presenting the casualty data for the Infantry regiments, Divisions, Corps, and the First American Army, four items are shown for each one: (a) The average daily casualty rate during the Meuse-Argonne offensive; (b) the frequency of various daily casualty rates, that is, how often each one occurred; (c) the maximal daily casualty rates with the proportionate part which occurred in the component unit of the Army, each Corps, and each Division; (d) important casualty rates on days either preceding or following the maximal casualty day.

43. Infantry Regiments.---

a. Frequency of casualty rates.---The summation curve, Fig. 69, shows how often casualty rates greater or less than a certain one, occurred in infantry regiments. Thus, in 97.15% of the infantry regimental combat days in question (both in line and in reserve of the Divisions operating on the line) the loss was 100 per 1000 or less; and further the loss was greater than 100 per 1000 of the regimental strength in 2.85% (100.0%---97.15%) of the days.


Fig. 69.---Summation curve showing how often various regimental daily casualty rates occurred during major operations in the American Expeditionary Forces in 1918. (3)

The Figure also shows how often casualty rates between certain ones occurred. Thus a rate between 90 and 100 per 1000 of the regimental strength occurred in 0.83% (97. 15%---96.32% ) of the battle days in question.

b. Average casualty day.---It is apparent then that high casualty day rates were comparatively infrequent. The average daily casualty rate, which was 20.36 per 1000 for the days in question, was also low. Since the average includes the days spent in the reserve of the Divisions which were on the front line, it is necessarily lower than it would have been for regiments in action only, and should probably be raised to approximately 25.00 per 1000 when comparing regimental losses with those in Divisions and Corps in combat.

c. Maximal casualty days.---An estimate of the medical personnel and equipment required for infantry regiments in severe combat cannot be based with safety on the most frequent daily casualty rates, or even on the average one, but should be based upon rates during severe combat.

In 1.74% (100%-98.26%) of the combat days the casualty rates were greater than 150 per 1000 regimental strength; and in .8670, or 21 days, they were greater than 200 per 1000 of the regimental strength. Table 17 shows the regimental daily combat casualty rates greater than 150 per 1000 of the regimental strength, and it also shows the more important loss rates on five other days, either immediately preceding or following each maximal casualty day; i.e., on six consecutive days.


Table 17.---Maximal infantry regimental casualty day rates per 1.000 men with the five next highest on days either immediately preceding or following each one; i. e., the rates are shown for six consecutive days.

d. Estimated combat requirements.---After considering such daily rates as above it would apparently be expedient to make provision for a 15% casualty day for an infantry regiment in severe combat, with the following distribution of the total casualties:

Killed in action 2.4%, gunshot wounded 9.6%, gas wounded 3%.

44. Casualties by arms of service.---In the American Expeditionary Forces the casualty rate for the infantry was much greater than those for the other branches. The following table shows the relative standing of the casualty rates, infantry being taken as 100.(2)

Infantry 100.00
Machine Gun 70.12
Signal Corps 16.46
Tank Corps 15.85
Artillery 11.58
Engineers 9.15
Medical Department 8.54
Quartermaster Department 3.05
Cavalry 3.05
Ordnance 1.83
Aviation 1.83

45. Infantry Divisions.---

a. Frequency of casualty rates.---The summation curve, Fig. 70, shows how often during this experience casualty rates in Infantry Divisions greater or less than any specified one occurred. Thus in 98.80% of the infantry divisional combat days in question, the loss was 50 per 1000 or less of division strength, and in 1.2% (100.0%--98.8%) of the days it was greater.

A casualty rate between 40 and 50 per 1000 of the divisional strength occurred in 1.28% (98.80%---97.52%) of the battle days.

b. Average casualty day.---The average divisional casualty rate for all of the combat days in line in question was 9.61 per 1000, or slightly less than 1 .

c. Maximal casualty day.---Fig. 71, shows the divisional battle days with casualty rates greater than 60.0 per 1000, of the divisional strength. The graph also shows what part of the divisional casualty rate occurred in each of its component regiments, both in line and in reserve. The regimental rates are based here upon the divisional strength, so that the sum of the several component rates are equal to the divisional one.

The following table shows the more important loss rates on five days either immediately preceding or following each maximal one; i.e., on six consecutive days.


Table 18.---Maximal infantry divisional casualty day rates per 1000 men with the five next highest ones on days immediately preceding or following: i, e., the rates are shown for six consecutive days.

d. Estimated combat requirements.---Since a casualty day of approximately 6% or greater occurred not infrequently in Divisions in severe combat, it would apparently be expedient to make provision for a 6% casualty day for infantry divisions in severe combat with the following distribution: Killed in action, 0.96%, gunshot wounded 3.84% and gas wounded 1.20%.


Fig. 70.---Summation curve showing how often various divisional daily casualty rates occurred during major operations in the American Expeditionary Forces in 1918. 3

Fig. 71.---The daily casualty rates on the five maximal casualty days for divisions (not including the 2nd) in line (not in reserve) in the First American Army during the Meuse-Argonne and selected divisions in the Aisne-Marne and Somme Offensives; also the average daily casualty rate for the divisions during those engagements. (3)

NOTE: The above rates for the component regiments of the divisions are based upon the divisional strength (22,000) and are consequently only 11.36% of those in Table 17 for the infantry regiments which are computed upon the regimental strength (2,500).


46. Army Corps.---

a. Frequency of casualty rates.---The summation curve, Fig. 72, shows how often casualty rates in Army Corps greater or less than any certain one occurred during the Meuse-Argonne offensive. As stated on the graph, casualty rates greater than 20 per 1000 of corps strength, occurred in 1.95% (100%-98.05%) of the battle days; and that rate or less in 98.05% of the combat days. Casualty rates between 15 and 20 per 1000 of the Corps strength occurred in 1.97% (98.05%-96.08% ) of the battle days.

b. Maximal rates in each Corps.----Fig. 73 shows the maximal rate which occurred in each Corps, and also the average for each one during the period of the Meuse-Argonne offensive. The graph also shows what part of the corps rate occurred in each of its component divisions, the latter rates being based upon corps strength. It is apparent that in each instance the greater part of the maximal rate occurred in one division. If two divisions in line had been engaged in equally severe combat, the total corps maximal rate would have been nearer 30 than approximately 20 per 1000 corps strength.

The following table shows the maximal casualty days for the Corps in question, and also the more important loss rates on four days immediately preceding or following each such day; i. e., on five consecutive days.


Table 19.---Maximal daily casualty rates in the Corps of the First American Army during the Meuse-Argonne offensive with the four next highest rates on either immediately preceding or following days; i. e., on five consecutive days.

Fig. 72.---Summation curve showing how often various Corps daily casualty rates occurred in the five Corps of the First American Army, including the divisions with the French as a Corps, during the Meuse-Argonne Operation, Sept. 26 to Nov. 11, 1918. (3)

Fig. 73.---The maximal daily casualty rate for each Corps of the First American Army during the Meuse-Argonne Sept. 26 to Nov. 11, 1918, and also the average rate for each one during the entire period. (3)

NOTE: The above rates for the component divisions of the Corps are based upon the Corps strength on the specified dates and are consequently less than those shown in Table 18 for the divisions, which are computed upon the divisional strength.


c. Estimated combat requirements.---If an estimate is based then upon the American Expeditionary Forces experience modified as suggested above it would apparently be expedient to make provisions for a 3% casualty day for Army Corps when engaged in severe combat, with the following distribution: Killed in action 0.48%, gunshot wounded 1.92%, and gas wounded 0.60%.

47. First American Army.---

a. Frequency of casualty rates.---The summation curve, Fig. 74, shows how often casualty rates, greater or less than a certain one occurred in the First American Army during the Meuse-Argonne offensive. The lowest daily casualty rate recorded was 1.05 per 1000 Army strength, and the highest, 7.75 per 1000.


Fig. 74.---Summation curve showing how often various Army daily casualty rates occurred among American Troops in the First American Army during the Meuse-Argonne Operation, Sept. 26 to Nov. 11, 1918.

b. Daily casualty rates.---The following table shows the Army casualty rates for each day during the operations:


Table 20.---Casualty rates for the First American Army on each day during the Meuse-Argonne Offensive.

c. Maximal casualty rates.---Fig. 75 and Table 21 shows the maximal Army rates, with the part which occurred in each corps, based upon Army and not Corps strength.


Table 21.---Maximal daily casualty rates per 1000 strength in the First American Army during the Meuse-Argonne offensive.

Fig. 75.---The daily casualty rates on the three maximal casualty days for the First American Army in the Meuse-Argonne Sept. 26 to Nov. 11, and also the average daily casualty rate during the same period.

The above rates for the component Corps of the Army are based upon the army strength on the specified dates and are consequently less than those shown in Table 19 for the Corps, which are based upon the Corps strength on the same date.


d. Estimated combat requirements.---On each maximal casualty day, the greater part of the casualties occurred in two Corps out of a total of four or five. If the Corps in line had been equally engaged, the Army maximal rate would have been nearer 15.00 per 1000 than 7.75. If an estimate then is based upon the American Expeditionary Forces experience as modified, it would seem to be expedient to provide for a 1.5% casualty day for a Field Army when engaged in severe combat with the following distribution of the casualties: Killed in action 0.24%, gunshot wounded 0.96%, gas wounded .30%.

48. Summary.---

a. Estimated combat requirements.----The following table gives a summary of the daily casualty rates as suggested above, under discussion of each unit, to be used as a basis for estimating the requirement for medical personnel and equipment on severe combat days.


Table 22.---Casualty rates per 100 (%) of unit strength suggested as a basis for estimating the necessary medical relief on severe combat days, as determined by the American Expeditionary Forces experience.

b. Average casualty rates.----Fig. 76 shows the average casualty rates for infantry regiments, infantry divisions, Army Corps, and the First American Army during the Meuse-Argonne offensive.

The average rates for infantry regiments and infantry divisions referred to under the discussion of those organizations include casualty days in engagements other than in the Meuse-Argonne, and consequently differ from the ones shown here.

Since the average rate for infantry regiments includes the days for the regiments in divisional reserve some estimated increase should be made when their rate is compared with those for other units for which only combat days are included. It is suggested that for such a purpose, the average daily regimental casualty rate be increased from 17.96 to an estimated 20.00 per 1000 regimental strength.

O. BATTLE CASUALTIES---TRANSPORTATION REQUIRED.

It is difficult to secure accurate information in regard to the character of transportation required for battle casualties. During or after a severe combat day there will seldom if ever be enough litter bearers to carry all of the gassed and gunshot wounded to the Collecting Station, even if such was desirable. Consequently a certain number of the less severely gassed or gunshot wounded must walk to advanced relief stations. After the Collecting Station is reached, the more severely wounded must be transported as "Recumbent", while the less severely wounded can be carried as "Sitters".


Fig. 76.---Average daily casualty rates per 1000 strength, for the period from Sept. 26-Nov. 11, 1918, incl., for the total American Expeditionary Forces, First American Army, corps of the First American Army, divisions in line (not in reserve )of the First American Army, and infantry regiments of all divisions in line (including those in divisional reserve).

*This rate (20.00) estimated for infantry regiments in combat only.


The following data, which have been assembled from the medical records of the American Expeditionary Forces, may be of some assistance in arriving at an approximate estimate of the problem.

49. Gunshot wounded.---

a. Location.---It is suggested that the location of the gunshot wound, varied as the available information suggests, be used as a general index of the gravity of the wound and of the transportation required. Wounds of the soft tissue of the extremities when associated with extensive destruction of tissue, hemorrhage, shock, etc., are obviously more serious than fractures of the corresponding long bones associated with little destruction of tissue. Probably, however, the deaths which occurred in such cases can be used as an indication of the proportion of the wounds of relatively unimportant tissues which are very severe.

When a soldier had two or more wounds, the preference is given in the following tables to the location associated with the highest average fatality rate. The table shows how often wounds involving various tissues occurred, and also the fatality rate for each group. Wounds involving the abdominal or pelvic organs occurred in 11.07 per 1000 cases, with a fatality rate of 66.80% ; while those of the soft tissues of the lower extremity (not involving important blood vessels or nerves) occurred in 333.86 per 1000 wounds, with a fatality of 6.09%. Apparently then it may be said with safety that a patient of the latter class was a much less serious transportation problem than one of the first named group.


Table 23.---Location of battle gunshot wounds in the American Expeditionary Forces with rates per 1000 total cases; and percentage fatality rate of each location.(16)

The fatality rate from artillery wounds was 7.03% and from small arms 4.82%.(16)

Artillery missiles caused 70% of the gunshot wounds among American troops during the World War as compared with 10% among the Union troops during the Civil War .(2)

The following table shows that the general location of the gunshot wounds was very much the same during the two wars.(2)


Table 24.---Estimation of evacuation requirements for patients wounded by gunshot missiles. Number in each 1000 total wounded.

b. Transportation.----In Table 24 all cases which ultimately died are listed as "Littered", or "Recumbent"; and the remainder, that is, the cases which recovered, as "Littered" and "Recumbent", or as "Walkers" and "Sitters," according to the location of the wounds. The division of the recovery cases into the "Littered" and "Recumbent" and "Walkers" and "Sitters", such as in the case of those involving the soft tissues of the lower extremity, was based upon the location and fatality rate of the group.

50. Gunshot and Gas Wounded.---The medical records show the number of fatalities from war gases (1.73%) but no other information which may be used as a basis in estimating the transportation requirements. Consequently any such estimate must be based upon general experience and supposition.


Table 25.---Estimation of evacuation requirements for patients wounded by war gases and gunshot missiles, separately and combined; rates per 1000 cases.

Note: The above totals are calculated by multiplying the rates from the gunshot wounded by three, the ones from war gases by one, adding the results, and dividing the sum by four. This procedure is based upon the assumption that in severe combat war gases cause 25% and gunshot missiles 75% of the total wounded. (As calculated from Fig. 68, it would be 23.81% and 76.19%).


In the table above the data in regard to gunshot wounds are from Table 24, but those for the war gas wounds are as stated.

According to Table 25, 47% of the wounded would be littered to the Collecting Station and 53% could walk; 43% would be transported to the Hospital stations as "Recumbent" and 57% as "Sitters"; and to the Evacuation Hospital, 43% again would be transported as "Recumbent" and 57% as "Sitters."

Colonel Alexander N. Stark, who was Chief Surgeon of the First American Army, American Expeditionary Forces, says in his report of the Meuse-Argonne offensive, that 42% of the patients evacuated were carried "Prone" and 58% as "Sitters". He includes in his total, however, the sick as well as wounded.(2)

P. BATTLE CASUALTIES---DISPOSITION OF CASES IN THE COMBAT AREA.

The experience in the American Expeditionary Forces may be helpful in forming an approximate estimate of the relative number of the sick and wounded to be evacuated from the Combat Area. When considering this phase of the problem, the patients may be divided into two groups: (a) Long duration cases, and (b) short duration cases. The first group will require evacuation to the base and general hospitals, and second may be hospitalized somewhere within the Army Area.

The time limit for the latter group will depend to some extent upon the existing conditions. It is subdivided into: (a) cases that return to duty, and (b) those that die within a few days.

As stated elsewhere, the medical records from the American Expeditionary Forces show the duration of treatment both of cases that recovered and those that died. The value of this information is reduced materially because: (a ) Patients in some instances were retained for treatment in the Combat Area, and no records were made for them; and (b) trivial cases, which were sent to the base hospital, lost unnecessary time.

51. Percentage of short duration cases. [Note: The method of computing the data in this section is explained on pages 140 to 143.]---The following table shows the percentage of patients which: (a) Returned to duty, or (b) died during the first 10 days, treatment in hospital.


Table 26. Percentage of short duration cases, as reported in the American Expeditionary Forces, leaving hospital each day from the first to the tenth.

The maximal day of treatment shows the longest time of treatment, but the group includes all those of shorter duration. Thus, the maximal 5 days group includes cases leaving hospital at any time up to the end of the fifth day. Of the cases of diseases and nonbattle injury sent to hospital in the American Expeditionary Forces, 22.74% returned to duty in 5 days or less, and .25% died; while of the gunshot cases, 5.20% returned to duty in 10 days or less, and, and 6.36% died, etc.

If all duty cases requiring treatment for four days or less are retained within the Division area, 18.91% of the sick, 7.94% of the gassed, and 1.09% of the gunshot wounded will be so held. Further, if duty cases requiring treatment from 5 to 10 days are hospitalized within the Corps or Army Area, the group will consist of 19.14%, (38.05%-18.91%) of the sick, 11.83% (19.77%-7.94%) of the gassed cases, and 4.11% (5.20%-1.09%) of the gunshot cases. Obviously the selection of sick cases whose treatment will require a maximum of 4 days, 10 days etc., must depend upon the judgment of those selecting the cases.

The percentages can be applied to any casualty day for any unit. Assume a Division of 20,000 men with a daily casualty rate of 60 per 1000 men, and with a daily sick rate of 4.20 per 1000. Then there will be:

Casualties 1200
Killed (16%) 192
Gassed (20%) 240
Gunshot wounded (64%). 768
Sick 84

Assume further that all cases returning to duty in 10 days or less, and that all cases dying within 5 days or less are held within the Army Area. Then each day the number so held and evacuated from the above group will be:

Total (a) Duty in 10 days or less (b) Death in 5 days or less (c) Evacuated
(a)-[ (b) + (c)]
Sick 84 31.96 .21 51.83
Gassed 240 47.45 1.70 190.85
Gunshot wounded. 768 39.94 36.63 691.43

Total

1092 119.35 38.54 934.11

Nontransportable cases held within the front area will reduce still further, at least temporarily, the number to be evacuated.

52. Hospitalization of short duration cases.---

a. Accumulation in hospital to the end of any period of cases returned to duty or dying during that length of time.---The data in Table 27 from the American Expeditionary Forces records shows how short duration duty and death cases accumulate when the number of days of treatment varies from 1 to 5, 1 to 8, 1 to 10, etc. The table is based upon the same assumption as elsewhere in considering hospital populations; that is, that there is an average daily admission rate during the periods with a definite number going out each day. It answers question such as; "After six combat days by an infantry division with an average daily admission rate from diseases and nonbattle injuries of 4.20 per 1000, and from casualties of 50 per 1000 how many duty and death cases of six day or less duration will there be?" Since Table 27 is based upon an average daily admission rate of 1.00. the daily number of sick and wounded must first be determined.
Rate per 1000 Strength One-day cases Six-day cases
Disease & nonbattle injuries 4.20 20,000 84 504
Casualties 50.00 20,000 1,000 6,000
Killed (16%) 160 960
Gassed (20%) 200 1,200
Gunshot wounded (64%) 640 3,840

Table 27 shows how short duration cases accumulate in hospital.


Table 27.---The accumulation in hospital to any day from the first to the tenth of patients who will return to duty or die on that day or before. Admission rate from each cause is 1.00 per day.

Then in reply to the above question, the accumulation of the 6-day or less cases in hospitals, which must be based upon each day's admissions, would be:

At the close then of the six combat days with the admission rates as specified, the account of a division of 20,000 men for the sick and wounded combined would be:
Total to be accounted for (6504---960) 5544.00
Evacuated 4971.39
One to six-day cases in hospital in Army Area 316.68
.........Duty cases 213.84
.........Death cases 102.84
One to six-day cases which have left hospital 255.93
.........Duty cases (360.21-213.84) 146.37
.........Death cases (212.40-102.84) 109.56

b. Accumulation in hospital to the end of any period of cases returning to duty or dying during that length of time:---But the above question may be worded as follows, "After six combat days by an infantry division of 20,000 men with an average daily admission rate from diseases and nonbattle injuries of 4.20 per 1000 and from casualties of 50 per 1000, how many duty and death cases of 10 days or less duration will there be?" Obviously since the duration here is 10 days or less instead of 6 or less, there will be more cases retained within the Army Area. The answer to the question can be computed from the basic data in the following table.


Table 28.---The accumulation in hospital to any day from the first to the tenth of patients who will return to duty or die on the tenth day or before. Admission rate from each cause is 1.00 per day.

Since this table, as the previous one is based upon a daily admission rate of 1.00, the daily number of sick and wounded must first be determined. The table from page 129 is repeated here.
Rate per 1000 Strength One-day cases Six-day cases
Disease & nonbattle injuries 4.20 20,000 84 504
Casualties 50.00 20,000 1,000 6,000
Killed (16%) 160 960
Gassed (20%) 200 1,200
Gunshot wounded (64%) 640 3,840

Then the accumulation in hospital in six days of 1 to 10 day cases, based upon each days admissions, would be:
Duty cases Death cases Total
Sick 84 x 1.56) 131.04 84 x .02)..... 1.68 132.72
Gassed (200 x .89) 178.00 (200 x.04) ....8.00 186.00
Gunshot wounded (640 x .27) 172.80 (640 x .22) 140.80 313.60

Total

..............481.84 ...............150.48 632.32

The total 1 to 10 day cases during six days treated in the front area, and the ones to be evacuated, based upon the total for six days, would be:

Total in six days (a)

Duty in 10 days or less (b)

Deaths in 10 days or less (c)

Evacuated in six days (a)-[(b)+(c)]

Sick 504 (38.05%) 191.77 (.40% ) 2.02 310.21
Gassed 1200 (19.77%) 237.24 (1.10%) 13.20 949.56
Gunshot wounded 3840 (5.20% ) 199.68 (6.36% ) 244.22 3396.10
5544 628.69 259.44 4655.87

At the close then of the sixth combat day with the admission rates as specified and with the 1 to 10 day cases treated in the front area, the account of a division of 20,000 men for the sick and wounded combined would be:
Total to be accounted for (6504-960) ...... 5544.00
Evacuated 4655.87
One to ten day cases in hospital in Army Area . 632.32
..........Duty cases 481.84
..........Death cases 150.48
One to ten day cases which have left hospital 255.81
..........Duty cases (628.69-481.84) 146.85
..........Death cases(259.44-150.48) 108.96


Section Two, Chapter IV

Contents