Document peEnpddew64QqnE8k1RBOk617

.'.aciroftflii.'.'u.ii Fluor'.Utf 4. iposurM?. M! 5 Mi" I 3.30 1.56 4-3 :.3a 3.>3 1.00 l i.u 0.48 I'M >.75 :) '..34 0.7S 3.s4 1.54 Methodology in 4.00 1.60 il i.: 0.71 id 2.10 0.94 -'ii 3.92 1.57 Industrial Health THOMAS F. MAN'CVSO, M.D., M. ?.H. l.sr 0.57 ;4 2.2'. OSS 50 1.36 0.75 35 1.08 0.79 2.62 1.05 I Studies AN'D ELIZABETH J. COULTER, PhJD. COLVMBl'S, OHIO 18 1.81 o.n 3.35 2.87 ` 1.54 1.15 The Demographic Approach 1.56 0.62 ; 2.06 033 1.13 0.47 iSi 0.73 132 0.53 n 1.95 0.79 U is 0.37 0.35 1.93 0.77 .8 1323 0.49 3.49 1.40 4 2.73 U9 4 4.11 1.77 0 2.14 0.66 8 0.76 0.30 1.15 0.43 0.71 0.34 139 0.76 i 1.28 0.50 0.50 0J0 There are many opportunities tor investi pattern of disease distribution, as derived gators in the field of industrial health to from vital statistics and other sources, repre 2.10 03* 4 0.67 037 design and conduct studies to evaluate the sent an initial approach in epidemiologic relationship between chronic illness and methods relating to industrial populations. occupation. This area of study poses, how- Such studies involve geographic areas, sex, 3.33 7.73 1.35 3.09 some complex and difficult problems color, age, country of birth, and socioeco I which require varied approaches. nomic status. 1.75 0.71 I In 2 papers consideration will be given Demographic comparisons help to deter .5 2.64 5.32 0.50 1.06 3.33 0.20 .3 2.31 8J2 0.50 1.13 3.33 0.20 to the application of epidemiology to in dustrial populations within the framework of toe demographic and prospective or cohort aPproaches. This paper is concerned with toe demographic approach, which is of basic importance in industrial health studies. Demographic studies identifying charac teristics of the general population and the Submitted for publication Aug. 17, 1962. Formerly, Giiet, Division ot Industrial Hygiene \Pr- Mancuso), and Chief Statistician, Division of hai Statistics (Dr. Coulter), Ohio Department mine geographic and population character istics associated with mortality or morbidity due to specific diseases. Urban-rural com parisons, for example, which indicate rela tively high mortality in urban areas or in particular metropolitan counties represent initial clues to environmental influences, such as industry and air pollution, from which more refined epidemiologic studies can be derived. They further help to identify sex, color, nativity, and age groups with relatively 76 0t Health. 77 high risk of specific diseases or causes ot i 516 ARCHIVES OF ESVIROSMESTAl. //, TablLe 1T.----AAv-erage Annual Aae-Adiusted Death Rates* per 100,000 Population n White Male and While Poiwlcj'Rcsidcnls of Each of the Eight Metropolitan C,!^y '' fr.0~ nlGGY-- .....,,/ Xcopiasms :s-cA Yw* Neoplasm Site (and Cause of d- Geographic Are* State of Ohio 8 Metropolitan counties Cuyahoga FrankUa Hamilton Lucas Mahoning Montgomery Stark Summit All Malienunt N'eopliisms (140-205) i All Malignant Neoplasms Except Those of llreast & IteproductiTe Organs ISO-205) i Esophagus >150) 141X2 155X3 I67.M> 147.79 162,07 1.K 1C2.CO 136.G? 139.12 137.15 134.34 148.41 159.92 14130 156.03 147.11 155.97 129.08 132.45 130.76 3X5 4X6 6.10 3X2 4.98 5X3 4.92 4.12 3.68 3.41 Stomach '151) 11.69 12.60 16.59 9X8 9.26 11.75 16.31 9.51 20.85 10X0 Papcreas (157) Larrni U61) Whits .Males 7.07 2X0 7.56 2X9 8.23 3.27 7.08 3.01 7.17 3X0 7.43 2.84 9.18 2.95 6.46 2.08 7.10 2.28 6.91 2.09 tJjns. !Q 3tnj -<c txi 127| A* <isi <1.13 *X1 38JS 1> V . _S> Beider A Other Criosrr Or?M USD f i.H 4/7 IS* : A> w* -V 3* *x ji ** i- 4.W U< 4.47 3. 4.97 5.16 <27 3.13 4.42 4.77 White Females 123 State of Ohio I 140XC 70J36 0.60 5.6S 3.92 0.16 SJ4 151 1. 8 Metropolitan counties 145.25 73.18 0.69 6X0 4.04 0.17 S.70 134 Cuyahoga Franklin Hamilton Lucas Mahoning Montgomery Stark Summit 148.7J 151.75 155X0 140.14 140.29 134.51 125.34 142.77 70.08 74X5 73.71 67X4 S0X3 64,26 65.38 76.07 0.84 0.601 0.401 0.62| 0X9| 0.681 1.031 0.61 | 8.14 4X1 5.13 4X1 30.32 4.19 4X4 6.00 4.12 0.08 | 4.11 4.41 0X0 j 543 4.41 0.20| 4.0 4.42 0.18] 4J9 3.54 0X0] 7XJ 2.SS 0.10 j <04 2.42 0.14 1 ta 4.93 0.34] 424 - > * 381 t* t 1M 131 lit V' l. 136 1X3 1.46 1.42 13 . % . - ' ' t TVs^wbiie popclat: . ? ' ' 91M IwSiLa. Japsoese. C `Mortality rates were computed on tho Intsis of mortality data for the specified type of malignant neoplasm for white eu> 4 ` | O* cwcTifoiiun ue and white lenities, classifiedby 5-year -ce croup*, available in the Ohio Department of Health, and population data obuM t {upa'-atieti or me 1 avenging population counts given by tho doeonnul United States censuM* of 1950 and 1960 for the corresponding set. cok*,a4 7;-' . I Vwgi^sjof Irmpha; age groups. Age adjustments were made by the direct method on the basts of the age distribution of the total populuiao tt* United States as shown by the 1960 Unucti States census. - 5 <WH ; He 4ofh me is based death. Such data provide a basic reference source for studies of industrial workers which, in turn, require similar classification by population characteristics. In this manner, it i? possible to identify ami eliminate the effect a particular variable. such as age. may have on mortality patients observed in soec.dc industries. blortuUiy rates of in dustrial workers by population characteris tics. such as sex, color, and a^e. help to determine the possible effect of selective employment practices relative to such popu lation characteristics on the morbidity and mortality experience of a given plan: >,* . industry. 5 Geographic Location In studying geographic differences as law epidemiologic leads in industrial hrahv studies, a first consideration is urhan-rurti comparisons. These may provide initial c.Vi relative to industrial and agricultural {.: iations and furnish the basis for fur.?*epidemiologic efforts and other ty'!*r -* studies pertaining to air pollution. #oci<v.nomic factors, and medical care. F6. April. 'Ate & i y o ,> ~ J ' \tru*r amarJ rfts% <ix to cancer while, fore;. v*w>-.?te male res: :onl arras o:* Or.:c. ap- \ ears. tf-gtility rales w 4srt of mortality **vo Jk-j-arnrieni o: 7***iia:or.. C;Aiii K V-yi. a.->5 }:A~ sx the IO5.5 V.h j.rc'aur- rrg --j.r: h -w < :;-.e tW 4' VVW 0: 2own >V - r... ?* -n Vrcr-:' 7 vi> ! nr. ** Mmcuc :icn Due to :n Counties X .) Code) L Broncnus. V- T->e*v.-a '.2. 3!S ) 36.02 +0.47 +1.26 +2.79 +5.53 +3.91 41.13 36.2] 36.36 31.39 5.3+ 5.70 6.19 5.63 +6.4+2S 7.23 .21 i i i \ chite male? .a obtained by .. color, and ".on ol the Manx o: ;es a? basfc ial heaith rban-rurai nirial dues -ra! popu.r further tvt'ts of . socioec^- ;[iTi:ODOLOGY--i.\DLSTRIAL iiHALTH STL'OILS 517 \!,uipant Xcflpiasms of Selected Sites Antony of dice, 55-61 Years of Age, 1950-1960 Sidney tiSO) Bladder b Other Criaarv Organs OSU 4Brain c Other Parts of CNS 093) Leukemia A Aleukemia () 7$. $ 4.05 6.40 6.63 +.07 4.44 6.34 7.30 +.05 3.36 3.73 3.35 5.97 4.03 5.07 3.5S 4.47 3.61 4.97 5.56 4A7 3.23 4.42 4.77 6.92 5.96 6.S8 7.15 6.75 6.40 6.47 7.70 7.59 6.S9 6-76 7.67 8.16 6.35 6.72 7.90 1.72 1.53 3.98 4.32 1.81 2.60 4.22 5.05 135 1.34 4.47 5.33 Ul 2.43 1.80 2.69 4.30 3.42 5.15 5.25 1.97 1.43 I1S.26l 1.96 1.03 1.46 1.42 4.00 2.77 3.43 3.83 4.40 4.91 4as 4.98 0.35 2.29 521 5.16 ) Tbe oonvbite population group was defined to include Stgoes, Indians. Japanese, Chinese, and other nonwhite races. The metropolitan counties were defined as those with cities of 100,000 population or more in 1950. i Neoplasms of lymphatic and hematopoietic tissues are is eluded. i The death rate is baaed on frequency of less than 10 deaths. Our observations in Ohio cancer studies have emphasized the importance of such urban-rural comparisons, particularly bv population subgroups. These studies have shown, for example, that (1) an "urban'' influence existed for certain cancer sites and (2) lung cancer was relatively high among the foreign-born of both urban and rural areas ('Figure).-'9-30 Other studies have also emphasized urban-rural and related geo graphic considerations.10-25-22-45 Comparisons of cancer mortality in small geographic areas such as counties mav pro vide further epidemiologic leads. Table 1 provides county comparisons of mortality due to specific cancer sites among white male and female residents of individual metropoli tan counties of Ohio for the period 1950 1960. The table reveals individual county variation in cancer mortality for the specific sites considered. Lung cancer mortality among white males was, for example, rela tively high in Hamilton County and relatively low in Summit County. Stomach cancer mortality among white males was, on the other hand, relatively high in Cuyahoga and Mahoning counties and relatively low in Franklin, Hamilton, and Montgomery Coun ties. These observations serve as a basis for further epidemiologic efforts. In individual county comparisons it must also be realized that a single industry may not appreciably affect the specific disease rate in a county just as a high incidence of a specific disease in a single work department IO *0 3 0 Average annual age-adjusted death rates due to cancer or the lung among native while, foreien-born white, and aartvc MTt nonwhite male residents of urban and rural areas of Ohio, who died in the state, ages 25-64 years, 1947-1951. reMMK 'Mortality rates were computed on the vwrrz basis of mortality data available in the Ohio Department of Health for the speci fied population, classified by age groups 25 44. -r-ja and 55-64 years, and population NOR WHITE data ir. :he 1950 U.5. Census for the cor responding population groups. Age adjust- ,-.;r.:s were made by the direct mrthod on the basis of the age distribution of the total population of continental United otitfs as shown by the 1950 United States Census. File nonwhite population group- was defined to include N'egroes. Indians, Japanese. Gtinese. and other nonwhr.ite races. The nat'ive-'born -group included persons xrr in ibe Unitesd State? or any of its territories or possssessions. The classification c: urban and rural areas was based on the definition of such areas used in the 1950 United i-tates Census. ** .Umvti.v--Coulter ol$ LXi 7 RO.\ .'! r.XT.-ii. HC.-iLT/-/ mav not arfect the rate for the disease of a large industry because oi the dilution factor. This point was demonstrated in a study of the lung cancer mortality rate in Lake County, Ohio, although previous studies 2T'S0 had demonstrated an abnormal mortality rate for lung cancer in a chromate plant located in that countv. When a large group of industries are located in a circumscribed area, such as a county, and are engaged in the same essential manufacturing processes or prod ucts, it can be expected, however, that the disease partem cf the county will redec: this concentration in the area if an environmental factor is involved. Residence Duration In studying diseases with long latent periods, it should be noted that the current place of residence of population groups under study may not be the place of environ mental exposure and therefore consideration must be given to long-term residence histories. This applies to the following types of current residence: death residence, place of retirement, place to which an individual moves for special medical or surgical care, place to which there has been significant migration because of attractive climate and job opportunities, and places within a city to which there have been population shifts with respect to race, nationality, and socio economic status. These types of shift in residence, singly or in combination, bring about changes in the age distribution of the population, as well as in the types oi illness and death, which thereby affect the iiiness and mortalitv pattern of a given locality. Residence history data are of basic impor tance in studying the possible association of such variables as occupation, atmospheric pollution, smoking, morbidity, and mortality. There are various methods of exploring residence duration in epidemiologic studies. One is to obtain residence history data for specific peculation groups, by such means as checking city directories and employment records and querying reiatives, friends, and other associates. This approach car. provide information, for example, or. the duration of residence in the county of residence at death as shown on the death certificate. Another approach is to compare place of residence at death with the birthplace as specified on the death certificate, if one can assume that an individual born in a given county who died in that county lived a major portion of his life there. Both of these approaches are of assistance in comparing mortality rates due to long-term illness in different geographic areas as they relate to duration of residence, which is important because of the latent pe riod between initial exposure and develop ment of the chronic disease leading to death. In considering residence data it is essential to keep in mind 2 points: (1) the inter relationships of residence and migration patterns with other factors such as socio economic phenomena, employment practices, and medical care and (2) geographic varia tions in the distribution of chronic illnesses depending on the latent period of the par ticular disease. Sex In industrial health studies, sex compari sons are essential in epidemiologic con sideration. When there is employment selection by sex within an industry, the dis tribution of iiiness by sex may provide the means of excluding certain departments and exposures, in this way refining the epi demiologic study. In some industries and occupations, women represent the predominant labor force. Under such circumstances, patterns of long-term illness or mortality in femaie emplovee populations may be studied in com parison with similar data on women in other industries ana women not employed in industry. It is aiso of special interest to consider whether the occupation or pattern of living of the husband affects the illness and mortalitv experience of his wife. Some cancer mortality data for Ohio shown in Tables 1 and 2 reveal characteristic natrerns of sex oinerence in cancer mortality. The tables indicate, for examoie. that cancer mortality was generally higher among males Vo!. 6. Apr.7. 1905 S1"' 0THOPOLOJy~/.V/.'C. females of each colt Simon to both sexes a fences were more mar, than for others. The re ^her hand, a consider^ T,BtE Z.-A-.cmcc Annua \'cpp!<zs'`is &T 0/<c..aea * Eight >/,?) Neoplasm Site (A Cause c MsJifnAnt neoplasms. inclu-iine neo ' * hematopoietic tissues (140-005 Boocs* cavity k pharynx (MO-US) Digestive organs * peritoneum (1.` Esophagus (ISO) Stomach 03!) SniaJ! intestine, inc). duodenum Large intestine, except rectum 0 Rectum 054) Biliary passages & liver, stated a Liver, secondary <t unspecified l Pancreas 037) Peritoneum (158) Unspecified digestive organa (lit Respiratory system Kose, nasal cavities, middle ear. Larynx 051) LuUg. bronchus, k trachea 062. Mediastinum 064) Thoracic organs, secondary 085 Breast dr genitourinary organs (17: Breast 070) Cervix uteri (171) Corpus uteri (172) Other k unspecified parts of uteOrary, faDopian tube, A broad : Otter k unspecified female gen;: Prostate 077) Testis (173) Ocher k unspecified male genitA Kidney (ISO) Bladder <t other urinary organs Other k unspecified sites (190-199'. Melanoma ofskin 090) Otter malignant neoplasm of s Eye 092: Brain k other parts of nervous: Thyroid grand (194) Other endocrine glands U9i) Bone 096) Connective tissue 097) Secondary k unspecified maiic- nodes . :9SOther <\r unspecified sites 099; k hematopoietic tiss'.. lycpnxtis k hematopoietic rs ' Mrr:i!::y rates were computed "aired Sis'.es a *newn r.y ;ne '.W* The ne-nwhife ponumticr. i The metropolitan Mur.:;e >| Tin- de'ith rate is based os a ;r- sl Mancuso--Ccutter `fc-U-TH at death, Another !l^ence at n the vnat an who died n of his are of ates due ographic Sjidence, .rent pedeveiopo death, essential is interdgration :s socio'ractices, 'C variaiilnesses the par- ompariic conloyment t*"* dis- the s and ne epi* pations, : labor trns of feir-ie n :on'.- ^TnOnOLOCY-lSIKSTRUL HE.4i.TH STU'iLS 519 than females of each color for cancer sites common to both sexes and that these dif ferences were more marked for some sites than for others. The females had, on the other hand, a considerable mortality due to cancer of the breast and reproductive organs, particularly in middle age. Individual metro politan county variations by sex shown in Table 1 may be useiul in selecting those cancer sites among the male population which Table 2.--Average Annual Age-Adjusted Death Rates" per 100.000 Population Due to Malignant Seoflasms of Specified Sites Among IVhiie and Xou'vhitc v Male and female Residents of the Eight Metropolitan Counties of Ohio,% Ages 25-64 Years, 1950-1960 ' i. I Neoplasm Site ( Cause 0/ Death Code) ^jaiignftDt neoplasms, including neoplasms of lymphatic ; 4 hematopoietic tissues (140-205) Buccal cavity pharynx (140-148) 1 Digestive organs peritoneum (130-159) $ophsgus 050) I Stomach (151) , Small intestine, inch duodenum (152} Large intestine, except rectum (153) Rectum CM) Biliary passages & liver, stated as primary site (155) Liver, secondary A unspecified (156) Pancreas (157) Peritooeum (15S) Unspecified digestive organs (159) Respiratory system (160-165) Nose, nasal cavities, middle ear, & accessory sinuses (160) Larynx (Ml) Long, bronchus. A trachea (162, 363) Mediastinum (164) ^Thoracic organs, secondary 065) Breast genitourinary organs (170.181) Breast (170) Cervix uteri (171) | Corpus uteri (172) ' Other A unspecified parts of uterus (173.174) . Ovary, fallopian tube. broadligament (175) Other unspecified female genital organs (176) PrenateOTT) I Testis (178) Other unspecifiedmale genital organs(179) Kidney (tW) Bladder other urinary organs (IS!) Other unspecified sites (190-199) Melanoma of sWa (190) Other malignant neoplasm of shin f19!) Eye 0921 Brain other parts of nervous system .193) Thyroid aland (194) Other endocrine giands (195) Bone U96) Connective tissue (197) Secondary unspecified malignant neoplasm of lymph nodes 198' Oti-er -*. unspecified sites (199) Lymphatic hematopoietic tissues '200-2(15) Lymphatic hematopoietic tissues i200-203. 20Ji Leukemia aleukemia :2) Total 156.2$ 3.99 47.94 3.36 10.51 0.46 14.32 6.34 3.01 2.99 6.11 0.55 0.29 26.18 0.32 1.56 23.82 0.32 0.16 48.08 1S.06 8.73 0.54 3.97 6.55 0.43 2.86 0.64 0.15 2.95 .WO 35.75 1.15 0.57 0.16 5.27 o.er C.45 1)13 0.53 0.26 ' 5.66 14.31 5 <** 6.09 Males White NbD'.efelte 153.33 213.5S 6.55 53.36 4.86 12.60 0.54 13.95 7.35 2.39 3.22 7.56 0.65 0.24 44.81 037 2.89 40.87 0.49 0.19 15.43 0.2S -- -- -- -- -- 4.92 1.46 0.28 4.07 4.44 17.76 1.52 0.67 0.16 6.84 0.55 0.53 1.26 0.51 0.26 6.96 56.30 14.59 29.03 0.54 | 13.10 7.23 3.79 6.71 9.97 0.54 f 0.80 58.78 1.04 3.06 54.06 0.32 $ 0.30 { 25.59 0.09! -- -- -- -- 14.53 0.19! 0.73 i 4.47 5.53 15.72 0.35 i 0.7! | 0 4.21 0.61 0.65! 1.09 0.62 5 0.23 5 5.43 37.44 10.14 7.30 10.05 17.22 10.60 6 62 Temales White Nonwhite 145.25 208.93 1.43 37.43 0.69 6.30 02)7 14.50 S.U 3.44 2.28 4.04 0.49 0.21 6.39 0.21 0.17 5.70 0.17 0.14 75.48 35.49 14.98 1.01 6.82 12.99 0.78 -- __ _ 151 1.60 53.29 0.95 0.39 0.1S 4.13 0.59 0.34 0.75 0.51 0.24 1.86 60.06 3.37 13.40 0.48 J 1722 7.95 3.SI 3.59 9.00 022 5 1.02 10.08 0.22 } 0.85 8.75 0.26 ! 0 107.03 32.27 36.31 1.59 17.22 11.71 1.39 -- -- -- 1.94 4.60 17.82 0.37 5 1.J5 0 2.S5 1.17 0.57 i 1.25 0.73 0.46 { 5.0S 11.23 0.20 3.05 9.17 12.08 7,50 4.45 ' Mortality rsmw computed op :s* basis o' T.cn-..:y data available i;i ;he Ohio Department 0/ ficitb tor residents oi the 8 metropolitan counties 0/ Ohie. classified by sex. coior. and 5-year see srni:r>s. and population da:a obtained by arersgiae poouiation tic:.* ctv*r. by thedeecr--tai United Stares tsnruses of 19V and :W for the correspond'." srx. color, and .ige croups ir. tr.is eccPipa.c '.-.rr-.i. Age adjustments were made i.7 the dtrect me-.rod or. me basis ol the age distribution of the rota! population of me "curd States ar shown by the i960 United S:at* Census. Tae mnv.-.nste population croup tvas defined Include Xen:*?. Indians, Japanese. Chinese, and oiner nonwhite race*. : Tit? iv.-nropG'.imr: vounties were defined as those vitr. cities o! IWA*) population 0: mere ir. ivOO. I The deaih rate is based or. a frequency of less than in deaths. .'/iinbn.fc--Coulter . 520 ARCHIVES OF EXVIROXMEXTAL HF..-U.T-, Table 3.--Average Annual Death Rates* per 100.000 Population Due to Malignant Xeopln, of the Eight Metropolitan Counties of Ohio,* 25-64 Years o/ ^ Neoplasm Site (A Cause of Death Code) AH malignant neoplasms, including neoplasms of lymphatic A- hemato poietic tissues (140-205) Esophagus GSd Stomach (151) Pancreas (15?) Larva* U61) LutJg, bronchus, A trachea (162,163) Kidcey (ISO) Bladder A other uricary organs (161) Brain A- other parts of nervous system (193) LeuXemia A aleukemia (XK) 25-64 iM.n 4.S3 12.41 7.49 .66 0.51 4.03 4.41 6.76 7.26 25-29 16.44 0 (MS { 0.1S } 0 0.T7 o.is i o.oe $ 1.31 3.0* 30-34 19.50 0 0.S1 0.34 } 0 2.07 0.17 } 0.17 J 2.19 2.S2 35-39 34.30 Age Group 40-44 66.3$ 4S-49~""` 13.01 0.41 V 1.66 0.52 5 0.29 J 5.27 0.S3 0.17 J 4.17 3.65 0.63 4.32 3.69 0.$$ 14.22 3.19 1.25 4.70 5.26 3.24 3.0? 6.2! 2-55 33.60 3.45 2.75 7.3$ 5.31 l; j i * Mortality rates were computed on the basis o( mortality data available in the Ohio Department of Health and population obtained by averaging population counts given by the decennial United States censuses of 1900 and I960. Age adjustments were made by the direct method on the basis of the age distribution of the total population of the United States as shown by the i960 United States census. t The definition of" white" males excluded Negroes, Indians. Japanese. Chinese and persons of other nonwhite races. {The metropolitan counties were defined as those with cities of 100,00 popcintlon or more in 1950. ! The death rate is based on a frequency of less than 10 deaths. merit further study, particularly in reference to occupational or other environmental ex posures peculiar to males. Color Color comparisons serve to separate some of the variables to be considered and to delineate avenues of investigation in epi demiologic studies. Although the environ ment of nonwhites has been identified to a certain extent, the epidemiologic potential of this population group has not been ade quately explored. In considering chronic illness, one must keep in mind, in addition to the latent period of the disease, the pattern of migration and employment of nonwhites over the years. Xonwhites may be affected to a different ex tent than whites bv selection for certain types of work, shifts of jobs, and exposures within the same industry over a long period of time, as well as by changes to other in dustries. This infiuences consideration of disease development in the nonwhite race and emphasiies the importance of duration of employment in industrial population stud ies. j nese factors provuie indications ot differences in cultural and living patterns that may merit further study by other epioemioloeic methods. Table 2 illustrates color variations bv specific cancer site for each sex as found in cancer studies of the S metropolitan counties in Ohio for the period 1950-1960. The coior differences were particularly marked for cancer of the digestive organs and peri toneum, and tor cancer of certain reproduc tive organs. Marked nonwhite excesses were observed, for example, for cancer of the esophagus and stomach for each sex; for cancer of the cervix, and other and un specified parts of the uterus, excluding the corpus uteri, among females, and for cancer of the prostate among the males. A nonwhite excess for cancer of the lung was also found for each sex. On the other hand, relatively higher cancer mortality among the whites than among the nonwhites of each sex was found for melanoma of the skin and cancer of the brain and other parts of the nervous system. An earlier study' revealed differences in cancer mortality among native and foreign-bom subgroups of the white popula tion which brought out the importance of studying nativity in coior comparisons for specific types of cancer, such as cancer of the lung and of the brain and central nervous system."0 Further refinements showing coior as we!! as sex and nativity differences by i'oi 6, April, 2965 .'2 ubthodoi.ogy--i ,i fleeted Sites At,<n Jiassifcd by Age. 19fv 50-54 *35.94 : 90 js.OJ !<(. 4.70 I iO 10.07 55-59 :iS1.34 73.57 31.73 7s.4$ S#<7 Ida.iS lo.rr n.jo 73.15 13.90 individual counties r more specific envin The large variari specific causes by a the age cornpositio groups by geograph the necessity of at; mortality rates. Table 3 provide mortality by specific males residing in th of Ohio in the peri rates generally she with age which w: some sites than :o* for most of the can tor example, negii: age groups. 25-34 y ranee, of varying later years. Morta and aleukemia. an< other parrs o: the the other hand. hit. sites at ages 25-3marked rise with 'jtber sites studied There are a r.\ i. Vi' e<_ , crr.p-.oy- may rer.ee*. varia*.i> rent, and occupa- >4<vv r;. r'vCi **v: I v I I ( ) c t j r-THOD0L0CV~I.\rfSTRUL tiii. LTH STL'DILS 521 i crlatcd Sites Among White* Male Residents titsified by A$'- NSO-VM S>-54 2S.95 JS.05 <.to 67.42 J.S9 5.39 1C.M o.or -J9 351-34 uur 31.13 1S-4S ?.i: 106.63 10.3: 11-90 13.13 13.90 60-64 633.30 23.IS 61.01 31-40 11.29 157.39 13.S3 23.0S" :7.73 22.4S am 23-04 133.33 4 SO 12.60 7.36 40.37 4.07 4.44 6.64 7.30 individual counties may provide the basis for more specific environmental studies. Age The large variations in mortality due to specific causes by age and the differences in the age composition of various population groins by geographic areas have emphasized the necessity of age-specific morbidity and mortality rates. Table 3 provides age comparisons of mortality by specific cancer site among white males residing in the 8 metropolitan counties of Ohio in the period 1950-1960. Mortality rates generally showed a marked increase with age which was, however, greater for some sites than for others. Mortality rates for most of the cancer sites considered were, for example, negligible in relatively young age groups, 25-34 years, and assumed impor tance, of varying magnitude, in middle or later years. Mortality rates due to leukemia and aleukemia, and cancer of the brain and other parts of the nervous system, were, on the other hand, higher than for other cancer sites at ages 25-34 years and showed a less marked rise with age than for most of the other sites studied. There are a number of special factors involved in considering age in relation to industrial employment. For example, dif ferent age groups of industrial populations may reflect variations in duration of employ ment. and occupational diseases mav occur iS McnClts^~CrU.'!er in different age groups depending on the latent period of the disease. There may be a shift of mortality to the younger age groups when occupational cancer is involved. In considering age it is also important to r e the dates of installation of control measures and to determine whether particular groups of workers were exposed before or after such installations. Employment requirements and policies may affect age patterns as well as types of illness in industrial groups. The employment requirements of certain industries may be such that the young and very strong are selected and therefore constitute the pre dominant population. Older age groups may, on the other hand, be proportionately greater in other industries, such as those with rela tively large proportions of highly skilled craftsmen. Retirement, seniority, and sick leave policies and practices may also affect the age composition of the work force as well as the scope of the epidemiologic study. The age at entrance to exposure has some bearing on the age at onset of occupational diseases such as cancer. . Country of Birth Nativity has become recognized as an im portant population characteristic for consid eration in epidemiologic and biostatistical studies.14-1--''-^:i"37 A point of interest with reference to conditions following mi gration of the foreign-born is the tendency observed for specific nativity groups to settle in particular areas in which they often live and work together, possibly associating with a specific industry or occupation group, and maintain the cultural characteristics of their nativity group. The migration patterns of the foreign-bom make it important to consider any industrial and occupational concentration of migrating groups in relation to their geo graphic location and other environmental characteristics. The long latent period for some diseases, such as cancer, makes it necessary to consider factors in both the country of birth and the dace to which migration occurred in terms of circumstances which prevailed as long as ARCH/1'ES OF EXI "IRQXUgXT u 20 or 30 years or more ago. Table 4 pro vides an exploration of this point in com tU'ALTfj Table 4.--Average Annual Age-Addled Rates for 100.000' Population Among Sc/eetfd paring lung cancer mortality among native and immigrant populations of Cuyahoga Xativc H'hitc Male and Immigrant Population Groups Residing in Ohio,'the Eight Metropolitan Counties and Cuyahoga County, Ohio, the county which had the highest white male lung cancer mortality County iw Ohio, and the Male Population of Enoland and IVales, and Italy, Ages 25-64, Selected Years Around 1956 ` rate of any of the S$ counties of Ohio in the period -19*4-7-1951 .-9 Table 4 shows that the mortality rate due to lung can cer among male immigrates to Cuyahoga County from England and Wales was con siderably lower than among the male popula tion of England and Wales but slighdv higher than the death rate for this cause among the native white males of Cuyahoga County and considerably higher than the lung cancer death rate among the native white male residents of Ohio. The lung cancer mortality rate among male immigrants from Italy to Cuyahoga County was, on the other hnnd^similar to the male lung cancer rate in Italy but considerably lower than the lung cancer dearb rate among native white Population Group <St Period ot Years Native-born white males residing in Ohio, 1947.195) S metropolitan counties of Ohio, 194M9M Cuyahoga County, Ohio 1947-1951 Immigrants from all foreign countries re siding in Ohio. 1947-1951 S metropolitan counties ot Ohio, 1947-1951 Cuyahoga County, Ohio, 1947.195! Cuyahoga County residents who immi grated (torn: England * Wales, 1947-1951 Italy. 1947-1951 Resident population oi: England and Wales. 1950 Italy. 1951 Arerace Annual Age-Adjusted Death P.*i* per lOO.OW population. Ages25*64 2CA5 25.47 28.75 35.67 35.31 35.11 31.75 IS.61 55.48 16.36 male residents of Cuyahoga County and approximately the same as the lung cancer mortality rate among the native white male residents of Ohio. This is of interest in comparison with the results of a study by Eastcott of cancer mortality among native and immigrant groups in Xew Zealand.15 Socioeconomic Comparisons Just as geographic exposure, race, ot country of birth may influence chronic * Mortality rates for England and Wales were computed oa the basis of mortality data supplied by the World Health Organ, ization and population data in United Nations publications: all other rates were computed on the basis of mortality dau available is the Ohio Department of Health and population data in the i960 United Slates Census or estimated on the oasis of data in this Census by methods described on page Si of Ref. enact 30. Aft adjustments were made by the direct method on the basis of tbe age distribution of the total population of the United States as shown by the-5950 United States Cenju*. The nonwhite population group was defined to include Negroes, Indians. Japanese, Chinese, and other sonwhite races. The native-born group included persons born in the United States or any of its territories or possessions. Metropolitan counties were defined as those wttb cities of 100.000 population or more in 1950. illness and mortality patterns, so may the socioeconomic status of a given population. Efforts have been directed toward deter mining the association of socioeconomic husband for married women. The other type of approach is to compare over-all mortality in geographic areas of different socio factors and mortality in studies conducted economic characteristics, such as census in England and Wales and the United States tracts, classified by one or more specific for a number of years." - characteristics like rent or income. Both Two approaches have been used in pro approaches can, of course, be used in study ing mortality by specific cause of death. viding socioeconomic comparisons of mor tality. One is to compare mortality rates by In industrial health studies socioeconomic socioeconomic characteristics of the indi factors represent an index, to a certain vidual. such as occupation, occupation and extent. <>; the employment and exposure o: broad social class, and occupation of the the individual worker, since occupational classifications are reiared to wages. This may * References 1*5. 5. 10. 51-34. 36. 56 provide a means not otherwise available oi Vol 6, Apr;:, ;C^ <u { ,,,'HeVOLOGY-lX t-mdlv sorting or dir f *: -nil groups by type ot . * ;<pure. Laborers an\ ,,ell as management m h ' The influence of wor ,;2l;,brf)s or the family 1 fusing, medical care. ' s.nnf.<ts that industrial i : cXtend their concern to ' patterns of the lair.mes Sumtns . This paper provides a : the demographic appro?. : health of industrial pc [ mention was made of the on geographic location, ir ami county residence, du sex, color, age. country economic factors. Parti made to the interrelatio factors. Conclus Demographic studies ence data on diseases an lion groups of different geographic location anologic clues which are f-. ing more refined invest industrial groups. \\ e wish to express ou: technical and reiared assistsVeigel. Chief. Mrs. Elizafc. members of the Division oi Thomas F. Mancuso. M.; cunationai Health. Grad"; Health, Pittsburgh. Pa. REFERS, 1. Backgrounds of Social Papers Published tv the V New York, iP-ij. 2. Ciocco, A., and P-- Sickness as a Cause ot": Review, r 3. Cou.'ter. Z. an- t Wai Statistics Ky C-r.=-_= " Ass. 54:750-~sP'. ]--zr- Morraiirv Trends in cLn- 13 Xcr.:;i;;--C:u::- \r. Death Selected * Male :.o. the <Q -e A.-.nun V'.'-Vid :h Rate - JCo.-X'O .5 <7 -A75 7S. 6* si.n >i.:s :$.6l :.v<s 6.56 on -ith Orgao. jidicacions: tafi:y data population o the basis $1 of Re/. *ct method ' '-'Poo of :r.siiS. groes. 'es. The i State} counties on or more ter type .ortaiitv $oc:o- census specific . Both . study* .th. certain atior.ai is :nay :b!e of i i } ;OL'OL v --ISi. i .'i A/.:/. U.7H 523 broadly sorting or differentiating occupajiona! groups by type of job. department, and exposure. Laborers and skilled groups as as management may be grossly dif ferentiated. The induence of work income on living standards of the family, such as nutrition, housing, medical care, and education, also suggests that industrial health studies should extend their concern to illness and mortality patterns of the families of workers. Summary This paper provides a brief discussion of the demographic approach to studying the health of industrial populations. Specific mention was made of the importance of data on geographic location, including urban-rural and county residence, duration of residence, ex, color, age, country of birth, and socio economic factors. Particular reference was made to The interrelation of some of these factors. Conclusion Demographic studies provide basic refer ence data on diseases and deaths in popula tion groups of different characteristics and geographic location and furnish epidemi ologic clues which are fundamental in guid ing more refined investigations of specific industrial groups. We wish to express our appreciation for the technical and related assistance of Mr. William H. Yeigel, Chief. Mrs. Elizabeth Everett, and other members of the Division of Vital Statistics. Thomas F. Mancuso, M.D.. Department of Oc cupational Health, Graduate School of Public Health. Pittsburgh. Pa. REFERENCES 1. Backgrounds of Social Medicine: A Group of Papers Published by the Miibar.k Memorial Fund. Xew York, mi. 2. Ciocco, A^ and Perron. D.: Statistics ers Sickness as a-Cause of Povetry: An Historical Review. 1. Hist. Med. :2'.42-60 -Nc. 1) 1957. 5. Coulter. E. J.. and Gurairick. L.: Analysis of Vital Statistics by Census Tract. J. Arret. Statistic Ass. -.730-740, 1959. Cutler. S. T.. and Eierer. F.: End Results and Mortality Trends in Cancer: I. End Results in *' .'-'encase--Catthcf Cancer, National Cancer Institute Monograph No. 6. U.S. Department of Health, Education, and Welfare, Public Health Service, 1961. 5. Dcnsen, P. M.; D'Alonzo, C- A., and Munn. M. G.: Op]x>rtunities and Problems in the Study of Chronic Diseases in Industrv, T. Chron. Dis. 1:231-252 (March) 1955. '' 6. Densen, P. M.: Epidemiology in Industrial Health, in Transactions Publication No. 30, the 21st Annual Meeting of The Industrial Hygiene Founda tion. Pittsburgh, Pa., Industrial Hygiene Founda tion, 1956. 7. Dom, H. F.: The Statistical Approach to the Epidemiology of Cancer, in Proceedings of the Second National Cancer Conference, New York. American Cancer Societv, 195-1, Vol. II. pp. 1103 1117. ' S. Dorn, H. F.: Mortality Rates and Economic Status in Rural Areas, Public Health Rep. 55 -.3-12, 1940. 9. Dorn, H. F.: The Changing Incidence of Cancer Throughout Lift, Bull. N.Y. Acad. Med. 31:717-725, 1955. 10. Dom, H. F., and Cutler, S. J.: Morbidity from Cancer in the United States, Public Health Monograph No. 56 of the U.S. Department of Health, Education and W'el fare, Public Health Sendee, U.S. Government Priming Office, 1959. 11. Downes. J.: Chronic Diseases Among Spouses, Milbank Mem. Fund Quart. 25:334-358, 1947. ' 12. Dublin, L. I., and Vane, R. I.: Causes of Death by Occupation, LT.S. Bureau of Labor Statistics Bulletin No. 507, U.S. Government Printing Office, 1930. 13. Eastcott, D. F.: The Comparative Mortality- Experience from Cancer of Certain Sites Between the Indigenous Non-Maori Population and Im migrants from Great Britain, extract from Report of the B.E.C.C. (N.Z.) Branch, Cancer Registra tion Scheme, Wellington, New Zealand, National Health Institute, 1954. 14. Gafafer, W. M.: Sickness Absenteeism Among Male and Female Industrial Workers. 1937 3946, Inclusive. Public Health Rep. 62:1538-1541 1947. ' 15. Gilliam. A. G.: Epidemiology in Occupational Medicine, Arch. Environ. Health 3:281-287. 1961. 16. Gilliam. A. G.: Opportunities for Application of the Epidemiologic Method to Study of Cancer, Amer. J. Public Heaith 43:1247-1257. 1953. 17. Gordon. T.; Crittenden. M.. 3nd Kaerszel. 'A .: End Resuits and Mortality Trends in Cancer: -L Cancer Mortality Trends in the United States I5v-.9cc. Nation?.! Cancer Institute Monograph No. 6. L.S. Department of Health, Education and Wel tare. Public Health Sendee. 1963. ie. Gover. M.: Trend of Recorded Cancer Mortality in the Death Registration States of 1900. 1 t ! 524 .{/ft.///: '!.< i'! U/iA t ./. U.Tii from 1900 to 1935, U.S. Public Health Service. Public Health Bulletin Xo. 24S, 1939. 19. Griswold, M. H.; Wilder, C. S.: Cutler. S- }., and Pollack. E. S.: Cancer in Connection 1935 1951, Hartford, Conn., Connecticut State Depart ment of Health, 1955. 20. Haenszel, W.; Marcus, S. C., and Zinunerer, E. G.: Cancer Mortality in Urban and Rurni Iowa, Public Health Monograph Xo. 37. U.S. Department of Health, Education and Welfare, Public Health Service. 1956 21. Haenszel, W.: Cancer Mortality Among the Foreign-Born in the United States. I. Xar. Cancer Inst. 26:37-132, 1961. ' 22. Heller, .1. R.: Cutler. S. and Haenszel. W. M.: Some Observations on the Epidemiology of Cancer in the United States, J.A.M.A. 159:1628 1634, 1955. 23. Hueper, W. C.: Occupational Tumors and Allied Diseases, Springfield, 111., Charles C Thomas, Publisher, 1942. 24. Hueper, W. C.: Environmental Factors in the Production of Human Cancer, in Cancer, London, Butterworth & Co. Ltd.. 1957, Vol. 1, Chap. II, pp. 404-496. 25. Levity M.; Haenszel, W.; Carroll, B. E.; Gerhardt, P. R.; Handy, V. H., and Ingraham, S. C.: Cancer Incidence in Urban and Rural Areas of Xew York State, 1. Xat. Cancer Inst. 24:1243 1257. 1960. .\ ` 26. Lombard, H. L., and tearing, C. R.: Cancer Studies in Massachusetts: 3. Cancer Mortality in Nativity Groups, ]. Prevent. Med. 3:343-361. 1929. 27. Mancuso, T. F.: Occupational Cancer Survey in Ohio, in Proceedings of Public Health Cancer Association of America, October, 1949. pp. 56-70. 28. Mancuso, T. F., and Hueper. W. C.: Occupa tional Cancer and Other Health Hazards in a Chromate Plant: A Medical Appraisal; I. Lung Cancers in Chromate Workers, Jndustr. Med. Surg. 20:358-363, 1951. 29. Mancuso, T. F.: Macfsrlane. E. M., and Porterfield, J. D.: The Distribution of Cancer Mortality in Ohio, Amer. 1. Public Health 45:58-70. 1955. 30. Mancuso, T. F., and Coulter, E. J.: Cancer Mortality Among Xative White. Foreign-Bom White, and Xonwhnc Malt Residents oi Qhi0Cancer of the Lung. Larynx. Bladder, and Central Nervous System, I. Xat. Cancer Inst. 20:79-iq; 1958. " 31. Moriyama. I. M.. and Gumlnick, L.: Qc. cupational and Social Cass Differences in MonaVmin Trends and Differentials in Mortality, \*ew York, Milhank Memorial Fund, 1955, pp. 61-73. 52. Ptitno, M. E.: Geographic Stu<iv of Car,cPr Prevalence Within an Urban Population, J>u)>|;c Health Rep. 69:705-715, 1954. 33. Patno. M. E.: Mortality and Economic Levd in an L'rban Area. Public Health Rep. 75:841-5^ I960. '' 34. Perrott. G. St- L, and Collins, S. D.: Relation of Sickness to Income and Income Change in \i) Surveyed Communities: Health and Depress!,^ Studies Xo. 1. Method of Study and Genera! Result* ior Each Locality, Public Health Rep. 50:595-6" 1935. 35. Proceedings of Conference on Epidemiology Research in Occupational Health, I. Occup. Med. 4:562-650. 1962. 36. `flie Registrar-General's Decennial Supply, mem, England and Wales, 1931, Part IJa. Occujs- tiottal Mortalitv, London. His Majesty's Stationer* Office, 1938. ' '* 37. Steiner. P. E.: Cancer: Race and Geography. Baltimore, The Williams & Wilkins Company, 1954. 38. Stocks, P.: Regional and Local Difference in Cancer Death Rates. Studies on Medical ani Population Subjects Xo. 1. General Register Office. London, Her Majesty's Stationery Office. 1947, p. 46. 39. Stockwell. E. G.: Socio-Economic Status tud Mortality in the United States: Renew of the Literature. Public Health Rep. "6:1081-1086, 1961. 40. U.S. Xational Office of Vital Statistics: Cancer Mortality in the United States, in Vital Statistics Special Reports. Vol. 32. Xos. 2, 3, 4, :,j>5 S, 1950. -M. U.S. Xational Office oi Vital Statistics: Death Rates for Selected Causes by Age, CcA r. and Sex: United States and Each Stats. 1949-Uh\. in Vital Statistics Special Reports, Vol. 4, X.-. 16-26. 1959. i % \ i \ S t 4 i I i i Josep. editor at training His 0 been prir neglected the pauc; rnent me: diseases i since he j< t`sin- group pi o: anhri As hi :i::c 'r.surc HE.-LTk T ar^oiuiivntat Fluonoe v. es :o M?. M; 2.5 M,-M> *2 3. 1.26 jv 4.22 1.43 :i 2.2A 1.00 :s LA 0.44 !- . >4 :0 0.75 2 3-M 1.54 .45 4. 1.60 1.73 0.71 5 2.10 0.34 .j* 3.92 1.57 27 1.67 0.67 A 3.21 OSS 50 I.S6 0.75 '55 1.96 0.79 a 2.62 1JM IS 1.SI 0.72 3-35 1.61 52 2.37 US 1.56 O.lfi ) 2.06 043 US 0.47 U2 0.73 s 1-32 0.33 II 1.36 0.79 -2 OS? oas IS 1.93 0.77 3 1233 0.49 '2 3.49 1.40 * 2.73 1.19 A 4.41 1.77 0 2.74 0.S6 i 0.76 OJO 2 US 0.4S 7 o.ri 0.34 1.39 0.76 6 1216 0J0 7 0.50 0231 a 2.10 0.84 4 0.6* 02T7 . 3.34 us 7.73 3-09 1.75 0.71 . .3 2.64 . 5.32 0-W 1.06 3253 0.20 .2 2.31 4252 0.50 1.13 3253 02 -- 76 i Methodology in Industrial Health Studies The Demographic Approach THOMAS F. MAN'QUSO. M.D., M.P.H. AND ELIZABETH J. COULTER. Ph.D. COLUMBUS. OHIO There are many opportunities tor investi gators in the field of industrial health to design and conduct studies to evaluate the relationship between chronic illness and occupation. This area of study poses, how ever, some complex and difficult problems which require varied approaches. In 2 papers consideration will be given to the application of epidemiology to in dustrial populations within the framework of die demographic and prospective or cohort approaches. This paper is concerned with dte demographic approach, which is of basic importance in industrial health studies. Demographic studies identifying charac teristics of the general population and the Submitted for publication Aug. 17, 1962. formerly, Grief, Division of Industrial Hygiene (Dr. Mancuso), and Chief Statistician, Division oi 'ital Statistics (Dr. Coulter), Ohio Department Health. pattern of disease distribution, as derived from vital statistics and other sources, repre sent an initial approach in epidemiologic methods relating to industrial populations. Such studies involve geographic areas, sex, color, age, country of birth, and socioeco nomic status. Demographic comparisons help to deter mine geographic and population character istics associated with mortality or morbidity due to specific diseases. Urban-rural com parisons, tor example, which indicate rela tively high mortality in urban areas or in particular metropolitan counties represent initial clues to environmental influences, such as industry and air pollution, from which more refined epidemiologic studies can be derived. They further help to identify sex, color, nativity, and age groups with relatively high risk of specific diseases or causes of 516 aRaimis of estiroxmextai. ,n Table l.--Avcrapr AnntuU Aac-Adiiuled Death Raid* Per 100,000 PopulatiQ^ n H'wlf Male and U'lute Femalef Residents of Each of the Eujht Metropolitan c,!u*{- ' r;:dxOLOG^ -- Neoplasm Site (and Cause of r>- Geographic Am State of Ohio 8 Metropolitan coaaties Cuyahoga FrantsUs Hamilton Lucaa Mahoning Montgomery Stark Summit All Malifimnt Neopliismj 0 40006) ) All Malienant Neoplasms Eu-cpt Tlioee of Ilreast A- Heiiroductlre Organs !140-165, IS0-305) i Fsophaxus ,150) 141.02 1S5J3 167JO HT. 163.07 153.60 1G3.00 136.G2 139.13 137.15 134.34 148.41 159.93 141J0 156.03 147.11 155.97 129.06 132.45 130.76 345 446 6.10 3J3 4.96 5J3 4.92 4.12 3.68 3.41 Stomach '131) 11.69 13.60 16.59 948 946 11.75 16.31 941 1045 1040 Pascreus (157) Larynt (161) White Males 7.07 240 7.56 249 843 347 7.08 341 7.17 340 7.43 244 9.18 2.95 6.46 2.08 7.10 3.18 6.91 2.09 Lnr.j. * `rU. . !0* 3Log U* 4l *341 41.11 3&JI 34J> BUii-ier * Other Urinary Organs tlSi) ( .; . i.** . j* - ij* - - a* i.ta A* jr.l $ t :.. tS %* ** . 4.44 <.47 161 4.97 116 417 113 4.77 White Females 1J3 State of Ohio 140.20 70J6 OJO 5.68 3.92 0.16 AM i l.0 8 Metropolitan counties Cuyahoga Franklin Hamilton Lucas Mahoning Montgomery Stark Summit 145JS 73.18 0.69 640 i.Qi 0.17 a* 146.71 151.75 155JO 140.14 146J9 134.51 125J4 143.77 76.06 74J5 73.71 67J4 S0J3 64.36 65J8 7647 044 0.601 0.401 0.631 049 | 0.661 1.031 041 | 8.14 441 5.13 441 1042 4.19 444 6.00 4.12 0.081 All 4.41 040| AS 4.41 040 | AO 4.42 0.13) 44? 3.54 040 J ;s 2JS 0.10 | 4.64 2.42 0.14 | as 4.93 0411 iSA A - . Ui ' ; u \m IM t* - ad 144 IcO l.ffl 146 )sa l.6 l.C ITS ' j7WN*wbue popolst; **uan Ji&scn. C 'Mortality rates computed on tlio luisis of mortality data (or th# specified typo of malignant neoplasm for white eu* 4' t* r-eUTpoli--n couc and white females, classified by Syear croups, available in tbe Ohio Department of Health. and population data oiiuM w ^ if'sUjm putm'-atioo or me I areragmg population counts given by the Ooecnniii United States censuses of 1950 and I960 for the correspondIn* set, ' | * wytems of lymphs; age rww- Age adjustments were nude by tbedirect method oo tb*basis of th* at*distribution otthe total populutan v< tu >. <U<4 United St3tes as shown by the i960 United Sum census. ' ,v } f2a doth raie is based death. Such data provide a basic reference source tor studies of industrial workers which, in turn, require similar classification by population characteristics. In this manner, it is possible to identify and eliminate the effect a particular variable, such as age. may have on mortality patients observed in specihe industries. Mortality rates of in dustrial workers by population characteris tics. such as sex, color, and a>e. help to determine the possible cried uf selective employment practices relative tu such popu lation characteristics on the morbidity and mortality experience cf a given pi:,;;: <* industry. Geographic Location In studying geographic differences as lus epidemiologic leads in industrial hci>;V studies, a first consideration is urban-tv* comparisons. These may provide initial elsr* relative to industrial and agricultural iations and iurnish the basis ior fur1-*epidemiologic efforts and other lyj-c* studies pertaining to air pollution. nomic factors, and medical care. fV,:'. 6. April. i*tS Ft l. > i. 1; ^ ,i c j ; 'jar* -Sac to cancer wivir while, /ore*. vwwS.'.te male res: .'jy*.* 4rra5 of Ohio. *<w* >var?. l?u V-rtaJitv rases w %sm t{ mortality 7a l^artnert of ^fjcgniiaSior.. C:2i::. K v>5 r.:-t&.r -* she jOA) ^'.5 j.-v-ha-Jrwri wtr? ma.i; r; -W **.* I J." i-&>4 jr.vcliti.-.r. c: a* .v ; "* ;--f t. -:t: ^>-w Wi' nr. * Usr..*. --. :?! -. 7T ^a-7 ' fyp 'i I,- :icn Due M .) Code) L:y>*. Rrencnus, V Truth1.-a JC2. !) 5 { j 1 36.02 40.67 41.2$ 42.79 <5.53 43.9! 41.13 36.21 36.36 34.39 S.34 5.70 6.19 5.63 6.45 42 t.a .-21 jrhitt rakit' .1 obtained by color, and .ion of the 'ii:TiiODOLOOY--lSDCSTRIAL UliALTH STL'OILS 517 \(,uicnant Xeopiasms of Selected Sites Among of Ohio. 25-64 Years of Age, 1950-1960 ffdoey oso> Bladder <fc Other Urinary Or?a:)i 7IS1) Brain it Ocher Parti of CNS ;i93> Leukemia dr Aleukemia (204) 3.76 4.05 6.40 6.63 4.07 4.44 6.64 7*0 4.05 4.47 6.92 7*9 3.36 3.61 3.96 6*9 3.7S 4.97 (35 6.76 3*5 5.16 7.15 7.67 5.97 4-5? 6.73 8.16 4.03 3.13 6.40 6*3 5.07 4.42 6.47 6.72 3*8 4.77 7.70 7.90 1.72 1.33 3.96 4*2 1*1 1.60 4.13 5.05 1*S 1.34 4.47 5*3 1*1 1*0 4*0 5.15 2.48 1.69 3.42 5.15 1.97 1.96 430 4.40 1.46 1.03 2.77 4*1 1*1 1.46 3.43 4*5 14* 1.42 3*3 4.S6 o.ss 2*9 5*1 5.16 t Tbe ooBwbite population croup ms defined to iaclode N'epoo. Indian. Japanese, Cbifioe, and other nonwhlte races. ; Tbe metropolitan counties were defined u tbose with cities of100,000 population ormen ia 1M0. | Neoplasms of lymphatic and hematopoietic tissues are in eluded. Tbe death rate is based on a frequency of less than 10 deaths. Our observations in Ohio cancer studies have emphasized the importance of such urban-rural comparisons, particularly by population subgroups. These studies have shown, for example, that (l) an "urban'' influence existed for certain cancer sites and (2) lung cancer was relatively high among the foreign-bom of both urban and rural areas ('Figure).*'9-30 Other studies'have also emphasized urban-rural and related geo graphic considerations.20'2'-22'45 Comparisons of cancer mortality in small geographic areas such as counties may pro vide further epidemiologic leads. Table 1 provides county comparisons of mortality due to specific cancer sites among white male and female residents of individual metropoli tan counties of Ohio for the period 1950 1960. The table reveals individual county variation in cancer mortality for the specific sites considered. Lung cancer mortality among white males was, for example, rela tively high in Hamilton County and relatively low in Summit County. Stomach cancer mortality among white males was, on the other hand, relatively high in Cuyahoga and Mahoning counties and relatively low in Franklin, Hamilton, and Montgomery Coun ties. These observations serve as a basis for further epidemiologic efforts. In individual county comparisons it must also be realized that a single industry may not appreciably affect the specific disease rate in a county just as a high incidence of a specific disease in a single work department ''ll :es a> bash: >.1 beaith :rban-rurai nirial clues jral ;>opti.r further tvots ot . socioeco- Average annual age-adjusted death rates due to cancer of the lung among native white, foreism-bom white, ana turret MU nonwhire male residents of urban and rural areas of Ohio, who died in the state, ages 25-64 years, 1947-1951. Mortaiiry rates were computed on the rmtCMK HM basis of mortality data available in the Ohio Department o: Health for the speci fied population, classified by age groups 25 44. -?-:4, and 55-64 years, and population WWJT* data :r. the 1956 U.3. Census for the tor- responding population groups. Age a-djust- r.-.;r.:s were made by the direct rrahed or e ie to ao *o the :;.si$ oi the age distribution of the total population of continental United aart aa ee,ooo wovutww States as shown by the 1950 United States Census. The nonwhite population group was denned :c include Xerroes, Indians. Japanese. Gtinese. and other nonwhite races. The native-born -group included persons xt-r. ir. the United States or any of its territories or possessions. The classification c: urban and rural ars?~> was based on the definition of such areas used in the 1950 United States Census. '- Mottease--Coulter V';' 1v.v j - - 51$ .-i/wri/;-'.' or- health mav not affect the rate tor the disease of a residence in the county of residence at death hirsje industry because of the dilution factor. as shown on the death certificate. Another This point was demonstrated in a study of approach is to compare place of residence at the lung cancer mortality rate in Lake death with the birthplace as specified on the County, Ohio, although previous studies 27,29 death certificate, it one can assume that an had demonstrated an abnormal mortality rate individual bom in a given county who died for lung cancer in a chromate plant located in that county lived a major portion of his in that county. When a large group of life there. Both of these approaches are of industries are located in a circumscribed area, assistance in comparing mortality rates due such as a county, and are engaged in the same to long-term illness in different geographic essential manufacturing processes or prod areas as they relate to duration of residence, ucts. it can be expected, however, that the which is important because of the latent pe disease partem of the county will reflect this riod between initial exposure and develop concentration in the area if an environmental ment of the chronic disease leading to death. factor is involved. In considering residence data it is essential Residence Duration In studying diseases with long iatent to keep in mind 2 points: (1) the inter relationships of residence and migration patterns with other factors such as socio periods, it should be noted that the current economic phenomena, employment practices, place of residence of population groups under study may not be the place of environ mental exposure and therefore consideration and medical care and (2) geographic varia tions in the distribution of chronic illnesses depending on the latent period of the par must be given to long-term residence ticular disease. histories. This applies to the following types of current residence: death residence, place Sex of retirement, place to which an individual moves for special medical or surgical care, place to which there has been significant migration because of attractive climate and iob opportunities, and places within a city to which there have been population shifts with respect to race, nationality, and socio economic status. These types of shift in residence, singly or in combination, bring about changes in the age distribution of the population, as well as in the types of illness and death, which thereby affect the illness and mortality pattern of a given locality. Residence history data are of basic impor tance in studying the possible association of such variables as occupation, atmospheric pollution, smoking, morbidity, and mortality. There are various methods of exploring residence duration in epidemiologic studies. One is to obtain residence history data for specific population groups, by such means as checking city directories and employment records and querying relatives, friends, and In industrial health studies, sex compari sons are essential in epidemiologic con sideration. When there is employment selection by sex within an industry', the dis tribution of illness by sex may provide the means of excluding certain departments and exposures, in this way refining the epi demiologic study. In some industries and occupations, women represent the predominant labor force. Under such circumstances, patterns of long-term illness or mortality in female employee populations may be studied in com parison with similar data on women in other industries and women not employed in industry. It is aiso of special interest to consider whether the occupation or pattern of living of the husband affects the illness and mortality experience of his wife. Some cancer mortality data for Ohio shown in Tables 1 and 2 reveal characteristic patterns of sex difference in cancer mortality. other associates. This approach car. provide The tables indicate, for example, that cancer information, for example, on the duration of mortality was generally higher among males V01. 6. Apr:!. J96S Sr' .rijO:-OLO<.y. \!l.` '. females of each col. to both sexes r. ei were more nr, lh,n for others. The .e hand, a considerabi Xoiia*n> Site (A Cause < \taJfp,*ot ?fP;a5nl5- incluiioe 3K ` * hemawP^ietic tissues U-JO-Xi gureal easier <fc pharynx mo-MS; PijestK* organs & peritoneum (I ? Esophagus (140) stomach am Small intestine, incl. duodenum Largeintestine, except rectum o Rectum 05*) Biliary passage A liter, stated s Lirer. secondary A unspecified ( Ptsofas (J4D Peritoneum (1SS) Oospeeifled digestive organs (lif Rmpiratofy system (190-jM) Most, nasal cavities, taiddie ear. Larynx (161) Lung, bronchus. A uachea (1G2. Mediastinum (16*) Thoracic organs, secondary <jk Breast A- genitourinary organs ti': Breast (170) Cervix uteri (in) Corpus uteri (TO Other A unspecified parts ofuteOrarr. fallopian tube. & broad : Other A unspecified female gen:: Prostate (ITT) Testis (ITS) Other A unspecified male gentta Sidney (ISO) Bladder A- other urinary organs Other A unspecified sites (ISO-199; Meianome ofskin (190) Other malignant neoplasm of si Eye'592: Brain A ocher parts of nervous: Thyroid giand (!94> Other eodocrtne giandj (195) Bone `196) Connective tissue .597) Secondary A unspecified nuiiy nodes .;!>$ brspocified sites '199' Lym-rati: A henwioroiefie tsssi. Lymp.-.ni: & hematopoietic LsuccCis A aieukemia :2H) * Mortality r.vej were romputet-rtot V. v.ai OVaf.V.O; >( C'T.IO. .;a.; rapt/- a:--, as* aiyis-.r.*--.? ` ('sued States as snon-a r-y in* - The nc-ntrr.'.te ponumtmv. rr-u : The metrono.'ttats ur.;:c <* i Tiiv otrito n.te is based ot. - ;o- .1/>- .'if.'iVr `i'.i.a.TH at <Jeath, Another idence at n the mat an vho died a of his -5 are of ates due graphic evidence, '.rent pedeveiopo death, essential ie interigration s socioractices, ic variaiiinesses the par- ompariic conloyment t1-- dis- the s and ie epi- pations, : labor r.T;S Ot :'e;naie :r. :on> to :n res: to - Ohio r:?..:ry. ranee: males si0HOr>OLOCV^!M>CSTHlAL HEALTH STL'/'IES 519 [han females of each color for cancer sites common to both sexes and that these dif ferences were more marked for some sites than for others. The females had, on the other band, a considerable mortality due to cancer of the breast and reproductive organs, particularly in middle age. Individual metro politan county variations by sex shown in Table 1 may be useful in selecting those cancer sites among the male population which Table ---Average Annual Age-Adjusted Death Rates* per 100.000 Population Due to Malignant feoplasms of Specified Sites Among White and Xonn-hite* Male and Female Residents of the Eight Metropolitan Counties of Ohio,* Ages 25-64 Years, 1950-1960 "' i ij Neoplasm Site (.& Cause of Death Code) Total Males Whitt Non'wbJi* Females White Noorrhtte jdalignant neoplasms. including neoplasms of iymphatie ^ hematopoietic tissues (H0-20J) Bucoal cavity A pharynx (140.148) p'gestive organs A peritoneum < 120-150) Esophagus uso) Stomach (131) Small intestine. Inch duodenum (132) Large intestine, except rectum (133) Rectum (134) Billarv passages A liver, stated as primary site (135) User, seeoodary <t unspecified (133) Pancreas (157) Peritoneum (158) Unspecified digestive organs (138) Respiratory system (16W65) Heat, nasal cavities, middle ear, A accessory sinuses (160) Larynx (181) Lone- bronchus. * trachea (182. 163) Mediastinum (164) ^Thoracic organs. secondary (185) Breast A genitourinary ortans U7W81) Breast (170) Cervix uteri (171) Corpus uteri (172) Other it unspecified parts of uterus (173.174) Ovary, fallopian tube. & broad ligament (173) Other A unspecified female genital organs (176) Prostate (ITT) Tatis (173) Other & unspecified male genital organs (179) Sidney UW) Bladder other urinary organs (1S1) Other A unspecified sites (190-199) Melanoma of shin (190) Other malignant neoplasm of skin (191) Eye (19J) Brain A other parts of nervous system (193) Thyroid eland (190 Other endocrine glands i!93) Bone `196) Connective tissue fl97"> Secondary A unspecified malignant neoplasm of lymph poces 1SS0;r** t; unspecified sites '199> Lymphatic <t hematopoietic rusur* '200-2031 Lyxpcatie A hcaatcpoieuc ttxxue* ;200-203,203) Leukemia 4; ilsikin"i :2C>4} 146.23 3.99 47.94 3.36 1041 0.46 14.32 6.34 3.01 2.99 6.11 045 029 26.18 022 126 2322 042 0.16 48.08 1S.06 S.73 0.54 3.97 MS 0.43 246 0.64 0.13 2.93 .120 13.75 1.13 0.57 0.16 3.27 0.67 0.43 1.03 0.33 0.26 5.68 MJ1 S.22 6.09 133.33 6-53 53.36 4.68 12.60 0.44 13.96 7.33 239 342 7.58 0.6$ 0.24 44.SI 047 249 4047 0.49 0.19 15.43 0.28 _ _ -- _ -- 4.95 1.46 046 4.07 4.44 17.76 US 047 0.16 6.94 0.5% 0.33 1.26 041 046 i.43 17.44 10.14 7.30 213.5S 6.98 8840 1449 29.03 0.34 | 13.10 743 3.79 6.71 9.97 041 { 040 33.78 1.04 3.08 M46 0.32 i 0.30 4 23.39 0.09 { ___ _ _ _ _ 1448 0.19 \ 0.73 1 4.47 343 15.72 0.53 ) on i 0 4.21 0.61 | 0.63 i 1.09 0.62 * 0.23 5 1003 1742 10.60 6 82 143.23 1.43 37.43 0.09 6.30 047 1440 5.11 3.44 2.28 4.04 0.49 0.21 649 041 0.17 3.70 0.17 0.14 73.48 33.49 14.98 1.01 6.82 12.99 0.7S -- ___ _ 141 1.00 13.29 0.95 049 o.is 4.13 0.09 044 0.75 041 044 5.05 It 43 6.20 5.05 2C8.93 146 60.06 347 13.40 0.4S i 1742 7.9$ 341 349 9.00 042| 1.02 10.08 0.22 } 043 8.75 046 { 0 107.03 32.27 3641 149 1742 11.71 149 -- __ _ 1.94 4.60 17.82 0.37} 1.1$ 0 2.SS 1.17 0.67 { 1.25 0.73 0.46 i 9.17 12.08 740 4.45 * Mortality rata war* computed cr hast* o( mcnsfity da.ta available i: i the Ohio Dapicrrmen; of Health for residents of the 8 Tcetrouo'itaa count** otOhio. classified be s. color, and .'..year at stri;-*. isd ;>opuinoo data obtained by arersftar peculation ~-*r. .\c -.fte-deeennuil United States censuses of jf>fv and iSr*' tor the ccrrefPondinc set. rotor. arid -?e croupe it. this eec- .Tipaic Age adjustments were mace iy mo d;rt memod or. the basts of cite at distribution of tnt total population of we Tcrfd State a* shows hr the lSon "aired Star*? Ccnsns. 'Tn* mnv.-.nte population croup Ci- defined include Secrt-es. Indians. Japanese. Chinese. and other tionwhtu race*, t The metropolitan eountie* were defined rs tno* vitr. of i(,`.<><r. po;n;iat;on or more in j$io. * the dents rate ts based or. a frequency of iess than tn oeatns. Me.ncuse--C<i.;f.-r t \ I i t 520 ARCHI! US OF HXl'lRO.X.'.lEXTAL \l ETH OPO LOO'--.' Table J.--Aternge Annual Death Rates* per 100.000 Population Due to Malignant A'eoptn, of the Eight Metropolitan Counties of Ohio.* 26-64 Years of ^ of X-h'fcd Sites .Ami>n Chiss-.ner. oy Age. jv.v- - Age Group Neoplasm Site (a Cause of Death Code) 25-64 25-29 3044 35-39 40-44 *5^9TM"""" 50-54 55-59 All malignant neoplasms, including 1M.J1 15.44 19.50 34.30 58.3$ 123.01 228.95 381.34 I ' ' neoplasms of lymphatic A hemato poietic tissues (540-205/ Esophagus (350) 4 .S3 0 0 0.41 { 0.63 -.3.24 T.sO :3.:r Stomach (151) Pancreas 057) Larva.* 061) Lung, bronchus. A trachea (162,163) 12.51 2.M 40.51 0.4$ | 0.1$ | 00.77 0.SI 044 i 0 2.07 1.88 0.52| 0.29 5.27 4.32 3.59 O.S$ 1443 $.07 641 2.55 33.60 : j 15.05 12.50 4.70 3-..:3 Is 5$ s.ir ICs.ss l K-'daev u$0) 4.03 0.1$ \ 0.1* $ 0.M 2.19 3.45 J 5.S9 :r." !Biadder A other urinary organs (181) 4.41 0.08 i 0.17 l 0.17 1 1.25 2.75 $.. Brsin A other parts of nervous system 6.78 Ul 2.19 4.17 4.70 73% 13.15 0931 Leukeati* A aleukemia (204) 7.28 3.04 2.32 3.8$ 5.26 531 10.07 13.90 * Mortality rates were computed oo tbe basis of mortality data available is tb Ohio Department of Health and population obtained by avenging population counts given by thedecensial United States censuses of IfiSO and i960. Ace adjustments were made by tbe direct method on the basis of the ace distribution of the total population of the United States as shown by the 19 United States census. t The definition of "white" males excluded Negroes. Indians, Japanese. Chinese and persons of other nonwhite races, j The metropolitan counties were defined as those with cities of 100,000 population or more in 1050. | The death me is based oo a frequency of less than 10 deaths. individual counties : more specific envin merit further study, particularly in reference Table 2 illustrates color variations bv 1 to occupational or other environmental ex specific cancer site for each sex as found in The large variat: specific causes by a posures peculiar to males. Color Color comparisons serve to separate some cancer studies of the 8 metropolitan counties in Ohio for the period 1950-1960. The color differences were particularly marked for cancer of the digestive organs and peri ] the age cornpositio groups by geograph the necessity of ac mortality rates. of the variables to be considered and to delineate avenues of investigation in epi demiologic studies. Although the environ toneum, and for cancer of certain reproduc tive organs. Marked nonwhite excesses were observed, for example, for cancer of the j Table 3 ,provid< mortality by specific males residing in th ment of nonwhites has been identified to a certain extent, the epidemiologic potential of this population group has not been ade quately explored. esophagus and stomach for each sex; for cancer of the cervix, and other and un specified parts of the uterus, excluding the corpus uteri, among females, and for cancer of Ohio in the peri rates generally she with age which w; some sires than fee In considering chronic illness, one must of the prostate among the males. A nonwhite for most of the can keep in mind, in addition to the latent period of the disease, the pattern of migration and employment of nonwhites over tbe years. Xonwhites may be affected to a different ex excess for cancer of the lung was also found for each sex. On the other hand, relatively higher cancer mortalitv among the whites than among the nonwhites oi each sex was , j tor example, negif age groups. 25-3-* v tance. of varying later years. Mona' tent than whites by selection tor certain found for melanoma of the skin and cancer ; ami aleukemia, an. i types of work, shifts of jobs, and exposures oi the brain and other parts of the nervous , other parts of the within the same industry over a long period system. An earlier study revealed differences the other hand, hit; i of time, as well as by changes to other in dustries. This inffuences consideration of disease development in the nonwhile race and emphasizes the importance of duration in cancer mortality among native anJ foreign-bom subgroups of the white popula tion which brought out the importance of sites at ages 25-3tr.arked rise with . other site? studied There are a tv. of employment in industrial population stud studying nativity in coior comparisons for ies. These 'actors provide indications of specific types of cancer, such as cancer of differences in cultural and living patterns the lung and of the brain and central nervous ? that may merit further study by other epi system.50 Further refinements showing coior > demiologic methods. as well as sex and nativity differences by i'oi 6, April, 2963 *'- .jftQDCLGGY--lXL'L'STRlAL HEALTH STUDIES 521 . ejected Sites H'm'J.-? Male Residents ffcSSw by Age. 19S0-JM) 5>-M 25.95 i.ro . 67.12 5.J9 lC.rS io.o: S5-J9 3SI-34 132*7 31.13 3S.-4S m: 103.S3 io.s: 11.90 13.13 13.90 0044 *3330 AtC - Adjusted 23-04 153.23 23.IS 01.01 31.40 11.39 167.39 li.SS 23.0S" ir.n 22.4$ 4.SO 12.00 7.56 2.S9 4057 4.07 4.44 6.S4 750 individual counties may provide the basis for more specific environmental studies. Age The large variations in mortality due to specific causes by age and the differences in the age composition of various population roifps by geographic areas have emphasized the necessity of age-specific morbidity and mortality rates. Table 3 provides age comparisons of mortality by specific cancer site among white males residing in the 8 metropolitan counties of Ohio in the period 1950-1960. Mortality rates generally showed a marked increase with age which was, however, greater for some sites than for others. Mortality rates for most of the cancer sites considered were, for example, negligible in relatively young age groups, 25-34 years, and assumed impor tance, of varying magnitude, in middle or later years. Mortality rates due to leukemia and aleukemia, and cancer of the brain and other parts of the nervous system, were, on the other hand, higher than for other cancer sites at ages 25-34 years and showed a less marked rise with age than for most of the other sites studied. There are a number of special factors involved in considering age in relation to industrial employment. For example, dif ferent age groups of industrial populations may reflect variations in duration of employ ment. and occupational diseases may occur Mcncusc--Coulter in different age groups depending on the latent period of the disease. There mav be a shift of mortality to the younger age groups when occupational cancer is involved. In considering age it is also important to r t the dates of installation of control measures and to determine whether particular groups of workers were exposed before or after such installations. ' Employment requirements and policies may affect age patterns as well as types of illness in industrial groups. The employment requirements of certain industries may be such that the young and very strong are selected and therefore constitute the pre dominant population. Older age groups may, on the other hand, be proportionately greater in other industries, such as those with rela tively large proportions of highly skilled craftsmen. Retirement, seniority, and sick leave policies and practices may also affect the age composition of the work force as well as the scope of the epidemiologic study. The age at entrance to exposure has some bearing on the age at onset of occupational diseases such as cancer. . Country of Birth Nativity has become recognized as an im portant population characteristic for consid eration in epidemiologic and biostatistical studies.13"1--0*-'-3"-37 A point of inttrest with reference to conditions following mi gration of the foreign-born is the tendency observed for specific nativity groups to settle in particular areas in which they often live and work together, possibly associating with a specific industry or occupation group, and maintain the cultural characteristics of their nativity group. The migration patterns of the foreign-bom make i: important to consider any industrial and occupational concentration of migrating groups in relation to their geo graphic location and other environmental characteristics. The long latent period for some diseases, such as cancer, makes it necessary to consider factors in both the country oi birth and the place to which migration occurred in terms of circumstances which prevailed as lent: as .ARCH/1 'LS OF F..Vf'IROSMii,,YTAL WiAijrtj 20 or 30 years or more ago. Table 4 pro* vide? an exploration of this point in com paring lung cancer mortality among native and immigrant populations of Cuyahoga County, Ohio, the county which had the highest white male lung cancer mortality Table 4--Average Annual Aye-Adjusts & . Rates per 100.000 Pofiliation Among Stioa-d S* PaotpiruelatiWonkitGe roMupasleReasniddinIgmminigrOanhtio,3'a/,qe/'. Eight Metropolitan Counties and Cuyahoga County in Ohio, <id the Male Population of England and fT'afc;, and Italy, Ages 25-64, Selected Tears .4round'1950* rate of rtnv of the SS counties of Ohio in the period 1947-1931."* Table 4 shows that the mortality rate due to lung can cer among male immigrants to Cuyahoga Countv from England and Wales was con siderable' lower than among the male popula tion .of England and Wales but slightly higher than the death rate for this cause among the native white males of Cuyahoga County and considerably higher than the lung cancer death rate among the native white male residents of Ohio. The lung cancer mortality rate among male immigrants from Italy to Cuyahoga County was, on the other hand^similar to the male lung cancer rate in Italy but considerably lower than the Population Group & Period of Yews Native-born white miles reridlcj in Ohio. 1WM941 a metropolitan counties of Ohio, lMr-is.5! Cuyahoga County. Ohio 1W7.19M Immigrants from ail foreign countries re siding in Ohio. 1947-lSM S metropolitan counties of Ohio. 1WT-1SJ1 Cuyahoga County. Ohio, 1MM941 Cuyahoga County residents who immi grated from: England A Wain. 1947-1951 Italy. 194MB51 Resident population of: England and Wales. 1940 Italy. 1951 armee AnniJ A|<-Adjtutee Dtath R*,t Wf 10U.CCO Fopul.ltioi. A?es2i-J4 *45 y.t- ss.rs 36.6* 3441 3S.1I 31.75 18.61 54.48 1646 lung cancer death rate among native white male residents of Cuyahoga County and approximately the same as the lung cancer mortality rate among the native white male residents of Ohio. This is of interest in comparison with the results of a study by Eastcott of cancer mortality among native and immigrant groups in New Zealand.13 Socioeconomic Comparisons Just as geographic exposure, race, or country of birth may influence chronic * Mortality rates tor Enfland and Wales were computed on the bastsofmortaltty datasupplied by the World Health Organ ization and population data in United Nations publications: all other rates were computed oo the basis of mortality <jat-a available in the Ohio Department of Health and population data is the 1940 United States Census or estimated on the oasis of data in this Census by methods described on page Si of Re*, erase* 30. Aft adjustments were mad* by the direct method on the basis of the age distribution of the total population of the United States as shown by the 1940 United States Census. The nonwhite population group was defined to include Negroes, Indiana. Japanese. Chinese, and other noavhite races. The native-born group included persons born in the United States or any of its territories or possessions. Metropolitan counties were defined as those with cities of 100.000 population or more in 1940. illness and mortality patterns, so mav the socioeconomic status of a given population. Efforts have been directed toward deter mining the association of socioeconomic factors rsnc mortality in studies conducted in England and Wales and the United States for a number of years." .. Two approaches have been used in pro husband for married women. The other type of approach is to compare over-all mortality in geographic areas of different socio economic characteristics, such as census tracts, classified by one or more specific characteristics like rent or income. Eoth approaches can, of course, be used in study ing mortality by specific cause of death. viding socioeconomic comparisons of mor tality. One is to compare mortality rates by socioeconomic characteristics of the indi vidual. such as occupation, occupation and broad social class, and occupation of the In industrial health studies socioeconomic factors represent an index, to a certain extent, of the employment and exposure o: the individual worker, since occupational classifications are related to wages. This may References 1-3. 3. 10. 3i-S4. 36. 39. provide a means not otherwise available of Sol. 6, April. ;c/>; 34 ,,.rfi~oLO'rry^i:" 1 * ilv sorting or <lir \a 'CVcrooP^ytypeot " raj-urc- Laborers * ;",,'as management m * -.minted. , *'The influence of wor ,.rd< of the faiimy *. rare:. that industrial i their concern *0 ^.ents of the taimues Surr.ma This paper provides a the demographic appro, health of industrial, pc ! petition was made oi the on geographic location, m i county residence, du ** color, age- country gnomic factors. Parti- to the interreUuo factors. Conclus Demographic studies ; ^ce data on diseases an uon groups of different geographic location an alogic clues which are ft jng more refined invest Industrial groups. i \Y wish to express ou: technical and related assistsVcigel. Chief. Mrs. ElizarBicnbers of the Division o: Thomas F. Mancuso. M-. ajpationai Heaiih. Grad-.:; Health. Pittsburgh. Pa. i _ R2FERE. j 1. Backgrounds of Social Papers Published by the M New York, 19--1. 2. Ciocco, A., and Perr i Sickness as a Cause ot r Review. T Hist. Med. 3. Coulter. . an: vur ! Vital Statistic* by C=r.su* 7 1 Ass. 34:730-741;, tt'-c 4. Cutler. 5. an: Zt--r Mortality Trends ir. Car. to .Census:--7.':;..v- " : ,-rv; . hi.-," -t I'.Ti! , -T.OL"-jL'jJ ` --!X`-L V r,H. < H L l~S 523 ' vc' Death Sifted ' Male :?. lac '(1 :.*> p.sre xo u:or.. .A.rs .^sr ss.n n.:i S..61 MS 6.26 'JIlpUfKj 00 .aith Orgao. jtlicatiOQS'. rtality data population >n the bojta fSJ of Ref- method - 'tioo of tr.siis, groe. Tes. The i States counties on or more '-iZ type .`.rtailty >oc:o- census specific . Both ; study- .th. :riontic certain J.t.or.ai ' i; rnay .bit o: jjroatlly sorting: or differentiating occupa Cancer. Yational Gmcer Institute Monograph Yo. tional groups by type of job. department, and eNpo$ure. Laborers and skilled groups as u'ell as management may be grossly dif ferentiated. The infiuence of work income on living 6. U.S. Department of Health, Education, and Welfare, Public Health Service, 1961. 5. Dcnsen, P. M.; D'AJonzo, C. A., and Munn. M. G.: Opi>ortunitics and Problems in the Study of Cironic Diseases in Indusirv, T. Chron. Dis. 1:231-252 (March) 1955. '' . -ctandarcis of the family, such as nutrition, 6. Densen, P. M.: Epidemiology in Industrial housing, medical care, and education, also Health, in Transactions Publication Yo. 30, the 21st iI suggests that industrial health studies should Annual Meeting of The Industrial Hygiene Founda tion. Pittsburgh, Pa., Industrial Hygiene Founda extend their concern to illness and mortality tion. 1955. patterns of the families of workers. 7. Dom, H. F.: The Statistical Approach to the Epidemiology of Cancer, in Proceedings of the Summary This paper provides a brief discussion of Second Yational Cancer Conference. Yew York. American Cancer Societv, 1954, Yo!. II. pp. 1105 1117. * the demographic approach to studying the 8. Dom, H. F.: Mortality Rates and Economic health of industrial populations. Specific mention was made of the importance of data on geographic location, including urban-rural and county residence, duration of residence, Status in Rural Areas, Public Health Rep. 55:3-12, 1940. 9. Dom, H. F.: The Changing Incidence of Cancer Throughout Life, Bull. Y.Y. Acad. Med. 31:717-725, 1955. sex. color, age, country of birth, and socio economic factors. Particular reference was made to The interrelation of some of these factors. 10. Dom, H. F., and Cutler. S. J.: Morbidity from Cancer in the United States, Public Health Monograph Yo. 56 of the U.S. Department of Health, Education and Welfare, Public Health Sendee, U.S. Government Printing Office. 1939. Conclusion Demographic studies provide basic refer 11. Downes. J.: Chronic Diseases Among Spouses, Milbank Mem. Fund Quart. 25:334-358 1947. ence data on diseases and deaths in popula tion groups of different characteristics and geographic location and furnish epidemi ologic clues which are fundamental in guid ing more refined investigations of specific industrial groups. 12. Dublin, L I., and Vane. R. I. : Causes of Death by Occupation, U.S. Bureau of Labor Statistics Bulletin Yo. 507, U.S. Government Printing Office, 1930. 13. Eastcott, D. F.: The Comparative Mortality Experience from Cancer of Certain Sites Between the Indigenous Yon-Maori Population and Im We wish to express our appreciation for the technical and related assistance of Mr. William H. Veigel, Chief. Mrs. Elizabeth Everett, and other members of the Division of Vital Statistics. Thomas F. Mancuso, M.D.. Department of Oc cupational Health, Graduate School of Public Health. Pittsburgh, Pa. migrants from Great Britain, extract from Report of the B.E.C.C. (Y.2.) Branch, Cancer Registra tion Scheme, Wellington, Yew Zealand. Yational Health Institute, 1954. 14. Gafafer, \V. M.: Sickness Absenteeism Among Male and Female Industrial Workers. 1937 1946, Inclusive. Public Health Rep. 62:1338-1 Ml. 1947. REFERENCES 15. Gilliam, A. G.: Epidemiology in Occupational Medicine, Arch. Environ. Health 3:28!-2S7. 1961. 1. Backgrounds of Social Medicine: A Group of Papers Published by the Miibank Memorial Fund. .Yew York, 1941. 16. Gilliam, A. G.: Opportunities for Application of the Epidemiologic Method to Study of Cancer. Ante:. .1. Public Health 43:1247-1257. 1953, 2. Ciocco, A_, and Perron. D.: Statistics cn 17. Gordon. T.; Crittenden. M.. and Kaer.szel. Sickness as a-Cause of Poverty: An Historical 'A .: End Resuits and Mortality Trends in Cancer: Review. J. Hist. Med. 12:42-60 Yc. I) 193.". !I. Cancer Mortality Trends in the United States 1. Co-civ*:. E. J.. and Gurair.itk. L.: Analysis e: `&3v-;9cc. Yational Cancer Institute Monograph \ i;al Statistics by Cense* Tract. T. Arret. Statistic. Yo. 6. L.5. Department of Health, Education and A>* 54:730-740. 1959. ' Welfare. Public Health Service. 1961. -. Curler. $ T.. and E-ierer. ?.: End Results and \Z. Gover. M.: Trend of Recorded Cancer Mortality Trends in Carter: I. End Results in Mortality in the Death Registration States of 1900. '.icncnso--Coulter | 1 t \ AHi.tli: !.S "/ !R'. u- :ia.\J.7ii i 524 from 1900 to I955, U.S. Public Health Service. Public Health Bulletin Xo. 245, 1939. 19. Gnsu-olrf, M. H.; Wilder, C S.; Cutler. S. }., and Pollack. E. S.: Cancer in Connecticut 1935 1951, Hartford, Conn., Connecticut State Depart ment of Health, 1953. 20. Haenstel, W.; Marcus, S. C-. and Zimnierer, E. G.: Cancer Mortality in Urban and Rural Iowa, Public Health Monograph Xo. 37, U.S. Department of Health, Education and Welfare, Public Heaitii Service. 1956. 21. Haensael, W.: Cancer Mortality Among the Foreign-Born in the United States. /. Nat. Cancer Inst 26:57-152, 1961. 22. ffeffer, J. R.: Curler. 3. ant.' W. M.: Some Observations on the Epidemiology o; Cancer in the United States, J.A.M.A. 159:1628 1634. 1955. ' White, and Xomv'mu Male Residents of Ohio- Cancer ot the Lung. Larynx. Bladder, and Central Nervous Svstein, f. Nat. Cancer Insc. 20:79.{q$ 1958. ' ' ' 31. Morivnvnn. I. M.. and Gumlmck, L.: Qc. cupntional and Social Class Differences in Mortal^,., in Trends and Differentials in Mortality, York, Milhank Memorial Fund, 1955, pp. 61-73. 32. Patno. M. E.: Geographic Study ot Car,tfr Prevalence Within an Urban Population, Public Health Rep. 69:703-715. 1934. 35. Patno. M. E.: Mortality and Economic Levej in an Urban Area. Public Health Rep. TJ :8-4| I960. '' 34. Perron. G. St. and Collins, S. D.: Rela^ ot Sickness to Income and Income Change in it> Surveyed Communities: Health and Depress!,*, Studies Xo. 1. Method of Study and General Result* i i 4 l t t> i Toseo: editor Qt training- V His o been prtr. neglected the pauci ment nte, diseases i 23. Hneper, W. C.: Occupational Tumors avid for Each Locality, Public Health Rep. 50:595-622, since he y Allied Diseases, Spring-field, 111., Charles C Thomas. Publisher, 1942. 24. Hueper, W. C.: Environmental Factors in 1935. 35. Proceedings ot Conference on Epidomiofor> Research in Occupational Health, I. Occup. i ? the Production of Human Cancer, in Cancer, 4:562-650, 1962. London, Butterworth & Co. Ltd., 1957, Vol. i. 36. The Registrar-General's Decennial Supply Chap. II. pp. 404-496. 25. LeviarM.; Haensrel, W.; Carroll, B. E,; Gerhardt, P. R.-, Handy, V. H.. and Ingraham. S. C.: Cancer Incidence in Urban and Rural Areas ment, England and Wales. 1931. Part Hi. Occup*. tional Mortalitv. London, His Majesty's Stationery Office. 1938. '' 37. Steiner. P. E.: Cancer: Race and Geography. i of Xew York State, T. Nat Cancer Inst. 24:1243* Baltimore, The Williams & Wilkins Company, lpcj 1257. 1960. ' ., ` 26. Lombard, H. L., and Dearing, C. R.: Cancer ' Studies in Massachusetts: 3. Cancer Mortality in Nativity Groups, J. Prevent. Med. 3:343-361. 1929. 27. Mancuso, T. F.: Occupational Cancer Survey in Ohio, in Proceedings of Public Health Cancer Association of America, October, 2949, pp. 56-70. 28. Mancuso, T. F., and Hueper, W. C.: Occupa tional Cancer and Other Health Hazards in a Chromate Plant: A Medical Appraisal; 1. Lung Cancers in Chromate Workers, Industr. Med. Surg. 20:358-363, 1951. 29. Mancuso, T. F.; Mactarlane. E. M.. and Porterfield, J. D.: The Distribution of Cancer Mortality in Ohio, Amer. 1. Public Health 45:58-70. 38. Stocks, P.: Regional and Local Difference in Cancer Death Rates, Studies on Medical atvi Population Subjects No. 1. General Register Office, London, Her Majesty's Stationery' Office, 1947. ft 46. 3Q. Stockwfcll. E. G-: Socio-Economic Status aid Mortality in the United States: Review of ihLiterature. Public Health Rep. 76:1081-1086, 19t*L 40. U.S. National Office of Vital Statistic: Cancer Mortality in the United States, in Vital Statistics Special Reports. Vol. 32, Nos. 2, 3. 4, S, 1950. il. U.S. National Office of Vital Statistics: Death Raies tor Selected Causes by Age, Cekr. and Sex: United States and Each State. 194<MV:'.. % i 1955. 30. Mancuso, T. F., and Coulter, E. J.: Cancer Mortality Among Native White. Foreign-Bom in Vital Statistics Sj-tcial Reports, Vol. 4&, S>. 16-26. 1939. LY.r.k group p: ot armrt r.ruentr.?. As h: i