The Question of Circumcision

Cancer, Volume 13, Issue 3: Pages 442-445, May-June 1960.

Accuracy of Determination of Circumcision Status


CIRP logo Note:

This file does not include three tables and one figure.

THE RELATIVE immunity of Jewish women and, to a lesser extent, of Moslem women to cancer of the cervix has been attributed to the fact that Jewish and Moslem men are routinely circumcised.5 One statistical study that presented information on the circumcision status of the husbands by inquiring from the wives and, if uncertain, also from husbands, reported a higher percentage of control patients exposed only to circumcised husbands than for the cervical cancer patients. However the differenc was not as large as would be expected if the low cancer rate amongst Jewish women is solely due to their contact with their circumcised husbands. Two other studies in which only wives were questioned revealed no difference in this aspect.1,3 It, therefore, seems pertinente to review the manner in which circumcision data can be obtained and with what degree of reliability they may be regarded.


The study has been carried out in Los Angeles, Calif,, and in New York, N.Y. One hundred male and 100 female clinic and ward patients, all more than 30 years of age, were interviewed in each locality. In Los Angeles, the patients were interviewed at the White memorial Hospital; in New York, at Memorial Hospital and Bellevue Hospital in Manhattan and at the Veterans Administration Hospital in Brooklyn. In the case of the male patients, the patient was first asked as to his circumcision status and then carefully examined by a physician.


Among the female patients in Los Angeles, 36% did not understand the term circumcision or did not know the circumcision status of their husbands (Table 1). Twenty-four per cent stated that their husbands were circumcised, and 40% of the patients stated that they thought their husbands were not circumcised. Eighteen per cent of the patients were Negroes. Three per cent had a college education, 15% had attended high school, and the remainder had attended only grammar school.

Among the female patients in New York, 38% did not understand the term circumcision or did not know the circumcision status of their husbands, 22% thought their husbands were circumcised, and 40% believed that their husbands were not circumcised. The patients included 13% Negroes. Six percent had a college education, 21% had attended high school and the remainded had attended only grammarschool.

Of the men interviwed in Los Angeles, 83% were white, and 17% were Negroes. Eighty-five percent were born in the United States, 9% were born in Mexico, and the others were born in European countries. Twenty-three per cent had attended high school, 8% had attended college, and the remainder had attended only grammar school. Or had no education. Five per cent did not understand the term circumcision (Table 2). Sixty-six per cent of the patient stated that they were not circumcised and upon physical examination by one of us (E.L.W.) were found to be uncircumcised. Fourteen per cent stated that they were circumcised but physical examination showed them to be uncircumcised. Eleven per cent stated that they were circumcised but physical examination showed them to be circumcised. Nine of these 11 patient were circumcised after the age of 5, and all were circumcised prior to age 30. Four per cent of patients stated that they were not circumcised but examination showed them to have a short foreskin.

The male patients in the New York series included 13% Negroes. Only 61% were born in the United States; 11% were born in Italy, and 9% in Britain, with the remainder born in other European countries. Twenty-six per cent had gone to high school, and 9% to college; the remainder had attended only grammar school. Seventeen per cent of these patients did not understand the term circumcision. This is a larger percentage of unknowns than that found in the Los Angeles series, possibly because of the higher percentage of foreign born individuals in the New York series. Upon examination by one of us (S.L.) 66% were found to be uncircumcised when they so had stated. Six per cent were found to be uncircumcised when the patient stated that he was. One per cent was found to be circumcised when the patient declared that he was not, and 10% of the patients were found to be circumcised, in agreement with the patient's statement.

The following scheme may make it easier to compare the type of foreskin from study to study. The Los Angeles and New York data were quite similar in this regard though obtained by different observers. Fifty-two per cent of the Los Angeles men and 62% of the New York group not claiming circumcision had foreskins that nearly covered the glans (type A, Table 3). Twenty-three per cent of the former and 17% of the latter had foreskins that covered or nearly covered the glans (type A, Table 3). Twenty-three percent of the former and 17% of the latter had foreskins that covered at least one-half of the glans (type B). In 20% and 13% respectively the foreskin was at in "rest position" beyond the sulcus but after forward traction covered at least one-half of the glans without compressing it (type C). In 5% of the Los Angeles patients and 8% of the New York patients, the foreskin was beyond the sulcus and could not be extended over the glans without compressing it (type D). It would appear that these individuals had naturally short foreskins. About one-half of the group had foreskins that extended beyond the glans when pulled forward (type C) stated that their foreskins in earlier life had extended over the glans but that as they got older the foreskin retracted. In some instances the patients stated that they actively attempted to pull the foreskin back beyond the sulcus.


The review of these data indicates a sizable error in ascertaining the circumcision status by interviewing men, an error that is likely to be larger when the information is obtained from women. In the Los Angeles study 36% of the women did not know the circumcision status of their husbands. This clinic population should have been similar to the population studied by Jones et al.,3 who reported only one-third as many unknowns. This may reflect a difference in the manner in which the question was asked. Our present percentage of unknowns is similar to the percentage of unknowns that we reported in a previous investigation,5 which showed that in a female clinic population, one-third to one-half of the patients were not familiar with the circumcision status of their husbands. There may, of course, be even additional errors in the answers of those women who state that they do know; this would seem apparent from the fact that more women state that there husbands are circumcised than is apparent from data obtained by physical examination.

The extent to which data obtained from women can lead to possible erroneous conclusions has been reviewed by us another report.6

The inadequacy of the data based solely on interviews with women is further emphasized by the errors which appear in the answers of men as to their own circumcisions status. Dun and Buell have recently made observations similar to those found here.2 They emphasized the existence of naturally short foreskins and the fact that some surgical circumcisions are not complete. Lilienfeld and Graham have also reported a large error in reporting by men.4 Thirty-five per cent reported they were circumcised, and examination showed that, in fact, 44% were circumcised. This percentage of circumcision seems rather high when compared to what we have found in New York and Los Angeles. In the present study 24% of 200 male patients were unable to correctly state their circumcision status.

What is more, it appears that even physicians may be unable to tell with certainty whether the patient has at some past date undergone a surgical excision of the foreskin. This was found to hold true particularly for patients with type B and type C foreskins. From the practical point of view, it may make little difference whether the husband has been historically considered circumcised or not. The important consideration is the length of the foreskin in terms of its smegma-collecting capacity. It is with these thoughts in mind that we have introduced the classification illustrated in Fig. 1. This classification has been tested on 100 men in Los Angeles and 100 men in New York, 2 groups that were comparable by race, education, circumcision history, and circumcision status by conventional physical examination. These 2 groups were also found to be comparable by the classfication introduced, despite the fact that the observations were made by two different observers. This finding seems to demonstrate that this classification is a reproducible scheme by which different observers may obtain comparable results.

CIRP logo Note:

Later research has established that smegma is not carcinogenic.

It is recommended that future studies on cancer of the cervix employ a scheme descriptive of the penile prepuce such as the classification presented here. In individuals with type C it is important to consider the age at which the prepuce began to retract. If this occurred when the individual was past the age of 40 this type represents a situation of potential smegma collection identical with type A or type B. If however, the foreskin began retracting spontaneously or with manipulation by the individual before he was 20 years of age, the situation is similar to that of type D. If circumcision in fact protects against cancer of the cervix, patients with cancer of the cervix should have a higher proportion of husbands with type A and type B foreskins thando matched controls.

In summary, the present study demonstrated that in assessing whether circumcision or short foreskins tend to prevent cancer of the cervix, interviews with women alone are ineffective. If interviews with men are added to those interviews with women who do not specifically know the circumcision status of their husbands, the results are improved but still not perfect. It seems probable that questioning men, particularly those in the higher income groups and with better education, would result in more reliable answers. However, since cancer of the cervix occurs more commonly in the lower income groups, studies of these population groups are most pertinent. Since it has been shown that even upon questioning men a substantial error can be introduced, the most accurate way, though a difficult one, is to examine the husband or husbands of women with cancer of the cervix and control patients and to describe the type of foreskin in line with a scheme such as described in this study. Such a survey, though time consuming and difficult, can contribute further to our knowledge of the etiology of cervical cancer.


  1. Data on circumcision in men have been obtained from men and women. Knowledge of circumcision status tends to be better among the men than among the women.
  2. Even among the men who were questioned as to their circumcision status, a sizable error is noted.
  3. To obtain a definitive answer on circumcision status, a physical examinationof the men is mandatory.
  4. A scheme for recording the type of foreskin has been presented and is recommended as a basis for comparisonof future studies in this field.


  1. DUNHAM, L. J.; THOMAS, L. B.; EDGCOMB, J. H., and STEWART, H. L.: Some environmental factors an development of uterine cancers in Israel and New York City. [Abstr. in Program.] Seventh Cancer Concress, London, England, July 6-12, 1958; p. 281.
  2. DUNN, J. E., JR, and BUELL, P.: Association of cervical cancer with circumcision of sexual partner. J. Nat. Cancer Inst. 22: 749-764; 1959.
  3. JONES, E. G.; MACDONALD I., and BRESLOW, L.: Study of epidemiologic factors in carcinoma of the uterine cervix. Am. J. Obst. & Gynec. 76: 1-10; 1958.
  4. LILIENFELD, A. M., and GRAHAM, S.: Validity of determining circumcision status by questionnaire as related to epidemiological studies of cancer of the cervix. J. Nat. Cancer Inst. 21: 713-720, 1958.
  5. WYNDER, E. L.; CORNFIELD, J.; SCHROFF, P. D., and DORAISWAMI, K. R.: Study of environmental factors in carcinoma of the cervix. Am. J. Obstet & Gynec. 68: 1016-1052, 1954.
  6. WYNDER, E. L.; MANTEL, N., and LICKLIDER, S.D.: Statistical considerations on circumcision and cervical cancer. Am. J. Obstet & Gynecol. In press.

From the Section of Epidemiology, Division of Preventive Medicine, Sloan-Kettering Institute for Cancer Research, New York, N.Y.

In interviewing the female patients in Los Angeles, Calif., we were aided by Miss J. Barens and Miss R. Jacobs and in New York, N.Y, by Miss J. Roth.

Received for publication May 1, 1959.


The Circumcision Information and Resource Pages are a not-for-profit educational resource and library. IntactiWiki hosts this website but is not responsible for the content of this site. CIRP makes documents available without charge, for informational purposes only. The contents of this site are not intended to replace the professional medical or legal advice of a licensed practitioner.

Top   © 1996-2024 | Please visit our sponsor and host:  External link IntactiWiki.