Does age influence screening for colorectal cancer?

Does age influence screening for colorectal cancer?

Marie L. Borum


Colorectal cancer causes appreciable morbidity. It is the second most common cause of cancer death from malignancy in the USA [1, 2]. Intervention in those with localized disease results in better survival than intervention in individuals with more advanced disease [3]. US guidelines recommend annual rectal examinations (beginning at age 40), annual faecal occult blood (FOB) testing (beginning at age 50) and a flexible sigmoidoscopy every 3-5 years (beginning at age 50) [1, 4, 5].

There are few recommendations for colorectal cancer screening in older people, although the American College of Physicians has suggested that there may be little benefit in continuing flexible sigmoidoscopy beyond 70 years in patients who have been screened up to that point [1].

A study of elderly nursing home residents suggests that quality of life and life expectancy, rather than chronological age, should influence whether screening should take place [6].

This study evaluates whether patient age influences the use of screening methods by internal medicine resident physicians in an ambulatory care clinic.

Materials and methods


The George Washington University Medical Center has a 3-year internal medicine programme in which resident physicians undergo primarily hospital-based internal medicine and subspecialty training. In addition, there is a weekly internal medicine ambulatory care (outpatient) clinic. A preventive care educational programme includes education about cancer screening with verbal and written information for resident physicians. Cancer screening guidelines arc reinforced during these weekly clinics.

Colorectal cancer screening

The screening guidelines include annual rectal examinations, annual FOB and flexible sigmoidoscopy every 3-5 years. The results of these screening tests should be recorded in the medical records.

Data collection

The records of all patients who attended the resident physicians’ clinic between 1989 and 1994 were reviewed retrospectively. Patients included in the study were seen for a health maintenance evaluation and were judged to be at average risk for colorectal cancer. All patients were [is greater than or equal to] 50 years of age and required to undergo colorectal cancer screening as part of the clinical assessment.

If cancer screening was not done, the reasons for not adhering to the recommendations were to be documented. If cancer screening had been conducted by another physician or at another institution, this information was to be verified and documented.

Exclusion criteria included patients seen only for a specific problem, those at increased risk for colorectal cancer and those who were cared for primarily by an attending physician. Patients with symptoms suggestive of gastrointestinal disorder were also excluded.

This study was approved by The George Washington University Medical Center’s institutional review board.

Data analysis

The Epi Info Program (version 6) was used to develop a database and for analysis. Statistical significance was determined by using contingency tables, which generated [chi square] and P-values [7]. Statistical significance was confirmed using the Instat program, which employs Fisher’s exact test and two-tailed analysis to generate P-values [8, 9]. Statistical significance was set at P [is less than] 0.05.


There were 94 patients aged 50-60 years (54 women), 55 aged 61-70 years (26 women, 29 men) and 51 aged [is greater than or equal to] 71 years (30 women). Frequencies of rectal examinations, FOB testing and flexible sigmoidoscopies were determined for each age category and for men and women in each category.

Examination of the case notes of the 200 patients revealed that resident physicians performed 85 (42.5%) rectal examinations, 95 (47.5%) FOB tests and 21 (10.5%) flexible sigmoidoscopies. Of the 110 women studied, 41 (37.3%) had rectal examinations, 43 (39.1%) had FOB tests and 13 (11.8%) had flexible sigmoidoscopies. Of the 90 men studied, 44 (48.9%) had rectal examinations, 52 (57.8%) had FOB tests and eight (8.9%) had flexible sigmoidoscopies.

Significantly more rectal examinations were performed in patients aged 50-60 years than in those aged 61-70 years (P [is less than] 0.005) and [is greater than or equal to] 71 years (P [is less than] 0.00002). Patients who were aged 61-70 years had more rectal examinations than those aged [is greater than or equal to] 71 years (P [is less than] 0.08). There was significantly more FOB testing in patients aged 50-60 than in those aged [is greater than or equal to] 71 years (P [is less than] 0.02). There was no difference in the performance of flexible sigmoidoscopies with age category.

Men aged 50-60 years more frequently had rectal examinations than women in the same age group (P [is less than] 0.001). There was a statistically significant difference in the performance of FOB between men and women (P [is less than] 0.008).

More women aged 50-60 years had rectal examinations than women aged 61-70 years (P [is less than] 0.05) or [is greater than or equal to] 71 years (P [is less than] 0.008). Women who were aged [is greater than or equal to] 71 years frequently underwent less FOB testing than women aged 50-60 years (P [is less than] 0.002). Men aged 50-60 years had more rectal examinations than men aged 61-70 years (P [is less than] 0.001) or [is greater than or equal to] 71 years (P [is less than] 0.00001). Men aged 61-70 years had more rectal examinations than men aged [is greater than or equal to] 71 years (P [is less than] 0.04).

There were no significant difference in the comorbidities in the different age categories.


All patients evaluated by the resident physicians were independent, community dwellers seen in the ambulatory care clinic. None was from a nursing home. All were seen for health evaluation and maintenance.

The resident physicians did not adhere to colorectal cancer screening guidelines in patients in all age categories. Younger patients were statistically more likely to have rectal examinations than those who were older. This applied to both sexes, but men (in all age groups) were more likely to have screening tests than women.

The incidence or colorectal cancer rises in a linear fashion with increasing age. The role of screening for this disease in older people has not yet been firmly established. However, current US guidelines recommend annual rectal examinations and FOB tests in middle-aged and older people. In this survey, screening procedures (other than endoscopy) were performed less often in older subjects and in women. The differences could not be explained by co-morbidity and suggest screening bias in internal medicine residents.

Key messages

* In an survey of screening for colorectal cancer in those attending a clinic for health maintenance evaluation, the frequency of rectal examinations declined significantly with age.

* Patients over 70 were less likely to have occult blood testing than those in their fifties.

* Men were more likely to have rectal examinations than women and, in the over-70 group, men were more likely than women to have stools tested for occult blood.


[1.] US Preventive Services Task Force. Screening for colorectal cancer. In: Guide to Clinical Preventive Services, second edition. Baltimore, MD: Williams and Williams, 1996; 89-104.

[2.] Wingo PA, Tong T, Bolden S. Cancer statistics, 1995. CA Cancer J Clin 1995; 45, 8-30.

[3.] Ries LAG, Miller BA, Hankey BF et al. (eds). SEER Cancer Statistics Review, 1973-1991: tables and graphs. NIH Publication, no 94-2789. Bethesda, MD: National Cancer Institute, 1994.

[4.] American Cancer Society. Guidelines for Cancer-related Check Up: an update. Atlanta, GE: American Cancer Society, 1993.

[5.] Fleischer DE, Goldberg SB, Browning THE et al. Detection and surveillance of colorectal cancer. JAMA 1989; 261: 580-5.

[6.] Richardson JE Health promotion in the nursing home patient. J Am Board Faro Pract 1992; 5: 127-36.

[7.] Epi Info, version 6, Atlanta, GE: Centers tot Disease Control and Prevention, 1994.

[8.] InStat, version 1.13, San Diego, CA: GraphPAD Software, 1990.

[9.] Hirsh RE, Riegelman RK. Statistical First Aid–interpretation of health research data. Boston: Blackwell Scientific Publications, 1992.

Marie L. Borum

Department of Medicine, The George Washington University medical Center, 2150 Pennsylvania Avenue, NW. MFA Building, Suite 3-410, Washington, DC 20037, USA Fax: (+1) 202 994 7955

COPYRIGHT 1998 Oxford University Press

COPYRIGHT 2000 Gale Group