Current issues of ACP Journal Club are published in Annals of Internal Medicine

Quality Improvement

Computerized reminders increased cancer screening and counseling in primary care settings

ACP J Club. 1991 Nov-Dec;115:95. doi:10.7326/ACPJC-1991-115-3-095

Related Content in the Archives
Computer-based clinical decision support systems: a review
Improving physician compliance with preventive care protocols
Computerized medical records and preventive health care

Source Citation

McPhee SJ, Bird JA, Fordham D, Rodnick JE, Osborn EH. Promoting cancer prevention activities by primary care physicians. Results of a randomized, controlled trial. JAMA. 1991 Jul 24/31;266:538-44.



To evaluate the effectiveness of a computerized reminder system supplemented by patient education materials in promoting cancer prevention screening and counseling by primary care physicians in community-based practices in California, USA.


Randomized controlled trial, with an intervention of 1 year.


Fee-for-service, solo (55%) or small (< 5 physicians) group practices.


307 clinical faculty were sent recruitment letters describing the study. Only 1 physician per practice could be enrolled. 30 family physicians and 10 internists were eligible and willing to participate. 1 physician in the control group did not complete the study.


Hardware and software were installed by the investigators at the practices of the intervention group (n = 20). Orientation and training were provided for the staff. Pre-intervention data from the practice records on 7 screening procedures (stool occult-blood test, digital rectal examination, sigmoidoscopy, Papanicolaou smear, pelvic examination, mammography, and breast examination) and on assessment and counseling for diet and smoking were entered. Reports were generated from updated records reminding the physician and patient of previous cancer prevention activities and giving due dates for future assessments or counseling. Pamphlets on cancer prevention were also provided, as well as information about helping patients change behavior. There was no intervention for the control group (n = 20), other than collection of data for a comparable sample of patients.

Main outcome measures

Physician compliance with recommended cancer screening and counseling, based on a random sample of 60 of each physician's patients taken in each of the preintervention and intervention years. Patients were aged ≥ 40 years and had been enrolled in the practice for ≥ 1 year before a visit during the study period.

Main results

In regression analyses adjusted for physicians' preintervention performance and for differences in patient samples, incremental absolute differences in performance scores (in favor of the intervention group) were Papanicolaou smear, 30.7% (P = 0.01); smoking counseling, 17.3% (P = 0.03); stool occult-blood test, 14.5% (P = 0.001); dietary counseling, 13.9% (P = 0.001); dietary assessment, 12.3% (P = 0.01); pelvic examination, 11.8% (P = 0.002); rectal examination, 10.5% (P = 0.004); smoking assessment, 10.2% results of (P = 0.03); and breast examination, 8.7% (P = 0.03). No difference existed in sigmoidoscopy or mammography between groups.


Computerized reminders for primary care physicians and their patients increased physician performance of cancer screening and counseling.

Source of funding: National Cancer Institute.

Address for article reprint: Dr. S.J. McPhee, 400 Parnassus Avenue, A-405, San Francisco, CA 94143-0320, USA.


This study supports the conclusions of many university-based studies that computer reminders change physicians' behavior and shows that these conclusions also apply to private practice. The benefit of reminders is not limited to cancer prevention. In private practice, computer reminders about chemotherapy dosing were also effective (1). In university settings, reminders improved inpatient antibiotic prescribing (2), the follow-up and treatment of positive strep cultures (3), and various treatment, test monitoring, and prevention activities (4). With stand-alone computer programs, personnel time is required to register each patient and record each patient's encounter data. These time requirements were not fully discussed in this report, but data entry costs pose barriers to the widespread use of reminder systems. Data interchange standards, such as ASTM and HL7 (5), permit linkage of physicians' computers to data sources such as hospital registration systems and commercial laboratories, and can reduce or eliminate these barriers. These standards are already used by many reference laboratories and hospitals. Practical reminder systems will become widely available only if physicians insist that their diagnostic services and hospitals support these standards.

Clement J. McDonald, MD
Regenstrief Institute, Indiana UniversityIndianapolis, Indiana, USA.


1. Wirtschafter D, Carpenter JT, Mesel EA. A consultation extender system for breast cancer adjuvant chemotherapy. Ann Intern Med. 1979;90:396-401.

2. Pestotnik SL, Evans RS, Burke JP, Gardner RM, Classen DC. Therapeutic antibiotic monitoring surveillance using a computerized expert system. Am J Med. 1990;88:43-8.

3. Barnett GO, Winickoff R, Dorsey JL, et al. Quality assurance through automated monitoring system and concurrent feedback using a computer based medical information system. Med Care. 1978;16:962-70.

4. McDonald CJ, Hui SL, Smith DM, et al. Reminders to physicians from an introspective computer medical record. A two-year randomized trial. Ann Intern Med. 1984;100:130-8.

5. McDonald CJ, Martin DK Overhage JM. Standards for the electronic transfer of dinical data: progress and promises. Topics in Health Record Management. 1991;11(4):1-16.