Patient-carried card prompts vs computer-generated prompts to remind private practice physicians to perform health...

We compared the efficacy of a computer-generated prompt sheet placed on the front of patients' charts with a patient-carried prompt card to remind physicians to perform selected health maintenance items.

A randomized prospective single-blind study was performed in private practice offices in rural eastern North Carolina, with 28 family practitioners and nine general internists participating. Twenty-two physicians were randomized to the computer prompt group and 22 to the card prompt group. Four physicians in the computer group did not complete the study because of computer software problems, and three physicians did not complete the study because their limited staff was unable to enter patient data into the computer. Before the intervention, 20 patient charts (10 males and 10 females) from each physician were audited for the performance of influenza vaccinations, stool for occult blood, pap smears, breast examinations performed by the physicians, and mammograms. One year after the intervention was instituted, chart audits for the above five items were done again on 20 different patient charts for each physician.

There was a 7% increase in the performance of influenza vaccinations in the card group (17% vs 24%) compared with a 6% increase in the computer group (20% vs 26%). There was a 5% decrease in the performance of stool for occult blood in the card group (28% vs 23%) compared with a 1% increase in the computer group (30% vs 31%). There was an 11% decrease in the performance of pap smears in the card group (26% vs 15%) compared with a 3% increase in the computer group (23% vs 26%). There was a 2% decrease in the performance of breast examinations by the physician in the card group (35% vs 33%) compared with a 3% increase in the computer group (30% vs 33%). Finally, there was a 3% increase in the performance of mammograms in the card group (22% vs 25%) compared with an 11% increase in the computer group (15% vs 26%).

Our data show a greater increase in performance of health maintenance items in the computer-prompted group. The performance of stool for occult blood, pap smears, breast examinations performed by the physician, and mammograms were increased more in the computer-prompted group than in the card group. However, there was not a statistically significant difference after intervention for any of the audited health maintenance items for either the computer group or the card group. Overall, health maintenance measures were performed in only a minority of appropriate patients.