PURPOSE: Influenza and pneumococcal vaccines are underused. Systems approaches that incorporate administrative and organizational strategies are more successful than education of providers for improving vaccination rates. Little has been published on the long-term success and durability of such efforts. METHODS: We performed a 10-year time-series study to examine the durability and success of an ongoing, multifaceted, institution-wide influenza and pneumococcal vaccination program. The program was first implemented at the Minneapolis Department of Veterans Affairs (VA) Medical Center in 1987-88 following the demonstration that a clinic-based standing order policy was much more successful than provider education for improving vaccine delivery. The program ensures that vaccine is offered to all high-risk patients followed up at the medical center, promotes convenient access for patients, and facilitates efficient administration of vaccine. Specific elements include an annual mailing to patients, standing orders for nurses, walk-in clinics, and the use of standardized, preprinted documentation forms. Initially the program targeted high-risk outpatients for influenza vaccination. It was extended to include inpatients in 1989-90. Pneumococcal vaccinations were added to the program in 1994-95. Vaccination rates are estimated each year from surveys mailed to randomly selected patients, and vaccine utilization is monitored through pharmacy logs. RESULTS: The survey response rates have exceeded 75% each year. Influenza vaccination rates for all high-risk patients followed up at the medical center have increased from 58% following the 1987-88 vaccination season to 84% in 1996-97 (P [is less than] 0.001). Pneumococcal vaccination rates have also increased from 34% in 1994-95 to 63% in 1996-97 (P [is less than] 0.001). Vaccination rates are similar for inpatients and outpatients, but rates for high-risk patients [is less than] 65 years of age remain lower than for the elderly: 69% versus 89% for influenza, 1996-97 (P [is less than] 0.001); 52% versus 66% for pneumococcal, 1996-97 (P = 0.05). For elderly patients followed up at the medical center, influenza (89% versus 67%, P [is less than] 0.0001) and pneumococcal (66% versus 43%, P [is less than] 0.0001) vaccination rates significantly exceeded those for the state of Minnesota in 1996-97. The annual number of influenza vaccine doses dispensed has increased from 10,000 in 1987-88 to more than 22,000 in 1996-97; and more than 13,000 doses of pneumococcal vaccine have been administered from 1994-95 through 1996-97. CONCLUSION: This simple, multifaceted program that incorporates administrative and organizational strategies to enhance influenza and pneumococcal vaccination rates has been highly durable and successful over a 10-year peroid. Similar strategies if implemented in other settings may enhance vaccination rates for the millions of high-risk patients who have yet to be immunized. Am J Med. 1998;105:385-392. [C] 1998 by Excerpta Medica, Inc.