Background: Multiple dose measles vaccination intervention was applied in Gweru City, Zimbabwe, in 1990-96, following a single dose applied at 9 months of age during 1983-89. In the same periods in Bulawayo, only a single dose of measles vaccine was applied to children at 9 months of age. This study investigated the impact on measles transmission patterns of multiple dose measles vaccination strategy. Study design: Quasi-experimental community intervention applied in Gweru city with Bulawayo city as a control. The intervention included a single mass vaccination campaign carried out in1990 targeted at children aged 12-119 months irrespective of their vaccination status or disease history. Children born after 1990 were vaccinated at 9 months of age plus another single dose applied at any point between ages of 12 and 23 months (revaccination). Subjects: Measles cases were identified in both cities through surveillance. Results: Mean coverage rates for measles vaccine applied at 9 months of age were in 1983-89 85.7% and 84.6% in Bulawayo and Gweru respectively, while in 1990-96 they were 89.0% and 89.7%, respectively. In both periods the vaccine coverage rates were not significantly different in the two cities (p=0.464). In the 12-23 months age group, Gweru measles vaccination coverage rate in 1990 was 83% for single dose and 82.4% for second dose in 1990-96. Measles incidence rates in 1983-89 in both cities significantly declined and were not significantly different (p=0.898). Median incidence rates of measles in1990-96 were 131.0 and 19.0/100 000 population in Bulawayo and Gweru respectively and these were significantly different (p= 0.021). Bulawayo had measles epidemics in 1992, 1993, 1994 and 1996. In Bulawayo in 1993-96 vaccinated measles cases accounted for a median of 58% of all reported cases aged 10-119 months. Median incidence rates of measles among vaccine failures aged 10-119 months in 1993-96 in Bulawayo and Gweru were 419.1 and 13.1/100 000 population respectively and these incidence rates were significantly different (p= 0.021). In Bulawayo in 1993-96 cases aged 60-119 months accounted for a median of 56.5 % of all reported cases. Median incidence rates of measles among cases aged 60-119 months in 1993-96 in Bulawayo and Gweru were 869.9 and 26.9/100 000 population respectively and these rates were significantly different (p=0.021). Conclusion: Multiple dose vaccination strategies led to significantly reduced measles transmission in Gweru (compared to Bulawayo) in 1990-1996 by minimizing occurrence of vaccine failures and occurrence of cases in older children aged 60-119 months.