OBJECTIVE: To estimate the risk of survival of unaffected cofetuses of anomalous triplets. METHODS: Retrospective cohort study of triplets delivered in the United States from 1995 through 1998. Four triplet clusters were identified: cluster A (all members anomaly- free); cluster B (1 anomalous member); cluster C (2 anomalous members), and cluster D (all 3 members anomalous). We compared the risk for stillbirth and infant mortality among nonanomalous fetuses in clusters A, B, and C after adjusting for intracluster correlations. RESULTS: A total of 7,560 triplet clusters (98.9% ) were analyzed after excluding cluster D (1.1% ). The total still- birth rate was 20.9 (cluster A), 61.0 (cluster B), and 81.1 (cluster C) per 1,000 (P for trend <.001), and infant mortality rate was 56.4 (cluster A), 108.8 (cluster B), and 196.1 (cluster C) per 1,000 (P for trend <.001). Using cluster A as the referent category, the risk for stillbirth among anomaly- free clustermates climbed with increase in the number of siblings with anomalies in a dose- response pattern (adjusted odds ratio, 95% confidence interval 1.5, 0.7- 3.1, for cluster B; and 5.2, 1.4- 18.8, for cluster C; P for trend = .03). For infant mortality, the only rise in risk was in cluster C (3.3, 1.6- 6.7), whereas cluster B showed comparable risk with the referent category (0.8, 0.5- 1.4; P for trend >.05). CON- CLUSION: The presence of anomalous fetuses compromises the survival of normal cotriplets. These findings could prove useful for counseling affected parents and highlight the need for follow- up of normal coinfants of anomalous fetuses. OBJECTIVE: To estimate the risk of survival of unaffected cofetuses of anomalous triplets. METHODS: Retrospective cohort study of triplets delivered in the United States from 1995 through 1998. Four triplet clusters were identified: cluster A (all members anomaly-free); cluster B 1 anomalous member; cluster C (2 anomalous members), and cluster D (all 3 members anomalous). We compared the risk for stillbirth and infant mortality among nonanomalous fetuses in clusters A, B, and C after adjusting for intracluster correlations. RESULTS: A total of 7,560 triplet clusters (98.9%) were analyzed after excluding cluster D (1.1%). The total still-birth rate was 20.9 (cluster A), 61.0 (cluster B), and 81.1 (cluster C) per 1,000 (P for trend <.001), and infant mortality rate was 56.4 (cluster A), 108.8 (cluster B), and 196.1 (cluster C) per 1,000 (P for trend <.001). Using cluster A as the referent category , the risk for stillbirth among anomaly- free clustermates climbed with increase in th e number of siblings with anomalies in a dose-response pattern (adjusted odds ratio, 95% confidence interval 1.5, 0.7-3.1, for cluster B; and 5.2, 1.4-18.8, for cluster C; P for trend = .03). For infant mortality, the only rise in risk was in cluster C (3.3, 1.6- 6.7), whereas cluster B showed comparable risk with the referent category (0.8, 0.5-1.4; P for trend> The presence of anomalous fetuses compromises the survival of normal cotriplets. These findings could prove useful for counseling affected parents and highlight the need for follow-up of normal coinfants of anomalous fetuses.