Purpose: Our study assessed whether change in self-rated health is a stronger predictor of mortality than baseline self-rated health and the most recent self-rated health (prior to death or loss to follow-up) among disabled older women. Design and Methods: The Women's Health and Aging Study examined disabled older women at baseline and every 6 months for 3 years. During the follow-up period, 253 out of the 905 examined participants died. Cox regression models with time-dependent covariates were used. Results: After baseline characteristics were adjusted for, baseline self-rated health was not related to mortality. After covariates at the most recent observation and covariates measured only at baseline were controlled for, the most recent self-rated health was not associated with mortality either. After time-dependent covariates and covariates measured only at baseline were adjusted for, decline in self-rated health was significantly associated with increased mortality. Implications: Change in self-rated health is a stronger predictor of mortality than self-rated health at baseline and at the most recent observation. Older women with "fair" health are worse off if they are on a declining health trajectory than if their "fair" health is stable. Family caregivers and clinicians need to closely monitor change in self-rated health among disabled older women.