Multimorbid adults are more likely to have depression. However, existing data are mostly cross-sectional or retrospective with poor control of baseline depressive symptoms and a focus on long-term effects. This prospective study examined the short-term independent predictive association of multimorbidity with depressive symptoms. We collected baseline and three-month follow-up data from a population-based sample of 300 community-dwellers (aged 18–77) in Hong Kong. Multiple regression was used to examine the predictive association of baseline multimorbidity (two or more physical chronic conditions), relative to having one or zero conditions, with depressive symptoms in three months measured by the Center for Epidemiological Studies-Depression (CES-D, out of 60) scale. Multivariable adjustments were made for socio-demographics, baseline CES-D scores, and baseline self-perceived physical health status. A sub-analysis was conducted to compare multimorbid participants with monomorbid (one condition) ones. In our sample, 48 participants (16%) had multimorbidity. Adjusted analysis showed that on average, multimorbid participants had 2.71 (95% CI, 0.36–5.06, Cohen’s d = 0.128) more points in the CES-D scale at three-month follow-up than non-multimorbid participants (zero or one condition) did, which was independent of baseline CES-D scores, self-perceived physical health status, and socio-demographics. Compared with monomorbid participants, multimorbidity was associated with a similar difference of 2.92 (95% CI, 0.81–5.66, Cohen’s d = 0.220) points. Incremental R-square changes associated with the inclusion of multimorbidity were significant (P < 0.05). In conclusion, the effect of multimorbidity on depressive symptoms may take a shorter period to manifest than previously assumed. The mental health of adults with multimorbidity warrants more attention.