Many strategies have been used to improve pain management in institutionalized care settings, but there is no consensus on the effects of these methods. The study purpose was to compare the effect of a Pain Recognition and Treatment (PRT) protocol coupled with basic pain education (experimental group) versus basic pain education alone (control group) in (1) improving the pain management performance of registered nurses (RNs) and (2) reducing pain-related expressions of residents with dementia postintervention and at 3-month follow up. A double-blind cluster randomized controlled trial with a 3-month follow-up period was conducted with 195 residents of six dementia special-care units. The weekly pain management performance of RNs (e.g., use of pharmacologic and nonpharmacologic strategies, use of referral) was recorded and weekly average scores of the pain-related expressions of residents were assessed using the following: the Verbal Descriptor Scale (VDS), Pain Assessment in Advanced Dementia Scale (PAINAD), and the Cohen-Mansfield Agitation Inventory (CMAI). The generalized linear mixed model analysis showed that, after intervention, the experimental group had significantly more weekly nonpharmacologic pain relief strategies and weekly referrals for pain management than the control group. Residents in the experimental group had significantly fewer verbal and behavioral expressions of pain compared to those in the control group. However, the groups did not differ significantly in the use of pharmacological strategies or the agitated behaviors expressed by residents. The PRT protocol is effective and is recommended for routine use in residents with dementia to improve the quality of pain care.