In nursing homes (NHs) 80% of the patients have dementia, between 60%-80% exhibit behavioural disturbances (BPSD), and more than 60% have pain. Both pain and BPSD is more common in those with severe dementia. Since older persons with dementia have less communicative skills, suffer from more pain and exhibit more agitation, pain may be a contributing factor in these patients. More than 40% of patients with BPSD are treated with neuroleptics despite described side-effects. There is an urgent need to investigate the impact of individual pain management on BPSD in patients with dementia.

It was hypothesized that

• pain increase BPSD in patients with dementia
• individual pain treatment decrease BPSD in patients with dementia

We intend to conduct an eight weeks, multi-centre, rater-blinded, cluster randomized, and parallel-group trial, with follow up after four weeks. We will choose cluster randomizing by practical reasons. 920 NH patients were screened and 352 patients with moderate or severe dementia and BPSD were included. The treatment period is 8 weeks, with further follow after four weeks.

The primary outcome measure will be reduction in aggression and agitation as well as other items which are measured by means of CMAI (Cohen Mansfield Agitation Inventory). Secondary outcome are reduction in NPI-NH-subscale agitation/aggression and other items which are measured by the Neuropsychiatric Inventory, Nursing Home Version (NPI-NH). Further, we want to evaluate the concomitant use of acute medication. Additionally, Activities of Daily Living function (ADL) and MIni Mental State Examination (MMSE) will be used as secondary outcome measure.

Pain in patients with dementia will be assessed and followed by the MOBID-2 Pain Scale (secondary outcome measure). MOBID-2 Pain Scale is a staff-administered behavioural instrument for assessment pain in older persons with dementia with god validity and reliability (Husebo 2008, 2009). MOBID-2 is based on patient's pain behaviour in connection with standardised, guided movements of different body part, and pain behaviour related to internal organs, head and skin. Additionally, pain will be registered by pain diagnoses, -etiology, and -duration. The MOBID-2 score is derived from caregiver in a clinical bedside situation during morning care.

Adverse events will be recorded and evaluated throughout the study as the primary assessment of safety and tolerability.

Inclusion criteria: Patients of either gender, 60 years of age or older, living in a nursing home (NH) diagnosed moderate or severe dementia measured by the Functional Assessment Staging (FAST) and MMSE, and BPSD in form of agitation / aggression as measured by subscale of NPI-NH and CMAI.

Exclusion criteria: Patients without cognitive impairment and without BPSD. Patients with hepatic or renal failure or diseases that make it impossible to follow the study schedule.

Inclusion Criteria:

• Aged 65 and older
• Residing in the NHs for at least 4 weeks
• Dementia according to Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association 1994), FAST score > 4 (Hughes 1982).
• Clinically relevant BPSD, operationally defined as CMAI score ≥ 39 or higher or/and at least one week history of agitation or aggression (Koss 1997).
• Written, informed consent provided by the participant (if they have capacity) or assent (if they do not have capacity) and a written proxy informed consent from a legally authorized representative empowered to make health-related decisions for the potential study participant.

Exclusion Criteria:

• Clinician responsible for care, or study clinician considers that the patient suffers from any physical condition, which would make participation in the trial distressing or likely to increase suffering
• Advanced severe medical disease/disorder with expected survival less than 6 months or that could interfere with participation
• Psychosis or other severe mental disorder prior to dementia diagnosis;
• Severe aggression (≥8) on item 3 of the NPI subscale, with aggression as the predominant symptom
• Schizophrenia, schizoaffective disorder and bipolar disorder
• Uncontrolled epilepsy
• Severe liver impairment
• Renal failure, as measured by or equivalent to an estimated creatinine clearance of < 50mL/min/1.73m,
• Severe injury or anaemia (Hb < 8.5 mmol/l), comatose state, current enrolment in another experimental protocol.
• Known allergy or adverse reaction to Paracetamol, Morphine ret, Buprenorphine plaster or pregabalin