Does a nurse-led mental health liaison service for older people reduce psychiatric morbidity in acute general medical wards? A randomised controlled trial.
Baldwin R, Pratt H, Goring H, Marriott A and Roberts C;
Commented by , 23 Sep 2004
Objective
To determine the clinical effectiveness of a nurse-led mental health liaison service for older people by a randomised controlled trial.
Method
Subjects
All patients admitted to 4 acute medical wards in Tameside, Manchester, UK were screened for suitability 3-5 days after admission, using the 4-item Geriatric Depression Scale (GDS) and the 6-item Orientation-Memory-Concentration test (OMC).
Exclusion criteria were discharge within 3 days of admission, medical instability, profound sensory loss, or acute risk of self harm.
Intervention
Consenting subjects were allocated to either "usual care" (which could include referral to the local old age psychiatry team), or a multi-faceted nurse-led intervention. For depression, this comprised antidepressant medication, managing anxiety, problem-solving and adjusting to loss. For confusion, the intervention included education and addressing disorientation, nutrition and safety issues. Interventions were individually tailored lasting a maximum of 6 weeks.
Outcome Measures
The main outcome measures were the Health of the Nation Outcome Scale for Older People (HoNoS 65+), the 30-item GDS, and the Mini-Mental State Examination (MMSE) administered at baseline and after 6-8 weeks. Ratings were administered by research assistants blind to participants’ group.
Results
486 patients were screened for eligibility, of whom 264 met inclusion criteria. 111 of these declined to participate, leaving 153 to be randomised. 33 of these subsequently dropped out (due to death, exacerbation of mental illness or withdrawing consent) leaving complete data on 59 subjects and 61 controls. There were no significant differences in demographic data or illness types between subjects and controls.
At follow-up, there were no significant differences between intervention and control group on HoNoS 65+ or MMSE scores. The mean 30-item GDS fell slightly at 6 week follow-up in the control group, which reached statistical significance. However, secondary outcome measures of prescribed psychotropic medication, length of stay, or re-admission rates showed no difference in the intervention group.
Discussion
This is an ambitious study in a population with many variables that are difficult to analyse. There are a number of methodological limitations, acknowledged by the authors, including the power of the study, the high number of drop outs, the validity of the outcome measures; the adequacy of blinding at allocation, and the short follow-up duration of the study.
There is little literature addressing the effectiveness of psychiatric liaison services for older people, and the authors are to be complimented for being the first to attempt a randomised controlled trial of a generic service.
The largely negative results of this study should not deter clinicians from attempting to improve liaison old age psychiatry services, nor researchers from attempting to analyse the impact of new services.
Two thirds of admissions of medical wards in the United Kingdom are over 65, and half of these have a mental disorder (principally delirium, dementia and depression). There is a substantial body of literature suggesting these very common clinical problems are underdiagnosed and undertreated.
Future research may focus on whether targeted interventions from psychiatric services can reduce length of stay and re-admission rates, and whether a multi-disciplinary approach (as opposed to a nurse-led approach), leads to better outcomes.