Abstract
Background: Cognitive impairment is common in Parkinson's disease (PD) but has limited treatment options. Medication has shown some benefits but accompanied by risk of adverse events. We aimed to investigate effectiveness and feasibility of nonpharmacological interventions for people with PD and cognitive impairment on patient-centred outcomes. Methods: Systematic searches of five databases (MEDLINE, Embase, CINAHL, PsycINFO and Web of Science) were performed for studies evaluating nonpharmacological interventions for people with PD and cognitive impairment, reporting health-related quality of life, function (activities of daily living) or wellbeing outcomes, published up to 15 May 2023. Two reviewers independently assessed full-text articles and one reviewer extracted data, with a second reviewer reliability checking all data extraction. Randomised controlled trials (RCTs) were synthesised through meta-analysis using a random-effects meta-analysis with restricted maximum likelihood method pooled estimate and observational studies through narrative synthesis. Results: Eleven RCTs and three noncontrolled studies were included, studying a range of interventions: cognitive training, cognitive stimulation, cognitive rehabilitation, physical and cognitive exercise, goal management training, psychoeducation with mindfulness, broader rehabilitation programs and a psychological intervention. Feasibility was demonstrated. The majority showed effectiveness for their primary outcome. Meta-analysis showed no significant improvement in HrQoL (seven RCTs: pooled effect, standardised mean difference, -0.20 [-0.57-0.18]) or function (four RCTs: 0.08 [-0.36, 0.52]), and wellbeing measurement was infrequent and indirect. Quality of evidence was judged as very low, limiting the conclusions drawn. Conclusion: Whilst nonpharmacological trials for cognitive impairment in PD have shown promise, we found no evidence of effectiveness on HrQoL, function or wellbeing. However, this is based on very low-quality evidence from a small number of diverse studies, not powered for these outcomes. Feasibility of a range of interventions has been demonstrated in both PD-mild cognitive impairment and PD-dementia. There is a need for more robust, adequately powered studies.