A systematic review of loss of independence in Parkinson’s disease

Abstract

Functional dependency (needing help with basic ADLs) is an important outcome in Parkinson's disease (PD). "Death or dependency", as opposed to being alive and independent, is a useful dichotomous indicator of poor outcome. We aimed to systematically review the progression to dependency in PD and what factors predicted development of dependency. Comprehensive searches were performed to identify observational studies of dependency in PD with follow-up of at least 3 years. Other forms of parkinsonism and highly selected cohorts were excluded. Descriptive analysis of included studies was performed and outcomes over time were plotted by type of cohort (inception/non-inception). Independent prognostic factors were identified. There were insufficient data for meta-analysis. Of 15,154 unique references identified, 14 studies were included. Most studies were of low quality. There was heterogeneity in definitions of dependency and the measured risk of dependency at similar time-points. Risk of dependency in inception studies was about 10-25 % at 5 years and about 20-50 % at 10 years; and risk of "death or dependency" in the inception studies was about 15-40 % at 5 years and about 35-70 % at ten years. More bradykinesia and older age were associated with more dependency, but there was little evidence for other prognostic factors. Few high-quality data on dependency are available. Heterogeneity in study populations, methodology and outcome reporting made data synthesis difficult. Few prognostic factors have been identified. Further data from representative inception studies are necessary to better understand the progression of dependency in PD.

Frontotemporal dementia as a comorbidity to idiopathic normal pressure hydrocephalus (iNPH): a short review of literature and an unusual case

“…A recent meta-analysis by Peterson et al [22] found strong evidence for improvement in iNPH patients following shunt surgery for global cognitive measure (MMSE), verbal learning and memory, and psychomotor speed. The evidence for improvement was weaker for tests of executive function.…”

Development and validation of prognostic survival models in newly diagnosed Parkinson's disease

“…We selected potential baseline predictors (ie, measured at diagnosis) to include in the model based on previously reported associations, a previous analysis of the measures of motor impairment with best predictive value in the PINE study (in which measures of axial severity were better predictors of long‐term outcomes than other measures of motor function, including the postural‐instability‐gait‐difficulties/tremor‐dominant classification), and clinical knowledge. We limited the number of candidate predictors to avoid overfitting .…”

Predictors of functional dependency in Parkinson's disease

“…To avoid overfitting the model, we selected potential predictors measured at baseline (at diagnosis), which we thought were likely to be associated with both PD and dependency, on the basis of previously reported associations, a previous analysis of motor predictors in the PINE study, and clinical knowledge. We decided that age should be included in the model irrespective of statistical significance because of strong evidence from previous studies that it is associated with multiple prognostic outcomes.…”