Synthesizing complex data on functional decline in older people using partial least squares (PLS) analysis.

Area of Science:

• Gerontology
• Health Services Research
• Biostatistics

Background:

• Aging is associated with a decline in functional capacity, impacting community independence.
• Existing measures for functional decline lack a comprehensive understanding of temporal interactions.
• Emergency Department (ED) presentations for non-admission cases offer a unique cohort for studying functional changes in older adults.

Purpose of the Study:

• To explore the utility of partial least squares (PLS) analysis in understanding the interplay of functional decline measures over time.
• To identify relationships between instrumental activities of daily living (IADLs) and social/support factors in older adults post-ED discharge.

Main Methods:

• A cohort of 147 individuals aged 65+ presenting to an ED without admission were followed for 3 months.
• Data collected via telephone included home situation, community service use, gait aid use, carer status, falls, hospitalizations, and IADLs.
• Partial least squares (PLS) analysis was employed to examine longitudinal relationships between these variables.

Main Results:

• Instrumental activities of daily living (IADLs) naturally clustered into physical, mental, or combined domains.
• These IADL clusters strongly correlated with increased community support, gait aid use, carer involvement, and living alone.
• Relationships with hospitalizations were less clear, and falls were not well explained by the model.

Conclusions:

• Partial least squares (PLS) analysis provides a robust method for analyzing complex, longitudinal health data.
• This approach offers superior insight into the dynamics of functional decline over a 3-month period compared to conventional modeling.
• The findings highlight key factors associated with functional decline in older adults, informing targeted interventions.