Skip to main content
. 2024 Jul 17;12:1408641. doi: 10.3389/fpubh.2024.1408641

Table 2.

Multivariable multinomial regression model showing the association between social support and frailty transition categories in the total population.

Total
N = 907
Frailty transition categories
Stable non-frail Stable frail Improvement Worsening Death
Number of events N (%)
Social support
Strong 268 162 (60.4%) 34 (12.7%) 13 (4.9%) 25 (9.3%) 34 (12.7%)
Moderate 419 228 (54.4%) 67 (16.0%) 38 (9.1%) 42 (10.0%) 44 (10.5%)
Poor 220 100 (45.5%) 42 (19.1%) 15 (6.8%) 31 (14.1%) 32 (14.5%)
Crude model RRR (95% CI)
Social support
Moderate Reference 1.37 (0.85–2.22) 1.18 (0.75–1.85) 1.13 (0.74–1.73) 0.90 (0.54–1.50)
Poor Reference 1.67 (0.98–2.86) 0.84 (0.48–1.48) 1.09 (0.66–1.80) 1.28 (0.73–2.24)
Adjusted model RRR (95% CI)
Social support
Moderate Reference 1.06 (0.61–1.84) 1.11 (0.70–1.77) 1.05 (0.68–1.63) 0.82 (0.46–1.46)
Poor Reference 1.24 (0.66–2.35) 0.76 (0.42–1.37) 0.98 (0.58–1.66) 1.28 (0.67–2.45)

The percentages in the descriptive part of the table are row percentages.

The model is adjusted for age, gender, partner status, body mass index (BMI), Comparative Analysis of Social Mobility in Industrial Nations (CASMIN), self-rated health (SRH), polypharmacy, Charlson Comorbidity Index (CCI).