Fig. 2.

Standardized differences in 5-year cumulative all-cause mortality comparing moderate and high with low/null social engagement at baseline in pre-specified subgroups of residents in nursing homes in Madrid, Spain, 1998–1999 to 2009. Subgroup-specific risk differences (squares with area inversely proportional to the variance) and their 95% confidence intervals (horizontal lines) were obtained from spline-based survival models weighted by combined inverse probability weights and stratified by baseline level of social engagement and resident subgroup. Subgroup-specific weights were used to standardize cumulative mortality in each social engagement level and resident subgroup to the weighted distribution of baseline confounders in the entire resident subgroup, including age (65–74, 75–79, 80–84, 85–89, or ≥ 90 years), sex (women or men), educational level (less than primary; primary; or secondary or more), marital status (married, single, or widowed/divorced), facility ownership (public/subsidized or private), facility size (< 100, 100–299, or ≥ 300 beds), length of stay (0–1, 2–4, or ≥ 5 years), assigned caregiver (yes or no), frequency of external visits (monthly or less, weekly, or daily), number of chronic conditions (0–1, 2–3, or ≥ 4), and functional dependency (no, mild, or moderate)