**3. Results**

#### *3.1. Descriptive Characteristics, Including Change in Nutritional Status*

Out of 254 patients screened at the Falls and Fractures Clinic between October 2016 and December 2018, 106 (76% female) consecutive individuals with median age of 79 (Q1, Q3 72, 82) years were re-assessed at 6-month follow-up. Table 1 presents descriptive characteristics of this study sample. On attendance at baseline, polypharmacy and severe comorbidity were quite prevalent (67% taking ≥5 medications and 45% with a CACI ≥5). The median number of reported falls in the past year was 2 falls. Most individuals (92%) were osteopenic/osteoporotic, and 22% were sarcopenic. On attendance at follow-up, the median number of reported falls in the past 6 months decreased to 0 falls. Moreover, 91 (86%) individuals reported using vitamin D supplements and 5 (5%) protein supplements, 70 (66%) reported having an osteoporosis treatment, and 51 (48%) reported being physically active, as part of the post-fall care plans recommended.


**Table 1.** Descriptive characteristics of the study sample, including change in nutritional status.


**Table 1.** *Cont.*

Q1, Q3 = 25th, 75th percentile; CACI = Charlson age-comorbidity index; GDS = Geriatric Depression Scale; MNA= Mini Nutritional Assessment; BMI = body mass index; number of missing values: a 1, b 3, c 5, d 6, e 10, f 14 g 15, h 16, j 27; \*: *p* < 0.05 for paired *t*-test or Wilcoxon signed-rank test for the difference between baseline and follow-up values.

The prevalence of malnutrition or risk of malnutrition based on an MNA score <24/30 was 29% at baseline and 15% at 6-month follow-up. Most individuals maintained or improved nutritional status with at least 75% not having a decrease in MNA score (median change of 0.0 (0.0, 3.3) points, *p* = 0.001) (Table 1). Specifically, 73 individuals (69%) maintained nutritional status, including 65 (61%) who remained well-nourished and 8 (8%) who remained malnourished or at risk of malnutrition. Moreover, 20 individuals (19%) improved and 6 individuals (7%) deteriorated nutritional status, as illustrated in Figure 1.

**Figure 1.** Change in MNA score (ΔMNA) by baseline MNA score across subgroups of nutritional status change. One dot represents one individual. Six participants could not be represented due to missing value in MNA score at follow-up.

#### *3.2. Nutritional Status and Musculoskeletal Health at Baseline*

In age-sex adjusted analyses, a 1-point increase in baseline MNA score was associated with an increase of 0.02 m/sec (95% CI 0.00, 0.03, *p* = 0.022) in gait speed and of 0.14 points (95% CI 0.05, 0.23, *p* = 0.003) in SPPB score. Additional adjustment for baseline variables weakened the associations, which became non-significant (Supplementary Table S2).

#### *3.3. Changes in Nutritional Status and Musculoskeletal Health*

Table 2 compares clinical, biochemical and musculoskeletal measures between subgroups of nutritional status change using the maintained (well-nourished) nutritional status group as reference group. The reference group was associated with significant increase in BMI (*p* = 0.002), weight (*p* = 0.031), 25OHD levels (*p* = 0.013), gait speed (*p* < 0.001), SPPB score (*p* = 0.008) and decrease in CTx levels (*p* < 0.001). Individuals who improved nutritional status were associated with greater increase in BMI (*p* = 0.002) and weight (*p* = 0.006) compared to the reference group, and similar increase in SPPB score (*p* = 0.177). They were not associated with significant improvement in gait speed (*p* = 0.193), but their TUG time significantly decreased (*p* = 0.001). Those who maintained (malnourished/at risk) or deteriorated nutritional status were not linked to significant improvement in any of the variables.

In multiple linear regression analyses (Table 3), change in nutritional status over 6 months showed the strongest associations with SPPB. After adjusting for baseline and care plan covariates, a 1-point increase in MNA score over 6 months was associated with an increase of 0.20 points (95% CI 0.10, 0.31, *p* < 0.001) in SPPB score. In subgroup analyses, individuals who improved nutritional status had for 3.30 sec (95% CI −6.34, −0.26, *p* = 0.033) a larger decrease in time for the TUG test compared to the reference group. Conversely, those who deteriorated in nutritional status had a larger decrease in SPPB score by 1.74 points (95% CI −3.29, −0.20, *p* = 0.028).



follow-up values within each group; †: *p* < 0.05 for *t*-test or Wilcoxon rank-sum test for the difference between the improved, maintained (malnourished/at risk) or deteriorated group vs.

the maintained (well-nourished) nutritional status group (*ref*), respectively.



of medications—all continuous except sex), and care plan variables (osteoporosis treatment, vitamin D supplement use, protein supplement use, physical activity—all categorical); b β coefficient is the change in musculoskeletal outcome associated with a unit increase in nutritional status over 6 months (ΔMNA). Models also adjusted for baseline MNA score.
