Table 1: Participant characteristics


No recommendation
(N=40)		 Recommendation
(N=72)		 Overall
(N=112)		 P-value
Baseline age
Mean (SD) 76.1 (6.94) 74.7 (14.2) 75.2 (12.1) 0.848
Median [Min, Max] 76.0 [63.0, 91.0] 76.0 [0, 97.0] 76.0 [0, 97.0]
Missing 1 (2.5%) 1 (1.4%) 2 (1.8%)
Sex
Female 8 (20.0%) 7 (9.7%) 15 (13.4%) 0.326
Male 31 (77.5%) 62 (86.1%) 93 (83.0%)
Missing 1 (2.5%) 3 (4.2%) 4 (3.6%)
Race
non-White 2 (5.0%) 6 (8.3%) 8 (7.1%) 0.882
Unknown/NA 4 (10.0%) 11 (15.3%) 15 (13.4%)
White 34 (85.0%) 55 (76.4%) 89 (79.5%)
DBI group
0 23 (57.5%) 14 (19.4%) 37 (33.0%) <0.001
0.01-0.99 12 (30.0%) 21 (29.2%) 33 (29.5%)
1+ 5 (12.5%) 37 (51.4%) 42 (37.5%)
Initial DBI
Mean (SD) 0.382 (0.548) 1.21 (0.977) 0.916 (0.936) <0.001
Median [Min, Max] 0 [0, 2.08] 1.00 [0, 3.90] 0.700 [0, 3.90]
3-Month DBI
Mean (SD) 0.352 (0.586) 1.07 (0.972) 0.809 (0.918) <0.001
Median [Min, Max] 0 [0, 2.20] 0.890 [0, 3.90] 0.700 [0, 3.90]
Missing 0 (0%) 2 (2.8%) 2 (1.8%)
Provider type
ARNP 6 (15.0%) 12 (16.7%) 18 (16.1%) 0.408
PA 3 (7.5%) 0 (0%) 3 (2.7%)
Physician 26 (65.0%) 53 (73.6%) 79 (70.5%)
Unknown 5 (12.5%) 6 (8.3%) 11 (9.8%)
Missing 0 (0%) 1 (1.4%) 1 (0.9%)
Number of fall meds
Mean (SD) 3.15 (1.95) 5.08 (2.58) 4.40 (2.55) <0.001
Median [Min, Max] 3.00 [0, 7.00] 5.00 [1.00, 13.0] 4.00 [0, 13.0]
Missing 1 (2.5%) 1 (1.4%) 2 (1.8%)
Cardiovascular drug
No 12 (30.0%) 11 (15.3%) 23 (20.5%) 0.181
Yes 28 (70.0%) 61 (84.7%) 89 (79.5%)
Hypoglycemic drug
No 33 (82.5%) 51 (70.8%) 84 (75.0%) 0.393
Yes 7 (17.5%) 21 (29.2%) 28 (25.0%)
Opioid
No 36 (90.0%) 59 (81.9%) 95 (84.8%) 0.523
Yes 4 (10.0%) 13 (18.1%) 17 (15.2%)
Anticonvulsant drug
No 31 (77.5%) 43 (59.7%) 74 (66.1%) 0.163
Yes 9 (22.5%) 29 (40.3%) 38 (33.9%)
Anticholinergic
No 30 (75.0%) 43 (59.7%) 73 (65.2%) 0.267
Yes 10 (25.0%) 29 (40.3%) 39 (34.8%)
Sedative/hypnotic drug
No 38 (95.0%) 59 (81.9%) 97 (86.6%) 0.151
Yes 2 (5.0%) 13 (18.1%) 15 (13.4%)
Muscle relaxer
No 39 (97.5%) 60 (83.3%) 99 (88.4%) 0.0809
Yes 1 (2.5%) 12 (16.7%) 13 (11.6%)
Antidepressant
No 33 (82.5%) 33 (45.8%) 66 (58.9%) <0.001
Yes 7 (17.5%) 39 (54.2%) 46 (41.1%)
Other drug
No 35 (87.5%) 53 (73.6%) 88 (78.6%) 0.229
Yes 5 (12.5%) 19 (26.4%) 24 (21.4%)







Table 2: Stratified by recommendation acceptance



Recommendation not performed
(N=35)		 Recommendation performed
(N=37)		 Overall
(N=72)		 P-value
Baseline age
Mean (SD) 75.3 (14.9) 74.2 (13.6) 74.7 (14.2) 0.953
Median [Min, Max] 76.0 [0, 97.0] 75.0 [0, 92.0] 76.0 [0, 97.0]
Missing 1 (2.9%) 0 (0%) 1 (1.4%)
Sex
Female 3 (8.6%) 4 (10.8%) 7 (9.7%) 0.962
Male 30 (85.7%) 32 (86.5%) 62 (86.1%)
Missing 2 (5.7%) 1 (2.7%) 3 (4.2%)
Race
non-White 3 (8.6%) 3 (8.1%) 6 (8.3%) 0.995
Unknown/NA 6 (17.1%) 5 (13.5%) 11 (15.3%)
White 26 (74.3%) 29 (78.4%) 55 (76.4%)
NumberRecommendationsMade
Mean (SD) 1.97 (1.22) 1.84 (1.21) 1.90 (1.21) 0.897
Median [Min, Max] 2.00 [1.00, 6.00] 2.00 [1.00, 7.00] 2.00 [1.00, 7.00]
NumberReccomendationsAccepted
Mean (SD) 0 (0) 1.43 (1.07) 0.736 (1.05) <0.001
Median [Min, Max] 0 [0, 0] 1.00 [1.00, 7.00] 1.00 [0, 7.00]
Initial DBI
Mean (SD) 1.32 (0.965) 1.11 (0.992) 1.21 (0.977) 0.668
Median [Min, Max] 1.10 [0, 3.90] 0.860 [0, 3.30] 1.00 [0, 3.90]
3-Month DBI
Mean (SD) 1.32 (0.975) 0.835 (0.922) 1.07 (0.972) 0.112
Median [Min, Max] 1.05 [0, 3.90] 0.685 [0, 3.30] 0.890 [0, 3.90]
Missing 1 (2.9%) 1 (2.7%) 2 (2.8%)
Provider type
ARNP 6 (17.1%) 6 (16.2%) 12 (16.7%) 0.967
Physician 26 (74.3%) 27 (73.0%) 53 (73.6%)
Unknown 2 (5.7%) 4 (10.8%) 6 (8.3%)
Missing 1 (2.9%) 0 (0%) 1 (1.4%)
Cardiovascular drug
No 6 (17.1%) 5 (13.5%) 11 (15.3%) 0.913
Yes 29 (82.9%) 32 (86.5%) 61 (84.7%)
Hypoglycemic drug
No 24 (68.6%) 27 (73.0%) 51 (70.8%) 0.919
Yes 11 (31.4%) 10 (27.0%) 21 (29.2%)
Opioid
No 27 (77.1%) 32 (86.5%) 59 (81.9%) 0.588
Yes 8 (22.9%) 5 (13.5%) 13 (18.1%)
Anticonvulsant drug
No 20 (57.1%) 23 (62.2%) 43 (59.7%) 0.91
Yes 15 (42.9%) 14 (37.8%) 29 (40.3%)
Anticholinergic
No 22 (62.9%) 21 (56.8%) 43 (59.7%) 0.87
Yes 13 (37.1%) 16 (43.2%) 29 (40.3%)
Sedative/hypnotic drug
No 29 (82.9%) 30 (81.1%) 59 (81.9%) 0.981
Yes 6 (17.1%) 7 (18.9%) 13 (18.1%)
Muscle relaxer
No 30 (85.7%) 30 (81.1%) 60 (83.3%) 0.87
Yes 5 (14.3%) 7 (18.9%) 12 (16.7%)
Antidepressant
No 13 (37.1%) 20 (54.1%) 33 (45.8%) 0.355
Yes 22 (62.9%) 17 (45.9%) 39 (54.2%)
Other drug
No 26 (74.3%) 27 (73.0%) 53 (73.6%) 0.992
Yes 9 (25.7%) 10 (27.0%) 19 (26.4%)







Table 3: Stratified by DBI provided



DBI not provided
(N=25)		 DBI provided
(N=47)		 P-value
NumberRecommendationsMade
Mean (SD) 1.56 (0.768) 2.09 (1.36) 0.0404
Median [Min, Max] 1.00 [1.00, 4.00] 2.00 [1.00, 7.00]
NumberReccomendationsAccepted
Mean (SD) 0.840 (0.688) 0.681 (1.20) 0.477
Median [Min, Max] 1.00 [0, 2.00] 0 [0, 7.00]
DBI group
0 4 (16.0%) 10 (21.3%) 0.336
0.01-0.99 10 (40.0%) 11 (23.4%)
1+ 11 (44.0%) 26 (55.3%)
Initial DBI
Mean (SD) 0.970 (0.729) 1.34 (1.07) 0.087
Median [Min, Max] 0.800 [0, 2.50] 1.20 [0, 3.90]
3-Month DBI
Mean (SD) 0.777 (0.749) 1.21 (1.04) 0.0498
Median [Min, Max] 0.760 [0, 2.60] 1.00 [0, 3.90]
Missing 2 (8.0%) 0 (0%)
Baseline age
Mean (SD) 76.6 (5.61) 73.7 (17.1) 0.296
Median [Min, Max] 75.0 [69.0, 92.0] 76.0 [0, 97.0]
Missing 0 (0%) 1 (2.1%)





Proportions of recommendations accepted by group



Proportions of recommendations accepted based on different parameters
N Fall meds taken Recs made Recs accepted Proportion w/ recs accepted Proportion of recs accepted
Overall
Overall 72 5.1 1.9 0.7 51.4 43.4
Was DBI provided?
DBI not provided 25 5.1 1.6 0.8 68.0 61.3
DBI provided 47 5.1 2.1 0.7 42.6 33.9
DBI group
0 14 2.7 1.3 0.9 71.4 67.9
0.01-0.99 21 3.9 1.4 0.5 47.6 40.5
1+ 37 6.6 2.4 0.8 45.9 35.8
Number of recs
1 34 4.1 1.0 0.5 52.9 52.9
2 24 5.0 2.0 0.8 54.2 41.7
3 6 7.5 3.0 0.5 50.0 16.7
4 6 7.3 4.0 0.8 33.3 20.8
6 1 9.0 6.0 0.0 0.0 0.0
7 1 7.0 7.0 7.0 100.0 100.0





DBI not provided


Proportions of recommendations accepted based on different parameters
N Fall meds taken Recs made Recs accepted Proportion w/ recs accepted Proportion of recs accepted
Overall
Overall 25 5.1 1.6 0.8 68.0 61.3
DBI group
0 4 3.5 1.2 1.2 100.0 100.0
0.01-0.99 10 3.8 1.4 0.8 70.0 60.0
1+ 11 6.9 1.8 0.7 54.5 48.5



DBI provided


Proportions of recommendations accepted based on different parameters
N Fall meds taken Recs made Recs accepted Proportion w/ recs accepted Proportion of recs accepted
Overall
Overall 47 5.1 2.1 0.7 42.6 33.9
DBI group
0 10 2.3 1.3 0.7 60.0 55.0
0.01-0.99 11 3.9 1.5 0.3 27.3 22.7
1+ 26 6.5 2.7 0.8 42.3 30.4



Distribution of baseline DBI for those who received a rec






Baseline vs. 3-month DBI for those who had a rec accepted








Proportions of recommendations accepted based on drug classes taken
N Fall meds taken Recs made Recs accepted Proportion w/ recs accepted Proportion of recs accepted
Overall
Overall 72 5.1 1.9 0.7 51.4 43.4
Cardiovascular drug
No 11 3.2 1.8 0.5 45.5 30.3
Yes 61 5.4 1.9 0.8 52.5 45.8
Hypoglycemic drug
No 51 4.2 1.8 0.8 52.9 46.7
Yes 21 7.3 2.1 0.7 47.6 35.3
Opioid
No 59 4.7 1.8 0.8 54.2 46.5
Yes 13 6.9 2.5 0.5 38.5 29.5
Anticonvulsant drug
No 43 4.4 1.6 0.7 53.5 45.7
Yes 29 6.1 2.4 0.9 48.3 39.9
Anticholinergic
No 43 4.6 1.9 0.8 48.8 44.4
Yes 29 5.7 2.0 0.7 55.2 42.0
Sedative/hypnotic drug
No 59 4.8 1.8 0.7 50.8 43.1
Yes 13 6.5 2.2 1.0 53.8 44.9
Muscle relaxer
No 60 4.7 1.7 0.7 50.0 43.9
Yes 12 6.8 2.9 1.0 58.3 41.0
Antidepressant
No 33 3.8 1.5 0.7 60.6 51.5
Yes 39 6.1 2.2 0.7 43.6 36.5
Other drug
No 53 4.9 1.8 0.7 50.9 43.7
Yes 19 5.5 2.2 0.8 52.6 42.5






Exploring top drugs



## # A tibble: 116 × 2
##    Drug       Count
##    <chr>      <dbl>
##  1 amlodipine    33
##  2 metoprolol    28
##  3 losartan      28
##  4 tamsulosin    28
##  5 gabapentin    25
##  6 lisinopril    23
##  7 trazodone     13
##  8 carvedilol    11
##  9 duloxetine    10
## 10 cetirizine    10
## # ℹ 106 more rows
## # A tibble: 75 × 2
##    Drug        Count
##    <chr>       <dbl>
##  1 gabapentin     25
##  2 trazodone      13
##  3 duloxetine     10
##  4 cetirizine     10
##  5 loratadine     10
##  6 sertraline      9
##  7 bupropion       9
##  8 oxycodone       8
##  9 pregabalin      7
## 10 venlafaxine     7
## # ℹ 65 more rows







Proportions of recommendations accepted based on drug
N DBI Recs made Recs accepted Proportion w/ recs accepted Proportion of recs accepted
Overall
NA 72 1.2 1.9 0.7 51.4 43.4
Gabapentin
No 53 1.0 1.7 0.6 49.1 41.8
Yes 19 1.8 2.6 1.1 57.9 47.8
Trazodone
No 61 1.1 1.9 0.7 50.8 42.5
Yes 11 1.8 2.2 1.1 54.5 48.5
Loratadine
No 67 1.2 1.9 0.7 50.7 43.2
Yes 5 1.0 2.2 0.8 60.0 46.7
Duloxetine
No 63 1.1 1.8 0.8 54.0 45.9
Yes 9 1.9 2.4 0.4 33.3 25.9
Cetirizine
No 65 1.2 1.9 0.7 49.2 41.2
Yes 7 0.9 1.6 0.9 71.4 64.3
Glipizide/Glimeperide
No 70 1.2 1.9 0.8 52.9 44.6
Yes 2 1.1 1.0 0.0 0.0 0.0






Rec accepted models


## 
## Call:
## glm(formula = rec_performed_bi_n ~ AgeAtEnrollment + dbi_group + 
##     dbi_provided, family = binomial, data = dbi_rec)
## 
## Coefficients:
##                          Estimate Std. Error z value Pr(>|z|)  
## (Intercept)               5.62537    3.79619   1.482   0.1384  
## AgeAtEnrollment          -0.06469    0.04747  -1.363   0.1729  
## dbi_group0                1.42810    0.80579   1.772   0.0763 .
## dbi_group1+              -0.14556    0.63459  -0.229   0.8186  
## dbi_providedDBI provided -1.17479    0.56220  -2.090   0.0367 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 95.524  on 68  degrees of freedom
## Residual deviance: 86.073  on 64  degrees of freedom
## AIC: 96.073
## 
## Number of Fisher Scoring iterations: 4



## 
## Call:
## glm(formula = rec_performed_bi_n ~ AgeAtEnrollment + dbi_group + 
##     dbi_provided + dbi_group * dbi_provided, family = binomial, 
##     data = dbi_rec)
## 
## Coefficients:
##                                        Estimate Std. Error z value Pr(>|z|)  
## (Intercept)                             5.89258    3.96526   1.486   0.1373  
## AgeAtEnrollment                        -0.06315    0.04863  -1.298   0.1941  
## dbi_group0                             16.48248 1950.22368   0.008   0.9933  
## dbi_group1+                            -1.02521    0.97542  -1.051   0.2932  
## dbi_providedDBI provided               -1.93225    0.98684  -1.958   0.0502 .
## dbi_group0:dbi_providedDBI provided   -14.88472 1950.22393  -0.008   0.9939  
## dbi_group1+:dbi_providedDBI provided    1.49237    1.23577   1.208   0.2272  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 95.524  on 68  degrees of freedom
## Residual deviance: 83.477  on 62  degrees of freedom
## AIC: 97.477
## 
## Number of Fisher Scoring iterations: 16




DBI difference models


## 
## Call:
## lm(formula = dbi_diff ~ AgeAtEnrollment + dbi_group + dbi_provided, 
##     data = dbi_rec_performed)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.69594 -0.19307 -0.01757  0.15092  0.69611 
## 
## Coefficients:
##                           Estimate Std. Error t value Pr(>|t|)
## (Intercept)               0.536000   1.019101   0.526    0.603
## AgeAtEnrollment          -0.004048   0.012802  -0.316    0.754
## dbi_group0               -0.160655   0.157785  -1.018    0.317
## dbi_group1+               0.162033   0.168557   0.961    0.344
## dbi_providedDBI provided -0.094563   0.122514  -0.772    0.446
## 
## Residual standard error: 0.3427 on 30 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.1811, Adjusted R-squared:  0.07196 
## F-statistic: 1.659 on 4 and 30 DF,  p-value: 0.1855



## 
## Call:
## lm(formula = dbi_diff ~ AgeAtEnrollment + dbi_group * dbi_provided, 
##     data = dbi_rec_performed)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.72727 -0.23215 -0.00686  0.13989  0.69730 
## 
## Coefficients:
##                                       Estimate Std. Error t value Pr(>|t|)
## (Intercept)                           0.803926   1.102570   0.729    0.472
## AgeAtEnrollment                      -0.006865   0.013707  -0.501    0.620
## dbi_group0                           -0.282193   0.226093  -1.248    0.222
## dbi_group1+                           0.095331   0.222550   0.428    0.672
## dbi_providedDBI provided             -0.244085   0.243875  -1.001    0.325
## dbi_group0:dbi_providedDBI provided   0.274977   0.343853   0.800    0.431
## dbi_group1+:dbi_providedDBI provided  0.155534   0.309351   0.503    0.619
## 
## Residual standard error: 0.3507 on 28 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.1995, Adjusted R-squared:  0.02801 
## F-statistic: 1.163 on 6 and 28 DF,  p-value: 0.3535