Methodology

Objective: To examine the interaction between time and categorical variables (hhi_category, WIN_CAT, hearingaids1 and crosstab categories of WIN_CAT and hearingaids1) on cognitive outcomes (PACC scores), adjusting for covariates. General formula: Outcome (PACC) ~ time * Group + educ + gender + age + race + ADI_NATRANK + noise_censusblock2020_mean + (1 | ID)

Data Preparation: Baseline and follow-up data were transformed into long format.

Modeling: Applied linear mixed-effects models (LMM) using lmer from the lme4 package. Included a random intercept for participant ID to account for within-subject variability.

Covariate Adjustment: Adjusted for education, gender, age, race, ADI rank, and noise exposure.

Visualization:

Created regression plots for each model: Showed group-level intercepts and slopes over time for categorical predictors (hhi_category, WIN_CAT, hearingaids1 and crosstab categories of WIN_CAT and hearingaids1).

Findings:

All interaction terms showed small effect sizes and were not statistically significant. Group differences in intercepts were observed, but slopes over time remained consistent across groups. All results are summarized in a table for each model. Logistic regression between baseline paccscore bins and last followup pacc score bin was done and there is significant association between the two: - Baseline high PACC participants → 16.45× more likely to remain in high PACC at follow-up. - Baseline low PACC participants → 89% less likely to be in high PACC at follow-up. This suggests a strong association between baseline PACC and follow-up PACC status.

1 Table 1

For the longitudinal data to be used for the models After follow up 8, all 420 ID curated for Words in noise have 100% missing pacc score values, so they were removed and not displayed. For Hearing handicap group classification, mild to moderate and severe handicap were collapsed to one group of “handicap”.

Baseline Characteristics of Study Population
Total
(N=420)		 Impaired hearing
(N=179)		 Normal hearing
(N=241)		 P-value
age
Mean (SD) 72.5 (5.15) 73.9 (5.86) 71.4 (4.24) <0.001
Median [Min, Max] 71.6 [65.1, 94.7] 73.0 [65.4, 94.7] 70.5 [65.1, 87.2]
gender
Female 222 (52.9%) 69 (38.5%) 153 (63.5%) <0.001
Male 198 (47.1%) 110 (61.5%) 88 (36.5%)
educ
Mean (SD) 16.3 (2.48) 16.1 (2.43) 16.4 (2.51) 0.248
Median [Min, Max] 16.0 [10.0, 25.0] 16.0 [10.0, 24.0] 16.0 [10.0, 25.0]
Missing 7 (1.7%) 4 (2.2%) 3 (1.2%)
mocatots
Mean (SD) 24.9 (3.40) 23.9 (3.67) 25.7 (2.95) <0.001
Median [Min, Max] 26.0 [12.0, 30.0] 24.0 [12.0, 30.0] 26.0 [15.0, 30.0]
Missing 85 (20.2%) 35 (19.6%) 50 (20.7%)
PACC_BL
Mean (SD) -0.144 (0.758) -0.350 (0.852) 0.00901 (0.641) <0.001
Median [Min, Max] -0.0570 [-3.14, 1.42] -0.242 [-3.14, 1.42] 0.0487 [-2.38, 1.34]
Missing 107 (25.5%) 46 (25.7%) 61 (25.3%)
pacc_followup_1_arm_1
Mean (SD) -0.123 (0.747) -0.208 (0.594) -0.0615 (0.837) 0.0994
Median [Min, Max] -0.0748 [-8.42, 1.35] -0.230 [-1.78, 1.35] 0.00246 [-8.42, 1.07]
Missing 159 (37.9%) 70 (39.1%) 89 (36.9%)
pacc_followup_2_arm_1
Mean (SD) -0.153 (0.763) -0.414 (0.846) 0.0201 (0.652) 0.00142
Median [Min, Max] -0.0574 [-3.91, 1.30] -0.371 [-3.91, 1.07] 0.0382 [-2.11, 1.30]
Missing 277 (66.0%) 122 (68.2%) 155 (64.3%)
pacc_followup_3_arm_1
Mean (SD) -0.125 (0.785) -0.168 (0.699) -0.0979 (0.839) 0.68
Median [Min, Max] -0.0287 [-4.53, 1.13] -0.132 [-2.11, 0.754] 0.00659 [-4.53, 1.13]
Missing 336 (80.0%) 147 (82.1%) 189 (78.4%)
pacc_followup_4_arm_1
Mean (SD) -0.121 (0.737) -0.268 (0.882) -0.0333 (0.628) 0.173
Median [Min, Max] 0.00310 [-2.51, 1.08] -0.132 [-2.51, 1.08] 0.0505 [-1.85, 1.04]
Missing 326 (77.6%) 144 (80.4%) 182 (75.5%)
pacc_followup_5_arm_1
Mean (SD) -0.115 (0.846) -0.200 (0.884) -0.0619 (0.826) 0.494
Median [Min, Max] 0.0647 [-3.85, 1.15] 0.0854 [-2.74, 0.868] 0.0604 [-3.85, 1.15]
Missing 342 (81.4%) 149 (83.2%) 193 (80.1%)
pacc_followup_6_arm_1
Mean (SD) -0.0410 (0.758) -0.286 (0.927) 0.167 (0.507) 0.0442
Median [Min, Max] 0.0154 [-2.41, 1.38] -0.0795 [-2.41, 1.38] 0.0898 [-0.696, 1.26]
Missing 370 (88.1%) 156 (87.2%) 214 (88.8%)
pacc_followup_7_arm_1
Mean (SD) 0.0685 (0.618) 0.0273 (0.856) 0.0855 (0.522) 0.871
Median [Min, Max] 0.185 [-1.49, 1.07] 0.403 [-1.49, 1.00] 0.170 [-0.784, 1.07]
Missing 396 (94.3%) 172 (96.1%) 224 (92.9%)
pacc_followup_8_arm_1
Mean (SD) -0.115 (1.04) -0.316 (1.59) -0.0261 (0.807) 0.748
Median [Min, Max] 0.00155 [-2.33, 1.36] -0.150 [-2.33, 1.36] 0.00155 [-1.61, 1.02]
Missing 407 (96.9%) 175 (97.8%) 232 (96.3%)
ADI_NATRANK
Mean (SD) 48.5 (24.8) 47.6 (22.7) 49.1 (26.2) 0.518
Median [Min, Max] 48.0 [3.00, 100] 48.0 [4.00, 98.0] 47.0 [3.00, 100]
Missing 4 (1.0%) 2 (1.1%) 2 (0.8%)
wrat_rawword
Mean (SD) 63.1 (5.75) 62.4 (5.69) 63.7 (5.75) 0.0258
Median [Min, Max] 64.0 [31.0, 70.0] 63.0 [40.0, 70.0] 65.0 [31.0, 70.0]
phq_totscore
Mean (SD) 4.89 (5.07) 4.97 (5.52) 4.82 (4.66) 0.819
Median [Min, Max] 3.00 [0, 22.0] 3.00 [0, 22.0] 4.00 [0, 21.0]
Missing 158 (37.6%) 57 (31.8%) 101 (41.9%)
race
Black or African American 66 (15.7%) 18 (10.1%) 48 (19.9%) 0.00904
White 354 (84.3%) 161 (89.9%) 193 (80.1%)
hhi_category
Handicap 108 (25.7%) 77 (43.0%) 31 (12.9%) <0.001
No Handicap 312 (74.3%) 102 (57.0%) 210 (87.1%)
hhi_total_score
Mean (SD) 13.3 (16.0) 20.3 (18.8) 8.10 (11.0) <0.001
Median [Min, Max] 8.00 [0, 80.0] 14.0 [0, 80.0] 4.00 [0, 66.0]
PACC_class_sick
high PACC 259 (61.7%) 101 (56.4%) 158 (65.6%) 0.00963
low PACC 54 (12.9%) 32 (17.9%) 22 (9.1%)
Missing 107 (25.5%) 46 (25.7%) 61 (25.3%)
noise_censusblock2020_mean
Mean (SD) 52.3 (2.55) 52.4 (2.48) 52.3 (2.61) 0.803
Median [Min, Max] 52.4 [46.0, 59.3] 52.6 [46.0, 59.3] 52.2 [46.2, 58.4]
Missing 2 (0.5%) 1 (0.6%) 1 (0.4%)
Hearing Aid Use
No 322 (76.7%) 102 (57.0%) 220 (91.3%) <0.001
Yes 69 (16.4%) 62 (34.6%) 7 (2.9%)
Missing 29 (6.9%) 15 (8.4%) 14 (5.8%)
MOCA Category
high moca 178 (42.4%) 57 (31.8%) 121 (50.2%) <0.001
low moca 157 (37.4%) 87 (48.6%) 70 (29.0%)
Missing 85 (20.2%) 35 (19.6%) 50 (20.7%) 
## # A tibble: 1 × 6
##      mean median    sd range_min range_max n_observations
##     <dbl>  <dbl> <dbl>     <dbl>     <dbl>          <int>
## 1 0.00129 0.0942 0.703     -8.42      1.58           2074

2 Model 1: classified hhi

The outcome is the absolute pacc scores themselves at each follow up

model_hhiclassification <- lmer(
  pacc ~ time * hhi_category + educ + gender + age + race + ADI_NATRANK + noise_censusblock2020_mean + (1 | ID),
  data = Drives_WIN_HHI_longitudinal_long_clean,
  REML = FALSE
)
# View the summary of the model
summary(model_hhiclassification)
## Linear mixed model fit by maximum likelihood  ['lmerMod']
## Formula: 
## pacc ~ time * hhi_category + educ + gender + age + race + ADI_NATRANK +  
##     noise_censusblock2020_mean + (1 | ID)
##    Data: Drives_WIN_HHI_longitudinal_long_clean
## 
##      AIC      BIC   logLik deviance df.resid 
##   1874.0   1933.6   -925.0   1850.0     1056 
## 
## Scaled residuals: 
##      Min       1Q   Median       3Q      Max 
## -11.0589  -0.3915   0.0495   0.4363   7.8935 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  ID       (Intercept) 0.4049   0.6363  
##  Residual             0.1628   0.4035  
## Number of obs: 1068, groups:  ID, 393
## 
## Fixed effects:
##                               Estimate Std. Error t value
## (Intercept)                   0.889557   0.934925   0.951
## time                         -0.069118   0.016239  -4.256
## hhi_categoryNo Handicap      -0.018047   0.085648  -0.211
## educ                          0.028220   0.015171   1.860
## gender                        0.237542   0.072534   3.275
## age                          -0.032980   0.007211  -4.574
## raceWhite                     0.366903   0.107911   3.400
## ADI_NATRANK                  -0.002759   0.001657  -1.665
## noise_censusblock2020_mean    0.006952   0.013924   0.499
## time:hhi_categoryNo Handicap  0.028786   0.018127   1.588
## 
## Correlation of Fixed Effects:
##             (Intr) time   hh_cNH educ   gender age    racWht ADI_NA n_2020
## time        -0.019                                                        
## hh_ctgryNHn -0.050  0.202                                                 
## educ        -0.322  0.015 -0.117                                          
## gender      -0.204  0.001 -0.095  0.144                                   
## age         -0.496 -0.011 -0.017  0.014  0.079                            
## raceWhite   -0.115  0.003  0.136 -0.005  0.023 -0.007                     
## ADI_NATRANK -0.305  0.019 -0.061  0.254 -0.072  0.122  0.348              
## ns_cns2020_ -0.734  0.002  0.033  0.024  0.017 -0.108 -0.025  0.081       
## tm:hh_ctgNH  0.008 -0.896 -0.233 -0.015  0.005  0.011  0.007 -0.012  0.008

2.1 Model 1 table

Effect of hearing handicap status at baseline on pacc score
  pacc
Predictors Estimates CI p
(Intercept) 0.89 -0.94 – 2.72 0.342
time -0.07 -0.10 – -0.04 <0.001
hhi category [No
Handicap] 		-0.02 -0.19 – 0.15 0.833
educ 0.03 -0.00 – 0.06 0.063
gender 0.24 0.10 – 0.38 0.001
age -0.03 -0.05 – -0.02 <0.001
race [White] 0.37 0.16 – 0.58 0.001
ADI NATRANK -0.00 -0.01 – 0.00 0.096
noise censusblock2020
mean 		0.01 -0.02 – 0.03 0.618
time × hhi category [No
Handicap] 		0.03 -0.01 – 0.06 0.113
Random Effects
σ2 0.16
τ00 ID 0.40
ICC 0.71
N ID 393
Observations 1068
Marginal R2 / Conditional R2 0.122 / 0.748

2.2 Model 1 data vis

## Intercept for HHI Handicap: 0.8896
## Intercept for HHI No Handicap: 0.8715

3 Model 2: Words in noise

The outcome is the absolute pacc scores themselves at each follow up

model_WIN_cat <- lmer(
  pacc ~ time * WIN_CAT + educ + gender + age + race + ADI_NATRANK + noise_censusblock2020_mean + (1 | ID),
  data = Drives_WIN_HHI_longitudinal_long_clean,
  REML = FALSE
)
# View the summary of the model
summary(model_WIN_cat)
## Linear mixed model fit by maximum likelihood  ['lmerMod']
## Formula: pacc ~ time * WIN_CAT + educ + gender + age + race + ADI_NATRANK +  
##     noise_censusblock2020_mean + (1 | ID)
##    Data: Drives_WIN_HHI_longitudinal_long_clean
## 
##      AIC      BIC   logLik deviance df.resid 
##   1844.6   1904.1   -910.3   1820.6     1036 
## 
## Scaled residuals: 
##      Min       1Q   Median       3Q      Max 
## -11.0250  -0.3818   0.0507   0.4412   7.7702 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  ID       (Intercept) 0.4031   0.6349  
##  Residual             0.1653   0.4066  
## Number of obs: 1048, groups:  ID, 381
## 
## Fixed effects:
##                             Estimate Std. Error t value
## (Intercept)                 0.574339   0.950597   0.604
## time                       -0.047973   0.011779  -4.073
## WIN_CATNormal hearing       0.171807   0.080409   2.137
## educ                        0.026706   0.015342   1.741
## gender                      0.190457   0.075604   2.519
## age                        -0.030456   0.007535  -4.042
## raceWhite                   0.415370   0.109157   3.805
## ADI_NATRANK                -0.002354   0.001669  -1.410
## noise_censusblock2020_mean  0.007966   0.014070   0.566
## time:WIN_CATNormal hearing  0.003393   0.014984   0.226
## 
## Correlation of Fixed Effects:
##             (Intr) time   WIN_Ch educ   gender age    racWht ADI_NA n_2020
## time        -0.019                                                        
## WIN_CATNrmh -0.129  0.169                                                 
## educ        -0.318  0.006 -0.112                                          
## gender      -0.167  0.007 -0.248  0.163                                   
## age         -0.505  0.000  0.242 -0.017  0.011                            
## raceWhite   -0.130  0.010  0.142 -0.003 -0.004  0.031                     
## ADI_NATRANK -0.309  0.024  0.064  0.244 -0.088  0.131  0.360              
## ns_cns2020_ -0.723 -0.001 -0.024  0.039  0.020 -0.117 -0.027  0.075       
## t:WIN_CATNh  0.003 -0.786 -0.230 -0.004  0.003 -0.007  0.004 -0.018  0.019

3.1 Model 2 table

Effect of WIN_CAT status at baseline on pacc score
  pacc
Predictors Estimates CI p
(Intercept) 0.57 -1.29 – 2.44 0.546
time -0.05 -0.07 – -0.02 <0.001
WIN CAT [Normal hearing] 0.17 0.01 – 0.33 0.033
educ 0.03 -0.00 – 0.06 0.082
gender 0.19 0.04 – 0.34 0.012
age -0.03 -0.05 – -0.02 <0.001
race [White] 0.42 0.20 – 0.63 <0.001
ADI NATRANK -0.00 -0.01 – 0.00 0.159
noise censusblock2020
mean 		0.01 -0.02 – 0.04 0.571
time × WIN CAT [Normal
hearing] 		0.00 -0.03 – 0.03 0.821
Random Effects
σ2 0.17
τ00 ID 0.40
ICC 0.71
N ID 381
Observations 1048
Marginal R2 / Conditional R2 0.135 / 0.748

3.2 Data vis for model 2

## Intercept for Hearing Impairment: 0.5743
## Intercept for Normal Hearing: 0.7461

4 Model 3: Hearing aid use

The outcome is the absolute pacc scores themselves at each follow up

model_hearingaid_use <- lmer(
  pacc ~ time * hearingaids1 + educ + gender + age + ADI_NATRANK + noise_censusblock2020_mean + (1 | ID),
  data = Drives_WIN_HHI_longitudinal_long_clean,
  REML = FALSE
)
# View the summary of the model
summary(model_hearingaid_use)
## Linear mixed model fit by maximum likelihood  ['lmerMod']
## Formula: pacc ~ time * hearingaids1 + educ + gender + age + ADI_NATRANK +  
##     noise_censusblock2020_mean + (1 | ID)
##    Data: Drives_WIN_HHI_longitudinal_long_clean
## 
##      AIC      BIC   logLik deviance df.resid 
##   2834.7   2896.1  -1406.4   2812.7     1958 
## 
## Scaled residuals: 
##      Min       1Q   Median       3Q      Max 
## -12.1052  -0.4526   0.0496   0.4828   8.9091 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  ID       (Intercept) 0.4058   0.6370  
##  Residual             0.1323   0.3638  
## Number of obs: 1969, groups:  ID, 517
## 
## Fixed effects:
##                             Estimate Std. Error t value
## (Intercept)                 1.767669   0.809121   2.185
## time                       -0.043239   0.003713 -11.644
## hearingaids1                0.039626   0.084875   0.467
## educ                        0.017721   0.012707   1.395
## gender                      0.215460   0.062075   3.471
## age                        -0.034448   0.006216  -5.542
## ADI_NATRANK                -0.004415   0.001312  -3.365
## noise_censusblock2020_mean  0.005242   0.011928   0.439
## time:hearingaids1          -0.011489   0.009970  -1.152
## 
## Correlation of Fixed Effects:
##             (Intr) time   hrngd1 educ   gender age    ADI_NA n_2020
## time        -0.018                                                 
## hearingads1  0.053  0.079                                          
## educ        -0.372 -0.004  0.061                                   
## gender      -0.163  0.008  0.164  0.165                            
## age         -0.515  0.012 -0.138  0.060  0.047                     
## ADI_NATRANK -0.291  0.014  0.027  0.237 -0.086  0.107              
## ns_cns2020_ -0.746  0.003 -0.039  0.057 -0.024 -0.088  0.135       
## tim:hrngds1  0.006 -0.372 -0.187  0.004  0.002 -0.004 -0.003 -0.003

4.1 Model 3 table

Effect of hearing aid use status at baseline on pacc score
  pacc
Predictors Estimates CI p
(Intercept) 1.77 0.18 – 3.35 0.029
time -0.04 -0.05 – -0.04 <0.001
hearingaids1 0.04 -0.13 – 0.21 0.641
educ 0.02 -0.01 – 0.04 0.163
gender 0.22 0.09 – 0.34 0.001
age -0.03 -0.05 – -0.02 <0.001
ADI NATRANK -0.00 -0.01 – -0.00 0.001
noise censusblock2020
mean 		0.01 -0.02 – 0.03 0.660
time × hearingaids1 -0.01 -0.03 – 0.01 0.249
Random Effects
σ2 0.13
τ00 ID 0.41
ICC 0.75
N ID 517
Observations 1969
Marginal R2 / Conditional R2 0.099 / 0.779

4.2 Model 3 data vis

## Intercept for No Hearing Aid Use: 1.7677
## Intercept for Hearing Aid Use: 1.8073
# Create the labeled factor variable
Drives_WIN_HHI_longitudinal_long_clean <- Drives_WIN_HHI_longitudinal_long_clean %>%
  mutate(
    hearaid_win = case_when(
      hearingaids1 == 0 & WIN_CAT == "Normal hearing"   ~ 0,
      hearingaids1 == 1 & WIN_CAT == "Normal hearing"   ~ 1,
      hearingaids1 == 0 & WIN_CAT == "Impaired hearing" ~ 2,
      hearingaids1 == 1 & WIN_CAT == "Impaired hearing" ~ 3,
      TRUE ~ NA_real_  # Assigns NA if conditions are not met
    ),
    hearaid_win = factor(hearaid_win, levels = 0:3,  # Ensure it's a factor
                         labels = c("Normal Hearing & No Hearing Aid",
                                    "Normal Hearing & Hearing Aid",
                                    "Hearing Impairment & No Hearing Aid",
                                    "Hearing Impairment & Hearing Aid"))
  )

# Count occurrences at the baseline event with labeled categories
Drives_WIN_HHI_longitudinal_long_clean %>%
  filter(event_name == "baseline") %>%
  count(hearaid_win) %>%
  arrange(hearaid_win)  # Keeps categories in defined order
## # A tibble: 5 × 2
##   hearaid_win                             n
##   <fct>                               <int>
## 1 Normal Hearing & No Hearing Aid       209
## 2 Normal Hearing & Hearing Aid            7
## 3 Hearing Impairment & No Hearing Aid    99
## 4 Hearing Impairment & Hearing Aid       59
## 5 <NA>                                  174

5 Model 4: hearing aid use and WIN

model_hearaid_win <- lmer(
  pacc ~ time * hearaid_win + educ + gender + age + ADI_NATRANK + noise_censusblock2020_mean + (1 | ID),
  data = Drives_WIN_HHI_longitudinal_long_clean,
  REML = FALSE
)
# View the summary of the model
summary(model_hearaid_win)
## Linear mixed model fit by maximum likelihood  ['lmerMod']
## Formula: pacc ~ time * hearaid_win + educ + gender + age + ADI_NATRANK +  
##     noise_censusblock2020_mean + (1 | ID)
##    Data: Drives_WIN_HHI_longitudinal_long_clean
## 
##      AIC      BIC   logLik deviance df.resid 
##   1754.0   1827.8   -862.0   1724.0      994 
## 
## Scaled residuals: 
##      Min       1Q   Median       3Q      Max 
## -11.3287  -0.3983   0.0523   0.4344   8.0206 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  ID       (Intercept) 0.4151   0.6443  
##  Residual             0.1562   0.3952  
## Number of obs: 1009, groups:  ID, 369
## 
## Fixed effects:
##                                                      Estimate Std. Error
## (Intercept)                                          1.155102   0.969368
## time                                                -0.040010   0.009142
## hearaid_winNormal Hearing & Hearing Aid              0.190099   0.283220
## hearaid_winHearing Impairment & No Hearing Aid      -0.202484   0.093053
## hearaid_winHearing Impairment & Hearing Aid         -0.040110   0.114299
## educ                                                 0.026842   0.015885
## gender                                               0.214630   0.078253
## age                                                 -0.030075   0.007730
## ADI_NATRANK                                         -0.004196   0.001600
## noise_censusblock2020_mean                           0.006963   0.014437
## time:hearaid_winNormal Hearing & Hearing Aid        -0.105413   0.099449
## time:hearaid_winHearing Impairment & No Hearing Aid -0.025600   0.017563
## time:hearaid_winHearing Impairment & Hearing Aid    -0.004826   0.021884
##                                                     t value
## (Intercept)                                           1.192
## time                                                 -4.377
## hearaid_winNormal Hearing & Hearing Aid               0.671
## hearaid_winHearing Impairment & No Hearing Aid       -2.176
## hearaid_winHearing Impairment & Hearing Aid          -0.351
## educ                                                  1.690
## gender                                                2.743
## age                                                  -3.891
## ADI_NATRANK                                          -2.623
## noise_censusblock2020_mean                            0.482
## time:hearaid_winNormal Hearing & Hearing Aid         -1.060
## time:hearaid_winHearing Impairment & No Hearing Aid  -1.458
## time:hearaid_winHearing Impairment & Hearing Aid     -0.221

5.1 Model 4 table

Effect of hearing aid use and WIN_CAT status at baseline on pacc score
  pacc
Predictors Estimates CI p
(Intercept) 1.16 -0.75 – 3.06 0.234
time -0.04 -0.06 – -0.02 <0.001
hearaid win [Normal
Hearing & Hearing Aid] 		0.19 -0.37 – 0.75 0.502
hearaid win [Hearing
Impairment & No Hearing
Aid] 		-0.20 -0.39 – -0.02 0.030
hearaid win [Hearing
Impairment & Hearing Aid] 		-0.04 -0.26 – 0.18 0.726
educ 0.03 -0.00 – 0.06 0.091
gender 0.21 0.06 – 0.37 0.006
age -0.03 -0.05 – -0.01 <0.001
ADI NATRANK -0.00 -0.01 – -0.00 0.009
noise censusblock2020
mean 		0.01 -0.02 – 0.04 0.630
time × hearaid win
[Normal Hearing & Hearing
Aid] 		-0.11 -0.30 – 0.09 0.289
time × hearaid win
[Hearing Impairment & No
Hearing Aid] 		-0.03 -0.06 – 0.01 0.145
time × hearaid win
[Hearing Impairment &
Hearing Aid] 		-0.00 -0.05 – 0.04 0.826
Random Effects
σ2 0.16
τ00 ID 0.42
ICC 0.73
N ID 369
Observations 1009
Marginal R2 / Conditional R2 0.116 / 0.758

5.2 Model 4 data vis

## Intercept for Normal Hearing & No Hearing Aid (0): 1.1551
## Intercept for Normal Hearing & Hearing Aid (1): 1.3452
## Intercept for Hearing Impairment & No Hearing Aid (2): 0.9526
## Intercept for Hearing Impairment & Hearing Aid (3): 1.1150

Pacc score classification last follow up. Note PACC classification was done by first calculating the mean and sd of baseline pacc for the last available otdate. Those below and equal to -1 sd were called “low pacc” and those above were named “high pacc.

## Mean of last available PACC (excluding baseline): -0.2238
## SD of last available PACC (excluding baseline): 0.8653
## # A tibble: 2 × 2
##   last_PACC_class_sick Count
##   <fct>                <int>
## 1 high PACC              385
## 2 low PACC                53

Baseline pacc classification

## # A tibble: 2 × 2
##   PACC_class_sick_long_bl     n
##   <fct>                   <int>
## 1 high PACC_BL              342
## 2 low PACC_BL               107

#logiostic regression: predicting last pacc score bin using baseline pacc bin

## 
## Call:
## glm(formula = last_PACC_class_sick ~ PACC_class_sick_long_bl, 
##     family = binomial, data = filtered_data)
## 
## Coefficients:
##                                    Estimate Std. Error z value Pr(>|z|)    
## (Intercept)                          2.8016     0.2498  11.216  < 2e-16 ***
## PACC_class_sick_long_bllow PACC_BL  -2.2095     0.3495  -6.322 2.57e-10 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 265.90  on 369  degrees of freedom
## Residual deviance: 225.34  on 368  degrees of freedom
##   (178 observations deleted due to missingness)
## AIC: 229.34
## 
## Number of Fisher Scoring iterations: 5
logistic regression baseline pacc bin predicting last pacc follow-up bin
  last_PACC_class_sick
Predictors Odds Ratios CI p
(Intercept) 16.47 10.43 – 27.93 <0.001
PACC class sick long bl
[low PACC_BL] 		0.11 0.05 – 0.22 <0.001
Observations 370
R2 Tjur 0.138