Missing data imputation and exploratory analyses
Demographics | Physical health | Hearing | Cognition | Mobility and balance | Quality of life | Mental health | Social participation and support | Views on hearing loss |
Background
Why impute missing data?
Logistic regression and other statistical methods require complete cases. In our case, with 40+ variables of interest, dropping participants with any missing values leads to a sample size of 200+ cases instead of 500+. There is loss of power, and estimates from models will likely be biased. The goal of missing data imputation is to restore variance of the sample to make it better reflect characteristics of the population (i.e., give accurate estimates of standard errors, CI’s, and p-values).
MICE procedure
‘Multiple Imputation by Chained Equations’
Say there are three variables ‘A’, ‘B’, and ‘C’, each with some missing values. Choose variable ‘A’ to start filling in. Fill in missing values in ‘B’ and ‘C’ with temporary values, such as the mean.
Make an equation A ~ B + C, using observed cases of ‘A’ to predict missing values of ‘A’, including a random component. Move on to predicting missing ‘B’, using B ~ A + C and including predicted ‘A’, again including a random component. Move on to predicting missing ‘C’, using C ~ A + B and predicted ‘A’ and ‘B’. This is one cycle.
Choose the number of cycles, say 10 iterations. At the end of 10 iterations, an “imputed” dataset is formed with no missing values. Choose the number of imputed datasets to form (the same analysis will be conducted on each slightly different dataset, and the results will be pooled). Choose which variables should be used as predictors of other variables. Choose which method to use for predicting missing values, e.g., predictive mean matching. For derived variables such as summed scores, consider whether to impute components first and then calculate derived scores, or impute derived scores. Consider whether there should be an order of which variables to start imputing first.
Check whether the imputed data “converge” (the means and SEs stabilize after a few iterations, and are similar across imputed datasets). Check the distributions and characteristics of imputed data (e.g. integer-only values, floors and ceilings, plausibility), and their similarity to observed data.
1. Pre-imputation work
(i) Evaluate missingness
Variables of interest
Demographics: Age, Gender, Work, Volunteer, Live alonePhysical health: Rating, Multimorbidity score, Subjective vision (loss, rating aided), Subjective hearing (loss, rating aided, rating unaided)
Hearing health: Hearing device use, Hearing device type, Tinnitus past 7 days, 15i-SSQ mean score, SIM mean score, EmoCheQ mean score, HHIE-S total score, BPTA4
Cognitive health: Subjective cognitive impairment, CSRQ mean score
Mobility and balance: Mobility aid use, ABC (balance confidence) mean score
Quality of life: SWLS mean score, WHO overall QoL, WHO health QoL, WHO money, WHO four domains
Mental health: PHQ4, split into anxiety and depression subscores
Social factors: Loneliness, Social Network Index, Social Participation Frequency, Social Participation Types, Social Support MOS mean score
Views: Connections total correct score, Motivations mean score
% Missing data within participants
##
## The decimal point is at the |
##
## 0 | 00004444444444444444444444444444444444444999999999999999999999999999+193
## 2 | 22222222222222222222222222222222222222222277777777777777777777777777+16
## 4 | 000044444999993333333888888888888
## 6 | 222222222222222666666666666666666111111111111111111155555555555
## 8 | 0000000004444888833337777
## 10 | 221199
## 12 | 488
## 14 | 2055
## 16 | 47
## 18 | 65
## 20 |
## 22 | 5
## 24 | 2
## 26 |
## 28 |
## 30 |
## 32 |
## 34 |
## 36 |
## 38 |
## 40 |
## 42 | 9
## 44 |
## 46 |
## 48 |
## 50 |
## 52 |
## 54 |
## 56 |
## 58 |
## 60 | 6
## 62 |
## 64 | 6
## 66 | 87
## 68 |
## 70 | 8827
## 72 | 6
## 74 | 337
## 76 | 0
## 78 |
## 80 | 5
## 82 |
## 84 |
## 86 |
## 88 |
## 90 | 6
## 92 |
## 94 |
## 96 |
## 98 | 66Missing data patterns
Participants with <26% missing data (n=509): 
Participants with >42% missing data (n=18): 
(ii) Data cleaning
Rules
- For “single-item” variables (e.g. loneliness): NA’s are left as
is.
- For co-morbidity checklist: All responses other than “yes” (NA’s,
No, Don’t know) are treated as “no”.
- For composite variables: If a participant has “NA” for all items,
the composite score is also “NA”.
- WHOQOL-BREF: Follow their protocol for NA’s.
- For SIM, EmoCheQ, CSRQ, PHQ4, Views on motivations: Impute case
mean.
- For 15i-SSQ: Impute case mean within subscale.
- For HHIES: Impute case mode within subscale.
- For ABC: Recode NA’s to 0 (single participant).
- For Social Network Index: Recode NA’s to “never”.
- For Social Participation: Recode NA’s to “never”.
- For Social Support: Recode NA’s to “none”.
- For Views (Connections): Drop foil item; only correct items
credited.
- For Views (Motivations): Impute mean.
(iii) Evaluate missingness again
% Missing data within participants
##
## The decimal point is at the |
##
## 0 | 00000000000000000000000000000000000000000000000000000000000000000000+151
## 2 | 11111111111111111111111111111111111111111111111111111111111111111111+51
## 4 | 222222222222222222222222222222222222222222222
## 6 | 333333333333333333333333333333333333333333333333333333333
## 8 | 333333333333333333333333333
## 10 | 44444444444
## 12 | 55555
## 14 |
## 16 | 7
## 18 | 8
## 20 |
## 22 |
## 24 |
## 26 |
## 28 |
## 30 |
## 32 |
## 34 |
## 36 |
## 38 | 6
## 40 |
## 42 |
## 44 |
## 46 | 9
## 48 |
## 50 | 0
## 52 | 1
## 54 | 2
## 56 |
## 58 | 3333
## 60 | 444
## 62 |
## 64 | 6
## 66 |
## 68 | 8
## 70 | 8
## 72 |
## 74 | 0
## 76 |
## 78 |
## 80 |
## 82 |
## 84 |
## 86 |
## 88 | 66Missing data patterns
Participants with <20% missing data (n=509):
## # A tibble: 48 × 3
## variable n_miss pct_miss
## <chr> <int> <num>
## 1 Gender_bin 0 0
## 2 Age 2 0.393
## 3 Retired_bin 0 0
## 4 Volunteer 0 0
## 5 Lives_alone_bin 0 0
## 6 Phys_health_rating 7 1.38
## 7 Multimorbidity_score 0 0
## 8 Subj_vision_loss_bin 18 3.54
## 9 Subj_vision_rating_aided 1 0.196
## 10 Subj_hearing_loss_bin 36 7.07
## 11 any_hearing_device 0 0
## 12 study2_HA 0 0
## 13 Subj_hearing_rating_aided_REV 159 31.2
## 14 Subj_hearing_rating_unaided_REV 8 1.57
## 15 Tinnitus_past_wk_bin 0 0
## 16 ssq_speech_REV 46 9.04
## 17 ssq_spatial_REV 51 10.0
## 18 ssq_qualities_REV 3 0.589
## 19 SIM_mean 12 2.36
## 20 Emocheq_mean 8 1.57
## 21 hhies_emo_total 0 0
## 22 hhies_soc_total 0 0
## 23 HHIES_total 0 0
## 24 positive_SCI_bin 0 0
## 25 CSRQ_mean 0 0
## 26 Mobility_needs_bin 0 0
## 27 ABC_mean 0 0
## 28 SWLS_mean 0 0
## 29 WHOQOL_overall_qol 0 0
## 30 WHOQOL_health_qol 0 0
## 31 WHOQOL_Dom1_phys 0 0
## 32 WHOQOL_Dom2_psy 0 0
## 33 WHOQOL_Dom3_soc 1 0.196
## 34 WHOQOL_Dom4_env 0 0
## 35 WHO_money 0 0
## 36 PHQ4_anx_sum 1 0.196
## 37 PHQ4_dep_sum 1 0.196
## 38 Lonely_bin 1 0.196
## 39 Social_network_index 1 0.196
## 40 Soc_part_freq 1 0.196
## 41 Soc_part_types 1 0.196
## 42 Socsupp_mean 5 0.982
## 43 Connections_total 9 1.77
## 44 Motivate_mean 5 0.982
## 45 study1_HA_bin 78 15.3
## 46 study2_HA_bin 0 0
## 47 PTA4_better_ear 73 14.3
## 48 PTA4_asym 74 14.5Participants with >39%% missing data (n=18):
(iv) Justify discarding/retaining participants
Drop 2 cases that are fully missing
## [1] Progress Duration.in.seconds Finished
## <0 rows> (or 0-length row.names)## # A tibble: 46 × 3
## variable n_miss pct_miss
## <chr> <int> <num>
## 1 Gender_bin 0 NaN
## 2 Age 0 NaN
## 3 Retired_bin 0 NaN
## 4 Volunteer 0 NaN
## 5 Lives_alone_bin 0 NaN
## 6 Phys_health_rating 0 NaN
## 7 Multimorbidity_score 0 NaN
## 8 Subj_vision_loss_bin 0 NaN
## 9 Subj_vision_rating_aided 0 NaN
## 10 Subj_hearing_loss_bin 0 NaN
## 11 any_hearing_device 0 NaN
## 12 study2_HA 0 NaN
## 13 Subj_hearing_rating_aided_REV 0 NaN
## 14 Subj_hearing_rating_unaided_REV 0 NaN
## 15 Tinnitus_past_wk_bin 0 NaN
## 16 ssq_speech_REV 0 NaN
## 17 ssq_spatial_REV 0 NaN
## 18 ssq_qualities_REV 0 NaN
## 19 SIM_mean 0 NaN
## 20 Emocheq_mean 0 NaN
## 21 hhies_emo_total 0 NaN
## 22 hhies_soc_total 0 NaN
## 23 HHIES_total 0 NaN
## 24 positive_SCI_bin 0 NaN
## 25 CSRQ_mean 0 NaN
## 26 Mobility_needs_bin 0 NaN
## 27 ABC_mean 0 NaN
## 28 SWLS_mean 0 NaN
## 29 WHOQOL_overall_qol 0 NaN
## 30 WHOQOL_health_qol 0 NaN
## 31 WHOQOL_Dom1_phys 0 NaN
## 32 WHOQOL_Dom2_psy 0 NaN
## 33 WHOQOL_Dom3_soc 0 NaN
## 34 WHOQOL_Dom4_env 0 NaN
## 35 WHO_money 0 NaN
## 36 PHQ4_anx_sum 0 NaN
## 37 PHQ4_dep_sum 0 NaN
## 38 Lonely_bin 0 NaN
## 39 Social_network_index 0 NaN
## 40 Soc_part_freq 0 NaN
## 41 Soc_part_types 0 NaN
## 42 Socsupp_mean 0 NaN
## 43 Connections_total 0 NaN
## 44 Motivate_mean 0 NaN
## 45 study1_HA_bin 0 NaN
## 46 study2_HA_bin 0 NaNRetain 16 that mostly have age, audiogram, earlier measures
## Progress Duration.in.seconds Finished
## 45 58 413 0
## 77 19 247 0
## 136 58 308782 0
## 153 30 449 0
## 166 52 185 0
## 176 53 1693 0
## 187 53 518035 0
## 192 45 504174 0
## 228 13 96 0
## 246 64 1719 0
## 318 55 466 0
## 364 55 988 0
## 405 55 650 0
## 458 0 217 0
## 467 1 540 0
## 483 45 731 0
## 520 64 6724 0
## 522 70 2376 0## # A tibble: 46 × 3
## variable n_miss pct_miss
## <chr> <int> <num>
## 1 Gender_bin 2 11.1
## 2 Age 3 16.7
## 3 Retired_bin 2 11.1
## 4 Volunteer 2 11.1
## 5 Lives_alone_bin 2 11.1
## 6 Phys_health_rating 4 22.2
## 7 Multimorbidity_score 0 0
## 8 Subj_vision_loss_bin 4 22.2
## 9 Subj_vision_rating_aided 4 22.2
## 10 Subj_hearing_loss_bin 5 27.8
## 11 any_hearing_device 0 0
## 12 study2_HA 0 0
## 13 Subj_hearing_rating_aided_REV 10 55.6
## 14 Subj_hearing_rating_unaided_REV 5 27.8
## 15 Tinnitus_past_wk_bin 0 0
## 16 ssq_speech_REV 15 83.3
## 17 ssq_spatial_REV 13 72.2
## 18 ssq_qualities_REV 8 44.4
## 19 SIM_mean 13 72.2
## 20 Emocheq_mean 13 72.2
## 21 hhies_emo_total 15 83.3
## 22 hhies_soc_total 15 83.3
## 23 HHIES_total 15 83.3
## 24 positive_SCI_bin 15 83.3
## 25 CSRQ_mean 17 94.4
## 26 Mobility_needs_bin 17 94.4
## 27 ABC_mean 17 94.4
## 28 SWLS_mean 17 94.4
## 29 WHOQOL_overall_qol 18 100
## 30 WHOQOL_health_qol 18 100
## 31 WHOQOL_Dom1_phys 18 100
## 32 WHOQOL_Dom2_psy 18 100
## 33 WHOQOL_Dom3_soc 18 100
## 34 WHOQOL_Dom4_env 18 100
## 35 WHO_money 18 100
## 36 PHQ4_anx_sum 18 100
## 37 PHQ4_dep_sum 18 100
## 38 Lonely_bin 18 100
## 39 Social_network_index 18 100
## 40 Soc_part_freq 18 100
## 41 Soc_part_types 18 100
## 42 Socsupp_mean 18 100
## 43 Connections_total 18 100
## 44 Motivate_mean 18 100
## 45 study1_HA_bin 5 27.8
## 46 study2_HA_bin 0 0(v) Variable relationships
Correlations heatmap: Hearing
Correlations heatmap: Physical health
PCA
Note: PC1 and PC2 account for 36% of the total variance.
2. Imputation using MICE
Current method
- Use predictors that correlate 0.5 or better to impute missing data
in a variable (30 variables); if no such predictors, adjust threshold to
0.4 (3 variables), then 0.3, then 0.2 (2 variables)
pdf
- Eight variables are complete and do not require imputing; 2
variables have no predictors that correlate at least r = 0.2
(Subj_vision_loss_bin, Connections_total)
- MICE settings: total of 5 imputed datasets, with a maximum of 10 iterations per dataset, using predictive means matching to impute missing values
Check imputed data
PCA with standardized imputed data
PCA
Note: This is only 1 of 5 imputed datasets, but the plots are similar.
3. Predicting hearing aid adoption
Complete cases vs. imputed data
Note: All participants together; not filtered by hearing aid candidacy.
Regression output (complete cases, n=346)
Non-hearing aid users = 267
Hearing aid users = 79 (22.8%)
Note: Dropped Subj_hearing_rating_aided_REV & SWLS_mean due to VIF
> 5
## Term Estimate StdError z p OddsRatio CIlower CIupper
## 1 (Intercept) -24.2964 884.2360 -0.0275 0.9781 0.0000 0.0000 Inf
## 2 Gender_bin 0.5265 0.4564 1.1536 0.2487 1.6930 0.6921 4.1418
## 3 Age -0.0185 0.0346 -0.5336 0.5936 0.9817 0.9172 1.0506
## 4 Retired_bin 0.0571 0.5098 0.1121 0.9108 1.0588 0.3898 2.8757
## 5 Volunteer_rec 0.5477 0.2816 1.9448 0.0518 1.7293 0.9957 3.0031
## 6 Lives_alone_bin -0.2751 0.5566 -0.4943 0.6211 0.7595 0.2551 2.2608
## 7 Phys_health_rating 0.0966 0.3728 0.2590 0.7956 1.1014 0.5304 2.2868
## 8 Multimorbidity_score -0.0199 0.1206 -0.1652 0.8688 0.9803 0.7739 1.2417
## 9 Mobility_needs_bin -0.3204 1.1148 -0.2874 0.7738 0.7258 0.0816 6.4536
## 10 ABC_mean -0.0315 0.0261 -1.2055 0.2280 0.9690 0.9206 1.0199
## 11 Subj_vision_rating_aided 0.0508 0.2960 0.1717 0.8637 1.0521 0.5890 1.8795
## 12 Subj_hearing_loss_bin 17.0709 884.2210 0.0193 0.9846 25929968.0135 0.0000 Inf
## 13 Subj_hearing_rating_unaided_REV 1.0591 0.3850 2.7506 0.0059 2.8837 1.3559 6.1334
## 14 Tinnitus_past_wk_bin 0.1555 0.4104 0.3790 0.7047 1.1683 0.5227 2.6112
## 15 PTA4_better_ear 0.1214 0.0288 4.2225 0.0000 1.1291 1.0672 1.1945
## 16 PTA4_asym 0.0472 0.0153 3.0891 0.0020 1.0483 1.0174 1.0802
## 17 SSQ15i_mean_REV -0.6680 0.1855 -3.6002 0.0003 0.5128 0.3564 0.7376
## 18 Emocheq_mean -0.0308 0.2780 -0.1108 0.9118 0.9697 0.5623 1.6721
## 19 HHIES_total 0.0596 0.0386 1.5445 0.1225 1.0614 0.9841 1.1448
## 20 SIM_mean -0.1358 0.1165 -1.1662 0.2435 0.8730 0.6948 1.0968
## 21 positive_SCI_bin 1.1976 0.7118 1.6824 0.0925 3.3122 0.8207 13.3674
## 22 CSRQ_mean -0.1344 0.6195 -0.2170 0.8282 0.8742 0.2596 2.9440
## 23 PHQ4_anx_sum 0.7501 0.2500 3.0009 0.0027 2.1173 1.2972 3.4559
## 24 PHQ4_dep_sum -0.5725 0.3340 -1.7137 0.0866 0.5641 0.2931 1.0857
## 25 WHOQOL_overall_qol -0.4355 0.4717 -0.9233 0.3558 0.6469 0.2567 1.6306
## 26 WHOQOL_health_qol 0.3760 0.3228 1.1648 0.2441 1.4565 0.7736 2.7421
## 27 WHO_money 0.8039 0.2691 2.9869 0.0028 2.2343 1.3184 3.7865
## 28 Socsupp_mean 0.2161 0.2994 0.7216 0.4705 1.2412 0.6902 2.2320
## 29 Social_network_index -0.1362 0.1756 -0.7754 0.4381 0.8727 0.6185 1.2313
## 30 Soc_part_freq -0.2681 0.3671 -0.7303 0.4652 0.7648 0.3724 1.5706
## 31 Soc_part_types 0.0183 0.1582 0.1155 0.9080 1.0184 0.7469 1.3886
## 32 Lonely_bin -0.2511 0.6864 -0.3658 0.7145 0.7780 0.2026 2.9868
## 33 Motivate_mean 0.3051 0.2395 1.2743 0.2026 1.3568 0.8486 2.1694Regression output (imputed data, n=525)
## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -0.6224 0.3557 -1.7497 284.6201 0.0813 0.5367 0.2672 1.0776
## 2 Gender_bin 0.0549 0.0365 1.5051 444.0992 0.1330 1.0564 0.9835 1.1348
## 3 Age 0.0029 0.0026 1.1257 243.8879 0.2614 1.0029 0.9978 1.0080
## 4 Retired_bin -0.0276 0.0370 -0.7458 467.8971 0.4562 0.9728 0.9047 1.0459
## 5 Volunteer_rec 0.0396 0.0209 1.9002 383.6381 0.0582 1.0404 0.9986 1.0839
## 6 Lives_alone_bin -0.0302 0.0420 -0.7186 442.1674 0.4727 0.9703 0.8936 1.0535
## 7 Phys_health_rating -0.0096 0.0285 -0.3362 146.0093 0.7372 0.9904 0.9366 1.0473
## 8 Multimorbidity_score 0.0041 0.0077 0.5383 468.0730 0.5906 1.0041 0.9891 1.0194
## 9 Mobility_needs_bin -0.0296 0.0883 -0.3356 141.4920 0.7377 0.9708 0.8165 1.1543
## 10 ABC_mean -0.0034 0.0017 -2.0141 292.7450 0.0449 0.9966 0.9933 0.9999
## 11 Subj_vision_rating_aided 0.0079 0.0232 0.3393 470.0114 0.7345 1.0079 0.9631 1.0548
## 12 Subj_hearing_loss_bin 0.0980 0.0504 1.9432 292.4333 0.0529 1.1030 0.9992 1.2175
## 13 Subj_hearing_rating_unaided_REV 0.0968 0.0259 3.7421 302.7899 0.0002 1.1016 1.0471 1.1590
## 14 Tinnitus_past_wk_bin -0.0528 0.0351 -1.5057 418.2857 0.1329 0.9486 0.8855 1.0161
## 15 PTA4_better_ear 0.0078 0.0024 3.2462 49.0647 0.0021 1.0078 1.0031 1.0126
## 16 PTA4_asym 0.0032 0.0016 1.9854 70.8022 0.0510 1.0032 1.0001 1.0064
## 17 SSQ15i_mean_REV -0.0708 0.0143 -4.9653 115.1459 0.0000 0.9316 0.9059 0.9581
## 18 Emocheq_mean -0.0081 0.0219 -0.3680 249.1891 0.7131 0.9919 0.9503 1.0354
## 19 HHIES_total 0.0060 0.0029 2.0605 259.0164 0.0403 1.0060 1.0003 1.0118
## 20 SIM_mean -0.0021 0.0080 -0.2632 457.9060 0.7925 0.9979 0.9824 1.0137
## 21 positive_SCI_bin 0.0571 0.0604 0.9452 167.0927 0.3459 1.0588 0.9406 1.1918
## 22 CSRQ_mean -0.0489 0.0520 -0.9397 120.6915 0.3493 0.9523 0.8600 1.0545
## 23 PHQ4_anx_sum 0.0586 0.0200 2.9348 120.2604 0.0040 1.0604 1.0196 1.1027
## 24 PHQ4_dep_sum -0.0051 0.0218 -0.2348 433.3507 0.8145 0.9949 0.9533 1.0383
## 25 WHOQOL_overall_qol 0.0295 0.0343 0.8600 258.5346 0.3906 1.0299 0.9630 1.1016
## 26 WHOQOL_health_qol 0.0285 0.0240 1.1861 197.2200 0.2370 1.0289 0.9816 1.0785
## 27 WHO_money 0.0346 0.0199 1.7409 164.3832 0.0836 1.0352 0.9956 1.0764
## 28 Socsupp_mean 0.0025 0.0234 0.1076 435.4925 0.9143 1.0025 0.9576 1.0496
## 29 Social_network_index -0.0039 0.0145 -0.2693 171.3773 0.7880 0.9961 0.9682 1.0248
## 30 Soc_part_freq 0.0024 0.0285 0.0829 259.3816 0.9340 1.0024 0.9479 1.0600
## 31 Soc_part_types -0.0052 0.0127 -0.4107 440.2311 0.6815 0.9948 0.9704 1.0199
## 32 Lonely_bin 0.0063 0.0497 0.1265 171.8542 0.8995 1.0063 0.9129 1.1093
## 33 Motivate_mean 0.0212 0.0199 1.0665 466.0848 0.2867 1.0214 0.9824 1.0621Legend: Asterisks denote p < 0.05; crosses denote p [0.05, 0.1]
Accuracy of model predictions
Note: Not really a fair comparison as models and model coefficients are different between complete-cases and imputed data.
## [1] "Complete cases, using median as boundary"## predicted
## actual 0 1
## 0 170 97
## 1 3 76## [1] "Accuracy: 0.711"## [1] "Sensitivity: 0.962"## [1] "Specificity: 0.637"
## [1] "Imputed data, using median as boundary"## predicted
## actual 0 1
## 0 255 167
## 1 8 95## [1] "Accuracy: 0.667"## [1] "Sensitivity: 0.922"## [1] "Specificity: 0.604"Grouping by HHIE-S status
Counts
0-8 no handicap, 10-24 mild to moderate, 26-40 severe
## HHIES
## UseHA None MildMod Sev <NA>
## 0 233 148 30 11
## 1 26 62 13 2
## <NA> 0 0 0 0## [1] 0.100 0.295 0.302 0.154Regression output (imputed data, n=525)
## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) 0.1676 0.4489 0.3734 222.4966 0.7092 1.1825 0.4905 2.8504
## 2 Gender_bin -0.0371 0.0432 -0.8589 197.5811 0.3915 0.9636 0.8854 1.0487
## 3 Age 0.0002 0.0029 0.0590 207.6885 0.9530 1.0002 0.9945 1.0059
## 4 Retired_bin -0.0246 0.0411 -0.5981 220.3189 0.5504 0.9757 0.9002 1.0576
## 5 Volunteer_rec 0.0308 0.0231 1.3355 212.1983 0.1832 1.0313 0.9856 1.0790
## 6 Lives_alone_bin 0.0354 0.0463 0.7638 208.4812 0.4458 1.0360 0.9462 1.1345
## 7 Phys_health_rating -0.0335 0.0313 -1.0695 216.5727 0.2861 0.9671 0.9095 1.0282
## 8 Multimorbidity_score 0.0096 0.0111 0.8626 217.4195 0.3893 1.0096 0.9879 1.0319
## 9 Mobility_needs_bin 0.0211 0.1124 0.1882 221.7430 0.8509 1.0213 0.8194 1.2730
## 10 ABC_mean -0.0015 0.0026 -0.5689 204.9233 0.5700 0.9985 0.9934 1.0036
## 11 Subj_vision_rating_aided -0.0214 0.0280 -0.7629 219.0210 0.4464 0.9788 0.9266 1.0340
## 12 Subj_hearing_loss_bin 0.0868 0.0516 1.6815 169.1639 0.0945 1.0907 0.9858 1.2068
## 13 Subj_hearing_rating_unaided_REV 0.0451 0.0307 1.4672 112.6230 0.1451 1.0461 0.9850 1.1110
## 14 Tinnitus_past_wk_bin -0.0482 0.0432 -1.1155 190.8091 0.2660 0.9529 0.8756 1.0371
## 15 PTA4_better_ear 0.0102 0.0031 3.2379 35.4834 0.0026 1.0103 1.0041 1.0164
## 16 PTA4_asym 0.0043 0.0030 1.4450 100.0494 0.1516 1.0043 0.9984 1.0102
## 17 SSQ15i_mean_REV -0.0538 0.0198 -2.7231 134.5648 0.0073 0.9476 0.9116 0.9851
## 18 Emocheq_mean -0.0061 0.0287 -0.2122 221.4369 0.8321 0.9939 0.9396 1.0514
## 19 HHIES_total 0.0041 0.0075 0.5435 216.6440 0.5874 1.0041 0.9895 1.0190
## 20 SIM_mean 0.0029 0.0093 0.3079 199.0190 0.7585 1.0029 0.9848 1.0214
## 21 positive_SCI_bin 0.1420 0.1035 1.3719 210.3131 0.1716 1.1526 0.9410 1.4118
## 22 CSRQ_mean -0.0344 0.0540 -0.6374 216.2184 0.5245 0.9662 0.8691 1.0741
## 23 PHQ4_anx_sum 0.0515 0.0232 2.2210 206.7393 0.0274 1.0528 1.0060 1.1018
## 24 PHQ4_dep_sum -0.0423 0.0306 -1.3825 203.2001 0.1683 0.9586 0.9028 1.0178
## 25 WHOQOL_overall_qol -0.0148 0.0379 -0.3900 221.7773 0.6969 0.9853 0.9148 1.0613
## 26 WHOQOL_health_qol 0.0053 0.0269 0.1977 211.4222 0.8435 1.0053 0.9537 1.0597
## 27 WHO_money 0.0440 0.0237 1.8560 211.7084 0.0648 1.0450 0.9976 1.0947
## 28 Socsupp_mean 0.0169 0.0292 0.5789 195.6673 0.5633 1.0170 0.9605 1.0769
## 29 Social_network_index -0.0096 0.0157 -0.6140 176.1308 0.5400 0.9904 0.9604 1.0214
## 30 Soc_part_freq -0.0126 0.0335 -0.3762 205.1765 0.7072 0.9875 0.9247 1.0545
## 31 Soc_part_types -0.0068 0.0141 -0.4822 201.4868 0.6302 0.9932 0.9662 1.0211
## 32 Lonely_bin -0.0114 0.0612 -0.1859 222.6215 0.8527 0.9887 0.8769 1.1147
## 33 Motivate_mean -0.0208 0.0236 -0.8808 216.2020 0.3794 0.9794 0.9351 1.0258## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -1.3803 0.5891 -2.3432 156.9007 0.0204 0.2515 0.0793 0.7980
## 2 Gender_bin 0.1133 0.0662 1.7120 170.8199 0.0887 1.1200 0.9837 1.2751
## 3 Age 0.0057 0.0044 1.2739 162.3914 0.2045 1.0057 0.9971 1.0144
## 4 Retired_bin -0.0491 0.0667 -0.7365 210.1858 0.4622 0.9521 0.8354 1.0851
## 5 Volunteer_rec 0.0408 0.0378 1.0786 199.8944 0.2821 1.0416 0.9673 1.1217
## 6 Lives_alone_bin -0.1328 0.0769 -1.7279 216.8097 0.0854 0.8756 0.7531 1.0181
## 7 Phys_health_rating 0.0199 0.0488 0.4076 115.2121 0.6843 1.0201 0.9270 1.1225
## 8 Multimorbidity_score -0.0005 0.0114 -0.0400 205.9110 0.9681 0.9995 0.9774 1.0221
## 9 Mobility_needs_bin 0.0036 0.1420 0.0255 57.5391 0.9797 1.0036 0.7598 1.3257
## 10 ABC_mean -0.0045 0.0024 -1.8928 187.6136 0.0599 0.9955 0.9908 1.0002
## 11 Subj_vision_rating_aided 0.0357 0.0381 0.9365 216.0106 0.3501 1.0363 0.9618 1.1167
## 12 Subj_hearing_loss_bin 0.0883 0.1125 0.7848 151.2191 0.4338 1.0923 0.8762 1.3618
## 13 Subj_hearing_rating_unaided_REV 0.1410 0.0454 3.1067 216.3181 0.0021 1.1514 1.0534 1.2586
## 14 Tinnitus_past_wk_bin -0.0363 0.0557 -0.6510 216.5418 0.5157 0.9644 0.8646 1.0756
## 15 PTA4_better_ear 0.0087 0.0037 2.3348 51.1203 0.0235 1.0087 1.0014 1.0161
## 16 PTA4_asym 0.0030 0.0022 1.3464 50.2730 0.1842 1.0030 0.9987 1.0073
## 17 SSQ15i_mean_REV -0.0707 0.0204 -3.4631 156.6088 0.0007 0.9317 0.8952 0.9698
## 18 Emocheq_mean 0.0094 0.0335 0.2819 201.3290 0.7783 1.0094 0.9453 1.0779
## 19 HHIES_total 0.0031 0.0053 0.5891 189.5581 0.5565 1.0031 0.9927 1.0136
## 20 SIM_mean -0.0192 0.0140 -1.3737 204.8236 0.1711 0.9810 0.9544 1.0083
## 21 positive_SCI_bin 0.0922 0.0836 1.1029 210.1273 0.2713 1.0966 0.9309 1.2918
## 22 CSRQ_mean -0.0429 0.1001 -0.4285 59.3905 0.6698 0.9580 0.7873 1.1657
## 23 PHQ4_anx_sum 0.0425 0.0334 1.2700 89.8299 0.2074 1.0434 0.9773 1.1140
## 24 PHQ4_dep_sum 0.0233 0.0332 0.7016 204.5717 0.4837 1.0236 0.9591 1.0924
## 25 WHOQOL_overall_qol 0.0633 0.0635 0.9977 89.1944 0.3211 1.0653 0.9407 1.2065
## 26 WHOQOL_health_qol 0.0393 0.0418 0.9403 156.3902 0.3485 1.0401 0.9583 1.1289
## 27 WHO_money 0.0262 0.0312 0.8397 162.6448 0.4023 1.0265 0.9657 1.0913
## 28 Socsupp_mean -0.0277 0.0393 -0.7051 210.4527 0.4815 0.9727 0.9006 1.0506
## 29 Social_network_index 0.0067 0.0246 0.2725 193.2635 0.7855 1.0067 0.9593 1.0565
## 30 Soc_part_freq 0.0233 0.0473 0.4935 144.9131 0.6224 1.0236 0.9329 1.1230
## 31 Soc_part_types -0.0084 0.0235 -0.3572 200.9060 0.7213 0.9916 0.9470 1.0384
## 32 Lonely_bin 0.0368 0.0819 0.4494 144.9796 0.6538 1.0375 0.8836 1.2181
## 33 Motivate_mean 0.0755 0.0359 2.1033 188.9104 0.0368 1.0784 1.0051 1.1570Legend: Asterisks denote p < 0.05; crosses denote p [0.05, 0.1]
Grouping by vision status
Counts
##
## 0 1 <NA>
## 215 290 20Regression output (imputed data, n=525)
## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -0.3592 0.5758 -0.6238 169.2982 0.5336 0.6982 0.2259 2.1584
## 2 Gender_bin 0.0209 0.0505 0.4137 176.5291 0.6796 1.0211 0.9249 1.1274
## 3 Age -0.0009 0.0040 -0.2238 81.9828 0.8234 0.9991 0.9913 1.0070
## 4 Retired_bin -0.0442 0.0538 -0.8226 174.0656 0.4119 0.9568 0.8610 1.0632
## 5 Volunteer_rec 0.0681 0.0298 2.2886 162.1638 0.0234 1.0705 1.0097 1.1349
## 6 Lives_alone_bin -0.1214 0.0625 -1.9427 174.7081 0.0537 0.8857 0.7836 1.0011
## 7 Phys_health_rating -0.0374 0.0400 -0.9352 170.2876 0.3510 0.9633 0.8907 1.0419
## 8 Multimorbidity_score 0.0157 0.0139 1.1297 176.4648 0.2601 1.0158 0.9885 1.0439
## 9 Mobility_needs_bin -0.0865 0.1339 -0.6461 166.7874 0.5191 0.9171 0.7054 1.1924
## 10 ABC_mean -0.0044 0.0035 -1.2668 124.4677 0.2076 0.9956 0.9888 1.0025
## 11 Subj_vision_rating_aided 0.0050 0.0363 0.1369 178.2075 0.8913 1.0050 0.9360 1.0791
## 12 Subj_hearing_loss_bin 0.1129 0.0712 1.5859 164.7247 0.1147 1.1195 0.9737 1.2872
## 13 Subj_hearing_rating_unaided_REV 0.0135 0.0436 0.3100 46.7996 0.7579 1.0136 0.9306 1.1040
## 14 Tinnitus_past_wk_bin -0.0348 0.0544 -0.6395 174.6155 0.5233 0.9658 0.8681 1.0745
## 15 PTA4_better_ear 0.0098 0.0046 2.1281 10.5086 0.0579 1.0098 1.0008 1.0190
## 16 PTA4_asym 0.0042 0.0025 1.7127 35.0549 0.0956 1.0042 0.9993 1.0091
## 17 SSQ15i_mean_REV -0.0505 0.0210 -2.4065 166.5813 0.0172 0.9508 0.9124 0.9907
## 18 Emocheq_mean 0.0027 0.0335 0.0818 155.7236 0.9349 1.0027 0.9390 1.0708
## 19 HHIES_total 0.0086 0.0046 1.8660 154.8177 0.0639 1.0086 0.9996 1.0178
## 20 SIM_mean 0.0043 0.0118 0.3613 152.6866 0.7184 1.0043 0.9813 1.0278
## 21 positive_SCI_bin -0.0793 0.0961 -0.8252 151.8305 0.4105 0.9238 0.7652 1.1152
## 22 CSRQ_mean -0.0433 0.0759 -0.5701 91.6701 0.5700 0.9576 0.8253 1.1112
## 23 PHQ4_anx_sum 0.0512 0.0275 1.8638 170.6497 0.0641 1.0525 0.9973 1.1108
## 24 PHQ4_dep_sum -0.0158 0.0358 -0.4418 165.5642 0.6592 0.9843 0.9176 1.0559
## 25 WHOQOL_overall_qol 0.0132 0.0517 0.2548 153.2040 0.7992 1.0133 0.9156 1.1213
## 26 WHOQOL_health_qol 0.0633 0.0360 1.7557 133.1752 0.0814 1.0653 0.9928 1.1432
## 27 WHO_money 0.0259 0.0268 0.9680 178.7429 0.3344 1.0262 0.9737 1.0816
## 28 Socsupp_mean 0.0269 0.0333 0.8080 173.7509 0.4202 1.0273 0.9624 1.0965
## 29 Social_network_index 0.0007 0.0200 0.0354 158.3002 0.9718 1.0007 0.9622 1.0407
## 30 Soc_part_freq 0.0727 0.0430 1.6908 175.1287 0.0927 1.0754 0.9885 1.1700
## 31 Soc_part_types -0.0175 0.0188 -0.9316 177.3309 0.3528 0.9827 0.9471 1.0195
## 32 Lonely_bin 0.0100 0.0740 0.1346 162.9869 0.8931 1.0101 0.8737 1.1677
## 33 Motivate_mean -0.0002 0.0292 -0.0071 179.0708 0.9943 0.9998 0.9442 1.0587## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -1.2524 0.4867 -2.5732 144.8155 0.0111 0.2858 0.1101 0.7420
## 2 Gender_bin 0.1062 0.0549 1.9338 210.8000 0.0545 1.1120 0.9986 1.2384
## 3 Age 0.0072 0.0036 2.0195 233.2851 0.0446 1.0072 1.0001 1.0144
## 4 Retired_bin -0.0444 0.0512 -0.8666 253.8329 0.3870 0.9566 0.8652 1.0575
## 5 Volunteer_rec 0.0293 0.0307 0.9529 172.8749 0.3420 1.0297 0.9696 1.0936
## 6 Lives_alone_bin 0.0649 0.0587 1.1056 252.8785 0.2699 1.0671 0.9511 1.1972
## 7 Phys_health_rating 0.0129 0.0413 0.3124 58.8488 0.7559 1.0130 0.9342 1.0984
## 8 Multimorbidity_score -0.0057 0.0097 -0.5831 239.4511 0.5604 0.9943 0.9756 1.0134
## 9 Mobility_needs_bin 0.0113 0.1283 0.0880 39.9255 0.9303 1.0114 0.7865 1.3005
## 10 ABC_mean -0.0022 0.0020 -1.0826 136.9480 0.2809 0.9978 0.9939 1.0017
## 11 Subj_vision_rating_aided -0.0023 0.0320 -0.0727 251.3298 0.9421 0.9977 0.9370 1.0623
## 12 Subj_hearing_loss_bin 0.0668 0.0715 0.9333 171.9545 0.3520 1.0691 0.9293 1.2299
## 13 Subj_hearing_rating_unaided_REV 0.1215 0.0350 3.4697 216.4681 0.0006 1.1292 1.0543 1.2094
## 14 Tinnitus_past_wk_bin -0.0453 0.0478 -0.9492 220.9441 0.3436 0.9557 0.8702 1.0496
## 15 PTA4_better_ear 0.0070 0.0030 2.3490 167.5039 0.0200 1.0070 1.0011 1.0130
## 16 PTA4_asym 0.0021 0.0021 0.9997 228.8703 0.3185 1.0021 0.9980 1.0062
## 17 SSQ15i_mean_REV -0.1049 0.0186 -5.6458 136.9607 0.0000 0.9004 0.8682 0.9338
## 18 Emocheq_mean -0.0108 0.0286 -0.3769 217.5983 0.7066 0.9893 0.9353 1.0463
## 19 HHIES_total 0.0088 0.0039 2.2754 150.6507 0.0243 1.0088 1.0012 1.0166
## 20 SIM_mean -0.0093 0.0114 -0.8155 186.1924 0.4158 0.9907 0.9689 1.0131
## 21 positive_SCI_bin 0.1016 0.0773 1.3132 150.9085 0.1911 1.1069 0.9513 1.2880
## 22 CSRQ_mean 0.0397 0.0740 0.5359 82.3046 0.5935 1.0405 0.9000 1.2029
## 23 PHQ4_anx_sum 0.0760 0.0284 2.6749 66.3161 0.0094 1.0790 1.0205 1.1407
## 24 PHQ4_dep_sum -0.0104 0.0292 -0.3552 190.0976 0.7229 0.9897 0.9346 1.0479
## 25 WHOQOL_overall_qol 0.0959 0.0497 1.9311 118.9498 0.0558 1.1006 0.9985 1.2133
## 26 WHOQOL_health_qol 0.0196 0.0326 0.6014 162.1710 0.5484 1.0198 0.9567 1.0871
## 27 WHO_money 0.0278 0.0284 0.9808 96.1692 0.3291 1.0282 0.9725 1.0870
## 28 Socsupp_mean -0.0203 0.0353 -0.5750 224.0691 0.5658 0.9799 0.9144 1.0501
## 29 Social_network_index 0.0049 0.0208 0.2374 125.3419 0.8127 1.0049 0.9648 1.0467
## 30 Soc_part_freq -0.0453 0.0406 -1.1173 134.1840 0.2659 0.9557 0.8826 1.0349
## 31 Soc_part_types 0.0026 0.0179 0.1455 245.9398 0.8845 1.0026 0.9680 1.0384
## 32 Lonely_bin -0.0350 0.0670 -0.5223 189.6238 0.6021 0.9656 0.8468 1.1011
## 33 Motivate_mean 0.0238 0.0283 0.8426 236.5224 0.4003 1.0241 0.9688 1.0825Legend: Asterisks denote p < 0.05; crosses denote p [0.05, 0.1]
Variable selection: Predicting HA adoption
Penalized logistic regression (LASSO) with 5-fold cross-validation was conducted across 100 lambda (penalty) values. The lambda value that led to the best-fitting model (with lowest cross-validation error) was 0.044, which corresponds to a model with 13 predictors:
Regression output (13 selected) (imputed data, n=525)
## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -0.7791 0.1798 -4.3326 459.4624 0.0000 0.4588 0.3225 0.6527
## 2 Gender_bin 0.0300 0.0327 0.9184 454.7251 0.3589 1.0305 0.9665 1.0987
## 3 Age 0.0025 0.0021 1.1975 352.0198 0.2319 1.0025 0.9984 1.0066
## 4 Volunteer_rec 0.0378 0.0180 2.0979 506.9027 0.0364 1.0385 1.0025 1.0758
## 5 Lives_alone_bin -0.0308 0.0390 -0.7916 485.7513 0.4290 0.9697 0.8983 1.0467
## 6 Subj_hearing_loss_bin 0.0772 0.0483 1.6001 275.9907 0.1107 1.0803 0.9827 1.1875
## 7 Subj_hearing_rating_unaided_REV 0.0965 0.0249 3.8754 276.5074 0.0001 1.1013 1.0489 1.1564
## 8 PTA4_better_ear 0.0083 0.0023 3.6845 63.9576 0.0005 1.0083 1.0038 1.0129
## 9 PTA4_asym 0.0035 0.0016 2.2278 67.5542 0.0292 1.0035 1.0004 1.0067
## 10 SSQ15i_mean_REV -0.0775 0.0122 -6.3466 260.8414 0.0000 0.9254 0.9036 0.9478
## 11 HHIES_total 0.0053 0.0024 2.2187 381.4252 0.0271 1.0053 1.0006 1.0101
## 12 PHQ4_anx_sum 0.0470 0.0153 3.0756 96.5966 0.0027 1.0481 1.0172 1.0800
## 13 WHO_money 0.0401 0.0181 2.2155 155.8128 0.0282 1.0409 1.0046 1.0785
## 14 Socsupp_mean 0.0169 0.0205 0.8228 424.5504 0.4111 1.0170 0.9770 1.0587Limiting the model to the 8 predictors with p < 0.05, all 8 predictors remain significant in a standard logistic regression model.
Regression output (8 selected) (imputed data, n=525)
## Term Estimate StdError t df p OddsRatio CIlower CIupper
## 1 (Intercept) -0.6185 0.1085 -5.7002 112.1901 0.0000 0.5388 0.4355 0.6664
## 2 Volunteer_rec 0.0379 0.0178 2.1249 503.9135 0.0341 1.0386 1.0030 1.0755
## 3 Subj_hearing_rating_unaided_REV 0.1129 0.0227 4.9790 291.6845 0.0000 1.1195 1.0708 1.1705
## 4 PTA4_better_ear 0.0104 0.0020 5.2938 67.5750 0.0000 1.0105 1.0065 1.0144
## 5 PTA4_asym 0.0039 0.0015 2.5424 96.9679 0.0126 1.0039 1.0010 1.0069
## 6 SSQ15i_mean_REV -0.0784 0.0124 -6.3279 200.0383 0.0000 0.9246 0.9024 0.9473
## 7 HHIES_total 0.0054 0.0024 2.2734 298.3260 0.0237 1.0054 1.0007 1.0102
## 8 PHQ4_anx_sum 0.0422 0.0152 2.7789 82.2740 0.0068 1.0431 1.0125 1.0746
## 9 WHO_money 0.0519 0.0172 3.0212 174.7458 0.0029 1.0533 1.0184 1.0894
Accuracy of model predictions
## [1] "Imputed data, 8 predictors, using median as boundary to classify 0 or 1"## predicted
## actual 0 1
## 0 253 169
## 1 10 93## [1] "Accuracy: 0.659"## [1] "Sensitivity: 0.903"## [1] "Specificity: 0.6"4. Special populations
The standard relationship, HA ~ BPTA4
Plots showing thresholds for HA adoption: BPTA4, Asymmetry, HHIES
- Of the three variables, BPTA4 is the best standalone predictor of HA
adoption.
- Legend: Red = HA adopters.
- Lines are thresholds for highest area-under-curve. Complete cases for BPTA4, Asymmetry, HHIES, n=436.
Comparing three predictors: BPTA4, Asymmetry, HHIES
- On their own, BPTA4 and HHIES help predict HA adoption. On its own, PTA4 asymmetry doesn’t.
## Analysis of Deviance Table
##
## Model 1: study2_HA_bin ~ 1
## Model 2: study2_HA_bin ~ PTA4_better_ear
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 435 424.45
## 2 434 351.11 1 73.345 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Analysis of Deviance Table
##
## Model 1: study2_HA_bin ~ 1
## Model 2: study2_HA_bin ~ PTA4_asym
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 435 424.45
## 2 434 420.80 1 3.6462 0.0562 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Analysis of Deviance Table
##
## Model 1: study2_HA_bin ~ 1
## Model 2: study2_HA_bin ~ HHIES_total
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 435 424.45
## 2 434 407.67 1 16.779 0.00004199 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1- When BPTA4 is the main predictor, adding HHIES doesn’t help predict HA adoption, while adding PTA4 asymmetry does help a little.
## Analysis of Deviance Table
##
## Model 1: study2_HA_bin ~ PTA4_better_ear
## Model 2: study2_HA_bin ~ PTA4_better_ear + HHIES_total
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 434 351.11
## 2 433 349.44 1 1.664 0.1971##
## Call:
## glm(formula = study2_HA_bin ~ PTA4_better_ear + HHIES_total,
## family = "binomial", data = compdata_3hear)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.36669 0.44332 -9.850 < 2e-16 ***
## PTA4_better_ear 0.10502 0.01517 6.921 0.00000000000449 ***
## HHIES_total 0.01952 0.01505 1.297 0.195
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 424.45 on 435 degrees of freedom
## Residual deviance: 349.44 on 433 degrees of freedom
## AIC: 355.44
##
## Number of Fisher Scoring iterations: 5## Analysis of Deviance Table
##
## Model 1: study2_HA_bin ~ PTA4_better_ear
## Model 2: study2_HA_bin ~ PTA4_better_ear + PTA4_asym
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 434 351.11
## 2 433 345.57 1 5.5349 0.01864 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Call:
## glm(formula = study2_HA_bin ~ PTA4_better_ear + PTA4_asym, family = "binomial",
## data = compdata_3hear)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.508478 0.456061 -9.886 < 2e-16 ***
## PTA4_better_ear 0.113766 0.014962 7.604 0.0000000000000288 ***
## PTA4_asym 0.022765 0.009159 2.486 0.0129 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 424.45 on 435 degrees of freedom
## Residual deviance: 345.57 on 433 degrees of freedom
## AIC: 351.57
##
## Number of Fisher Scoring iterations: 5Plot: better and worse ear PTA4, and HA status
Plot: BPTA4, HHIES, and HA status
“Red flag” variables
Continuous and ordinal variables (32) have been dichotomized according to these cut-offs:
- PTA4 asymmetry ≥ 20 dB HL
- HHIES ≥ 10
- PHQ4 (anxiety and depression) subscores ≥ 3
- Social participation of less than once per week in any
activity
- Mean social support of less than “some of the time” (< 2.5 out of
5 on average, with 3 being “some of the time”)
- Generally, any of the 26 WHOQOL-BREF responses that are in the worst 2 of 5 categories. There are a few exceptions, where it’s clear that the worst 3 categories are red flags according to the distribution (physical pain, medical treatment, money, information access, getting around, health services, transport, negative mood).
Other variables (6) that are already dichotomized are:
- Tinnitus within the last week
- Need for mobility aids
- Retired status
- Poor or fair aided vision
- Subjective cognitive impairment
- Being lonely
Note: List of motivations to seek help is available but will exclude from analyses for now. A red flag could be a self-report of being “not at all” or only “slightly” motivated to seek help.
Sample: n=449 who have audiograms available.
Four groups, using BPTA4 > 25.0 as the criterion for “poor PTA”:
## poor PTA
## adopted HA 0 1
## 0 260 105
## 1 22 62Distribution of BPTA4 in 4 groups
## $nonHA_poorBPTA
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 26.25 27.50 31.25 32.47 36.25 52.50
##
## $HA_poorBPTA
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 26.25 31.25 34.38 34.92 38.44 51.25
##
## $nonHA_goodBPTA
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.50 11.25 15.00 15.53 20.00 25.00
##
## $HA_goodBPTA
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 10.00 16.56 20.00 19.26 22.19 23.75
Counts of participants experiencing each red flag, ordered high to low
Note: Test of proportions conducted for first two columns; generally requires 5 positive cases per group.
## Variable nonHA_poorPTA HA_poorPTA nonHA_goodPTA HA_goodPTA chisq pvalue
## 1 HHIES_bin 67 44 84 15 0.6038 0.4371
## 7 PHQ4_anx_bin 30 27 93 10 3.2517 0.0713
## 3 Tinnitus_past_wk_bin 45 16 72 9 4.1804 0.0409
## 32 WHO_satisf_sex_bin 24 14 55 3 0.0000 1.0000
## 8 PHQ4_dep_bin 31 12 66 6 1.6101 0.2045
## 27 WHO_sleep_satisf_bin 23 12 55 7 0.0378 0.8459
## 9 Lonely_bin 27 10 44 6 1.5581 0.2120
## 37 WHO_neg_feelings_bin 19 10 41 3 0.0127 0.9103
## 14 WHO_phys_pain_bin 26 9 52 5 1.8905 0.1691
## 15 WHO_med_treatment_bin 23 9 46 1 0.9383 0.3327
## 23 WHO_money_bin 24 9 39 3 1.2249 0.2684
## 6 positive_SCI_bin 15 8 21 3 0.0003 0.9856
## 13 WHO_health_satisf_bin 21 8 27 3 0.9183 0.3379
## 35 WHO_access_health_bin 14 6 35 3 0.2083 0.6481
## 4 Mobility_needs_bin 5 5 13 1 0.2825 0.5950
## 24 WHO_information_bin 8 5 9 0 0.0000 1.0000
## 12 WHO_qol_bin 4 4 5 0 0.1580 0.6910
## 21 WHO_energy_bin 6 4 13 1 0.0000 1.0000
## 29 WHO_capacity_work_bin 6 4 20 1 0.0000 1.0000
## 34 WHO_living_place_bin 6 4 8 3 0.0000 1.0000
## 5 Poor_aided_vision_bin 1 3 8 0 1.1304 0.2877
## 10 MOS_socsupp_bin 12 3 9 4 1.3431 0.2465
## 17 WHO_meaningful_bin 8 3 9 0 0.1421 0.7062
## 18 WHO_concentrate_bin 3 3 6 0 0.0550 0.8146
## 26 WHO_get_around_bin 6 3 10 1 0.0000 1.0000
## 28 WHO_ability_ADL_bin 6 3 11 0 0.0000 1.0000
## 30 WHO_satisf_self_bin 6 3 16 1 0.0000 1.0000
## 2 Asymmetry_bin 5 2 8 4 0.0062 0.9371
## 11 CCHS_socpartfreq_bin 5 2 12 2 0.0062 0.9371
## 20 WHO_healthy_env_bin 1 2 1 0 0.2169 0.6414
## 22 WHO_body_appear_bin 7 2 14 2 0.3561 0.5507
## 31 WHO_satisf_relat_bin 13 2 10 2 2.9552 0.0856
## 16 WHO_enjoy_life_bin 5 1 5 1 0.3920 0.5312
## 19 WHO_feel_safe_bin 2 1 3 0 0.0000 1.0000
## 25 WHO_leisure_bin 8 1 12 2 1.7058 0.1915
## 33 WHO_satisf_support_bin 5 1 8 1 0.3920 0.5312
## 36 WHO_transport_bin 3 1 13 2 0.0000 1.0000Percentages of participants experiencing each red flag, ordered high to low
Note: Test of proportions conducted for first two columns; generally requires 5 positive cases per group.
## Variable HA_poorPTA_pc nonHA_poorPTA_pc nonHA_goodPTA_pc HA_goodPTA_pc chisq pvalue
## 1 HHIES_bin 71.0 63.8 32.3 68.2 0.6038 0.4371
## 7 PHQ4_anx_bin 43.5 28.6 35.8 45.5 3.2517 0.0713
## 3 Tinnitus_past_wk_bin 25.8 42.9 27.7 40.9 4.1804 0.0409
## 32 WHO_satisf_sex_bin 22.6 22.9 21.2 13.6 0.0000 1.0000
## 8 PHQ4_dep_bin 19.4 29.5 25.4 27.3 1.6101 0.2045
## 27 WHO_sleep_satisf_bin 19.4 21.9 21.2 31.8 0.0378 0.8459
## 9 Lonely_bin 16.1 25.7 16.9 27.3 1.5581 0.2120
## 37 WHO_neg_feelings_bin 16.1 18.1 15.8 13.6 0.0127 0.9103
## 14 WHO_phys_pain_bin 14.5 24.8 20.0 22.7 1.8905 0.1691
## 15 WHO_med_treatment_bin 14.5 21.9 17.7 4.5 0.9383 0.3327
## 23 WHO_money_bin 14.5 22.9 15.0 13.6 1.2249 0.2684
## 6 positive_SCI_bin 12.9 14.3 8.1 13.6 0.0003 0.9856
## 13 WHO_health_satisf_bin 12.9 20.0 10.4 13.6 0.9183 0.3379
## 35 WHO_access_health_bin 9.7 13.3 13.5 13.6 0.2083 0.6481
## 4 Mobility_needs_bin 8.1 4.8 5.0 4.5 0.2825 0.5950
## 24 WHO_information_bin 8.1 7.6 3.5 0.0 0.0000 1.0000
## 12 WHO_qol_bin 6.5 3.8 1.9 0.0 0.1580 0.6910
## 21 WHO_energy_bin 6.5 5.7 5.0 4.5 0.0000 1.0000
## 29 WHO_capacity_work_bin 6.5 5.7 7.7 4.5 0.0000 1.0000
## 34 WHO_living_place_bin 6.5 5.7 3.1 13.6 0.0000 1.0000
## 5 Poor_aided_vision_bin 4.8 1.0 3.1 0.0 1.1304 0.2877
## 10 MOS_socsupp_bin 4.8 11.4 3.5 18.2 1.3431 0.2465
## 17 WHO_meaningful_bin 4.8 7.6 3.5 0.0 0.1421 0.7062
## 18 WHO_concentrate_bin 4.8 2.9 2.3 0.0 0.0550 0.8146
## 26 WHO_get_around_bin 4.8 5.7 3.8 4.5 0.0000 1.0000
## 28 WHO_ability_ADL_bin 4.8 5.7 4.2 0.0 0.0000 1.0000
## 30 WHO_satisf_self_bin 4.8 5.7 6.2 4.5 0.0000 1.0000
## 2 Asymmetry_bin 3.2 4.8 3.1 18.2 0.0062 0.9371
## 11 CCHS_socpartfreq_bin 3.2 4.8 4.6 9.1 0.0062 0.9371
## 20 WHO_healthy_env_bin 3.2 1.0 0.4 0.0 0.2169 0.6414
## 22 WHO_body_appear_bin 3.2 6.7 5.4 9.1 0.3561 0.5507
## 31 WHO_satisf_relat_bin 3.2 12.4 3.8 9.1 2.9552 0.0856
## 16 WHO_enjoy_life_bin 1.6 4.8 1.9 4.5 0.3920 0.5312
## 19 WHO_feel_safe_bin 1.6 1.9 1.2 0.0 0.0000 1.0000
## 25 WHO_leisure_bin 1.6 7.6 4.6 9.1 1.7058 0.1915
## 33 WHO_satisf_support_bin 1.6 4.8 3.1 4.5 0.3920 0.5312
## 36 WHO_transport_bin 1.6 2.9 5.0 9.1 0.0000 1.0000Median, mean, cumulative red flags per group
## nonHA_poorBPTA HA_poorBPTA nonHA_goodBPTA HA_goodBPTA
## 3 2 2 2## nonHA_poorBPTA HA_poorBPTA nonHA_goodBPTA HA_goodBPTA
## 4.95 4.13 3.67 4.77
Plot: All: Frequency of red flags, by fewer or more red flags
The plot shows n=232 with 2 or fewer red flags, and n=217 with 3 or more red flags.
Plot: Poor BPTA only: Frequency of red flags, by fewer or more flags
Sample: n=167 with BPTA > 25 dB HL
## Red flag group
## HA group fewer_flags more_flags
## nonHA_poorBPTA 48 57
## HA_poorBPTA 34 28
Plot: All: BPTA4, number of red flags, and HA adoption
Logistic regression: Does the number of red flags affect HA adoption?
Sample: n=167 with BPTA4 > 25 dB HL
Outcome: 1=HA adoption
Predictors: BPTA4 and count of red flags are standardized using mean and
SD of this group
##
## Call:
## glm(formula = study2_HA_bin ~ PTA4_better_ear_z * sum_flags_z,
## family = "binomial", data = compdata_poorBPTA)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.5464 0.1648 -3.317 0.000911 ***
## PTA4_better_ear_z 0.3925 0.1696 2.315 0.020640 *
## sum_flags_z -0.1575 0.1723 -0.914 0.360696
## PTA4_better_ear_z:sum_flags_z -0.2072 0.1965 -1.054 0.291723
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 220.31 on 166 degrees of freedom
## Residual deviance: 211.69 on 163 degrees of freedom
## AIC: 219.69
##
## Number of Fisher Scoring iterations: 4Interactive plot: Which red flags are present?
Logistic regression: Which red flags matter?
Sample: n=167 with BPTA4 > 25 dB HL
Outcome: 1=HA adoption
Predictors: “dump everything in the pot” (37 variables)
## Term Estimate StdError z p OddsRatio CIlower CIupper
## 1 (Intercept) -0.9371 0.419 -2.2350 0.0254 3.92e-01 0.1722 0.891
## 2 PTA4_better_ear_z 0.4906 0.209 2.3458 0.0190 1.63e+00 1.0840 2.461
## 3 HHIES_bin 0.5444 0.465 1.1718 0.2413 1.72e+00 0.6934 4.285
## 4 Asymmetry_bin -0.3818 1.247 -0.3061 0.7595 6.83e-01 0.0592 7.866
## 5 Tinnitus_past_wk_bin -0.2221 0.461 -0.4817 0.6300 8.01e-01 0.3244 1.977
## 6 Mobility_needs_bin 1.3234 1.350 0.9806 0.3268 3.76e+00 0.2667 52.909
## 7 Poor_aided_vision_bin 0.6984 1.327 0.5263 0.5987 2.01e+00 0.1492 27.098
## 8 positive_SCI_bin -0.7690 0.985 -0.7808 0.4349 4.64e-01 0.0672 3.194
## 9 PHQ4_anx_bin 1.8205 0.597 3.0508 0.0023 6.18e+00 1.9174 19.887
## 10 PHQ4_dep_bin -0.9593 0.939 -1.0219 0.3068 3.83e-01 0.0608 2.413
## 11 Lonely_bin -1.6527 0.983 -1.6822 0.0925 1.92e-01 0.0279 1.314
## 12 MOS_socsupp_bin 1.1210 1.661 0.6747 0.4998 3.07e+00 0.1182 79.615
## 13 CCHS_socpartfreq_bin 0.1971 2.302 0.0857 0.9317 1.22e+00 0.0134 110.843
## 14 WHO_qol_bin 103.5685 9665.394 0.0107 0.9915 9.53e+44 0.0000 Inf
## 15 WHO_health_satisf_bin 0.0931 1.027 0.0906 0.9278 1.10e+00 0.1465 8.220
## 16 WHO_phys_pain_bin 0.3714 0.796 0.4664 0.6409 1.45e+00 0.3044 6.904
## 17 WHO_med_treatment_bin -0.5016 0.860 -0.5833 0.5597 6.06e-01 0.1122 3.267
## 18 WHO_enjoy_life_bin -18.9591 4258.709 -0.0045 0.9964 0.00e+00 0.0000 Inf
## 19 WHO_meaningful_bin 1.8146 2.199 0.8253 0.4092 6.14e+00 0.0825 456.738
## 20 WHO_concentrate_bin 156.7447 23025.618 0.0068 0.9946 1.18e+68 0.0000 Inf
## 21 WHO_feel_safe_bin -46.1462 43001.441 -0.0011 0.9991 0.00e+00 0.0000 Inf
## 22 WHO_healthy_env_bin 103.9311 35540.038 0.0029 0.9977 1.37e+45 0.0000 Inf
## 23 WHO_energy_bin -22.0040 11260.531 -0.0020 0.9984 0.00e+00 0.0000 Inf
## 24 WHO_body_appear_bin -20.8552 13437.994 -0.0016 0.9988 0.00e+00 0.0000 Inf
## 25 WHO_money_bin -0.4080 0.823 -0.4961 0.6199 6.65e-01 0.1327 3.333
## 26 WHO_information_bin 193.0461 24725.409 0.0078 0.9938 6.90e+83 0.0000 Inf
## 27 WHO_leisure_bin -187.5550 37055.050 -0.0051 0.9960 0.00e+00 0.0000 Inf
## 28 WHO_get_around_bin -120.2570 10923.684 -0.0110 0.9912 0.00e+00 0.0000 Inf
## 29 WHO_sleep_satisf_bin -0.4458 0.747 -0.5970 0.5505 6.40e-01 0.1482 2.767
## 30 WHO_ability_ADL_bin -31.4935 37251.321 -0.0008 0.9993 0.00e+00 0.0000 Inf
## 31 WHO_capacity_work_bin 171.1688 36868.784 0.0046 0.9963 2.18e+74 0.0000 Inf
## 32 WHO_satisf_self_bin 25.5251 17917.940 0.0014 0.9989 1.22e+11 0.0000 Inf
## 33 WHO_satisf_relat_bin -211.5296 25594.403 -0.0083 0.9934 0.00e+00 0.0000 Inf
## 34 WHO_satisf_sex_bin 0.6246 0.681 0.9168 0.3593 1.87e+00 0.4913 7.100
## 35 WHO_satisf_support_bin -0.8938 2.062 -0.4335 0.6647 4.09e-01 0.0072 23.283
## 36 WHO_living_place_bin -173.8374 23982.937 -0.0072 0.9942 0.00e+00 0.0000 Inf
## 37 WHO_access_health_bin 0.7039 0.865 0.8135 0.4159 2.02e+00 0.3708 11.021
## 38 WHO_transport_bin -18.6399 10793.690 -0.0017 0.9986 0.00e+00 0.0000 Inf
## 39 WHO_neg_feelings_bin 1.1361 1.232 0.9218 0.3566 3.11e+00 0.2782 34.876## Variable No_flag Flag Percent
## 1 HHIES_bin 56 111 66.5
## 2 Asymmetry_bin 160 7 4.2
## 3 Tinnitus_past_wk_bin 106 61 36.5
## 4 Mobility_needs_bin 157 10 6.0
## 5 Poor_aided_vision_bin 163 4 2.4
## 6 positive_SCI_bin 144 23 13.8
## 7 PHQ4_anx_bin 110 57 34.1
## 8 PHQ4_dep_bin 124 43 25.7
## 9 Lonely_bin 130 37 22.2
## 10 MOS_socsupp_bin 152 15 9.0
## 11 CCHS_socpartfreq_bin 160 7 4.2
## 12 WHO_qol_bin 159 8 4.8
## 13 WHO_health_satisf_bin 138 29 17.4
## 14 WHO_phys_pain_bin 132 35 21.0
## 15 WHO_med_treatment_bin 135 32 19.2
## 16 WHO_enjoy_life_bin 161 6 3.6
## 17 WHO_meaningful_bin 156 11 6.6
## 18 WHO_concentrate_bin 161 6 3.6
## 19 WHO_feel_safe_bin 164 3 1.8
## 20 WHO_healthy_env_bin 164 3 1.8
## 21 WHO_energy_bin 157 10 6.0
## 22 WHO_body_appear_bin 158 9 5.4
## 23 WHO_money_bin 134 33 19.8
## 24 WHO_information_bin 154 13 7.8
## 25 WHO_leisure_bin 158 9 5.4
## 26 WHO_get_around_bin 158 9 5.4
## 27 WHO_sleep_satisf_bin 132 35 21.0
## 28 WHO_ability_ADL_bin 158 9 5.4
## 29 WHO_capacity_work_bin 157 10 6.0
## 30 WHO_satisf_self_bin 158 9 5.4
## 31 WHO_satisf_relat_bin 152 15 9.0
## 32 WHO_satisf_sex_bin 129 38 22.8
## 33 WHO_satisf_support_bin 161 6 3.6
## 34 WHO_living_place_bin 157 10 6.0
## 35 WHO_access_health_bin 147 20 12.0
## 36 WHO_transport_bin 163 4 2.4
## 37 WHO_neg_feelings_bin 138 29 17.4Logistic regression (LASSO): Which red flags matter?
Sample: n=167 with BPTA4 > 25 dB HL
Outcome: 1=HA adoption
Predictors: Just the non-zero coefficients after a LASSO procedure
## 39 x 1 sparse Matrix of class "dgCMatrix"
## s1
## (Intercept) -0.55135222
## compdata_poorBPTA$PTA4_better_ear_z 0.31845195
## HHIES_bin 0.12974388
## Asymmetry_bin .
## Tinnitus_past_wk_bin -0.35908626
## Mobility_needs_bin .
## Poor_aided_vision_bin 0.65861829
## positive_SCI_bin .
## PHQ4_anx_bin 0.92101827
## PHQ4_dep_bin -0.32583407
## Lonely_bin -0.21008509
## MOS_socsupp_bin .
## CCHS_socpartfreq_bin .
## WHO_qol_bin 0.66125592
## WHO_health_satisf_bin .
## WHO_phys_pain_bin -0.21863008
## WHO_med_treatment_bin -0.20349245
## WHO_enjoy_life_bin .
## WHO_meaningful_bin .
## WHO_concentrate_bin 0.51731231
## WHO_feel_safe_bin .
## WHO_healthy_env_bin .
## WHO_energy_bin .
## WHO_body_appear_bin -0.01532856
## WHO_money_bin -0.19690630
## WHO_information_bin 0.31942751
## WHO_leisure_bin -1.03642901
## WHO_get_around_bin .
## WHO_sleep_satisf_bin .
## WHO_ability_ADL_bin .
## WHO_capacity_work_bin .
## WHO_satisf_self_bin .
## WHO_satisf_relat_bin -0.89449372
## WHO_satisf_sex_bin .
## WHO_satisf_support_bin .
## WHO_living_place_bin .
## WHO_access_health_bin .
## WHO_transport_bin .
## WHO_neg_feelings_bin .##
## Call:
## glm(formula = m2_formula, family = "binomial", data = compdata_poorBPTA)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.6729 0.3736 -1.801 0.07167 .
## PTA4_better_ear_z 0.4399 0.1914 2.298 0.02153 *
## HHIES_bin 0.4373 0.4171 1.048 0.29447
## Tinnitus_past_wk_bin -0.4701 0.4053 -1.160 0.24616
## Poor_aided_vision_bin 1.1522 1.2542 0.919 0.35829
## PHQ4_anx_bin 1.5630 0.4910 3.184 0.00145 **
## PHQ4_dep_bin -0.7870 0.7100 -1.109 0.26764
## Lonely_bin -1.2201 0.8092 -1.508 0.13160
## WHO_qol_bin 2.4351 1.3677 1.780 0.07501 .
## WHO_phys_pain_bin -0.2999 0.6406 -0.468 0.63971
## WHO_med_treatment_bin -0.7263 0.6661 -1.090 0.27554
## WHO_concentrate_bin 3.1845 1.5447 2.062 0.03925 *
## WHO_body_appear_bin -1.1005 1.4283 -0.771 0.44100
## WHO_money_bin -0.5484 0.6357 -0.863 0.38829
## WHO_information_bin 3.4437 1.8016 1.912 0.05594 .
## WHO_leisure_bin -4.0037 2.2248 -1.800 0.07193 .
## WHO_satisf_relat_bin -2.7878 1.6208 -1.720 0.08543 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 220.31 on 166 degrees of freedom
## Residual deviance: 173.64 on 150 degrees of freedom
## AIC: 207.64
##
## Number of Fisher Scoring iterations: 6Plot: Red flags experienced by four “quadrants” for HHIES/BPTA
Plot: What’s happening with anxiety?
Sample: n=167 with BPTA4 > 25
BPTA4 thresholds for HA adoption
Table of thresholds
## Threshold Sensitivity Specificity Accuracy
## 1 -Inf 1.00000000 0.000000000 NA
## 2 3.125 1.00000000 0.002739726 18.93
## 3 4.375 1.00000000 0.008219178 19.38
## 4 5.625 1.00000000 0.024657534 20.71
## 5 6.875 1.00000000 0.049315068 22.72
## 6 8.125 1.00000000 0.068493151 24.28
## 7 9.375 1.00000000 0.104109589 27.17
## 8 10.625 0.98809524 0.156164384 31.18
## 9 11.875 0.98809524 0.202739726 34.97
## 10 13.125 0.97619048 0.271232877 40.31
## 11 14.000 0.96428571 0.317808219 43.88
## 12 14.625 0.96428571 0.320547945 44.10
## 13 15.625 0.96428571 0.369863014 48.11
## 14 16.875 0.92857143 0.424657534 51.89
## 15 18.125 0.90476190 0.457534247 54.12
## 16 19.375 0.88095238 0.506849315 57.68
## 17 20.625 0.85714286 0.542465753 60.13
## 18 21.875 0.80952381 0.591780822 63.25
## 19 23.125 0.79761905 0.630136986 66.15
## 20 24.375 0.73809524 0.665753425 67.93
## 21 25.625 0.73809524 0.712328767 71.71
## 22 26.875 0.72619048 0.767123288 75.95
## 23 28.125 0.69047619 0.791780822 77.28
## 24 29.375 0.59523810 0.819178082 77.73
## 25 30.625 0.55952381 0.852054795 79.73
## 26 31.875 0.46428571 0.873972603 79.73
## 27 33.125 0.40476190 0.882191781 79.29
## 28 34.375 0.36904762 0.917808219 81.51
## 29 35.625 0.30952381 0.923287671 80.85
## 30 36.875 0.23809524 0.947945205 81.51
## 31 38.125 0.19047619 0.956164384 81.29
## 32 39.375 0.15476190 0.964383562 81.29
## 33 40.500 0.08333333 0.972602740 80.62
## 34 41.125 0.08333333 0.975342466 80.85
## 35 42.500 0.07142857 0.978082192 80.85
## 36 44.375 0.05952381 0.980821918 80.85
## 37 45.625 0.04761905 0.986301370 81.07
## 38 46.875 0.03571429 0.991780822 81.29
## 39 48.125 0.02380952 0.994520548 81.29
## 40 49.375 0.01190476 0.994520548 81.07
## 41 50.625 0.01190476 0.997260274 81.29
## 42 51.875 0.00000000 0.997260274 81.07
## 43 Inf 0.00000000 1.000000000 NA
Split by: Fewer or more red flags
Split by: Women or men
##
## Men Women
## 183 266
Split by: Younger or older
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 52.00 61.00 67.00 66.59 72.00 89.00 3
Split by: Retirement status
##
## 0 1 <NA>
## 168 281 0
Split by: Anxiety and depression
##
## 0 1 <NA>
## 275 160 14
##
## 0 1 <NA>
## 320 115 14
Split by: Money
##
## 0 1 <NA>
## 361 75 13
5. Review of factors affecting HA uptake
Knoetze et al. (2023)
Knoetze et al. (2023) Factors influencing hearing help-seeking and hearing aid uptake in adults: A systematic review of the past decade. Trends in Hearing.
- 42 peer-reviewed studies on adults with subjective or confirmed HL,
published between 2011-2022; all study designs are cross-sectional or
cohort, except 1 experimental and 1 RCT (Table 1)
- 38 studies on HA uptake; total of 93 audiological and 66 non-audiological factors (Tables 4 and 5)
- Better ear PTA is the most commonly measured audiological factor,
and generally positively associated with HA uptake (Table 4)
- Self-reported hearing disability (HHI) is 2nd most commonly measured
audiological factor, with half the studies finding a positive
association and the other half finding no association with HA
uptake
- Age is most commonly measured non-audiological factor; 2/3 of
studies find no association of age with HA uptake; the remaining 1/3
find a positive association (Table 5)
- Sex is 2nd most commonly measured non-audiological factor; 4/5 of
studies find no association of sex with HA uptake
- Combining information from previous reviews and present review,
there is a lot of evidence for hearing sensitivity being an important
factor for HA uptake, and some evidence for self-reported problems (eg.
activity limitation) and financial support (Table 7)
- Interesting: Newly identified non-audiological
factors for HA uptake are cognitive anxiety (eg. from communication
breakdown) and urban residency
- Interesting: Co-morbidities that increase risk of
hearing loss (diabetes, hypertension, stroke) were negatively associated
with HA uptake, possibly because having to deal with chronic
illness
- Gaps in research: Factors with little data and/or mixed results for HA uptake include: willingness to use HA, social interaction, cost concerns, general health attitude (Table 7), HA cost, and counselling style