Abstract
Background
Hearing loss (HL) and menopausal hormone therapy (conjugated equine estrogens [CEE] and/or medroxyprogesterone acetate [MPA]) are separately associated with cognitive decline and increased risk of incident cognitive impairment. Joint effects of HL and HT could be associated with additive or synergistic decline in global cognition and risk of incident cognitive impairment among postmenopausal women.
Methods
Using the Women’s Health Initiative (WHI) Memory Study, 7,220 postmenopausal women with measures of HL, global cognition (Modified Mini-Mental State Examination score), and cognitive impairment (centrally adjudicated diagnoses of mild cognitive impairment and dementia) from 1996 to 2009. Multivariable linear mixed-effects models were used to analyze rate of change in global cognition. Accelerated failure time models were used to evaluate time to incident cognitive impairment, stratified by HT.
Results
Within the CEE-Alone trial, observed adverse effects of CEE-Alone on change in global cognition did not differ by HL, and estimated joint effects of HL and CEE-Alone were not associated with incident cognitive impairment. Within the CEE+MPA trial, while HL did not independently accelerate time to cognitive impairment, the adverse effect of CEE+MPA on global cognition was heightened in older women with HL. Older women on CEE+MPA either with HL (time ratio [TR] = 0.82, 95% confidence interval [CI]: 0.71, 0.94) or with normal hearing (TR = 0.86, 95% CI: 0.76, 0.97) had faster time to cognitive impairment than those with normal hearing and placebo.
Conclusions
HL may accentuate the adverse effect of CEE+MPA, not CEE-Alone, on global cognitive decline, not incident cognitive impairment, among postmenopausal women on HT.
Keywords: Hearing loss, Cognitive impairment, Cognitive decline, Hormone therapy
Hearing loss (HL), a highly prevalent condition among older adults in the United States, is an understudied risk factor for cognitive decline and dementia (1). About one-third of women experience HL by age 50, and approximately two-thirds have HL by age 60 (2). Hypothesized mechanistic pathways underlying the association of HL with cognitive outcomes include effects of poor hearing on cognitive load, changes in brain structure and function, and/or reduced social engagement (3).
Menopausal hormone therapy (HT) in older women is also associated with increased risk for dementia (4,5). In the Women’s Health Initiative Memory Study (WHIMS), women randomized to either conjugated equine estrogen with medroxyprogesterone acetate (CEE+MPA) or CEE-Alone had steeper declines in global cognition and increased risk of probable dementia, compared to those randomized to placebo (4–7). In observational studies, menopausal HT is associated with HL, with older women with naturally occurring menopause and longer duration of oral HT having increased HL risk (8). The relationship between HT and HL is biologically plausible, yet the influence of estrogen on HL is complicated. Age-dependent HL may be due to decline in female sex hormones that are potentially protective of HL (9,10). Decreased estrogen production resulting from menopause coincides with the average age of HL onset among women (11,12). Low estrogen levels could impair hearing through reduction of cochlear blood flow, neuronal physiology, or bone metabolism in the otic capsule (13–15). Yet, it remains unclear whether postmenopausal HT and HL may jointly aggravate the risk of cognitive decline and dementia.
However, the influence of postmenopausal HT on auditory function is complex and remains to be investigated. Using WHIMS data from the randomized trial of HT in older postmenopausal women, we examined the estimated independent and joint effects of HL with HT on potentially faster declines in global cognition and shorter time to incident cognitive impairment. While the primary aim was to evaluate whether HT had additive effects on the associations of HL with change in cognitive outcomes, we also assessed the independent associations of HL with these outcomes.
Method
Participants
From 1995 to 1999, 7,220 women aged 65–80 years with completed self-reported HL measures enrolled in WHIMS, an ancillary study to the WHI Hormone Trials that investigated the effects of postmenopausal estrogen therapy on cognitive outcomes (16). Participants were enrolled in two parallel clinical trials based on hysterectomy status: 2,834 with prior hysterectomy were randomized to 0.625 mg conjugated equine estrogen (CEE-Alone) or matching placebo and 4,386 women with intact uteri were randomized to 0.625 mg CEE and 2.5 mg medroxyprogesterone acetate (MPA) or matching placebo. The CEE-Alone trial ended in 2004 (17) and the CEE+MPA trial ended in 2002 (18), although post-trial follow-up continued. Due to differences in sample characteristics by trial, the two trials were analyzed separately. WHIMS was coordinated by the WHIMS Clinical Coordination Center (Wake Forest University Health Sciences, Winston-Salem, NC). The National Institutes of Health and the institutional review boards for all participating institutions approved the WHI and WHIMS protocols and consent forms (16).
Self-Reported Hearing Loss
Baseline self-reported HL was rated as being mildly, moderately, and severely bothersome versus normal (not having the condition) during the past 4 weeks.
Modified Mini-Mental State Examination
A measure of global cognition, 3MSE (19), was administered at baseline and annually up to September 2007 by trained, certified WHIMS technicians, blinded to randomization assignment (6,20). Scores range from 0 to 100, with higher scores indicating better global cognition (19,21). Z scores were calculated for 3MSE and standardized to baseline 3MSE mean and standard deviation.
Incident Cognitive Impairment
Incident cognitive impairment is defined as an adjudicated classification of either mild cognitive impairment (MCI) or probable dementia up to 2009 (16). Detailed information on diagnostic classification has been published (4,5) and is included in the Supplementary Material. Briefly, participants scoring below age-adjusted 3MSE cut-points were administered a standardized, validated neurocognitive test battery and questionnaires. An informant answered questions related to cognitive and behavioral functioning, and a local clinician conducted a neuropsychiatric evaluation. Optionally, the clinician ordered a computerized tomography brain scan and blood tests before making a classification of no cognitive impairment, MCI or probable dementia using standardized criteria. The classification and all assessment data were reviewed centrally by an adjudication panel consisting of three board-certified specialists (two neurologists and one geriatric psychiatrist) experienced in diagnosis of dementias, who made final impairment classifications (4,16).
Statistical Analysis
Descriptive statistics for participants grouped by hearing status were compared using t-tests and chi-squared tests within each trial. We first examined two-way interactions between HT and HL by trial. Then, we created a categorical indicator with the following categories: (0) normal hearing and placebo; (1) any self-reported HL and placebo; (2) normal hearing and HT; (3) any self-reported HL and HT. All analyses were adjusted for baseline demographic and clinical characteristics, that is, age, college education (≥16 vs <16 years of education), marital status (married vs not), region (Northeast, South, Midwest, and West), annual household income (≥$50,000/year vs <$50,000/year), race (white vs nonwhite), smoking status (current, former, or never), presence of diabetes and hypertension, depressive symptoms (abbreviated Center for Epidemiological Studies-Depression scale score (22)), and 3MSE.
Associations of hearing loss and hormone therapy with level and change in global cognition
Baseline associations of HL and HT with global cognition were assessed using analysis of covariance. Mean differences in annual global cognition by hearing status were compared using mixed-effects models with an unstructured variance–covariance structure. Intercepts and slopes varied among individuals. The fixed effects included HL, HT, covariates, visits, two-way interactions between HT and HL with visits, and three-way interactions among HT, HL, and visits. We modeled each visit annually to account for practice effects from repeated 3MSE administrations. The interaction of HL with annual visits was used to examine baseline self-reported HL with annual rate of 3MSE change. For the combined effects of HL and HT on annual rate of change in global cognition, we included the categorical indicator as the fixed effect instead of HL and HT separately and included a two-way interaction between this categorical indicator with each annual visit. We truncated the follow-up interval to 10 years, due to few observations after 10 years.
Association of hearing loss and hormone therapy with incident cognitive impairment
We estimated Kaplan–Meier survival curves to portray differences in the cumulative incidence of any cognitive impairment by hearing status (23). We used an accelerated failure time model (24) with log-normal distributions to examine the association of HL with incident cognitive impairment by trial, since the estimated effects were not proportional over time. When estimating the joint effects of HL and HT with incident cognitive impairment, we used the categorical indicator.
Sensitivity Analyses
We conducted two sensitivity analyses. First, we excluded all participants with “ever stroke” to see if inferences changed, because stroke could confound the associations of HL with cognitive outcomes. Second, we repeated all analyses restricted to on-trial observations when HT was administered to see if inferences changed.
Results
Sample Characteristics
Table 1 shows the sample characteristics of the WHIMS participants (N = 7,220) by trial. In both trials, those with HL were older with more depressive symptoms and lower 3MSE scores than those with normal hearing. In the CEE-Alone trial, those with HL were less likely to be college graduates, to earn an income of ≥$50,000, to be current smokers, and more likely to have diabetes than those with normal hearing. In the CEE+MPA trial, those with HL were less likely to married and had fewer years on the study, on average (Table 1).
Table 1.
Sample Characteristics From the Women’s Health Initiative Memory Study (N = 7,220)
| Baseline Characteristics | CEE Alone Trial (N = 2,834) | CEE+MPA Trial (N = 4,386) | ||||
|---|---|---|---|---|---|---|
| Hearing Loss | Normal Hearing | p Value for Difference | Hearing Loss | Normal Hearing | p Value for Difference | |
| n = 860 | n = 1,974 | n = 2,230 | n = 2,156 | |||
| Age, mean (SD) | 71.9 (4.1) | 70.6 (3.7) | <.001 | 71.9 (3.9) | 70.5 (3.7) | <.001 |
| College graduate, n (%) | 187 (21.7) | 512 (25.9) | .030 | 448 (34.6) | 1,019 (33.0) | .467 |
| Income of $50,000+, n (%) | 122 (14.2) | 363 (18.4) | .023 | 271 (20.9) | 738 (23.9) | .108 |
| Married, n(%) | 427 (49.7) | 998 (50.6) | .684 | 657 (50.8) | 1,767 (57.2) | .001 |
| Diabetes, n(%) | 101 (11.7) | 182 (9.2) | .039 | 84 (6.5) | 160 (5.2) | .083 |
| Hypertension, n(%) | .077 | .893 | ||||
| Never hypertensive | 818 (57.2) | 1,070 (54.2) | 820 (63.4) | 1,971 (63.8) | ||
| Untreated hypertension | 78 (9.1) | 166 (8.4) | 104 (8.0) | 262 (8.5) | ||
| Treated hypertension | 280 (32.6) | 714 (36.2) | 360 (27.8) | 832 (26.9) | ||
| Smoking Status, n(%) | .004 | .493 | ||||
| Current Smoker | 46 (5.4) | 155 (7.9) | 77 (6.0) | 213 (6.9) | ||
| Former Smoker | 349 (40.6) | 738 (37.4) | 520 (40.2) | 1,233 (39.9) | ||
| Never Smoker | 448 (52.1) | 1,064 (53.9) | 682 (52.7) | 1,598 (51.7) | ||
| Region, n(%) | .112 | .568 | ||||
| Northeast | 249 (29.0) | 490 (24.8) | 354 (27.4) | 853 (27.6) | ||
| South | 187 (21.7) | 435 (22.0) | 260 (20.1) | 627 (20.3) | ||
| Midwest | 205 (21.7) | 489 (24.8) | 333 (25.7) | 738 (23.9) | ||
| West | 219 (25.5) | 560 (28.4) | 347 (26.8) | 874 (28.3) | ||
| Race, n(%) | .239 | .095 | ||||
| White | 726 (84.7) | 1,630 (82.9) | 1,173 (91.0) | 2,739 (89.0) | ||
| Nonwhite | 128 (14.9) | 334 (17.0) | 115 (8.9) | 333 (10.8) | ||
| Depressive symptoms, mean (SD) | 0.05 (0.14) | 0.03 (0.10) | <.001 | 0.03 (0.11) | 0.02 (0.09) | <0.001 |
| Total modified Mini-Mental State Examination (3MSE) score, mean (SD) | 94.4 (4.7) | 94.8 (4.6) | .031 | 95.4 (3.9) | 95.7 (4.1) | .030 |
| Categorical 3MSE Score, n (%) | .090 | .033 | ||||
| ≥95 | 511 (59.4) | 1,258 (63.7) | 877 (67.8) | 2,217 (71.7) | ||
| 90–94 | 235 (27.3) | 476 (24.1) | 303 (23.4) | 640 (20.7) | ||
| <90 | 114 (13.3) | 240 (12.2) | 114 (8.8) | 235 (7.6) | ||
| Follow-up time for cognitive impairment, in years, mean (SD) | 7.3 (2.8) | 7.5 (1.7) | .109 | 7.5 (2.6) | 7.8 (2.6) | .002 |
| Follow-up time for cognitive change, in years, mean (SD) | 9.2 (3.0) | 9.4 (3.0) | .078 | 9.3 (2.9) | 9.6 (2.9) | .007 |
Note: SD, standard deviation. There are 22 missing education, 432 missing income, 16 missing marital status, 81 missing hypertension, 98 missing current smoking status, 14 missing race, and 202 missing depressive symptoms before randomization.
Figure 1 features the Kaplan–Meier plot that shows the unadjusted cumulative incidence of cognitive impairment by hearing status and HT by trial. There were 209 incident cases of cognitive impairment in the CEE-Alone trial and 255 incident cases of cognitive impairment in the CEE+MPA trial. The CEE-Alone trial participants contributed 20912.6 person-years, while the CEE+MPA trial participants contributed 33525.1 person-years.
Figure 1.
Kaplan–Meier plot of the association of hearing loss (HL) and hormone therapy with incident cognitive impairment for CEE-Alone (N = 2,802) and CEE+MPA (N = 4,352) randomized controlled trials in the Women’s Health Initiative Memory Study. PBO, placebo, TRT, treatment.
Associations between hearing loss with and without hormone therapy and global cognition cross-sectionally and longitudinally
Supplementary Table 1 shows cross-sectional associations of HL and HT with baseline global cognition. HL (β= −0.09; 95% confidence interval [CI]: −0.11, −0.06) and HT (β= −0.06; 95% CI: −0.08, −0.04) were independently associated with lower baseline global cognition in the CEE+MPA, but not CEE-Alone, trial. HL did not modify the association of HT with baseline global cognition in either trial.
Figure 2 shows the results from mixed-effects models examining the association of HL and cognitive change over time. Supplementary Table 2 includes the mean change in standard deviations of 3MSE at each visit, compared with baseline, by hearing status and the respective 95% CIs. In the CEE-Alone trial, the number of significant associations was small and with no clear pattern. In the CEE+MPA trial, there were associations of HL with decreases in global cognition at each annual visit from 4 years since baseline onward (Figure 2).
Figure 2.
Fitted mean differences in global cognition since baseline by hearing status over time for CEE-Alone (N = 2,802) and CEE+MPA (N = 4,352) randomized controlled trials.
We evaluated whether HL modified the association of HT with time. We did not find any significant three-way interactions in either trial. The results of the estimated combined effects of HL and HT with cognitive change are given in Table 2. Although there was no evidence of associations of combined HL and HT with cognitive change in the CEE-Alone trial, we found estimated combined effects of HL and HT with cognitive change in the CEE+MPA trial (Table 2). In that trial, HL and placebo were associated with 3MSE declines, starting 5 years from baseline and declining up to 10 years from baseline with a consistent pattern in the last 3 years, when compared with those with normal hearing and placebo. Additionally, those with HL and CEE+MPA assignment had steeper 3MSE declines than those with normal hearing and placebo, starting 3 years after baseline with more progressive declines over time and reaching the greatest decline at 10 years of follow-up (Table 2).
Table 2.
Estimated Joint Effects of Hearing Loss (HL) and Hormone Therapy (HT) Assignment on Change in Standard Deviation in Modified Mini-Mental Examination Score by Randomized Controlled Trial Within the Women’s Health Initiative Memory Study (N = 7,220)
| Visit Year | Placebo | CEE-Alone | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Normal Hearing | Hearing Loss | Normal Hearing | Hearing Loss | |||||||
| No. With HL/No. at Risk | β | 95% CI | No. With HT/No. at Risk | β | 95% CI | No. With HL and HT/No. at Risk | β | 95% CI | ||
| 0 | REF | 429/2,802 | −0.022 | (−0.128, 0.084) | 959/2,802 | 0.039 | (−0.042, 0.121) | 428/2,802 | 0.002 | (−0.104, 0.108) |
| 1 | REF | 413/2,713 | 0.082 | (−0.019, 0.184) | 934/2,713 | −0.020 | (−0.098, 0.059) | 413/2,713 | −0.074 | (−0.175, 0.027) |
| 2 | REF | 385/2,543 | −0.002 | (−0.111, 0.107) | 874/2,543 | −0.048 | (−0.132, 0.037) | 382/2,543 | −0.145 | (−0.254, −0.036) |
| 3 | REF | 380/2,516 | 0.078 | (−0.041, 0.197) | 866/2,516 | −0.043 | (−0.135, 0.050) | 376/2,516 | −0.030 | (−0.150, 0.090) |
| 4 | REF | 345/2,361 | 0.020 | (−0.114, 0.154) | 831/2,361 | −0.065 | (−0.169, 0.038) | 358/2,361 | −0.134 | (−0.267, 0.000) |
| 5 | REF | 331/2,243 | −0.002 | (−0.152, 0.149) | 787/2,243 | −0.065 | (−0.181, 0.050) | 328/2,243 | −0.074 | (−0.224, 0.077) |
| 6 | REF | 310/2,145 | −0.046 | (−0.214, 0.123) | 740/2,145 | −0.105 | (−0.233, 0.024) | 322/2,145 | −0.112 | (−0.279, 0.056) |
| 7 | REF | 261/1,799 | −0.107 | (−0.297, 0.084) | 623/1,799 | −0.183 | (−0.329, −0.037) | 272/1,799 | −0.124 | (−0.313, 0.065) |
| 8 | REF | 191/1,363 | −0.213 | (−0.430, 0.004) | 475/1,363 | −0.131 | (−0.297, 0.034) | 197/1,363 | −0.138 | (−0.354, 0.078) |
| 9 | REF | 169/1,201 | −0.228 | (−0.468, 0.013) | 434/1,201 | −0.135 | (−0.318, 0.048) | 170/1,201 | −0.120 | (−0.360, 0.120) |
| 10 | REF | 107/771 | −0.163 | (−0.441, 0.115) | 272/771 | −0.218 | (−0.429, −0.007) | 121/771 | −0.125 | (−0.399, 0.149) |
| Visit Year | Placebo | CEE+MPA | ||||||||
| Normal Hearing | Hearing Loss | Normal Hearing | Hearing Loss | |||||||
| No. with HL/No. at Risk | β | 95% CI | No. with HT/No. at Risk | β | 95% CI | No. with HL and HT/No. at Risk | β | 95% CI | ||
| 0 | REF | 658/4,352 | −0.025 | (−0.101, 0.051) | 1,511/4,352 | −0.023 | (−0.081, 0.035) | 631/4,352 | −0.061 | (−0.138, 0.015) |
| 1 | REF | 636/4,213 | −0.008 | (−0.083, 0.068) | 1,459/4,213 | −0.018 | (−0.076, 0.040) | 608/4,213 | −0.041 | (−0.117, 0.035) |
| 2 | REF | 613/4,032 | −0.010 | (−0.092, 0.072) | 1,397/4,032 | −0.010 | (−0.073, 0.053) | 584/4,032 | −0.057 | (−0.139, 0.026) |
| 3 | REF | 597/3,978 | −0.044 | (−0.135, 0.046) | 1,384/3,978 | −0.042 | (−0.112, 0.028) | 569/3,978 | −0.120 | (−0.211, −0.028) |
| 4 | REF | 559/3,755 | −0.059 | (−0.162, 0.043) | 1,316/3,755 | −0.081 | (−0.160, −0.002) | 535/3,755 | −0.154 | (−0.258, −0.051) |
| 5 | REF | 543/3,600 | −0.121 | (−0.237, −0.006) | 1,262/3,600 | −0.070 | (−0.159, 0.018) | 513/3,600 | −0.151 | (−0.268, −0.034) |
| 6 | REF | 507/3,449 | −0.080 | (−0.210, 0.050) | 1,204/3,449 | −0.050 | (−0.149, 0.050) | 484/3,449 | −0.229 | (−0.360, −0.098) |
| 7 | REF | 417/2,898 | −0.109 | (−0.256, 0.039) | 1,034/2,898 | −0.044 | (−0.156, 0.069) | 390/2,898 | −0.205 | (−0.355, −0.055) |
| 8 | REF | 317/2,229 | −0.202 | (−0.371, −0.034) | 788/2,229 | −0.095 | (−0.223, 0.033) | 293/2,229 | −0.256 | (−0.426, −0.085) |
| 9 | REF | 295/2,058 | −0.216 | (−0.402, −0.030) | 740/2,058 | −0.123 | (−0.264, 0.018) | 265/2,058 | −0.328 | (−0.516, −0.139) |
| 10 | REF | 179/1,352 | −0.306 | (−0.520, −0.092) | 489/1,352 | −0.104 | (−0.265, 0.056) | 166/1,352 | −0.364 | (−0.581, −0.148) |
Note: Bolded values mean p < .05.
Association of hearing loss and hormone therapy with time to incident cognitive impairment
Table 3 shows the results from accelerated failure time models examining the estimated independent and joint associations of HL and HT with incident cognitive impairment in a pooled sample and by each trial. Among all WHIMS participants, there were no associations of HL with time to incident cognitive impairment. In the CEE-Alone trial, there were no independent associations of HL and HT with incident cognitive impairment (Table 3). When the independent association was examined within the CEE+MPA trial, CEE+MPA treatment was associated with earlier onset of cognitive impairment after covariate adjustment, while HL was not (Table 3). For the analyses of joint effects, CEE+MPA, independent of HL, was associated with earlier onset of cognitive impairment.
Table 3.
Association of Hearing Status (Any Hearing Loss Vs Normal Hearing) and Hormone Therapy Assignment With Time to Incident MCI/Dementia up to 13 Years in WHIMS
| Estimated Independent and Joint Effects | Unadjusted | Adjusted | ||
|---|---|---|---|---|
| Time Ratio | 95% CI | Time Ratio | 95% CI | |
| Overall sample (N = 7,220) | ||||
| Independent effects | ||||
| Hearing loss | 0.90 | (0.85, 0.95) | 0.95 | (0.88, 1.02) |
| Hormone therapy assignment | 0.95 | (0.90, 1.00) | 0.95 | (0.89, 1.02) |
| Joint effects | ||||
| Normal hearing, placebo | REF | REF | REF | REF |
| Hearing loss, placebo | 0.88 | (0.81, 0.95) | 0.92 | (0.82, 1.02) |
| Normal hearing, hormone therapy | 0.93 | (0.87, 1.00) | 0.93 | (0.86, 1.02) |
| Hearing loss, hormone therapy | 0.86 | (0.79, 0.93) | 0.91 | (0.82, 1.01) |
| CEE-alone trial (N = 2,774) | ||||
| Independent effects | ||||
| Hearing loss | 0.98 | (0.900, 1.057) | 1.00 | (0.89,1.12) |
| CEE assignment | 0.99 | (0.915, 1.061) | 1.04 | (0.93,1.16) |
| Joint effects | ||||
| Normal hearing, placebo | REF | REF | REF | REF |
| Hearing loss, placebo | 0.98 | (0.67, 1.62) | 0.98 | (0.83, 1.15) |
| Normal hearing, CEE | 0.99 | (0.81, 1.10) | 1.03 | (0.90, 1.17) |
| Hearing loss, CEE | 0.95 | (0.66, 1.32) | 1.04 | (0.88, 1.23) |
| CEE+MPA trial (N = 4,298) | ||||
| Independent effects | ||||
| Hearing loss | 0.85 | (0.78, 0.92) | 0.92 | (0.83, 1.01) |
| CEE+MPA assignment | 0.92 | (0.85, 0.99) | 0.88 | (0.80, 0.97) |
| Joint effects | ||||
| Normal hearing, placebo | REF | REF | REF | REF |
| Hearing loss, placebo | 0.82 | (0.71, 0.94) | 0.88 | (0.76, 1.01) |
| Normal hearing, CEE+MPA | 0.88 | (0.78, 0.99) | 0.86 | (0.76, 0.97) |
| Hearing loss, CEE+MPA | 0.76 | (0.66, 0.88) | 0.82 | (0.71, 0.94) |
Note: Bolded values mean p < .05. CEE, conjugated equine estrogen; MPA, medroxyprogesterone acetate.
Sensitivity Analyses
Our results remained robust across two sensitivity analyses: exclusion of participants with ever stroke (Supplementary Tables 3 and 4) and restriction of observations to on-trial ones (Supplementary Tables 5–7).
Discussion
To the best of our knowledge, the current study is the first to examine the joint effects of both HL and HT in postmenopausal women within the context of randomized clinical trials. Baseline self-reported HL, coupled with postmenopausal HT with CEE+MPA, but not CEE-Alone, was associated with declines in global cognition, not incident cognitive impairment, in women aged 65–80 years. Those with any HL, independent of CEE+MPA, had declines in global cognition, starting 5 years from baseline, and those with both HL and CEE+MPA had declines in global cognition, starting 3 years from baseline, when compared with those with normal hearing and randomized to placebo. In contrast, when examining the independent associations of HL and HT, there were no associations between HL and incident cognitive impairment in either trial. There was, however, an association between CEE+MPA and incident cognitive impairment, independent of HL. Estimated joint effects of HL and CEE+MPA on incident cognitive impairment showed that those randomized to CEE+MPA, either with HL or normal hearing, had a faster onset of cognitive impairment.
Differences in cognitive trajectories by HL and HT differed between the two trials. In both trials, women who reported HL before randomization were more likely to have lower baseline 3MSE scores than women who reported normal hearing, but this was more pronounced in those without prior hysterectomy. Poorer baseline cognition before HT initiation, CEE independent of MPA, was associated with greater total brain and hippocampal atrophy (25), which suggests increased brain atrophy may mediate the association of HT with poorer cognitive outcomes (26). HL could represent an additional insult to brain vulnerability, especially to women with low 3MSE scores who were already vulnerable to the effects of HT. The apparent joint effects of CEE+MPA and HL on cognitive decline could be due to worsening of HL over time in users of CEE+MPA, thus leading to a larger magnitude of decline in cognitive function due to greater HL.
In the CEE+MPA trial, we observed an association of HL with decline in global cognition, starting 5 years after baseline. This was consistent with findings from a previous study that reported the association of HL with longitudinal declines in global cognition (27–30). Also, we observed an association of HL and CEE+MPA with a decline in global cognition, starting 3 years after baseline. CEE+MPA is associated with procoagulant (31) and proinflammatory (32) markers, which, in turn, are associated with vascular disease outcomes. Vascular disease outcomes may trigger a cascade of downstream neuronal injuries resulting in dementia (33). However, in WHIMS, ischemic burden post-trial was independent of HT (25,34). Also, our results did not change when we excluded participants who ever had a stroke. We did not observe differences in rates of cognitive change in the CEE-Alone group among those with the same level of hearing status. In prior WHI studies, CEE-Alone in older women was associated with declines in global cognition (35,36). While it appears that the MPA is the driver of these associations, we cannot fully disentangle the effects of MPA from CEE in this trial, given the differences in baseline sample characteristics and sample sizes.
Additionally, we examined the association of HL with and without HT with incident cognitive impairment in both trials. We did not observe associations between HL and incident cognitive impairment. However, we found that CEE+MPA was independently associated with faster onset of cognitive impairment. A previous WHIMS study examining HT (either CEE-Alone or CEE+MPA) with cognitive decline and probable dementia found that CEE+MPA increased the risk for probable dementia in women aged 65 years and older (4). Observational studies have reported neuroprotective effects of estrogen (37), but in studies of cell culture systems (38), rat models (39), and cynomolgus monkeys (40), the combination of CEE+MPA may reverse the neuroprotective effects observed from CEE-alone.
There were several strengths and limitations of the study. The strengths were the large sample of postmenopausal women with randomized HT assignment, repeated measures of 3MSE, and adjudicated MCI and dementia outcomes using standardized criteria. Another limitation was the imprecise nature of the self-reported HL measure, which queried participants on how “bothersome” HL was. A woman’s rating of bothersome HL may reflect objective hearing loss or cognitive difficulties or psychosocial or environmental factors or a combination of these and other factors. However, the misclassification of self-reported HL would likely bias our findings toward null results.
In summary, we observed associations of HL with cognitive decline in the CEE+MPA trial, but not the CEE-Alone trial. In the CEE+MPA trial, HL was associated with decline in global cognition, starting 5 years after baseline. The estimated joint effects of HL and CEE+MPA on cognitive decline were apparent 3 years after baseline. HL may exacerbate the adverse effects of CEE+MPA on change in global cognition. Those on CEE+MPA, independent of hearing status, had faster onset of cognitive impairment. As we observed associations of HL with cognitive decline, a future direction is to evaluate the association of HL with change in brain volumes, since the change in brain volumes may mediate the association of HL with cognitive decline.
Supplementary Material
Acknowledgments
We would like to thank the following people—Program Office: (National Heart, Lung, and Blood Institute, Bethesda, Maryland) Jacques Rossouw, Shari Ludlam, Joan McGowan, Leslie Ford, and Nancy Geller. Clinical Coordinating Center: (Fred Hutchinson Cancer Research Center, Seattle, WA) Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg. Investigators and Academic Centers: (Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn E. Manson; (MedStar Health Research Institute/Howard University, Washington, DC) Barbara V. Howard; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A. Thomson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Iowa, Iowa City/Davenport, IA) Jennifer Robinson; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker; (University of Nevada, Reno, NV) Robert Brunner; (University of Minnesota, Minneapolis, MN) Karen L. Margolis. Women’s Health Initiative Memory Study: (Wake Forest University School of Medicine, Winston-Salem, NC) Mark Espeland.
Funding
Both NMA and SMR are supported by the Intramural Research Program of the NIH, National Institute on Aging. The WHI program is funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C, and HHSN268201600004C. The Women’s Health Initiative Memory Study was funded in part by Wyeth Pharmaceuticals, Inc, St. Davids, PA.
Author Contributions
N.M.A., M.A.E., J.A.D., S.R.R., F.R.L., and S.M.L. made substantial contributions to conception and design. M.A.E., S.R.R., and S.M.L. contributed to acquisition of data. All authors contributed to analysis and interpretation of data. N.M.A. drafted the article. All authors revised it critically for important content and approved of the final version to be published. All authors meet the criteria for authorship stated in the Uniform Requirements for Manuscripts Submitted to Biomedical Journals.
Conflict of Interest
None reported.
References
- 1. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. Lancet. 2017;390:2673–2734. doi: 10.1016/S0140-6736(17)31363-6 [DOI] [PubMed] [Google Scholar]
- 2. Agrawal Y, Platz EA, Niparko JK. Prevalence of hearing loss and differences by demographic characteristics among US adults: data from the National Health and Nutrition Examination Survey, 1999–2004. Arch Intern Med. 2008;168:1522–1530. doi: 10.1001/archinte.168.14.1522 [DOI] [PubMed] [Google Scholar]
- 3. Lin FR, Albert M. Hearing loss and dementia—who is listening? Aging Ment Health. 2014;18:671–673. doi: 10.1080/13607863.2014.915924 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Shumaker SA, Legault C, Rapp SR, et al. ; WHIMS Investigators. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women’s Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003;289:2651–2662. doi: 10.1001/jama.289.20.2651 [DOI] [PubMed] [Google Scholar]
- 5. Shumaker SA, Legault C, Kuller L, et al. ; Women’s Health Initiative Memory Study. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women’s Health Initiative Memory Study. JAMA. 2004;291:2947–2958. doi: 10.1001/jama.291.24.2947 [DOI] [PubMed] [Google Scholar]
- 6. Rapp SR, Espeland MA, Shumaker SA, et al. ; WHIMS Investigators. Effect of estrogen plus progestin on global cognitive function in postmenopausal women: the Women’s Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003;289:2663–2672. doi: 10.1001/jama.289.20.2663 [DOI] [PubMed] [Google Scholar]
- 7. Espeland MA, Rapp SR, Shumaker SA, et al. ; Women’s Health Initiative Memory Study. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women’s Health Initiative Memory Study. JAMA. 2004;291:2959–2968. doi: 10.1001/jama.291.24.2959 [DOI] [PubMed] [Google Scholar]
- 8. Curhan SG, Eliassen AH, Eavey RD, Wang M, Lin BM, Curhan GC. Menopause and postmenopausal hormone therapy and risk of hearing loss. Menopause. 2017;24:1049–1056. doi: 10.1097/GME.0000000000000878 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kilicdag EB, Yavuz H, Bagis T, Tarim E, Erkan AN, Kazanci F. Effects of estrogen therapy on hearing in postmenopausal women. Am J Obstet Gynecol. 2004;190:77–82. doi: 10.1016/j.ajog.2003.06.001 [DOI] [PubMed] [Google Scholar]
- 10. Kim SH, Kang BM, Chae HD, Kim CH. The association between serum estradiol level and hearing sensitivity in postmenopausal women. Obstet Gynecol. 2002;99(5, Part 1):726–730. doi: 10.1016/S0029-7844(02)01963-4 [DOI] [PubMed] [Google Scholar]
- 11. Svedbrant J, Bark R, Hultcrantz M, Hederstierna C. Hearing decline in menopausal women–a 10-year follow-up. Acta Otolaryngol. 2015;135:807–813. doi: 10.3109/00016489.2015.1023354 [DOI] [PubMed] [Google Scholar]
- 12. Hederstierna C, Hultcrantz M, Collins A, Rosenhall U. The menopause triggers hearing decline in healthy women. Hear Res. 2010;259:31–35. doi: 10.1016/j.heares.2009.09.009 [DOI] [PubMed] [Google Scholar]
- 13. Horner KC. The effect of sex hormones on bone metabolism of the otic capsule—an overview. Hear Res. 2009;252:56–60. doi: 10.1016/j.heares.2008.12.004 [DOI] [PubMed] [Google Scholar]
- 14. Laugel G, Dengerink H, Wright J. Ovarian steroid and vasoconstrictor effects on cochlear blood flow. Hear Res. 1987;31:245–251. doi: 10.1016/0378-5955(87)90194-8 [DOI] [PubMed] [Google Scholar]
- 15. Stenberg A, Wang H, Fish Jr, Schrott-Fischer A, Sahlin L, Hultcrantz M. Estrogen receptors in the normal adult and developing human inner ear and in Turner’s syndrome. Hear Res. 2001;157:87–92. doi: 10.1016/S0378-5955(01)00280-5 [DOI] [PubMed] [Google Scholar]
- 16. Shumaker SA, Reboussin BA, Espeland MA, et al. The Women’s Health Initiative Memory Study (WHIMS): a trial of the effect of estrogen therapy in preventing and slowing the progression of dementia. Control Clin Trials. 1998;19:604–621. doi: 10.1016/S0197-2456(98)00038-5 [DOI] [PubMed] [Google Scholar]
- 17. The Women’s Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA. 2004;291:1701–1712. doi: 10.1001/jama.291.14.1701 [DOI] [PubMed] [Google Scholar]
- 18. Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA. 2002;288:321–333. doi: 10.1001/jama.288.3.321 [DOI] [PubMed] [Google Scholar]
- 19. Teng E, Chui H. The modified Mini-Mental State Examination (3MS). Can J Psychiatry. 1987;41:114–121. [PubMed] [Google Scholar]
- 20. Henderson VW, Paganini-Hill A, Miller BL, et al. Estrogen for Alzheimer’s disease in women: randomized, double-blind, placebo-controlled trial. Neurology. 2000;54:295–301. doi: 10.1212/wnl.54.2.295 [DOI] [PubMed] [Google Scholar]
- 21. Tombaugh T, McDowell I, Kristjansson B, Hubley A. Mini-Mental State Examination (MMSE) and the Modified MMSE (3MS): a psychometric comparison and normative data. Psychol Assess. 1996;8:48. doi: 10.1037/1040-3590.8.1.48 [DOI] [Google Scholar]
- 22. Burnam MA, Wells KB, Leake B, Landsverk J. Development of a brief screening instrument for detecting depressive disorders. Med Care. 1988;26:775–789. doi:10.1097/00005650-198808000-00004 [DOI] [PubMed] [Google Scholar]
- 23. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457–481. doi: 10.2307/2281868 [DOI] [Google Scholar]
- 24. Orbe J, Ferreira E, Núñez-Antón V. Comparing proportional hazards and accelerated failure time models for survival analysis. Stat Med. 2002;21:3493–3510. doi: 10.1002/sim.1251 [DOI] [PubMed] [Google Scholar]
- 25. Resnick SM, Espeland MA, Jaramillo SA, et al. Postmenopausal hormone therapy and regional brain volumes: the WHIMS-MRI Study. Neurology. 2009;72:135–142. doi: 10.1212/01.wnl.0000339037.76336.cf [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Espeland MA, Tindle HA, Bushnell CA, et al. Brain volumes, cognitive impairment, and conjugated equine estrogens. J Gerontol A Biol Sci Med Sci. 2009;64A:1243–1250. doi: 10.1093/gerona/glp128 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Lin FR, Yaffe K, Xia J, et al. ; Health ABC Study Group. Hearing loss and cognitive decline in older adults. JAMA Intern Med. 2013;173:293–299. doi: 10.1001/jamainternmed.2013.1868 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Armstrong NM, An Y, Ferrucci L, Deal JA, Lin FR, Resnick SM. Temporal sequence of peripheral auditory processing with cognition in the Baltimore Longitudinal Study of Aging. J Gerontol A Biol Sci Med Sci. In press. doi: 10.1093/gerona/gly268 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Deal JA, Betz J, Yaffe K, et al. ; Health ABC Study Group. Hearing impairment and incident dementia and cognitive decline in older adults: the Health ABC Study. J Gerontol A Biol Sci Med Sci. 2017;72:703–709. doi: 10.1093/gerona/glw069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Deal JA, Sharrett AR, Albert MS, et al. Hearing impairment and cognitive decline: a pilot study conducted within the Atherosclerosis Risk in Communities Neurocognitive Study. Am J Epidemiol. 2015;181:680–690. doi: 10.1093/aje/kwu333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Rosendaal FR, Helmerhorst FM, Vandenbroucke JP. Female hormones and thrombosis. Arterioscler Thromb Vasc Biol. 2002;22:201–210. doi:10.1161/hq0202.102318 [DOI] [PubMed] [Google Scholar]
- 32. Cushman M, Legault C, Barrett-Connor E, et al. Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation. 1999;100:717–722. doi: 10.1161/01.cir.100.7.717 [DOI] [PubMed] [Google Scholar]
- 33. Zlokovic BV. Neurovascular pathways to neurodegeneration in Alzheimer’s disease and other disorders. Nat Rev Neurosci. 2011;12:723–738. doi: 10.1038/nrn3114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Coker LH, Espeland MA, Rapp SR, et al. Postmenopausal hormone therapy and cognitive outcomes: the Women’s Health Initiative Memory Study (WHIMS). J Steroid Biochem Mol Biol. 2010;118:304–310. doi: 10.1016/j.jsbmb.2009.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Espeland MA, Rapp SR, Manson JE, et al. Long-term effects on cognitive trajectories of postmenopausal hormone therapy in two age groups. J Gerontol A Biol Sci Med Sci. 2017;72:838–845. doi: 10.1093/gerona/glw156 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Espeland MA, Brunner RL, Hogan PE, et al. ; Women’s Health Initiative Study of Cognitive Aging Study Group. Long-term effects of conjugated equine estrogen therapies on domain-specific cognitive function: results from the Women’s Health Initiative study of cognitive aging extension. J Am Geriatr Soc. 2010;58:1263–1271. doi: 10.1111/j.1532-5415.2010.02953.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. LeBlanc ES, Janowsky J, Chan BK, Nelson HD. Hormone replacement therapy and cognition: systematic review and meta-analysis. JAMA. 2001;285:1489–1499. doi: 10.1001/jama.285.11.1489 [DOI] [PubMed] [Google Scholar]
- 38. Nilsen J, Brinton RD. Impact of progestins on estradiol potentiation of the glutamate calcium response. Neuroreport. 2002;13:825–830. [DOI] [PubMed] [Google Scholar]
- 39. Chesler EJ, Juraska JM. Acute administration of estrogen and progesterone impairs the acquisition of the spatial Morris water maze in ovariectomized rats. Horm Behav. 2000;38:234–242. doi: 10.1006/hbeh.2000.1626 [DOI] [PubMed] [Google Scholar]
- 40. Herbison AE, Horvath TL, Naftolin F, Leranth C. Distribution of estrogen receptor-immunoreactive cells in monkey hypothalamus: relationship to neurones containing luteinizing hormone-releasing hormone and tyrosine hydroxylase. Neuroendocrinology. 1995;61:1–10. doi: 10.1159/000126810 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.