Major depressive disorder, depressive symptoms, and bilateral hearing loss in hispanic adults

Journal of Affective Disorders 44 (1997) 189–195

Brief report

Major depressive disorder, depressive symptoms, and bilateral1hearing loss in hispanic adults

a , b , a , c*David J. Lee , Orlando Gomez-MarinaDepartment of Epidemiology and Public Health, University of Miami School of Medicine, P.O. Box 016069 (R-699), Miami, FL

33101, USAbDepartment of Otolaryngology, University of Miami School of Medicine, P.O. Box 016069 (R-699), Miami, FL 33101, USA

cDepartment of Pediatrics, University of Miami School of Medicine, P.O. Box 016069 (R-699), Miami, FL 33101, USA

Received 21 January 1997; revised 25 March 1997; accepted 25 March 1997

Abstract

Studies in non-Hispanic whites have documented higher rates of hearing loss in adults with depression versus thosewithout depression. Data from the Hispanic Health and Nutrition Examination Survey were used to investigate associationsbetween depression and bilateral hearing loss in Cuban-American, Mexican-American, and Puerto Rican adults 20–74 yearsof age. Pure tone thresholds were calculated by averaging thresholds obtained at 500, 1000, and 2000 Hz separately for eachear. Hearing loss was defined as bilateral pure tone average threshold levels greater than 25 db. Lifetime history of a majordepressive episode was assessed by the Diagnostic Interview Schedule, and depressive symptoms were assessed by theCenter for Epidemiologic Studies Depression Schedule. Prevalence of overall hearing loss was consistently lower inHispanics with a history of major depressive disorder than among those without such a history. Odds of hearing impairmentwas significantly greater among Puerto-Ricans reporting more depressive symptoms versus fewer symptoms. However, nosuch associations were found among Mexican-Americans or Cuban-Americans. In conclusion, despite the findings in PuertoRicans with depressive symptoms short of major depression, hearing impaired Hispanics overall do not appear at increasedrisk of major depressive disorder. 1997 Elsevier Science B.V.

Keywords: Hearing loss; Audiometry; Depression; Depressive symptoms; Hispanic Americans

1. Introduction psychiatric disturbance were first postulated in theearly part of this century (Cooper, 1976). More

Associations between hearing loss and severe recent studies, however, have failed to replicate thesefindings (Jones and White, 1990). Furthermore, thevalidity of these earlier studies have been challenged*Corresponding author. Tel.: 1 1 305 2436980; fax: 1 1 305on methodological grounds (Meadow-Orlans, 1985;2433384; e-mail: [email protected].

1 Prager and Jeste, 1993).An earlier version of this paper was presented at the 48th AnnualMeeting of the Gerontological Society of America. Inconsistent associations between depression and

0165-0327/97/$17.00 1997 Elsevier Science B.V. All rights reservedPII S0165-0327( 97 )00044-X

190 D.J. Lee, O. Gomez-Marin / Journal of Affective Disorders 44 (1997) 189 –195

less severe levels of hearing loss have also been City area (Gonzalez et al., 1985; Delgado et al.,reported. Yovell et al. (1995) found significantly 1990). Overall response rates for the physical exami-higher levels of hearing impairment among de- nation where audiometric data was collected werepressed patients versus community controls. In 61, 76, and 75%, respectively, for Cuban-Americans,another case-control study of 20 patients with bipolar Mexican-Americans, and Puerto Ricans (Delgado etaffective disorder or major depression, odds of al., 1990). Complete audiometric, sociodemographic,hearing loss was higher in cases versus controls (30 and depression data was collected on 420 Cuban-vs. 10%), although these differences were not American, 1626 Mexican-American, and 600 Puertostatistically significant (Prager and Jeste, 1993). One Rican adults aged 20–74 years.other case-control study found significantly greaterodds of hearing impairment at 250 Hz among 2.2. Control measuresdepressed patients versus controls (Watt, 1985). TwoEuropean studies of elderly living in the community Increased age and, to a lesser extent, male genderfound positive associations between depression and are associated with risk of hearing loss in Hispanichearing loss (Herbst and Humphrey, 1980; Carbal- adults (Lee et al., 1991). Sociodemographic mea-lese et al., 1993), while another European study sures most consistently associated with hearing lossfound no association between depression and hearing in Hispanics are: (1) marital status (not marriedloss (Thomas et al., 1983). Finally, one study of US versus married), (2) employment status (unemployedmale veterans reported an increased prevalence of versus employed), and (3) health insurance statusdepression in hearing impaired versus non-impaired (uninsured versus insured) (Lee et al., 1996).participants (27 vs. 16%), a difference which ap-proached statistical significance (P 5 0.06) (Mulrow 2.3. Depression measureset al., 1990).

With one exception (Mulrow et al., 1990), each of Depressive symptoms were assessed by the Centerthese studies was either conducted in samples com- for Epidemiologic Studies Depression (CES-D)prised of non-Hispanic white participants, or the scale. This scale has been validated in differentethnic / racial composition of the study participants populations (Radloff, 1977; Weissman et al., 1977),was not specified by the investigators. Approximate- including Hispanic populations (Roberts, 1980). Thely one-third of the sample in the Mulrow et al. 20-item CES-D assesses depressive symptoms which(1990) study was comprised of Hispanics. Unfor- have occurred in the previous 2 weeks. Range oftunately, ethnic-specific analyses were not described. scores on the CES-D is 0–60 with higher scoresIt is therefore presently unknown if depressed His- indicative of greater symptomatology.panics are more likely to be hearing impaired The factor structure of the CES-D scale has beencompared with non-depressed Hispanics. examined in the HHANES (Guarnaccia et al., 1989).

A three-factor structure was identified: affect /somatic, interpersonal, and positive affect. However,this latter factor, was not included in the present

2. Methods analysis due to poor internal consistency reliability(a 5 0.19–0.60). The internal consistency reliability

2.1. Study population and design of the remaining two factors in the three Hispanicsubgroups were: affect / somatic (a 5 0.87–0.91) and

The Hispanic Health and Nutrition Examination interpersonal (a 5 0.57–0.79). Range of scores forSurvey (HHANES), conducted in 1982–1984 by the these measures varied in the three subgroups due toNational Center for Health Statistics (NCHS), used a differences in the number of subscale items. Thecomplex multi-stage sampling design, obtaining a affect / somatic subscale was comprised of itemsrepresentative sample of Mexican-Americans from which assess depressed affect (e.g., depressed, sad),five southwestern states, Cuban-Americans from the and somatic complaints (e.g., poor appetite, sleepMiami area, and Puerto Ricans from the New York difficulties). The interpersonal difficulties subscale

D.J. Lee, O. Gomez-Marin / Journal of Affective Disorders 44 (1997) 189 –195 191

was comprised of items assessing problems such as Institute, 1995). Multiple logistic regression wasthe belief that people are unfriendly and that people used to estimate the odds of hearing loss in depres-dislike them. It was hypothesized that odds of sion subgroups after controlling for age, gender,hearing impairment would be greater among His- marital status, employment status, and health insur-panics who score higher on this scale than those who ance status. Prior to completing these multivariatescore lower, given previous research in non-His- analyses, interactions between age, gender and eachpanics which has documented higher levels of hear- depressive measure were examined. No significanting impairment among those reporting more social interactions were found.isolation (Weinstein and Ventry, 1982; Thomas et al.,1983; Mulrow et al., 1990).

The Diagnostic Interview Schedule (DIS) was 3. Resultsdeveloped to identify the presence of psychiatricdisorders in community surveys according to the Table 1 presents results for Cuban-Americans.criteria of the Diagnostic and Statistical Manual of The odds of hearing impairment was slightly lowerMental Disorders, 3rd edition (DSM-III) (Robbins et in Cuban-Americans with a DIS-DSM-III lifetimeal., 1981; Task Force on Nomenclature and Statis- diagnosis of a major depressive episode than thosetics, 1980). Only questions pertaining to the diag- without a lifetime diagnosis of depression. Odds ofnosis of depression were administered in the hearing loss was greater among Cuban-Americans inHHANES. Application of a computer algorithm to the upper tertile on the CES-D scale relative to theinterview responses produced either a diagnosis of lower tertile, but these differences failed to reachno major depressive episode, a lifetime history of a statistical significance. There was no differencemajor depressive episode, or bereavement (NCHS, between odds of hearing impairment among those1987). In the present study, bereaved individuals reporting fewer versus those reporting more symp-were coded as not having a lifetime episode of major toms on the CES-D affective /somatic subscale anddepressive disorder. the interpersonal difficulties subscales.

The results for Mexican-Americans are presented2.4. Hearing measures in Table 2. The odds of hearing impairment was

significantly lower among Mexican-Americans withAudiometric testing was conducted in the a lifetime diagnosis of a depressive episode than

HHANES using standardized methodology (NCHS, those with no history of depression (OR 5 0.01; 95%1988), including calibration specifications for au- CI 5 0.00–0.03). Although significant, it should bediometric equipment (American National Standards noted that the estimate of this odds ratio is unstableInstitute, 1970). Air conduction thresholds were because there were no cases of hearing loss amongobtained in sound treated rooms at 500, 1000, 2000, Mexican-Americans with a history of major depres-and 4000 Hz. Pure tone averages (PTAs) were sive episode. Odds of hearing loss did not differcalculated by averaging thresholds obtained at fre- significantly among Mexican-Americans reporting noquencies of 500, 1000, and 2000 Hz for each ear. symptomatology versus those reporting at least someOverall bilateral hearing loss was defined as a PTA depressive symptomatology on the CES-D measuregreater than 25 dBHL in both ears (American and subscales.Speech-Language-Hearing Association, 1981). The odds of hearing impairment was approximate-

ly three times lower in Puerto Ricans with a lifetime2.5. Statistical analysis history of a major depressive episode versus those

without such a diagnosis (see Table 3). Odds ofBecause of the multi-stage sampling design, all hearing impairment was significantly greater among

analyses were performed with adjustments for sam- those scoring in the middle (OR 5 2.59; 95% CI 5

ple weights and design effects using the SUDAAN 1.15–5.82) and upper (OR 5 4.58; 95% CI 5 1.95–statistical package for analysis of complex sample 10.75) tertile on the CES-D scale compared to thosesurveys (Delgado et al., 1990; Research Triangle in the lower tertile. Odds of hearing impairment was


Table 1Depression and overall bilateral hearing loss ( . 25 dBHL) among Cuban-American adults 20–74 years of age: data from the HispanicHealth and Nutrition Examination Survey, 1982–1984

a a b cDepression measure No. without hearing loss No. with hearing loss Adjusted odds ratio 95% CIdDIS: major

Depressive episode, no 368 40 1.00 referentDepressive episode, yes 11 1 0.83 (0.06–11.88)

dCES-D: total score0–3 212 23 1.00 referent4–9 76 6 0.70 (0.16–3.03)10–60 62 10 1.44 (0.60–3.47)

dCES-D: subscaleAffective symptoms, 0–1 187 21 1.00 referentAffective symptoms, 1–24 163 18 0.88 (0.36–2.15)Interpersonal difficulties, 0 275 30 1.00 referentInterpersonal difficulties, 1–15 75 9 1.08 (0.43–2.73)aTotal number of participants for each depression symptom varies due to item non-response.bOdds ratio comparing prevalence of hearing loss in those with versus those without depressive symptoms after adjustment for age, gender,martial status, employment status, and insurance status.cConfidence interval.dDIS, Diagnostic Interview Schedule; CES-D, Center for Epidemologic Studies Depression Scale.

Table 2Depression and overall bilateral hearing loss ( . 25 dBHL) among Mexican-American adults 20–74 years of age: data from the HispanicHealth and Nutrition Examination Survey, 1982–1984

a a b cDepression measure No. without hearing loss No. with hearing loss Adjusted odds ratio 95% CIdDIS: major



dCES-D: subscaleAffective symptoms, 0 624 32 1.00 referentAffective symptoms, 1–30 778 47 1.12 (0.42–2.99)Interpersonal difficulties, 0 1156 67 1.00 referentInterpersonal difficulties, 1–6 246 12 0.73 (0.31–1.71)aTotal number of participants for each depression symptom varies due to item non-response.bOdds ratio comparing prevalence of hearing loss in those with versus those without depressive symptoms after adjustment for age, gender,martial status, employment status, and insurance status.cConfidence interval.dDIS, Diagnostic Interview Schedule; CES-D, Center for Epidemologic Studies Depression Scale.

also significantly greater among Puerto Ricans re- 4. Discussionporting more affective symptoms and greater inter-personal difficulties compared to those reporting This is the first study to examine associationsfewer or no symptoms (OR 5 1.90; 95% CI 5 1.04– between depression and hearing loss in Hispanic3.46; OR 5 3.03; 95% CI 5 1.42–6.46). Americans. In addition, there are several strengths


Table 3Depression and overall bilateral hearing loss ( . 25 dBHL) among Puerto Rican adults 20–74 years of age: data from the Hispanic Healthand Nutrition Examination Survey, 1982–1984

a a b cDepression measure No. without hearing loss No. with hearing loss Adjusted odds raio 95% CIdDIS: major



dCES-D: subscaleAffective symptoms, 0 289 11 1.00 referentAffective symptoms, 1–30 273 15 1.90 (1.04–3.46)Interpersonal difficulties, 0 409 15 1.00 referentInterpersonal difficulties, 1–14 153 11 3.03 (1.42–6.46)aTotal number of participants for each depression symptom varies due to item non-response.bOdds ratio comparing prevalence of hearing loss in those with versus those without depressive symptoms after adjustment for age, gender,martial status, employment status, and insurance status.cConfidence interval.dDIS, Diagnostic Interview Schedule; CES-D, Center for Epidemologic Studies Depression Scale.

that this study has over previous research in non- postulate no reasonable mechanism to account for aHispanic populations. First, this is the largest study lower risk of hearing loss among Hispanics whoof those published which examines associations have experienced a major depressive episode at somebetween depression and hearing loss. Second, au- time during their lives. However, one possibility todiometric testing was completed under carefully account for the discrepancy between our findings andcontrolled and standardized conditions. This report is previous depression and hearing research is thatalso the first to test the hypothesis that odds of depressed cases in a case-control study are morehearing impairment is greater among those who have likely to be identified as hearing impaired because ofexperienced a major depressive episode sometime poor performance during audiometric testing. Suchduring their lives relative to those who have never testing requires both vigilance and cooperation whichexperienced a major depressive episode. Our results severely depressed patients may not be able tofailed to support this hypothesis. Rather, findings for provide. This hypothesis could be tested by utilizingthis study indicated that Hispanics who have ex- audiometric tests which do not require the vigilanceperienced a major depressive disorder sometime of the testee, such as brainstem evoked potentialduring their lifetime were less likely to be hearing audiometry (Markand, 1994) and otoacoustic emis-impaired. The pooled prevalence of hearing loss sions testing (Londsbury-Martin et al., 1995).among those with a history of major depressive Also, contrary to one of our hypotheses, we founddisorder across the three Hispanic groups was 1.6%, no consistent associations between depressive symp-while the pooled prevalence of hearing loss among tomatology and hearing loss among Cuban-Ameri-those with no history of a major depressive disorder cans and Mexican-Americans. However, odds ofwas 6.3%. hearing impairment was significantly greater among

In light of results from case-control studies which Puerto Ricans reporting greater symptomatologyhave documented increased odds of hearing impair- versus Puerto Ricans reporting lower levels ofment among non-Hispanics experiencing concurrent depressive symptomatology. The findings for bothepisodes of major depressive disorder (Watt, 1985; factor-derived CES-D subscales indicate that odds ofPrager and Jeste, 1993; Yovell et al., 1995), we can hearing impairment was significantly greater among


Puerto Ricans experiencing both greater depressed that associations between depressive symptom re-affect and greater interpersonal difficulties. However, porting and hearing loss in this Hispanic sub-groupthe odds ratio for the latter subscale was larger than was inflated due to differential non-response bias.for the former (i.e., OR 5 3.03 versus OR 5 1.90), One other important study limitation should alsosuggesting that odds of hearing loss is more strongly be noted. The ability to detect significant associa-associated with an increased risk of interpersonal tions in the present study was hindered by poordifficulties than with risk of depressed affect. statistical power. This low statistical power was due,

Puerto Ricans residing in New York have been in part, to the relatively small number of hearingshown in the HHANES (Guarnaccia et al., 1990) and impaired subjects in this study. The low number ofin several other studies to report higher levels of participants with moderate or greater hearing lossdepressive symptomatology than other Hispanic (i.e., . 40 dBHL) prevented us from testing ourgroups and non-Hispanics (Potter et al., 1995). In hypothesis that associations between depression andaddition, the rate of lifetime depressive disorder, as hearing loss would be stronger when using a higherassessed by the DIS, was highest among mainland cutpoint to define hearing impairment. Finally, thePuerto Ricans compared to rates for other Hispanic relatively small number of hearing impaired subjectssub-groups. These higher rates of depression and in our analyses led to unstable point estimates due todepressive symptomatology may be due, in part, to the absence of hearing loss cases among thosewillingness to report ‘socially undesirable’ symptoms relatively few Hispanics with major depressive disor-(Potter et al., 1995). Applied to the present findings, der.positive associations between CES-D assessed depre- As indicated earlier, this is the largest population-ssive symptoms and hearing loss may be significant based study which included clinical measures ofonly among Puerto Ricans because participants in hearing loss and depression. Given the issue of lowthis ethnic subgroup are more likely to ‘admit’ to statistical power in the present analysis, it may bedepressive symptoms than are respondents from the unreasonable to advocate replication of this studyother two Hispanic sub-groups. Associations be- using a similar design. Rather, the case-controltween hearing loss and depressive symptoms in these design may be more appropriate given that it is betterlatter two sub-groups could therefore be attenuated suited and more cost-effective for the examination ofby a response style which makes respondents less rare conditions. Such a case-control design is alsolikely to report ‘socially undesirable’ depressive ideally suited to include labor intensive hearingsymptoms. assessment methodologies such as brainstem evoked

Non-response bias to the physical examination potential audiometry and otoacoustic emissions test-where audiometric data was collected may have also ing. These methodologies, which are time consumingcontributed to the positive associations between and expensive to include in any study, would bedepressive symptoms and hearing loss in Puerto cost-prohibitive for population-based surveys similarRicans. Response rates to the examination portion of in design to the present study.the survey where audiometric data was collectedwere significantly higher in Puerto Rican adults, whoreported they had a history of trouble hearing (87%) 5. Conclusionversus those who denied such a history (79%)(NCHS, 1994). No significant differences in re- To summarize, this study found a non-significantsponse rates were noted for Cuban and Mexican- lower risk of bilateral hearing impairment among USAmericans. Unfortunately self-report data on depres- Hispanics with a history of major depressive disordersive symptoms was not collected during the inter- versus those with no such history. No significantview, so it is not possible to determine if depressed associations between depressive symptomatologyHispanics were more or less likely to report for the and hearing loss were noted among Cuban-Ameri-examination portion of the survey. Nevertheless, the cans and Mexican-Americans. However, odds offinding of greater response to the examination among hearing impairment was significantly greater amonghearing impaired Puerto Ricans raises the possibility Puerto Ricans reporting more depressive symptoms


analysis of otoacoustic emissions. Otolaryngol. Head Neckversus those reporting fewer or no depressive symp-Surg. 112 50–63.toms.

Markand, O.N., 1994 Brainstem auditory evoked potentials. J.Clin. Neurophysiol. 11 319–342.

K.P. Meadow-Orlans, Social and psychological effects of hearingAcknowledgments loss in adulthood: a literature review, in: H. Orlans (Ed.),

Adjustment to Adult Hearing Loss, College-Hill Press, Inc.,San Diego, CA, 1985.This work was supported by grant 1 R03 DC

Mulrow, C.D., Aguilar, C., Endicot, J.E. et al., 1990 Association02073-01 from the National Institue on Deafness and between hearing impairment and the quality of life of elderlyOther Communication Disorders and grant 1 R29 individuals. J. Am. Geratr. Soc. 38 45–50.AG124444-01A1 from the National Institute on National Center for Health Statistics, Public Use Data Tape

Documentation. Measures of Depression Ages 20–74 Years.Aging.Public Health Service. US Government Printing Office,Washington, 1987.

National Center for Health Statistics, Public Use Data TapeReferences Documentation. Hearing Ages 6 Months-74 Years. Public

Health Service. US Government Printing Office, Washington,1988.American National Standards Institute, American National Stan-

National Center for Health Statisics. M.L. Rowland, R.N. Forth-dard Specifications for Audiometers. American Standardsofer, Investigation of nonresponse bias: Hispanic Health andInstitute ANSI S3. 1970, 6–1969.Nutrition Examination Survey. DHHS Publ. No. (PHS) 94-American Speech-Language-Hearing Association 1981 On the1393, (Vital and Health Statistics, Series 2, no. 119), 1994.definition of hearing handicap. ASHA. 23 293–297.

Potter, L.B., Rogler, L.H., Moscicki, E.K., 1995 DepressionCarballese, C., Appollonio, I., Rozzini, R. et al., 1993 Sensoryamong Puerto Ricans in New York City: the Hispanic Healthimpairment and quality of life in a community elderly popula-and Nutrition Examination Survey. Soc. Psychiatry Psychiatr.tion. J. Am. Geratr. Soc. 41 401–407.Epidemiol. 30 185–193.Cooper, A.F., 1976 Deafness and psychiatric illness. Br. J.

Prager, S., Jeste, D.V., 1993 Sensory impairment in late-lifePsychiatry. 129 216–226.schizophrenia. Schizophr. Bull. 19 755–772.Delgado, J.L., Johnson, C.L., Roy, I., Trevino, F.M., 1990

Radloff, L., 1977 The CES-D scale: a self-report scale for researchHispanic Health and Nutrition Examination Survey: Methodo-in the general population. Appl. Psychol. Meas. 8 385–401.logical Considerations. Am. J. Public Health. 80 (Suppl) 6–10.

Research Triangle Institute, Software for Survey Data AnalysisJ.F. Gonzalez, T. Exxati, A.A. White et al., Sample design and(SUDAAN) Version 6.40. Research Triangle Park, Northestimation procedures. In: Plan and Operation of the HispanicCarolina, 1995.Health and Nutrition Examination Survey, 1982–84. National

Robbins, L.N., Helzer, J., Crougham, J. et al., 1981 The NationalCenter for Health Statistics; DHHS Pub No. (PHS) 85–1321;Institute of Mental Health Diagnostic Interview Schedule: its(Vital and Health Statistics Series 1, No. 19), 1985.history, characteristics and validity. Arch. Gen. Psychiatry. 38Guarnaccia, P.J., Angel, R., Worobey, J., 1989 The factor structure381–389.of the CES-D in the Hispanic Health and Nutrition Examina-

Roberts, R.E., 1980 Reliability of the CES-D scale in differenttion Survey: the influences of ethnicity, gender and language.ethnic contexts. Psychiatry Res. 2 125–134.Soc. Sci. Med. 29 85–94.

Task Force on Nomenclature and Statistics. Diagnostic andGuarnaccia, P.J., Good, B.J., Kleinman, A., 1990 A critical reviewStatistical Manual of Mental Disorders, 3rd ed. Americanof epidemiological studies of Puerto Rican mental health. Am.Psychiatric Association, Washington, DC, 1980.J. Psychiatry. 147 1449–1456.

Thomas, P.D., Hunt, W.C., Garry, P.J. et al., 1983 Hearing acuityHerbst, K., Humphrey, C., 1980 Hearing impairment and mentalin a healthy elderly population: effects on emotional, cognitive,state in the elderly living at home. Br. Med. J. 281 903–905.and social status. J. Gerontol. 38 321–325.Jones, E.M., White, A.J., 1990 Mental health and acquired hearing

Watt, J., 1985 Hearing and premorbid personality in paranoidimpairment: a review. Br. J. Audiol. 24 3–9.states. Am. J. Psychiatry. 142 1453–1455.Lee, D.J., Carlson, D.L., Lee, H.M., Ray, L.A., Markides, K.S.,

Weinstein, B.E., Ventry, I.M., 1982 Hearing impairment and social1991 Hearing loss and hearing aid use in Hispanic adults:isolation in the elderly. J. Speech Hear. Res. 25 593–599.results from the Hispanic Health and Nutrition Examination

Weissman, M., Sholomskas, D., Pottenger, M. et al., 1977Survey. Am. J. Public Health. 81 1471–1474.Assessing depressive symptoms in five psychiatric populations:Lee, D.J., Gomez-Marin, O., Lee, H.M., 1996 Sociodemographica validation study. Am. J. Epidemiol. 106 203–214.correlates of hearing loss and hearing aid use in Hispanic

Yovell, Y., Sackeim, H.A., Epstein, D.G. et al., 1995 Hearing lossadults. Epidemiology. 7 443–446.and asymmetry in major depression. J. Neuropsychiatr. Clin.Londsbury-Martin, B.L., Martin, G.K., McCoy, M.J., 1995 NewNeurosci. 7 82–89.approaches to the evaluation of the auditory system and current

Major depressive disorder, depressive symptoms, and bilateral hearing loss in hispanic adults

Documents

Transcript of Major depressive disorder, depressive symptoms, and bilateral hearing loss in hispanic adults