| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Netherlands Institute of Mental Health and Addiction, Utrecht, The Netherlands.
Address reprint requests to: Ron de Graaf, Monitoring & Epidemiology Department, Psychiatric Epidemiology Unit, Netherlands Institute of Mental Health and Addiction, Da Costakade 45, 3521 VS Utrecht, The Netherlands. Email: rgraaf{at}trimbos.nl
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: Five hundred twenty-three face-to-face interviews were carried out by persons with practical skills in communication with hearing-impaired people.
RESULTS: Of prelingually and postlingually deaf men, 27.1% and 27.7%, respectively, reported mental distress (scores on the General Health Questionnaire 
2), and among women these figures were 32.4% and 43.2%. These rates are higher than in the general population (men: 22.0%; women: 26.6%). Among the prelingual category, none of the demographic or hearing loss-related characteristics was associated with mental health status as measured by the General Health Questionnaire (GHQ). Of these variables, only additional impairment or serious illness was associated with the brief Symptom Checklist (SCL-8D). Among the postlingual category, female gender and equilibrium disturbance was associated with both the GHQ and SCL-8D, and additional impairment or serious illness with the SCL-8D. For both categories, the risk of mental distress also was higher in those with more communication problems, lower levels of self-esteem, and poorer acceptance of the hearing loss. Opportunities for identification in youth and social support were not associated with mental health.
CONCLUSIONS: Mental health status differs between the hearing-impaired and the general population, but not as much as is sometimes suggested. Mental distress is greater in those in certain categories of the hearing-impaired.
Key Words: mental health, • GHQ, • SCL-8D, • auditive handicap, • deaf people.
Abbreviations: GHQ = General Health Questionnaire;; SCL = Symptom Checklist;; OR = odds ratio.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
In the past, mental health care provisions were accessible inadequately to the deaf, because workers were unaware of the pervasive effects of deafness, unfamiliar with deaf culture and incapable of communicating with deaf people (7–9). This may have also contributed to a poorer state of mental health among people who are deaf. In the Netherlands, the deaf were dependent on social work services organized by deaf institutes for their pupils and ex-pupils. Thus, people who were deaf had to go "back to school" with their problems. Much has changed in the past 10 years. Social work agencies for the deaf have sprung up outside the deaf institutes, "deaf teams" set up by the Regional Institutes for Ambulatory Mental Health Care (RIAGGs) have opened up mental health care to deaf people, and psychiatric facilities specifically for the deaf were started (10). Initiatives to improve mental health care for the deaf have been undertaken in many other countries as well (11, 12).
Little research has been done on factors influencing mental health among people with hearing impairments and these factors differ between different categories. In this article, we examine demographic factors, characteristics of the auditory impairment (like severity and rapidity of the hearing loss, and accompanying disorders), and social and individual aspects (like communication problems, identification opportunities, integration with hearing-impaired, social support, self-esteem, and acceptance of hearing loss) that might be associated with the mental health status of deaf and severe hard-of-hearing adults. Analyses were performed separately for those with an age of onset before and after their third birthday.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
We chose to conduct face-to-face interviews, because we suspected (corresponding to Schroedel (13)) that a written questionnaire would elicit a poor response. Respondents could choose between being interviewed in sign language or spoken language (supplemented by gestures if necessary). To limit the duration of the interviews (averaging 91 minutes, SD = 31), we sent the more factual questions to the respondents to be completed beforehand. During the interviews, respondents were given the written questionnaire as an aid to comprehension. Virtually all (96%) respondents were interviewed in their homes. The fieldwork took place from June to October 1997.
Recruitment of Respondents
The following recruitment methods were applied (with percentages contacted through each method): advertisements in periodicals targeting the deaf and hard of hearing (18%); appeals directed to members of work groups, such as one involving people who are deaf who were reared orally (10%); appeals on the deaf pages of teletext (0.2%); personal recruitment by interviewers in a wider circle than their circle of acquaintances and in clubs for deaf and hard-of-hearing people (64%, of whom 5% were located through the text telephone directory); and snowball technique (by word of mouth or written) (8%). The number of people that lose their hearing at a later age is relatively small compared with the number that were born deaf or became deaf at an early age (1). Special care was taken to reach the former group in the recruitment. The research was introduced as a study on the general health and well being of deaf and severely hard-of-hearing people. Terms such as "psychological problems" were avoided, and all people considered severely hearing impaired were invited to take part. Exact nonresponse figures could not be obtained because of the way of recruitment, but the interviewers estimated the number of people who were personally approached but refused participation at 20% to 30%.
Questionnaire Design and Language Usage
Many prelingually deaf people have a comparatively limited comprehension of words and sentences. To minimize strain, we formulated brief, concrete questions and used concise measuring instruments. To avoid overly complicated or abstract language, we consulted a panel of deaf adults while developing the questionnaire. We also consulted members of the study’s advisory committee, who were professionals who had many contacts with deaf people, including one who was deaf. Examples of changes made as a result of their comments were a simpler language and the avoidance of statements phrased in the first person, which cause confusion in sign language. Only three respondents found the language level of the questionnaire too daunting. Their data have been omitted from the analyses. At the end of the interview, interviewers estimated the frequency with which the respondent asked for clarification on a five-point scale. In 49% of the cases, this was seldom, 29% occasionally, 14% regularly, 5% often, and 3% very often.
With the aid of a sign-language interpreter and a deaf mental health care professional, the questionnaire was also recorded in sign language on video (this may be obtained from the first author). This tape was used in instructing the interviewers to ensure uniform administration. It was not used in the interviews themselves, because a previous Dutch study had shown that respondents tend to request frequent clarification even if they were interviewed with the help of a questionnaire on video (14).
Selection and Training of Interviewers
Our decision to conduct face-to-face interviews required certain communication skills of the interviewers, and we, therefore, enlisted deaf and severely hard-of-hearing people for this task. Another key consideration in doing so was the role they could play in recruiting respondents. Sixteen interviewers were recruited through extensive contacts in the field, including one interviewer who was not hearing impaired. Selection criteria included practical skills in communicating with the deaf or hard of hearing and familiarity with the deaf community. Interviewing experience was not required, but we provided extensive interview training and intensive supervision during the field work period. Interviewers were told that the study took place among the general deaf population and that interviewers should not select individuals with mental health problems.
One drawback to using these interviewers is the possibility that they might know some of the respondents. Whenever that happened, another interviewer took over the interview. All respondents were informed, so that information would be kept confidential.
Variables Assessed
The questionnaire predominantly was made up of structured questions with standard-answer categories to be filled in by the interviewer. The General Health Questionnaire (GHQ) is a well-known instrument in epidemiological studies of mental health. (15) Because validated instruments for mental health status among people who are hearing-impaired do not exist, we used a second instrument to measure mental health distress.
The dependent variables were 1) mental health status. It was assessed by means of the GHQ. Given the nature of the research population, the 12-item brief version was used (Cronbach 
= 0.88). The scoring method recommended by Goldberg (15) was applied, whereby the weights 0, 0, 1, 1 are assigned to the four successive answer categories. The possible range of scores thus becomes 0 to 12, with a higher score indicating more mental health complaints. 2) The most common mental disorders are depression and anxiety. The brief eight-item Symptom Checklist-8D (SCL-8D) (16) was used (1 = not at all; 5 = extremely) (Cronbach = 0.89). It assesses the degree of depression and anxiety together in one questionnaire, and the possible range of scores is 8 to 40.
We choose these two instruments because it is hardly possible among this population to use more sophisticated instruments which measure mental disorders, like the Composite International Diagnostic Interview (CIDI), because many people who are deaf are limited in their understanding of words and sentences.
The most important independent variables were:
= 0.87) (1 = almost never; 4 = very often). This brief version, originally designed for the elderly, was used because it was better suited to our population than the longer list. = 0.83). was low at 0.47.
Statistical Analysis
The dependent variables were dichotomized. Bivariate associations (odds ratios and 95% confidence intervals) between the categories of the predictor variables and the dichotomized dependent variables measuring mental health were calculated, using the percentage of respondents scoring above the threshold value for mental health as the dependent variable. To test for linear trends, potential determinants were modeled as continuous variables. Then, multiple logistic regression analyses (method enter) were performed.
Because many interaction terms between age of onset (before vs after age 3) and the other independent variables were significant, all analyses were carried out separately for those with an age of onset before and after their third birthday (hereafter called prelingually and postlingually, respectively). In 4 of the 523 respondents, the age of onset was unknown; they were omitted from the analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
|
Prelingually deaf subjects rated their communication skills in speechreading and sign language higher than the postlingual category. Therefore, it is not surprising that the late-onset group reported communication problems more often. Prelingually deaf respondents rated their communication skills in speechreading higher than those in sign language. One reason for this may be our recruiting effort in a work group of people who are deaf and were reared orally. Thirty-0ne percent of the prelingual and 38% of the postlingual category reported hearing-impaired family members, and 29% and 9%, respectively, had had other opportunities to identify with other deaf adults before age 16. The degree of integration with other hearing-impaired people was reflected in four items: much more prelingually than postlingually deaf subjects had a hearing-impaired partner; they also more often had equal or greater numbers of deaf or hard-of-hearing than hearing friends; they less often belonged to an organization for the deaf or hard of hearing; but they more often visited gatherings of hearing-impaired people. Acceptance of hearing loss was assessed with three items: more postlingually than prelingually deaf subjects felt handicapped by their hearing loss; they also more often reported problems with their hearing impairment in daily life; but no differences were found in acceptance in the hearing world.
Mental Health
There is no generally accepted threshold value for the GHQ-12. We have based our assessments on the advised (20) and commonly used value of 2. One third of the respondents (33.5%) had total scores of 2 or higher: 30.2% of the prelingual and 38.4% of the postlingual category (
2 = 3.8; df = 1; p = .05). The mean score was 1.5 (SD = 2.4) and 2.4 (SD = 3.4), respectively.
Because the use of the SCL-8D is relatively new, we set a threshold value using the GHQ score as a reference point (threshold value of 2). This value was chosen so that the difference in percentages between false-positive and false-negative respondents was as small as possible. This was the case for an SCL-8D value of 13 (sensitivity = 0.75; specificity = 0.82). Of the respondents (36.4%) had total scores of 13 or higher: 29.2% of the prelingual and 46.9% of the postlingual category (2 = 16.9; df = 1; p < .0001). The mean score was 12.0 (SD = 4.6) and 13.7 (SD = 5.9), respectively. Above mentioned significant results continued to exist, when controlled for gender and age. Thus, those developing their hearing impairment after age three were more likely to report mental distress than those who developed it earlier. The correlation between the GHQ and SCL-8D total scores was high (0.73). Hereafter, scores above the threshold value of the GHQ and SCL-8D are defined as "mental distress."
Risk Factors for Mental Distress
We first studied bivariately) which variables had an apparent influence on mental health (Tables 3 and 4). None of the demographic characteristics were associated with the GHQ and SCL-8D score among the prelingual category. Among the postlingual category, gender was associated with both dependent variables, with women more likely than men to report mental distress. Those with a mean education less often scored above the threshold value of both dependent variables in comparison with the lowest educated group.
|
|
Several social and individual aspects were associated with mental health as measured by both dependent variables. Respondents with more communication problems, lower self-esteem, and less acceptance of their hearing loss were more likely to report mental distress, among both categories. No bivariate significant trends were found between mental health and the respondents’ identification opportunities, their social support or their integration with other hearing-impaired people. Interestingly, prelingually and postlingually deaf subjects with moderate social support reported mental distress measured by SCL-8D less often than those with low social support, but also less often than those with high support.
To identify the unique contributions of the independent variables to mental health, we performed multiple logistic regression analyses (Tables 3 and 4). Among the prelingual category, of the variables bivariately associated with the GHQ and SCL-8D score only self-esteem retained this association in the multivariate analysis. Gender now showed a significant association with the SCL-8D as well. Among the postlingual category, of the variables bivariately associated with the GHQ score, self-esteem and acceptance of the hearing loss retained this association in the multivariate analysis. Educational level and degree of integration with hearing-impaired people now showed a significant trend as well: those with a lower educational level were more at risk, and a high degree of integration with other hearing-impaired people increased the probability of belonging to the group with a poorer mental health status. Among the postlingual category, a number of variables bivariately associated with the SCL-8D score retained their association in the multivariate analysis: gender, presence of an additional impairment or a serious illness, communication problems, and self-esteem. For social support, again we found that those with moderate support reported mental distress measured by SCL-8D less often than those with low support, but also less often than those with high support.
Analyses by Subgroups
The above analyses included all respondents in the sample. We also performed analyses on certain subgroups. As we noted, mental health was not significantly associated with the subjective severity of the hearing loss. Among the 62% of respondents who knew the objective severity of their hearing loss, we also tested whether this factor might be related to mental health. There was no significant difference between those whose hearing loss was less than 90 dB or 90% and those with a greater loss in the percentage that scored above the GHQ and SCL-8D threshold value, among both the prelingual and postlingual category.
Speechreading skills and sign language mastery were examined for the subsample who were profoundly deaf. Such skills are less relevant for the hard of hearing, because they can still benefit considerably from their residual hearing. Respondents who scored 0 or 1 on Schein’s Hearing Scale (N = 411) were selected. Among the prelingual category, no significant differences in either mental health measure were apparent between those reporting lower or higher competence at speechreading. Among postlingually deaf subjects, however, those with a lower competence at speechreading more often reported mental distress than those with a higher competence (56% vs 38%; 2 = 5.0; df = 1; p = .03). In both categories no differences were found between those with a mediocre or an effective mastery of sign language.
The impact of sudden vs. gradual hearing loss was examined for those who were not congenitally deaf (N = 283). On mental distress measured by GHQ, no significant differences were found between those whose impairment was sudden and those with gradual hearing loss. However, on the SCL-8D, significant differences did emerge (38% vs 52%; 2 = 5.0; df = 1; p = .03). If the hearing loss was a gradual process, the chance was higher that one belonged to the group that experienced more depression and anxiety.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
It is not known how representative our respondents were of the entire population of severe hearing-impaired people. Although we applied a range of recruitment methods, we cannot rule out that they reached more people with higher education levels. People who do not read much will not have seen our advertisements. It also seems plausible that those with more education may more readily respond to calls for participation. It is also possible that the interviewers were a higher educated sample than the entire population of hearing-impaired people, and that their subsequent circle of acquaintances were as well. The proportion of respondents with higher educational levels was, therefore, high (25%), although still lower than the rate of 32% in the general population (21). We might also point out that the percentage of people of non-Dutch origin, a steadily growing part of the country’s deaf population (1), was extremely low. The proportion of female respondents (63%) was greater than that in the population. In view of the higher representation of better-educated respondents and the low representation of ethnic minorities, the mental health in the entire population of deaf and severe hard-of-hearing people may be worse than our study suggests. The immense difficulty of obtaining representative samples of deaf people for epidemiological research has been noted previously (22).
A greater percentage of hearing-impaired people were found to be in a poor state of mental health than the figures found in general population studies (23), although the figures are not as high as those suggested in the introduction. A prospective psychiatric epidemiologic study of 7076 members of the Dutch population (23) has found 22.0% of the men and 26.6% of the women to have GHQ scores of 2 or higher. In our study, 27.1% and 27.7% of prelingually and postlingually deaf men and 32.4% and 43.2% of prelingually and postlingually deaf women scored 2 or higher. It should be noted that the present study contained a relatively large share of respondents with additional impairments. Such conditions increase the chances of mental distress. The same may be said for several other factors such as having no paid job (50% of the sample vs 37% of the general population) and living alone (29% vs 18%) (20, 24). Because GHQ and SCL-8D do not measure former mental distress, we do not know how much the lifetime mental health status of the hearing-impaired differs from the general population.
Other researchers have also concluded that mental distress is more common in the deaf and severe hard of hearing than in the general population. Fellinger (25) studied 310 clients of a clinic for medical and psychiatric care and social work which targets the deaf and hard of hearing. Stress was reported by 80% of the women and 64% of the men, compared with 53% and 35% in a control group of 273 hearing clients seeing a doctor for a preventative medical check-up.
A number of the determinants we identified for mental distress or for feelings of depression and anxiety were no different from those among the general population. Both in our sample and in the general population (23, 24), mental distress is more prevalent in women than in men. The degree of objective and subjective hearing loss were not associated in our study with the GHQ and SCL-8D score, although a nearly significant trend was seen for a higher risk for mental distress (GHQ) with lower subjective degree of hearing loss among postlingually deaf subjects. Objective and subjective degree of hearing loss were not strongly intercorrelated. Other research similarly indicates that people with identical degrees of hearing loss can differ in their sense of being impaired by it (26).
Although one might hypothesize that postlingually deaf people would get on better in the hearing world, having been through (several) stages of language acquisition and emotional development (14), our study found them more likely to report mental distress than were the prelingually deaf, but among women only. Luey et al. (6) suggested that postlingually deaf people tend to miss spoken communication more than prelingually deaf people, and that they are more inclined to see deafness as a handicap and a loss. Our study also showed a distinct difference between the prelingually and postlingually deaf in their acceptance of their impairment. We found also that people who had gradually lost their hearing more often experienced depression and anxiety than those who had lost it more rapidly within a 2-year period.
We found that communication problems also give rise to mental distress, both among the prelingual and postlingual categories, although lower competence at speechreading was associated with mental distress among the postlingual category only. People who experience communication problems may avoid new contacts, and in the long-term this may increase their social isolation and reduce their quality of life (27, 28). Low self-esteem and acceptance of one’s hearing loss have emerged in our study as important predictors of mental distress, whereas low current social support and opportunities in youth to identify with deaf adults were far less salient predictors. Furthermore, we found in the multivariate, but not in the bivariate, analysis that a higher degree of integration with other hearing-impaired people was associated with higher GHQ scores among the postlingual category. Perhaps having problems, especially among postlingual deaf people, leads one to seek support from those in a similar position.
What significance do our findings have for prevention work and care provision aimed at the deaf and hard of hearing? Some predictive variables, such as the characteristics of the hearing impairment, are difficult or impossible to alter. What professionals can do, however, is to be aware that certain factors, among them female gender, postlingual and gradual hearing loss, and concomitant disorders, can constitute added risks for mental distress. Improving the communication skills to be applied in a variety of situations could well help to avert mental problems. Helping deaf and severely hard-of-hearing people to accept the impairment and to enhance self-esteem could likewise serve to reduce their mental distress.
In cross-sectional studies, the causality of associations is often not clear. In our study, for example, low self-esteem and acceptance of hearing loss can be determinants of mental distress, but they might also be the result of a negative mental health status. Thus, the results found here should be interpreted with caution. Prospective studies among the severely hearing-impaired will give more insight into factors influencing mental health status. In future research, it is also of importance to study the validity and reliability of instruments measuring mental health status among the hearing impaired.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication August 15, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. H. Kvam, M. Loeb, and K. Tambs Mental Health in Deaf Adults: Symptoms of Anxiety and Depression Among Hearing and Deaf Individuals J. Deaf Stud. Deaf Educ., January 1, 2007; 12(1): 1 - 7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Tambs Moderate Effects of Hearing Loss on Mental Health and Subjective Well-Being: Results From the Nord-Trondelag Hearing Loss Study Psychosom Med, September 1, 2004; 66(5): 776 - 782. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
