Nature, Causes, and Epidemiology of Hearing Loss
In this chapter, the term hearing loss, used by itself, denotes any or all levels of severity of hearing difficulty. These levels of hearing impairment comprise mild (26-40 decibel hearing level, dBHL), moderate (41-60 dBHL), severe (61-80 dBHL), and profound (81 dBHL or greater). The term deafness denotes profound hearing impairment (WHO 1991, 1997). Disabling hearing impairment in adults is defined as "a permanent unaided hearing threshold level for the better ear of 41 dB or greater; for this purpose, the hearing threshold level is to be taken as the better ear average hearing threshold level for the four frequencies 0.5, 1, 2, and 4 kHz." Disabling hearing impairment in children under the age of 15 years is defined as a permanent, unaided hearing threshold level for the better ear of 31 dB or greater; for this purpose, the hearing threshold level is to be taken as the better ear average hearing threshold level for the four frequencies 0.5, 1, 2, and 4 kHz.
Mathers and others (2003) estimate that in 2002, 255 million people worldwide had disabling hearing loss (moderate or worse hearing loss in the better ear). Those 192 million people with adult-onset loss (age 20 years and above) and 63 million people with childhood-onset loss make up almost 4.1 percent of the world's population and just over 40 percent of all people globally with hearing loss of any severity. The prevalence rates of adult-onset hearing loss were estimated by subtracting the prevalence rate for childhood onset (estimated in terms of prevalence in ages around 15 to 19). Numbers with childhood-onset hearing loss by cause have so far not been estimated separately but are included among sequelae of other diseases (for example, infectious diseases such as meningitis, otitis media, congenital conditions). It has been estimated that at least 50 percent of the burden of hearing loss could be prevented by primary, secondary, and tertiary preventive measures (Brobby 1989; WHO 1991).
Causes and Characteristics
Hearing loss is grouped according to International Classification of Diseases and Related Health Problems, 10th revision, version for 2003 (ICD-10) into conductive and sensorineural loss and other hearing loss, ICD-10 codes 90-91 (WHO 2003). The main causes are shown in table 50.6 according to the proportion that these contribute to the total burden (WHO 1986). Chronic otitis media (COM, as in ICD-10 codes H65-H67) includes chronic suppurative otitis media and otitis media with effusion. These forms of otitis media, together with some other middle ear diseases, such as perforation of the tympanic membrane, cholesteatoma, and otosclerosis, are the major causes of conductive hearing loss. In most WHO estimates of the burden of otitis media, the data are not disaggregated into acute and chronic otitis media.
[Table .]
Hearing loss is a chronic and often lifelong disability that, depending on the severity and frequencies affected, can cause profound damage to the development of speech, language, and cognitive skills in children, especially if commencing prelingually. That damage, in turn, affects the child's progress in school and, later, his or her ability to obtain, keep, and perform an occupation. For all ages and for both sexes, it causes difficulties with interpersonal communication and leads to significant individual social problems, especially isolation and stigmatization. All these difficulties are much magnified in developing countries, where there are generally limited services, few trained staff members, and little awareness about how to deal with these difficulties.
In addition to its individual effects, hearing loss substantially affects social and economic development in communities and countries. Ruben (2000), taking into account rehabilitation, special education, and loss of employment, estimated the cost to the U.S. economy in 1999 of communication disorders (hearing, voice, speech, and language disorders) at between US$176 billion and US$212 billion (2004 dollars; 2.5-3 percent of the gross national product of the United States in that year). Hearing loss accounted for about one-third of the prevalence of these communication disorders.
Risk Factors
Occupations exposed to high levels of noise or ototoxic chemicals are also at risk, and noise exposure potentiates chemical ototoxicity in some cases (Fechter 1995; Morata 1998). Certain lifestyles (for example, use of personal stereos, noisy toys, firecrackers) and hobbies (for example, hunting) are also linked to levels of noise exposure that can cause hearing loss (Berglund and others 2000; Goelzer, Hansen, and Sehrndt 2001). Smoking may be a risk factor for high-frequency hearing loss, adding to the effect of noise (Mizoue, Miyamoto, and Shimizu 2003). Other risk factors include poverty, poor access to health care, poor hygiene, and overcrowding, all of which can lead to upper respiratory tract infections, otitis media, and other infections that may cause hearing loss, such as measles and meningitis. Detailed risk factors and indicators have been developed for neonates and infants (Joint Committee on Infant Hearing 2000); these include conditions that should require admission to a neonatal intensive care unit, stigmata of syndromes causing hearing loss, positive family history, craniofacial anomalies, certain in utero and post-natal infections (cytomegalovirus, herpes, rubella, syphilis, toxoplasmosis, meningitis), hyperbilirubinemia, conditions requiring prolonged mechanical ventilation or oxygenation, persistent otitis media with effusion, and others.
Ototoxic medications, low birth-weight, and low Apgar scores have also been cited as risk factors for neonates (Vohr and others 2000). Offspring of consanguineous marriages have a significantly higher incidence of autosomal recessive diseases, including hearing impairment. Such diseases are an important cause in communities where consanguinity is common (Shahin and others 2002; Zakzouk 2000). Certain ethnic groups (First Nations peoples such as Inuit and North American Indians, as well as Australian Aboriginal people) appear to be at higher risk of developing COM (WHO 1998).
Age, Geographic, and Gender Burdens
The prevalence of disabling hearing impairment that increases markedly with age is mainly related to the effect of presbycusis. The current shortage of data, particularly in developing countries, prevents accurate assessment of the global distribution of the burden and causes.
Male-to-female ratios of age-standardized adult-onset prevalence rates were found to be greater than 1 in most studies in all WHO regions (Mathers, Smith, and Concha 2005). This finding may be related to occupational noise-induced hearing loss, which differentially affects men.
Mortality
Barnett and Franks (1999) have found evidence that adults with postlingual onset of deafness have higher mortality than nondeaf adults. A 10-year longitudinal analysis of participants (age 55 to 74 years) in the U.S. National Health and Nutrition Examination Survey I found that, at baseline, hearing loss predicts mortality; relative risk = 1.17 (Mui and others 1998). Other studies have reported that the association disappears after controlling for age, and in any case, any relationship that may exist is too small to appear in published WHO estimates of deaths by cause (WHO 2004a, annex table 2) and by years of life lost, or YLLs (Mathers, Smith, and Concha 2005), in any region. A small number of deaths (4,000 globally in 2002) are recorded for otitis media (WHO 2004a), but these deaths are mainly due to infective complications and, hence, are not directly caused by hearing loss.
Years Lived with Disability and DALYs
Data on years of life lived with disability (YLDs) and DALYs are available only for adult-onset hearing loss. The disease model used, the assumptions and methods used for calculation, and the disability weights are described elsewhere (Mathers, Smith, and Concha 2005). Total global YLDs for adult-onset hearing loss in 2001 are estimated to be 25.87 million, or 4.7 percent of total YLDs attributable to all causes, which makes hearing loss a leading cause of YLDs. Because YLLs are taken to be zero for all regions, the DALY figures are identical to the YLD figures. The most comprehensive data available are for all adult-onset hearing loss (WHO 2004b; Mathers, Smith, and Concha 2003). Fewer data on the burden are available at present for childhood hearing loss and specific causes.
Interventions
Effective interventions include screening programs, education, surgery, medications, and assistive devices.
Population-based Interventions
Neonatal or early infant hearing screening is important because early identification of hearing loss (before 6 months of age, with early intervention) is associated with significantly better language development and may lead to better school and occupational performance than that of children identified after 6 months with early intervention (Keren and others 2002; Yoshinaga-Itano and others 1998). Implementation of neonatal hearing screening raises from 20 to 80 percent the numbers of children with normal development of language, compared with children whose hearing loss is detected later (Yoshinaga-Itano and Gravel 2001). Early identification of hearing impairment can reduce the median age of identification of hearing impairment from between 12 and 18 months to 6 months or less. Universal neonatal hearing screening is highly sensitive, but depending on the test method used, it may result in many false positives (which may increase parental anxiety and lead to unnecessary follow-up tests and interventions). It has a low positive predictive value. Some screening protocols may decrease false-positive rates (Kennedy and others 2000). Universal neonatal hearing screening has been endorsed in developed countries (Joint Committee on Infant Hearing 2000), although some experts urge caution (Paradise 1999); however, it is expensive to implement and, for most developing countries, is not yet an option. Hearing screening targeted at high-risk neonates is generally used in developing countries that do any type of neonatal screening, but screening may fail to detect 50 percent or more of cases of impairment (Lutman and Grandori 1999). Neonatal screening programs will not detect the 10 to 20 percent of cases of permanent childhood hearing impairment that starts later in life and for which later surveillance is needed (Grote 2000). No publications were found that have addressed the DALY burden that might be avoided by implementing neonatal hearing screening.
A recent WHO meeting of experts on noise-induced hearing loss (WHO 1998) concluded that exposure to excessive noise is the major avoidable cause of permanent hearing impairment worldwide. They agreed that, in developing countries, occupational noise and urban environmental noise are increasing risk factors for hearing impairment. Experts attending the meeting recommended that all countries implement national programs for prevention of noise-induced hearing loss, including effective hearing conservation. However, there are no published reports yet on the effectiveness of such programs in developing countries. The United States has produced a guide to hearing conservation programs in the workplace (Franks, Stephenson, and Merry 1996). It advises how to appraise programs by assessing the completeness of their components and by evaluating both the individual audiometric data for threshold shift and the group data for variability compared with a nonexposed population. Even in developed countries, there have been few, if any, clinical trials and little convincing evidence of the efficacy of occupational hearing conservation programs (Dobie 1995).
Personal Services
Chronic suppurative otitis media is one of the most common causes of hearing impairment in developing countries. Opportunities for prevention arise at all levels of national health systems, particularly in the community and at the primary level through primary ear and hearing care (PEHC) (WHO 1998). Appropriate health promotion measures include breastfeeding, immunization, adequate nutrition, personal hygiene, improved housing, reduced overcrowding, and adequate access to clean water. Primary health care workers can be given appropriate training and basic equipment for early detection and management of chronic suppurative otitis media, but the effectiveness and cost-effectiveness of this intervention in developing countries has not yet been assessed.
Although WHO does not currently recommend treating what is commonly called chronic middle ear infection with antibiotics at the primary level (WHO 2000), evidence suggests that antibiotics, especially topical quinolones, are more effective and cost-effective than ear toilet alone (Acuin, Smith, and Mackenzie 2000). WHO is reviewing these recommendations (WHO 2004b). New methods of delivery of effective but expensive topical antibiotics may lower the cost in poor communities, but treatment failure may be due to a high reinfection rate attributable to poor hygienic conditions. To be effective as public health measures, interventions need to be implemented on a large scale, with good coverage of the targeted population (van Hasselt and van Kregten 2002). Ear surgery plays an essential part in the prevention of further hearing impairment and, sometimes, in the improvement of hearing.
Services at the secondary level of intervention include provision of hearing aids in developing countries, which should assign priority to children with moderate or severe hearing loss, followed by adults (Arslan and Genovese 1996; WHO 2004c). However, even though globally about 6 million hearing aids are dispensed annually (WHO 1999), there have been no published randomized, controlled trials of the effectiveness of hearing aids in reducing hearing disability in developing countries and few trials in developed countries.
A randomized trial of amplification in 194 U.S. veterans showed significant improvements in communication, cognition, and social and emotional function, plus significant alleviation of depression, with hearing aids compared with controls (Mulrow and others 1990). No significant differences were observed in clinical effectiveness and cost-effectiveness between newer hearing aids that use digital signal processing and those that do not—in particular, analog-based aids (Parving 2003; Taylor, Paisley, and Davis 2001). Digital signal processing aids are not affordable for most people in developing countries. Over-the-counter hearing aids that can be purchased and used without prior training are commonly available in some developing countries. Those aids were found not to meet the prescription gain requirements of the majority of elderly clients who usually purchased them (Cheng and McPherson 2000).
Learning to use a hearing aid and developing "hearing tactics"are also important. Random assignment to a course for new hearing aid users significantly reduced the handicap compared with controls not assigned (Beynon, Thornton, and Poole 1997). Lack of compliance in use is a substantial problem everywhere among elderly and child users, including in developing countries (Furuta and Yoshino 1998; Sorri, Luotonen, and Laitakari 1984). Thus, measuring coverage without taking into account actual usage is not enough to assess alleviation of the burden.
Cochlear implants are provided to children and adults with severe and profound bilateral deafness on the basis that known short-term outcomes in auditory receptive skills (Richter and others 2002) will translate through various medium-term outcomes into greater social independence and quality of life (the social and quality outcomes have not yet been tested in a trial or observational study) (Summerfield and Marshall 1999). Cochlear implantation is beneficial in prelingually and postlingually deaf children (Makhdoum, Snik, and van den Broek 1997) and, when accompanied by aural (re)habilitation, leads to higher rates of mainstream placement in schools and lower dependence on special education support services (Francis and others 1999). Multichannel implants are superior to single-channel implants (Cohen, Waltzman, and Fisher 1993) and are more beneficial when implanted in young children (Richter and others 2002). There has been no economic analysis of cochlear implants in developing countries, and such interventions are currently not a priority in most parts of the developing world (Berruecos 2000; WHO 2004c; Zeng 1995).
Intervention Cost and Cost-Effectiveness
All the data on the costs and cost-effectiveness of interventions related to hearing loss (including school-age screening, treatment of COM, surgical interventions, hearing aids, and cochlear implants) come from developed countries. Although they can be summarized quite readily, it is not clear whether and how they relate to the costs that would be experienced in developing countries.
