Lucien J. C. Anteunis

Specialised treatment based on cognitive behaviour therapy versus usual care for tinnitus: a randomised controlled trial

BACKGROUND Up to 21% of adults will develop tinnitus, which is one of the most distressing and debilitating audiological problems. The absence of medical cures and standardised practice can lead to costly and prolonged treatment. We aimed to assess effectiveness of a stepped-care approach, based on cognitive behaviour therapy, compared with usual care in patients with varying tinnitus severity. METHODS In this randomised controlled trial, undertaken at the Adelante Department of Audiology and Communication (Hoensbroek, Netherlands), we enrolled previously untreated Dutch speakers (aged >18 years) who had a primary complaint of tinnitus but no health issues precluding participation. An independent research assistant randomly allocated patients by use of a computer-generated allocation sequence in a 1:1 ratio, stratified by tinnitus severity and hearing ability, in block sizes of four to receive specialised care of cognitive behaviour therapy with sound-focused tinnitus retraining therapy or usual care. Patients and assessors were masked to treatment assignment. Primary outcomes were health-related quality of life (assessed by the health utilities index score), tinnitus severity (tinnitus questionnaire score), and tinnitus impairment (tinnitus handicap inventory score), which were assessed before treatment and at 3 months, 8 months, and 12 months after randomisation. We used multilevel mixed regression analyses to assess outcomes in the intention-to-treat population. This study is registered with ClinicalTrials.gov, number NCT00733044. FINDINGS Between September, 2007 and January, 2011, we enrolled and treated 492 (66%) of 741 screened patients. Compared with 247 patients assigned to usual care, 245 patients assigned to specialised care improved in health-related quality of life during a period of 12 months (between-group difference 0·059, 95% CI 0·025 to 0·094; effect size of Cohens d=0·24; p=0·0009), and had decreased tinnitus severity (-8·062, -10·829 to -5·295; d=0·43; p<0·0001) and tinnitus impairment (-7·506, -10·661 to -4·352; d=0·45; p<0·0001). Treatment seemed effective irrespective of initial tinnitus severity, and we noted no adverse events in this trial. INTERPRETATION Specialised treatment of tinnitus based on cognitive behaviour therapy could be suitable for widespread implementation for patients with tinnitus of varying severity. FUNDING Netherlands Organisation for Health Research and Development (ZonMW).

A prospective longitudinal study on radiation-induced hearing loss

Recent clinical reports indicate that patients receiving radiotherapy that includes the auditory system in the treatment volume are likely to develop an irradiation-induced hearing loss. Although the early, presumed reversible, conductive hearing impairment due to secretory otitis media following radiotherapy is a sequela well known by radiation oncologists, permanent hearing loss, both conductive and sensorineural, is believed to be rare. A prospective study was performed enrolling patients receiving postoperative radiotherapy for unilateral parotid tumors. Audiometric results prior to irradiation were compared with those obtained 2 years later. Up to 50% of the patients (9 of 18) developed a clinically relevant hearing loss in the irradiated ear, both conductive and/or sensorineural. The contralateral ear remained unaltered. The hearing loss was permanent in 6 patients (33%) and affected their quality of life. The data suggest that changes occur in the inner ear as well as in the auditory nerve and auditory brain stem with conventional irradiation schemes with daily fractions of 2 to 2.5 Gy with a total dose of 50 Gy.

Mild Hearing Impairment Can Reduce Verbal Memory Performance in a Healthy Adult Population

We studied to what extent immediate and delayed recall in an auditory verbal learning paradigm was affected by basic information processing speed (digit copying) and hearing acuity (average hearing acuity at 1, 2 and 4 KHz at the better ear). A group of 453 individuals in the age between 23 and 82 years with no overt hearing pathology was recruited from a larger study of cognitive aging (Maastricht Aging Study, MAAS). After controlling for age, sex, educational level, and processing speed it was found that a mild to moderate hearing loss predicted lower verbal memory performance. Auditory administered verbal memory tests can underestimate true memory performance, particularly in older individuals with unknown hearing status.

Prevalence rates of otitis media with effusion from 0 to 2 years of age: healthy-born versus high-risk-born infants

In a prospective-longitudinal study, prevalence rates of otitis media with effusion (OME) were analysed in 150 healthy-born and 100 high-risk-born infants, aged 0-2 years. In order to determine OME, otoscopy and tympanometry were performed at 3-monthly intervals beginning at term date. The Maastricht Otitis Media with Effusion Study (MOMES) algorithm was used to standardize the diagnosis. The distribution of relevant background characteristics was similar in both groups except for gestational age and birth weight, which were significantly lower (P < 0.001) in the high-risk-born group. The ratio of unilateral:bilateral OME was 2:3. Prevalence rates of OME were strongly associated with age (P < 0.001). During the first months of life, OME prevalence rates increased rapidly in both groups, but did not differ significantly. However, from the age of 6 months on, OME prevalence rates of the high-risk group became significantly higher compared with the normal group (P < 0.05). The peak prevalence of OME (59% in the high-risk group versus 49% in the normal group) was observed around the age of 10 months. Although gradually decreasing prevalence rates were noted in both groups with aging, the differences between high-risk and normal infants remained, at least up to 24 months. In conclusion, OME is a very prevalent, age-dependent disorder during infancy, especially in high-risk infants. Peak prevalence of OME was found in the second half of the first year of life.

Response shift in the measurement of quality of life in hearing impaired adults after hearing aid fitting

Reason for the study: Response shift is the change in the meaning of ones self-evaluation of a target construct, like quality of life (QOL). The objective of this study was to investigate whether response shift in the measurement of generic and specific QOL occurred in persons with a relatively mild health condition. For this purpose hearing impairment was used as a research model. Major findings: Response shift effects were observed in the scores on the dimensions of hearing related QOL. In the scores on overall hearing related QOL, and in the scores on the generic control items, no response shift occurred. Conclusions: This study showed that response shift effects can take place in a relatively mild condition as well. The occurrence of response shift in QOL ratings over time could have large implications for the estimation of the effectiveness of medical interventions and for the use of these estimations in cost-effectiveness analyses. After a successful treatment the conventional change could be an underestimation of the effectiveness of the treatment, although it has also been argued that psychological adaptation is a welcome capacity of human beings, and that then-test changes do no justice to this capacity.

Risk factors of otitis media with effusion during infancy.

UNLABELLED Associations of possible risk factors with prevalence of otitis media with effusion (OME) were prospectively studied in a cohort of 250 infants, aged 0-2 years. In order to determine OME, otoscopy and tympanometry were performed at 3-monthly intervals beginning at term date. Eighteen epidemiologically relevant features were inventoried by means of standardized questionnaires. Multivariate analysis controlled for possible confounding factors. Prevalence of OME was most strongly associated with age (P-value < 0.001). Other factors significantly associated with the prevalence of OME (P-value < 0.05) were gestational age, birth weight, breastfeeding, day-care attendance, number of siblings, season, and parent-reported ear infection, hearing loss, mouth breathing and common cold. No significance was found for gender, date of birth, passive smoking, family history of otitis media, parental socio-economic status and histories of snoring and consultation of a physician. IN CONCLUSION both intrinsic and extrinsic factors appear to play an important role in the prevalence of OME. Some of the risk factors appeared to be time-dependent.

Societal effects of hearing aid fitting among the moderately hearing impaired.

The objective of this study was to evaluate hearing aid fitting from a societal viewpoint, rather than the more traditional patient perspective. The effects of hearing aid fitting on generic quality of life (EuroQol), social functioning (SF36), auditory disability, productivity at paid and unpaid labour, and medical consumption, were assessed in a prospective study among 80 moderately hearingimpaired first-time hearing aid applicants. The study showed that hearing aid fitting solved problems with paid employment, but did not seem to affect unpaid work. Use of medical services remained relatively stable after first-time hearing aid fitting. The Amsterdam Inventory proved to be a useful questionnaire to assess the effects on hearing disability. No effects of hearing aid fitting on generic quality of life could be determined with the EuroQol, while hearing aid fitting did lead to an improvement in one aspect of generic quality of life; namely social functioning. El objetivo de este estudio fue evaluar la adaptación de auxiliares auditivos desde un punto de vista social, más que desde la perspective tradicional del paciente. Los efectos de la adaptación de auxiliares auditivos en la calidad genérica de vida (EuroQWol), funcionamiento social (SF36), discapacidad auditiva, productividad en el trabajo remunerado o no remunerado, además de gastos médicos, fueron evaluados en un estudio prospectivo con 80 pacientes con impedimentos auditivos moderados a quienes se les adaptó por primera vez un auxiliar auditivo. El estudio mostró que la adaptación de auxiliares auditivos resuelve problemas en los empleos remunerados pero no parece afectar el trabajo sin remuneración. El uso de servicios médicos permaneció relativamente estable después de la primera adaptación. El cuestionario Amsterdam probó ser un instrumento útil para evaluar los efectos en la discapacidad auditiva. No se pudieron determinar los efectos de la adaptación de auxiliares auditivos en la calidad genérica de vida por medio del EuroQol, mientras que la adaptación de auxiliares auditivos condicionó la mejoría en un aspecto de la calidad genérica de vida, que es concretamente, el funcionamiento social.

Tinnitus: a cost study.

Objectives: The aim of this study was to examine the costs of tinnitus in The Netherlands from a health care and a societal perspective. Furthermore, the impact of disease characteristics and demographic characteristics on these costs were examined. Methods: A bottom-up cost of illness study was performed, using the baseline data on a cost questionnaire of a randomized controlled trial investigating the (cost) effectiveness of an integral multidisciplinary treatment for tinnitus versus care as usual. Mean yearly costs were multiplied by the prevalence figure of tinnitus for the adult general population to estimate the total cost of illness of tinnitus to society. Because cost data usually are not normally distributed, a nonparametric bootstrap resampling procedure with 1000 simulations was performed to determine statistical uncertainty of the cost estimates per category. Several questionnaires measuring disease and demographic characteristics were administered. The impact of disease characteristics and demographics on costs was investigated using a multivariate regression analysis. Results: Total mean societal cost of illness was &OV0556;6.8 billion (95% confidence interval: &OV0556;3.9 billion–&OV0556;10.8 billion). The larger part of total cost of illness was not related to health care. Total mean health care costs were &OV0556;1.9 billion (95% confidence interval: &OV0556;1.4 billion–&OV0556;2.5 billion). Significant predictors of both health care costs and societal costs were tinnitus severity, age, shorter duration of tinnitus, and more severe depression. Conclusion: The economical burden of tinnitus to society is substantial, and severity of tinnitus is an important predictor of the costs made by patients.

Willingness to Accept versus Willingness to Pay in a Discrete Choice Experiment

OBJECTIVES Our main objective was to compare willingness to accept (WTA) and willingness to pay (WTP) in a discrete choice experiment on hearing aid provision. Additionally, income effect and endowment effect were explored as possible explanations for the disparity between WTA and WTP, and the impact of using a WTA and/or WTP format to elicit monetary valuations on the net benefit of the new organization of hearing aid provision was examined. METHODS Choice sets were based on five attributes: performer of the initial assessment; accuracy of the initial assessment; duration of the pathway; follow-up at the ear, nose, and throat specialist; and costs. Persons with hearing complaints randomly received a WTP (costs defined as extra payment) or WTA (costs defined as discount) version of the experiment. In the versions, except for the cost attribute, all choice sets were equal. RESULTS The cost coefficient was statistically significantly higher in the WTP format. Marginal WTA was statistically significantly higher than marginal WTP for the attributes accuracy and follow-up. Disparity was higher in the high educational (as proxy for income) group. We did not find proof of an experience endowment effect. Implementing the new intervention would only be recommended when using WTP. CONCLUSIONS WTA exceeds WTP, also in a discrete choice experiment. As this affects monetary valuations, more research on when to use a payment or a discount in the cost attribute is needed before discrete choice results can be used in cost-benefit analyses.

A Longitudinal Study of the Validity of Parental Reporting in the Detection of Otitis Media and Related Hearing Impairment in Infancy

A total of 150 full-term and 66 pre-term infants were selected at birth and prospectively examined at three-monthly intervals from birth until the age of 27 months. Parental reports of middle ear infection and/or hearing impairment were obtained prior to otoscopic and audiometric evaluation. The relationships between parental reports and the diagnoses acute otitis media (AOM), otitis media with effusion (OME) and hearing impairment (HI), were assessed in terms of sensitivity, specificity and positive and negative predictive values, using the data obtained during the follow-up period. AOM and OME were diagnosed using otoscopy and tympanometry. Hearing was assessed by conditioned orientation reflex audiometry. HI was defined as averaged thresholds > or = 20 dB compared with age-appropriate response levels. Despite the repeated feedback which parents received on the middle ear condition and hearing of their infants, the majority fail to recognize the presence of AOM, OME and HI. The limited sensitivity of parental reporting should discourage both researchers and clinicians from using it as a diagnostic or monitoring instrument, but they should not disregard parental concern when confronted with it.