|  Home  |  Home VUmc  |   Print this page  |
Betrokken en zorgvuldig - Kennis maakt ons betervisual


Current projects

  1. National Longitudinal Study on Hearing (NL-SH)

  2. PredictEAR- Predicting older persons’ behavior with regard to their hearing: Do new predictors and subgroup effects hold the key to success?

National Longitudinal Study on Hearing (NL-SH)

 Period:  November 2006 – present


Mariska Stam, PhD, is co-project leader of the National Longitudinal Study on Hearing (NL-SH). The NL-SH, also internationally known as the Netherlands Longitudinal Study on Hearing, is an ongoing prospective cohort study implemented via the Internet. The NL-SH is a unique cohort, as it includes adults ranging from 18 to 70 years at baseline, and hearing impairment is the central topic under investigation.

Hearing status is determined by the online digit-triplet speech-in-noise test (known as the National Hearing Test: www.hoortest.nl) and also by self-reported measures such as the Amsterdam Inventory for Auditory Disability and Handicap. The main aim of the NL-SH is to identify the impact of hearing ability on activity and participation of adults in a large range of important life domains. Questionnaires administered via the internet are used to examine psychosocial health (distress, depression, anxiety, somatization, loneliness and self-efficacy), work characteristics (job demands and control, need for recovery, sick leave), medication use, and health care utilization.

Data collection started in 2006. Since 2011 a 5-year follow-up measurement round is performed (referred to as NL-SH 2), and since 2016 10-year follow-up data (NL-SH 3) is being collected. Follow-up measurements are planned for every five years.

More than 10 scientific papers have already been published based on the NL-SH data. We are participating in the Geoscience and Health Cohort Consortium (GECCO). In addition, the NL-SH results are frequently disseminated via (inter)national newspapers, magazines, websites, and conferences. Our recommendations are incorporated in general practice and hearing health care. For more information and to read the NL-SH results: www.hooronderzoek.nl.

For the summary of NL-SH results:


PredictEAR- Predicting older persons’ behavior with regard to their hearing: Do new predictors and subgroup effects hold the key to success?


 Projectcode:  WC 2013-005
 Project leader:  M. Pronk, PhD
 Grant:  Oticon Foundation.
 Period:  1 Dec 2012 – present



Marieke Pronk, PhD, project leader. Prof. Sophia Kramer, PhD, senior researcher. Niek Versfeld, PhD, senior researcher. Prof. Dorly Deeg, PhD, senior researcher.


Hearing loss is one of the most common and most disabling chronic conditions in older adults. However, the numbers of older adults seeking hearing help and taking up hearing aids remain disappointingly low. Recent literature reviews convincingly show that the factors that cause this are still largely unknown. Moreover, no study ever investigated the possibility of subgroup-specific effects and very few studies applied appropriate statistics.


1) What are new relevant predictors of hearing help-seeking behaviour (first-time help seeking for hearing problems and first-time hearing aid uptake) among older adults and what is their predictive strength relative to that of established predictors?

2) Do subgroups exist which each have their unique set of relevant predictors?
We aim to develop (subgroup-specific) prediction models that explain a considerable amount of variance in help seeking and hearing aid uptake. We will disseminate the prediction models to the scientific field and to clinical practice.



Help Seeking (HS) Study: Observational cross-sectional study, patient-control set-up.

Hearing Aid Uptake (HAU) Study: Prospective observational study, patient-control set-up.

Samples and data collection

In the HS study, predictors of help seeking will be investigated in older (55 years and over) respondents who report hearing problems. Respondents who never sought formal hearing help (i.e., the ‘non-help seekers’) were recruited from the Longitudinal Aging Study Amsterdam (n=141). ‘Help-seekers’ who were about to seek hearing help for the first time were recruited at hearing aid dispensers (n=64) and general practices (GPs; n=5).
All respondents were asked to fill in a questionnaire and do a hearing test (speech-in-noise test) via the telephone (Smits et al., 2004).

In the HAU study, predictors of hearing aid uptake will be investigated in an older (55 years and over) sample of 380 respondents who just heard from their hearing health care provider (ENT doctor or hearing aid dispenser) that they are eligible for one or two hearing aids such that health care reimbursement is provided. 270 respondents were recruited via hearing aid dispensers and 110 respondents via ENT departments of regional hospitals (total of 380). Shortly after the appointment in which the respondents heard from their ENT doctor/ hearing aid dispenser they needed a hearing aid (t0), the respondents were asked to fill in a questionnaire and do a hearing test (digit triplet speech-in-noise test via the telephone [Smits et al., 2004]). Their audiograms were requested from the particular health care provider. Five weeks (t1) and four months (t2) after inclusion in the study, respondents’ status regarding hearing aid uptake was assessed to determine whether they had started a hearing aid trial or not (‘uptaker’ vs. ‘non-uptaker’, respectively).

Measures-Dependent variables

HS study: Sought professional help (hearing health care) for the first time (yes/no).

HAU study: Tried out (a) hearing aid(s) (yes/no). Measured at t2, so 4 months after having received advice from an ENT doctor/hearing aid dispenser to get 1/2 hearing aid(s).

Measures-Independent variables (potential predictors)

For both studies, 24 factors were assessed through a questionnaire and objective hearing tests. These factors are: self-reported hearing status, objective hearing loss (digit triplet speech-in-noise test [HS & HAU study] and PTA0.5,1.2 kHz [HAU Study only]), age, level of education, coping strategies, mastery, attitudes towards HAs (benefits and barriers) and hearing loss (passive acceptance), and social pressure and evaluation of hearing aids by others stage of behavior change, duration of hearing loss, discrepancy between views of the patient and the health professional on the patient’s hearing loss (HAU-study only), self-efficacy of hearing aid use (HAU-study only), technology use (HAU-study only), occupational status, partner status, ethnicity (HAU-study only), cognition, comorbidity, previous use of hearing self-tests or having had work-related hearing assessments, amount of social interaction (network size and degree of participation).

Statistical Analyses

Multivariate prediction models will be built using logistic regression models. The relevance of the new predictor variables will be identified relative to that of the established predictor variables. The presence of subgroups effects will be tested via interaction terms and stratified analyses. If possible, prediction rules will be developed.


Data collection and data entry for both studies have been completed. Currently the project is in the stage of data cleaning and analysis.

HS study: 141 non-help seekers and 69 help-seekers (64 via hearing aid dispensers, 5 via GPs) could be included. This give a total sample size of n=210.

HAU study: 380 respondents (270 via hearing aid dispensers, 110 via ENT doctors) could be included, giving a total sample size of n=380. Of these 380, 246 (65%) tried out a hearing aid after 4 months, and 134 (35%) did not.

Manuscript in preparation: Predictors of overestimation and underestimation of hearing problems in older persons.


© VU medical center Privacy | Disclaimer | Copyright | Webredactie