Skip to main content
NCBI home page
Journal of the American Academy of Audiology logoLink to Journal of the American Academy of Audiology
. 2025 Jul 1;36(4):248–260. doi: 10.3766/jaaa.240108

Establishing Linguistic Categories for Hearing Aid Experiences in Adults: A Modified Delphi Study

Ilze Oosthuizen *,†,✉, Vinaya Manchaiah *,†,‡,§,‖, Stefan Launer #,**, De Wet Swanepoel *,†,‡
PMCID: PMC12575430  PMID: 40474536

Abstract

Background:

Recent advancements in automated natural language processing (NLP) methods and tools have enhanced the efficacy and accuracy of quantitative analysis of natural language data. NLP offers significant potential for audiology by providing valuable insights from open-text responses about users’ lived hearing aid experiences.

Purpose:

This study aimed to establish linguistic categories pertinent to the experiences of adults using hearing aids, with the ultimate goal of developing a specialized text processing module to facilitate natural language analysis of textual data on hearing aid experiences.

Research Design:

A modified electronic Delphi (e-Delphi) design was employed.

Study Sample:

A panel of 16 audiology experts from seven countries.

Data Collection and Analysis:

Two survey rounds were conducted. In round 1, experts rated categories from the Linguistic Inquiry and Word Count (LIWC) software and categories from principal component analysis of open-ended text data on hearing aid experiences. Experts also responded to open-ended prompts regarding categories derived from qualitative studies on hearing aid experiences. Responses were condensed and refined into items for rating during round 2. In round 2, experts reconsidered their ratings in light of the group answers of round 1. All ratings used a four-point Likert scale of importance. Measures of central tendency, levels of dispersion, and Cronbach’s α reliability coefficients were conducted in both rounds for comparative purposes.

Results:

The open-ended section generated 26 items. Consensus was met on 53 linguistic categories encompassing social (e.g., social situations, social support), emotional (e.g., positive tone, negative tone), cognitive (e.g., cognitive processes, causation), lifestyle (e.g., work, leisure), hearing aid–related (e.g., sound quality, use and handling), and service delivery–related dimensions.

Conclusions:

The linguistic categories identified provide a foundation for developing a customized LIWC text processing module tailored to the analysis of hearing aid experiences. Future research is needed to refine and validate the custom text processing module.

Clinical Relevance Statement:

The findings provide a framework to enable the development of a customized hearing aid LIWC dictionary. Such a tool may enhance clinicians’ ability to explore hearing aid outcomes and improve patient-centered care by integrating real-world experiences into clinical practice.

Keywords: hearing aid experiences, natural language analysis, modified e-Delphi method, user perspective, hearing aid outcomes

INTRODUCTION

Hearing loss affects a significant portion of the population (World Health Organization, 2021), with considerable consequences for communication, social interaction, overall health, and quality of life (Lin et al, 2019; Borre et al, 2023). To mitigate these effects, hearing aids are commonly employed as the first‐line clinical management option (Chisolm et al, 2004; Ferguson et al, 2017). Evidence consistently shows that hearing aids can reduce the negative consequences of hearing loss and effectively enhance hearing-specific health-related quality of life and listening ability in adults with hearing loss (Ferguson et al, 2017). For example, hearing aid use improves communicative interactions, strengthens social connections, and reduces the risk of social isolation and loneliness (Lockey et al, 2010; Weinstein et al, 2016; Ferguson et al, 2017; Hall et al, 2023).

In addition, hearing aid use can support mental and cognitive well-being by lowering cognitive load and reducing listening-related fatigue (Hornsby, 2013; Picou et al, 2013; Holman et al, 2019) and may even lower the risk of dementia (Livingston et al, 2020; Lin et al, 2023). Hearing aids also enhance quality of life by allowing individuals to enjoy activities like music and social gatherings (Kitterick and Ferguson, 2018; Greasley et al, 2020; Chundu et al, 2021). Furthermore, they improve safety through awareness of environmental sounds and signals (Vercammen et al, 2020), boost workplace productivity and job satisfaction, and enhance socio-emotional well-being by increasing self-esteem and social engagement and mitigating the risk of depression and anxiety (Chisolm et al, 2007; Mener et al, 2013; Chundu et al, 2021).

Hearing aid outcomes are measured to demonstrate the success of hearing aid use, to assist clinicians in setting individualized goals and providing guidance, and to demonstrate the cost-effectiveness of hearing aids to funding agencies, improving accessibility for users (Gatehouse, 2001; Bray and Nilsson, 2002). Clinically, hearing aid outcomes are typically evaluated in terms of hearing aid use, benefit, and satisfaction (Humes, 1999) using objective, behavioral, and self-reported methods. Objective methods include real-ear measurements; behavioral assessments include, for example, speech perception in quiet and noise; and self-reported methods, often referred to as patient-reported outcome measures (PROMs), rely on standardized tools such as the International Outcome Inventory for Hearing Aids (IOI-HA; Cox and Alexander, 2002). Despite the benefits like enhanced generalizability and consistency across test-retest scenarios, they also have significant limitations. Hearing aid users may not have evaluated all the dimensions targeted by the questions, which raises concerns about the reliability of their responses to certain items. Neutral or nonresponses are also common, complicating the interpretation of results (Kochkin, 2000). Moreover, hearing aid PROMs often require higher reading levels to comprehend questionnaire items, suggesting that those with lower education levels may not easily understand these PROMs (Oosthuizen et al, 2023).

In addition, many existing outcome instruments may be viewed as outdated and lack the sensitivity needed to capture the nuanced and dynamic nature of contemporary hearing aid use, especially in light of rapid technological advancements. Typically, hearing aid owners complete PROMs during follow-up appointments with their clinician, several days or even weeks after the situations targeted by the questionnaire have occurred (Humes, 1999; Whitmer et al, 2016). This delay can introduce recall bias, particularly when capturing fluctuating perceptual experiences (Kahneman and Krueger, 2006; Stull et al, 2009).

Open-ended questions have increasingly been used as an alternative and/or supplementary method for exploring hearing aid experiences in clinical and research settings. These questions allow respondents to provide detailed and personalized feedback, offering deeper insights into their experiences. For example, open-ended questions have been employed to explore reasons for hearing aid uptake (Knoetze et al, 2024), to gather experiences of hearing aid users and audiologists regarding hearing aid use and fitting for music (Greasley et al, 2020), as well as to identify perceived benefits and challenges associated with hearing aid use experienced by hearing aid users (Tyler et al, 1983; Stephens and Meredith, 1991; Convery et al, 2011; Avierinos et al, 2024). More recently, smartphone-connected hearing aids have enabled hearing aid owners to report their experiences in near real time through hearing aid mobile applications. This method, also known as “experience sampling” or ecological momentary assessment (EMA), can minimize recall bias by reducing the time between the experience and the report (Fourie et al, 2024). Allowing listeners to express their experiences in their own words, such as through an open-text/comment field, ensures that responses are independent of predefined survey question wording. This textual data can then be accessed by clinicians, integrated with technical or acoustic data from the hearing aid (e.g., datalogging), and used to provide more personalized and responsive hearing care (Vercammen et al, 2023).

Responses to open-ended questions can be analyzed using qualitative and quantitative methods. Traditionally, qualitative methods have been mainly applied to open-text data, which is tedious and unrealistic to scale for large datasets. However, recent advancements in automated natural language processing (NLP) methods and tools have enhanced the efficacy and accuracy of quantitative analysis of natural language data. Tools such as the Linguistic Inquiry and Word Count (LIWC; Pennebaker et al, 2015; Boyd et al, 2022) allow researchers to extract meaningful information about social, psychological, and cognitive processes from text. Language analysis using LIWC can generate quantitative scores for various dimensions or categories, which can then be correlated with other measures of interest, such as overall scores on hearing aid PROMs. LIWC has been shown to provide reliable insights into aspects like attentional focus, emotionality, social relationships, thinking styles, and individual differences (Tausczik and Pennebaker, 2010; Pennebaker et al, 2015).

Automated analysis techniques have been increasingly applied in audiology. For instance, Bennett et al (2021) used topic modeling to identify themes from online consumer reviews about hearing aid use. Manchaiah et al (2021) analyzed 1,378 consumer reviews using LIWC to explore the relationship between linguistic features and self-reported hearing aid benefit and satisfaction. Their findings showed that higher engagement in social processes, positive emotions, hearing, and work-related dimensions were associated with higher benefit satisfaction ratings. In contrast, negative emotions, time awareness, and financial concerns were linked to lower hearing aid benefit and satisfaction. More recently, Vercammen and colleagues (2023) conducted a retrospective, exploratory analysis of EMA data generated from a hearing aid smartphone app. They applied cluster analysis, an automated text analysis method that allows the processing of large amounts of text data in a time-efficient way while at the same time aiming for similar goals as conventional, manual qualitative text analysis methods (Sbalchiero, 2018). Their results indicated that nearly 60 percent of participant reports related to speech intelligibility in challenging situations and sound quality were viewed positively, whereas almost 40 percent of the textual data on hearing aid management was described in negative terms. Thus, NLP offers significant potential for audiology by providing valuable insights from open-text responses about lived hearing aid experiences.

This study aims to establish a comprehensive framework of linguistic categories relevant to the lived experiences of adults already using hearing aids, leveraging expert consensus through a modified Delphi method. These categories will form the foundation of a specialized LIWC dictionary (or text processing module), enabling natural language analysis of textual data related to hearing aid experiences, including perceived benefits and satisfaction. This approach can offer an alternative method to complement the standardized PROMs currently employed in clinical and research settings, potentially advancing patient-centered care. These insights can serve as independent knowledge sources for hearing aid users, hearing-health-care professionals, and hearing aid manufacturers.

METHODS

This study employed a modified electronic Delphi (e-Delphi) survey method to identify linguistic categories relevant to hearing aid experiences by leveraging preselected linguistic categories from the LIWC software (Boyd et al, 2022), dimensions identified through Meaning Extraction Method (MEM) and principal component analysis (PCA) of open-ended text on hearing experiences (Oosthuizen et al, 2024), and systematically derived dimensions from qualitative studies on hearing aid experiences (Oosthuizen et al, 2022). Delphi studies often employ multiple rounds of surveys to obtain feedback from a panel of informed or expert individuals regarding their opinions on specific issues or concerns until consensus has been reached (Hsu and Sandford, 2007b; Vázquez-Ramos et al, 2007). Modified Delphi studies differ from classical techniques in allowing modifying items to be discussed with panel members (Keeney et al, 2011). This study’s design was modified by including preselected items and open-ended questions in the first round. An e-Delphi approach was also appropriate for this study because participants were recruited from worldwide geographic locations. The modified e-Delphi survey process used in this study was adapted from recommendations provided by Vázquez-Ramos et al (2007) and Keeney et al (2011). The Faculty of Humanities Research Ethics Committee, University of Pretoria (Ref.: HUM034/0923), approved the study before recruiting participants. The study was conducted over approximately 6 months, from January to July 2024, allowing for sufficient time for panel recruitment (January to February 2024), completion of two survey rounds, including granting extensions to panelists upon request to complete the surveys (round 1: February to March 2024; round 2: May to June 2024), and analysis of responses. The duration was deemed appropriate to balance the thoroughness of expert consensus development with the practical constraints of participant availability.

Panel Selection

Group size in Delphi studies is determined by the need to facilitate meaningful consensus through group dynamics rather than statistical power (Okoli and Pawlowski, 2004). Thus, the Delphi method does not require a representative sample for statistical purposes. Instead, the quality and expertise of the panel are prioritized over numerical representativeness (Powell, 2003). Panel selection for this study involved purposive sampling, targeting 23 professionals with extensive experience in providing or researching hearing aids for adults and having a substantial publication record on the topic. A target sample of 23 experts was chosen to account for possible attrition between rounds and stay within the recommended number of 10–15 panel members for a homogenous panel (Hsu and Sandford, 2007b).

Survey Development

Recruitment and communication with the panel members were conducted via the first author’s e-mail account. Data and information from the panelists were collected through the cloud-based Qualtrics Research Suite, which the University of Pretoria hosted. Due to the absence of prior research identifying pertinent linguistic categories for creating a customized LIWC text processing module for audiology, two survey instruments were developed explicitly for this study. A pilot test was conducted with an expert panelist, whose minor grammatical suggestions were incorporated, and the responses were included in the main study data. Participants had to provide electronic informed consent before completing the online survey.

Round 1 Instrument

The first survey aimed to obtain demographic information about the panel members and generate items for inclusion and rating in round 2. Participants were asked to provide their name, contact information (preferred e-mail address), and demographic information such as age, gender, geographic location, professional field of practice, employment setting, professional responsibilities, extent of professional publications related to adult hearing aid users and hearing aid outcomes, and perceived level of experience related to hearing aid intervention for adults.

After the demographic section, the survey presented preselected items, including (1) 26 linguistic dimensions/categories from the general LIWC software dictionary (Pennebaker et al, 2015; Boyd et al, 2022) pertinent to hearing aid experiences as identified by a recent publication (Manchaiah et al, 2021), and (2) six dimensions/categories identified from MEM and PCA of open-ended text on hearing experiences (Oosthuizen et al, 2024). Panelists were requested to review and rate these items using a Likert scale of importance (4 = absolutely necessary, 3 = very important, 2 = somewhat important, 1 = not important at all) based on their relevance for inclusion in a custom audiology LIWC text processing module. Development of this module aims to enable the natural language analysis of textual data related to hearing aid experiences, such as perceived benefits and satisfaction.

An open-ended section followed, where panelists reviewed 29 dimensions derived from a systematic review of qualitative studies on hearing aid experiences of adult hearing aid owners during and after fitting (Oosthuizen et al, 2022). Panelists were asked to provide feedback on the relevance of these dimensions for inclusion in the custom hearing aid experiences LIWC text processing module. They were also encouraged to suggest any additional categories they considered essential for the module.

Round 2 Instrument

In the second survey, panelists were provided with a summary of the expert group’s responses alongside their responses from the first round. Panelists were asked to reconsider their initial answers in light of the aggregated group statistics (mean, median, and mode). They could either modify their responses or retain their original answers.

The survey also provided a brief overview of the items (categories and subcategories) generated from the open-ended sections of round 1. The remainder of the response section asked the panelists to evaluate the listed items using a Likert scale of importance (4 = absolutely necessary, 3 = very important, 2 = somewhat important, 1 = not important at all), indicating how essential they believed each item was for inclusion in a custom audiology LIWC text processing module.

Procedure and Analyses

Round 1

The response rate and demographic information of the expert panelists were summarized using descriptive statistics. Items from the open-ended section were condensed and coded by the principal author using Microsoft Excel to remove redundancy and align items with the preselected categories. After this initial condensation, the authors met to further refine the items by removing duplicates and redundant entries. They reached a consensus on which items aligned with the preselected categories and identified any new items for inclusion in the round 2 instrument. Items suggested by fewer than three panelists were excluded.

Panelists were given 2 weeks to complete the round 1 survey, with e-mail reminders sent after 1 week and then 2 to 3 days before the closing date to encourage timely responses. Extensions were granted to three panelists who needed additional time. Group means, standard deviations, and frequency distributions for each preselected item were calculated using SPSS statistics software (IBM Corporation, version 29). Additional analyses conducted in SPSS included Cronbach’s α reliability coefficients to assess internal consistency, measures of central tendency (median and mode), and levels of dispersion (interquartile range [IQR]).

Round 2

Panelists who completed the round 1 survey were e-mailed their individual responses from round 1 and the group statistics from round 1 along with a survey link to complete the round 2 survey. They were given 2 weeks to respond, with the same e-mail reminder schedule as in round 1. Two panelists were granted an extension to allow them to complete this round. All analyses conducted in round 1 were repeated after round 2 for comparative purposes.

Consensus Definition

In this study, consensus was defined as homogeneity or consistency of opinion among the panelists (Graham et al, 2003). An item was considered to have reached consensus if at least 70 percent of the panel members rated it as “very important” or “absolutely necessary” (i.e., a score of 3 or 4 on the Likert scale; Hsu and Sandford, 2007a, 2007b; Xue et al, 2018; Meibos et al, 2019). For items that did not reach this level of agreement, measures of central tendency (mean, median, or mode) and a measure of dispersion (IQR) were used to assess the final inclusion of the item. A small IQR (≤1.0), as observed in similar Delphi studies (Rayens and Hahn, 2000; Meibos et al, 2019), and a high median score indicated a high level of consensus among the experts. Conversely, a large IQR and a low median score indicated a low level of consensus. Items not meeting the consensus criteria after round 2 were excluded from the final list.

RESULTS

This study aimed to establish linguistic categories relevant to hearing aid user experiences, focusing on expert consensus to develop a framework for future applications in clinical and research contexts.

Participants

Of the 23 individuals invited, 16 responded to the round 1 survey, yielding a response rate of 69.6 percent. Table 1 summarizes the characteristics of the panelists. Participants included professionals from the United States, United Kingdom, Australia, the Netherlands, Canada, Germany, and Switzerland. Most of the panel members were female (63 percent). All participants reported working in the field of audiological research, with 14 working full-time and two part-time. The majority (n = 15/16) were employed in university settings, while others worked in research organizations (n = 3), industry (n = 1), and public corporations (n = 1). Note that participants were able to select more than one employment setting. The panelists’ publication records varied, with the majority (n = 10) having published more than 20 articles on hearing aids in the adult population. When rating their perceived knowledge and experience in hearing aid intervention for adults, eight panelists identified themselves as having advanced knowledge (applied theory) and eight as experts (recognized authority).

Table 1.

Characteristics of Expert Panelists for the Modified Delphi Study (n = 16)

Characteristics n
Gender
 Women 10
 Men 6
Geographic location: country
 United States 4
 United Kingdom 4
 Australia 3
 Netherland 2
 Canada 1
 Germany 1
 Switzerland 1
Professional field of practice
 Research 16
Employment status
 Full-time 14
 Part-time 2
Employment setting(s)a
 University 15
 Research organization 3
 Other (industry = 1; public corporation = 1) 2
Approximate number of publications related to hearing aids in adult population
  1–5 2
  6–10 3
  11–15 1
  >20 10
Rating of perceived knowledge and experience in hearing aid intervention in adult population
  Advanced (applied theory) 8
  Expert (recognized authority) 8
Open in a new tab
a

Participants could select more than one option.

Round 1

In round 1, the first author condensed and coded responses from the open-ended section to eliminate redundancy and identify items that aligned with or differed from the preselected categories. After this initial condensation, the authors met to refine and remove duplicate or redundant items. They reached a verbal consensus on items for inclusion in the round 2 instrument. This process generated a total of 26 items related to hearing aid experiences, comprising six categories and 20 subcategories.

Round 2

In round 2, 12 panel members participated, resulting in a response rate of 75 percent (12/16). Using the previously outlined consensus criteria, 53 items were identified as meeting the necessary thresholds after round 2, including 21 of 26 general LIWC linguistic categories, all six PCA-derived categories, and all 26 items from the open-ended section. Five items from the general LIWC linguistic categories did not achieve final consensus: Authenticity, I-Words, Past Focus, Present Focus, and Cognitive Processes: Discrepancy. However, some general LIWC items did gain consensus between rounds, such as Cognitive Processes: Differentiation and Motives: Allure (see Table 2). Only one item, Cognitive Processes: Discrepancy lost consensus. Most of the general LIWC items maintained stable consensus across rounds. No items from either the PCA-derived categories (see Table 3) or the open-ended section (see Table 4) changed consensus status between rounds.

Table 2.

Consensus Results on Preselected Items from the General LIWC Linguistic Categories

Percent Agreementa
Item Round 1 (16 participants) Round 2 (12 participants) Between-Round Status
Consensus met
 Drives: achievement 941 1000 Stable
 Cognition 691 751 Gained
  Cognitive processes 820 751 Stable
  Insight 820 831 Stable
  Causation 881 1000 Stable
  Differentiation 691 830 Gained
 Affect—positive tone 941 1000 Stable
 Affect—negative tone 941 1000 Stable
 Social processes 941 1000 Stable
  Social behavior 1000 1000 Stable
   Communication 941 1000 Stable
  Social referents 881 1000 Stable
 Motives: allure 691 830 Gained
 Perceptual: auditory 1001 1000 Stable
 Perceptual: space 752 830 Stable
 Time 692 830 Gained
 Physical: health 881 1001 Stable
 Lifestyle: work 941 1000 Stable
 Lifestyle: leisure 1001 1000 Stable
 Lifestyle: home 941 1000 Stable
 Lifestyle: money 821 921 Stable
Consensus not met
 Authenticity 441 251 Stable
 I-words 632 581 Stable
 Past focus 501 251 Stable
 Present focus 632 581 Stable
 Cognitive processes: discrepancy 820 581 Lost
Open in a new tab
a

Consensus level for each item was equated with items having at least 70 percent of the panel members rating an item as “very important/absolutely necessary” (i.e., a score of 3–4 on a Likert scale). The IQR for each item is indicated in superscript.

Table 3.

Consensus Results on Preselected Items from the Categories Generated from the PCA of Text Responses to an Open-Ended Question Related to Hearing Experiences

Percent Agreementa
Item Round 1 (16 participants) Round 2 (12 participants) Between-Round Status
Consensus met
 Life change 1001 1001 Stable
 Social situation 1001 1000 Stable
 Quality of life 1000 1000 Stable
 Impact and speech understanding 1000 921 Stable
 Phone, work, socialize 941 921 Stable
 Music and environmental sounds 941 921 Stable
Open in a new tab
a

Consensus level for each item was equated with items having at least 70 percent of the panel members rating an item as “very important/absolutely necessary” (i.e., a score of 3–4 on a Likert scale). The IQR for each item is indicated in superscript.

Table 4.

Consensus Results on Items (Categories and Subcategories) Derived from Expert Panelists’ Suggestions in the Open-Ended Section in Round 1

Round 2 (12 participants)
Item Percent Agreementa
Consensus met
 Hearing and communication 1000
  Hearing 1000
  Conversation 1001
  Self-perceived benefit 1000
  Self-perceived satisfaction 1000
 Psychosocial functioning 1001
  Social situation 1001
  Socio-emotional 921
  Social support 921
 Hearing aid related 1001
  Use and handling 1000
  Sound quality 1000
  Care and maintenance 921
  Feedback 831
  Cost 821
  Connectivity 752
  Appearance and design 752
 Audiological service delivery 1001
 Impact on daily life 1000
  Quality of life 1000
  Work/occupation 1001
  Education 1001
 Listening situation specific 1001
  Music 1001
  Environmental awareness 1001
  Noisy listening situations 1001
Open in a new tab
a

Consensus level for each item was equated with items having at least 70 percent of the panel members rating an item as “very important/absolutely necessary” (i.e., a score of 3–4 on a Likert scale). The IQR for each item is indicated in superscript.

To assess the internal consistency of the items rated within each competency area, we calculated Cronbach’s α reliability coefficients for the general LIWC and PCA items after rounds 1 and 2. The PCA items demonstrated acceptable internal consistency in round 1 (Cronbach’s α = 0.745) and round 2 (Cronbach’s α = 0.833). The general LIWC items exhibited high internal consistency in round 1 (Cronbach’s α = 0.891). However, in round 2, the assumptions for calculating Cronbach’s α were violated, as three LIWC items (Affect Positive Tone, Affect Negative Tone, and Social Behavior) showed zero variance. This lack of variance aligns with the high level of consensus among panelists in scoring these items.

DISCUSSION

This study aimed to identify linguistic categories relevant to hearing aid experiences through a modified Delphi method. These linguistic categories are essential for the development of a custom LIWC text processing module tailored to natural language analysis of textual data related to adult hearing aid experiences. The diverse demographics and professional backgrounds of the expert panelists contributed to a comprehensive Delphi study, ensuring a well-rounded perspective on hearing aid interventions.

Consensus was reached for 53 linguistic categories applicable to adult hearing aid experiences (see Table 5). These categories align with factors identified in recent systematic reviews on hearing aid benefit and satisfaction, reflecting the multifaceted nature of lived experiences of hearing aid users (Oosthuizen et al, 2022; Mothemela et al, 2024). The panel agreed on categories encompassing social, emotional, cognitive, lifestyle, hearing aid–related, and service delivery–related dimensions. The alignment of these categories with existing literature underscores their relevance and applicability. For example, the consensus on LIWC categories like Affect (positive tone and negative tone), Social Processes, including Social Behavior—Communication, and Social Referents, resonates with previous research emphasizing the emotional and social factors affecting hearing aid use (Singh et al, 2015; Weinstein et al, 2016; e.g., Ferguson et al, 2017; Holman et al, 2023). The LIWC category of Affect conceptualizes Positive Tone and Negative Tone subcategories as reflecting sentiment (i.e., an attitude or mental feeling toward something) rather than just emotion per se (Boyd et al, 2022). Reports have indicated that a positive attitude among hearing aid users fosters perseverance and assertiveness in daily hearing aid use (Lockey et al, 2010; Gallagher and Woodside, 2018; Chundu et al, 2021), whereas negative emotions such as neuroticism have been found to have a negative association with hearing aid satisfaction (Singh et al, 2015).

Table 5.

Summary of Items That Met Consensus Across Different Survey Sections

Section Item
General LIWC categories Drives: achievement
Cognition
 Cognitive processes
  Insight
  Causation
  Differentiation
Affect—positive tone
Affect—negative tone
Social processes
 Social behavior
  Communication
 Social referents
Motives: allure
Perceptual: auditory
Perceptual: space
Time
Physical: health
Lifestyle: work
Lifestyle: leisure
Lifestyle: home
Lifestyle: money
PCA categories Life change
Social situation
Quality of life
Impact and speech understanding
Phone, work, socialize
Music and environmental sounds
Expert panel suggestions Hearing and communication
 Hearing
 Conversation
 Self-perceived benefit
 Self-perceived satisfaction
Psychosocial functioning
 Social situation
 Socio-emotional
 Social support
Hearing aid related
 Appearance and design
 Use and handling
 Connectivity
 Sound quality
 Feedback
 Care and maintenance
 Cost
Audiological service delivery
Impact on daily life
 Quality of life
 Work/occupation
 Education
Listening situation specific
 Music
 Environmental awareness
 Noisy listening situations
Open in a new tab

Furthermore, existing literature highlights the significant negative impact of hearing loss on psychosocial functioning, whereas improved psychosocial functioning is often the most commonly perceived benefit from hearing aid use (Oosthuizen et al, 2022). Hearing aid use has also been positively impacting feelings of depression and social isolation (Kalluri and Humes, 2012; Mener et al, 2013; Weinstein et al, 2016; Ferguson et al, 2017). The LIWC categories related to Social Processes parallel these findings, whereas the Social Referents category aligns with the impact of Social Support, for example, positive support from family or significant others, which is a key factor in predicting and influencing hearing aid use and satisfaction (Lockey et al, 2010; Laplante-Lévesque et al, 2013; Guerra-Zúñiga et al, 2014; Hickson et al, 2014; Singh et al, 2015).

Four general LIWC categories moved into the consensus range between rounds 1 and 2: Cognition, Cognitive Processes—Differentiation, Motives—Allure, and Time, suggesting their potential significance in understanding the nuances of hearing aid users’ experience narratives. The Cognition category is a broad category that captures various ways people think or refer to their thinking (Boyd et al, 2022). Including LIWC subcategories of Cognitive Processes, specifically Insight, Causation, and Differentiation, can potentially demonstrate how users think about their hearing aids and whether they perceive benefits and/or residual hearing difficulties. These LIWC cognition categories together with the LIWC Motives—Allure category might also link to motivational factors, that is, the cognitive processes that drive and direct behavior, as included in the Capability, Opportunity, Motivation, Behavior (COM-B) model (Michie et al, 2011). This model has been successfully applied in audiology research to describe and understand the daily experiences of adult hearing aid users (Maidment et al, 2019; e.g., Gomez et al, 2022; Oosthuizen et al, 2022).

The LIWC Time category is likely relevant to daily hearing aid experiences, particularly in terms of hearing aid usage. Hearing aid use, defined as the number of hours per day that a hearing aid is worn, is a key construct of hearing aid outcomes (Humes, 1999; Solheim et al, 2012). It is considered a strong indicator of hearing aid performance, with positive associations reported between daily usage time and both hearing aid benefit (Solheim et al, 2012; Wang et al, 2022) and satisfaction (Korkmaz et al, 2016; Wang et al, 2022).

Compared with the general LIWC categories, all the hearing aid experience–related categories (synthesized from the open section in round 1) reached consensus after just one round (i.e., round 2). Although the general LIWC framework provides a broad analysis of language use, particularly in the discipline of social psychology, it lacks the specificity required to address the unique challenges and experiences associated with daily hearing aid use. Typically, LIWC analysis distinguishes between function words and content words, with greater emphasis on content words such as personal pronouns and articles. This distinction can reveal several different psychological processes, such as the increased use of first-person singular “I” pronouns tracking lower status in dyadic social interactions (Campbell and Pennebaker, 2003; Pennebaker, 2011). However, in this study, the expert panel did not consider the LIWC category of I-Words (i.e., personal pronouns) necessary for inclusion in the custom LIWC text processing module. This may reflect the differing priorities of various health-care disciplines, such as social psychology versus audiology.

Including categories like Hearing Aid–Related subcategories, Audiological Service Delivery, and Listening Situation-Specific will provide a more targeted analysis, enabling researchers and clinicians to explore the complexities of user experiences more effectively. Hearing Aid–Related categories can reflect device-specific factors that impact user experiences in modern lifestyles. Factors like streaming, connectivity, sound quality, smartphone applications, ease of use, and comfort increasingly influence hearing aid outcomes, particularly in terms of perceived benefit and satisfaction (Meyer et al, 2014; Bennett et al, 2020; Picou, 2020). The impact of modern hearing aid features, such as wireless connectivity, that increase the likelihood of hearing aid benefits (Picou, 2020) may be better captured by including these categories, because various existing outcome instruments may be outdated and lack sensitivity to technological advancements in hearing aids.

Furthermore, the quality of service and support the hearing care professional provides can significantly influence hearing aid owners’ experiences. Confidence in the clinician and patient- and family-centered care has been shown to positively affect hearing aid experiences and outcomes (Laplante-Lévesque et al, 2010; Singh et al, 2015; Ekberg et al, 2023). The introduction of virtual forms of clinical support, such as mobile applications and virtual access to hearing care professionals, has also been shown to support improved hearing aid outcomes (Ferguson et al, 2016; Ross et al, 2022).

The results indicate that categories from the different survey sections participants rated have potential overlap (see Table 5), that is, general LIWC categories, categories from PCA, and hearing aid experience–related categories and subcategories (as derived from the round 1 open-ended section). For instance, a social dimension is evident in the overlap among the following categories: LIWC Social Processes, LIWC Social Behavior, LIWC Social Behavior—Communication, LIWC Social Referents, PCA category Social Situation, and hearing aid experience–related subcategories of Social Situation and Social Support. In addition, several categories show some overlap in a lifestyle dimension, for example, LIWC Lifestyle Work, PCA categories of Life Change, Quality of Life, and Phone, Work and Socialize, and hearing aid experience–related category of Impact on Daily Life with its subcategories. Similarly, a hearing dimension is apparent with overlapping categories such as LIWC Perceptual Auditory and hearing aid experience–related subcategories of Hearing, Conversation, Music, and Environmental Awareness. Future research should aim to meaningfully merge these overlapping categories when creating a custom hearing aid experiences LIWC dictionary.

The diverse categories identified emphasize the multifaceted nature and interplay between various factors that impact hearing aid experiences for adults. Figure 1 presents an example of the interaction between some categories from different sections that the experts rated. This finding aligns with the call to approach hearing health holistically (Campos and Launer, 2020; Saunders et al, 2021) and is similar to the dimensions identified by experts in Humes et al (2024). That is, to consider hearing health or auditory wellness within the context of healthy living, recognizing that improved hearing (e.g., through appropriately fitted hearing aids) can positively impact socio-emotional well-being (e.g., enhanced social interaction) and cognitive well-being (e.g., reduced perceived listening effort and resultant fatigue, cognitive health), ultimately leading to improved overall well-being (Campos and Launer, 2020; Vercammen et al, 2020; Saunders et al, 2021; Sarant et al, 2023; Humes et al, 2024; Timmer et al, 2024).

Figure 1.

Figure 1.

Open in a new tab

Graphical presentation of examples of interplay between categories from different sections from general LIWC categories, PCA categories, and hearing aid experience–related categories and subcategories.

Limitations and Future Research

The modified Delphi method used in this study is considered an appropriate methodology for obtaining expert opinions on a specific topic where consensus is lacking or for concept or framework development, when participants are geographically dispersed, and when anonymity in response can encourage unbiased feedback (Okoli and Pawlowski, 2004; Keeney et al, 2011). Indeed, this method offered several advantages to this study. Specifically, it enabled the collection of expert consensus from a geographically diverse panel and the development of a framework of linguistic categories (see Table 5) for the future development of a customized hearing aid LIWC text processing module. Despite the small sample size, it was still within the recommendation for a homogenous group of experts in a Delphi approach (Hsu and Sandford, 2007b). Thus, this approach allowed for integrating varied perspectives, enhancing the study’s findings with insights from different cultural and professional contexts. The iterative nature of the Delphi process ensured that the consensus reached was well considered and reflective of the panel’s collective expertise. Furthermore, the use of statistical analyses, including measures of central tendency, dispersion, and internal consistency (Cronbach’s α), ensures that the consensus was robust and objectively evaluated. These measures align with those used in previous studies with similar design (e.g., Keeney et al, 2011; Xue et al, 2018; Meibos et al, 2019).

However, some limitations should be acknowledged. One potential bias stems from the panel composition, because all participants reported working in research settings, with none actively engaged in a clinical field of practice. This may have caused a research-centric perspective on the results. This may limit the immediate applicability of these categories to clinical settings. Therefore, future studies could replicate or contrast these findings by considering a more extensive and diverse panel that incorporates the perspectives of clinicians actively engaged in practice and stakeholders such as hearing aid users. This may enable subgroup analyses, enhancing the robustness and generalizability of findings across different audiology contexts.

In addition, results may have been affected by some design limitations stemming from the phrasing of the instrument questions and the possibility of conformity pressure in group decision-making, a recognized limitation of Delphi methods (Woudenberg, 1991; Stone Fish and Busby, 1996). Furthermore, linguistic categories relevant to the early stages of the hearing aid journey, such as help-seeking, uptake, and acclimatization, were not necessarily captured in the consensus items. Future studies should consider including linguistic categories that reflect these initial stages to provide a more comprehensive understanding of the user experience across the hearing management journey, especially for first-time users. Moreover, the identified categories may not fully capture regional or cultural variations in linguistic expressions of hearing aid experiences, which could limit the broader relevance.

This study is nested in a larger NLP research project with a future aim of developing a custom hearing aid experiences LIWC text processing module for automated NLP and exploring potential use to determine hearing aid experiences. By enabling more precise and detailed analysis of natural language data from hearing aid users, this approach can offer a novel, patient-centered complement to standardized PROMs, with potential applications for clinicians, researchers, and hearing aid manufacturers. Therefore, the linguistic dimensions identified in this study will serve as a structural framework for developing such a customized hearing aid LIWC text processing module. However, refinement and merging of the identified categories should be conducted because some dimensions have potential overlap. Future studies should focus on creating and refining such a custom dictionary, following the method developed by Pennebaker, Boyd, and colleagues (Pennebaker et al, 2015; Boyd et al, 2022). This process will include five steps. The first step is synchronization, word collection, and generation by a dictionary creation committee (DCC) of experts in audiology for each category. The second step is the judge rating phase: reviewing of each category’s word list by the DCC, rating in terms of “goodness of fit,” and discussing possible disputes. The third step is base rate analyses: once a working version of the dictionary is constructed from the DCC’s ratings, the MEM (Boyd et al, 2022) will be used to determine how frequently dictionary words are used in corpus of texts to determine possible words to be excluded. The fourth step is psychometric analyses: each linguistic category will be separated into its constituent words; each word will be quantified as a percentage of total words; all the words for each category will be used to compute internal consistency statistics (e.g., Cronbach’s α, Kuder Richardson Formula 20) for each category as a whole. Words detrimental to a category’s internal consistency will be added to a list of words for omission from the final dictionary. The fifth step is the refinement phase: the first four steps will be repeated and reviewed by the DCC.

Once a final draft dictionary is created, future longitudinal studies will be needed to explore its application in real-world clinical settings, which could help address potential limitations in the current study’s expert panel composition. Future studies should actively involve clinicians engaged in practice alongside researchers and stakeholders, such as hearing aid users. This will offer valuable insights into the efficacy of the linguistic categories to reflect clinical realities and patient experiences and indicate the need for further refinement. Future research should determine how applicable such a draft dictionary will be across different countries and demographic cohorts, addressing potential cultural and linguistic diversity limitations, thus ensuring its relevance for diverse populations. Furthermore, a future research application of a custom hearing aid LIWC dictionary could involve analyzing large-scale, real-world datasets collected through EMA methods, such as textual or voice-recorded feedback provided by hearing aid wearers in a smartphone-connected hearing aid app. This approach could also aid in training machine learning algorithms, enabling hearing technology to better anticipate and respond to user needs.

Conclusions

This study identified 53 linguistic categories relevant to hearing aid experiences using a modified Delphi method, laying the groundwork for developing a custom LIWC text processing module tailored to hearing aid experiences. The consensus on these categories reflects the multifaceted nature of hearing aid use, encompassing social, emotional, cognitive, and lifestyle dimensions. The identified categories align with existing literature, reinforcing their relevance and potential applicability in research and clinical practice. Future research should focus on developing and validating a custom hearing aid LIWC dictionary, following established methods for dictionary creation and psychometric analysis.

Acknowledgments

The authors thank the expert panelists for their valuable contributions and insights in this study.

Abbreviations

DCC

dictionary creation committee

e-Delphi

electronic Delphi

EMA

ecological momentary assessment

IQR

interquartile range

LIWC

Linguistic Inquiry and Word Count

MEM

Meaning Extraction Method

NLP

natural language processing

PCA

principal component analysis

PROM

patient-reported outcome measure

Footnotes

Any mention of a product, service, or procedure in the Journal of the American Academy of Audiology does not constitute an endorsement of the product, service, or procedure by the American Academy of Audiology.

REFERENCES

  1. Avierinos RJ, Beukes E, Manchaiah V, Oosthuizen I, Le Roux T, Swanepoel DW. (2024) Meaningful life changes following hearing aid use: a qualitative user perspective. Int J Audiol 2024:1–10. [DOI] [PubMed] [Google Scholar]
  2. Bennett RJ, Kosovich EM, Stegeman I, Ebrahimi-Madiseh A, Tegg-Quinn S, Eikelboom RH. (2020) Investigating the prevalence and impact of device-related problems associated with hearing aid use. Int J Audiol 59(8):615–623. [DOI] [PubMed] [Google Scholar]
  3. Bennett RJ, Swanepoel DW, Ratinaud P, Bailey A, Pennebaker JW, Manchaiah V. (2021) Hearing aid acquisition and ownership: what can we learn from online consumer reviews? Int J Audiol 60(11):917–926. [DOI] [PubMed] [Google Scholar]
  4. Borre ED, Kaalund K, Frisco N, et al. (2023) The impact of hearing loss and its treatment on health-related quality of life utility: a systematic review with meta-analysis. J Gen Intern Med 38(2):456–479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boyd RL, Ashokkumar A, Seraj S, Pennebaker JW. (2022) The Development and Psychometric Properties of LIWC-22. Austin, TX: University of Texas at Austin. [Google Scholar]
  6. Bray V, Nilsson M. (2002) Assessing hearing aid fittings: an outcome measures battery approach. In: Valente M., ed. Strategies for Selecting and Verifying Hearing Aid Fittings. New York, NY: Thieme Medical Publishers. [Google Scholar]
  7. Campbell RS, Pennebaker JW. (2003) The secret life of pronouns: flexibility in writing style and physical health. Psychol Sci 14(1):60–65. [DOI] [PubMed] [Google Scholar]
  8. Campos JL, Launer S. (2020) From healthy hearing to healthy living: a holistic approach. Ear Hear 41(Suppl 1):99S–106S. [DOI] [PubMed] [Google Scholar]
  9. Chisolm TH, Abrams HB, McArdle R. (2004) Short- and long-term outcomes of adult audiological rehabilitation. Ear Hear 25(5):464–477. [DOI] [PubMed] [Google Scholar]
  10. Chisolm TH, Johnson CE, Danhauer JL, et al. (2007) A systematic review of health-related quality of life hearing aids: final report of the American Academy of Audiology task force on the health-related quality of life benefits of amplication in adults. J Am Acad Audiol 18(2):151–183. [DOI] [PubMed] [Google Scholar]
  11. Chundu S, Allen PM, Han W, Ratinaud P, Krishna R, Manchaiah V. (2021) Social representation of hearing aids among people with hearing loss: an exploratory study. Int J Audiol 60(12):964–978. [DOI] [PubMed] [Google Scholar]
  12. Convery E, Keidser G, Hartley L. (2011) Perception of a self-fitting hearing aid among urban-dwelling hearing-impaired adults in a developed country. Trends Amplif 15(4):175–183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cox RM, Alexander GC. (2002) The International Outcome Inventory for Hearing Aids (IOI-HA): psychometric properties of the English version. Int J Audiol 41(1):30–35. [DOI] [PubMed] [Google Scholar]
  14. Ekberg K, Timmer BH, Francis A, Hickson L. (2023) Improving the implementation of family-centred care in adult audiology appointments: a feasibility intervention study. Int J Audiol 62(9):900–912. [DOI] [PubMed] [Google Scholar]
  15. Ferguson M, Brandreth M, Brassington W, Leighton P, Wharrad H. (2016) A randomized controlled trial to evaluate the benefits of a multimedia educational program for first-time hearing aid users. Ear Hear 37(2):123–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ferguson MA, Kitterick PT, Chong LY, Edmondson‐Jones M, Barker F, Hoare DJ; Cochrane ENT Group . (2017) Hearing aids for mild to moderate hearing loss in adults. Cochrane Database Syst Rev 2017(9):CD012023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fourie C, Mahomed-Asmail F, Oosthuizen I, Manchaiah V, Vercammen C, Swanepoel DW. (2024) Hearing aid benefit in daily life: a qualitative ecological momentary assessment study. Front Audiol Otol 2:e1397822.. [Google Scholar]
  18. Gallagher NE, Woodside JV. (2018) Factors affecting hearing aid adoption and use: a qualitative study. J Am Acad Audiol 29(4):300–312. [DOI] [PubMed] [Google Scholar]
  19. Gatehouse S. (2001) Self-report outcome measures for adult hearing aid services: some uses, users, and options. Trends Amplif 5(3):91–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gomez R, Habib A, Maidment DW, Ferguson MA. (2022) Smartphone-connected hearing aids enable and empower self-management of hearing loss: a qualitative interview study underpinned by the behavior change wheel. Ear Hear 43(3):921–932. [DOI] [PubMed] [Google Scholar]
  21. Graham B, Regehr G, Wright JG. (2003) Delphi as a method to establish consensus for diagnostic criteria. J Clin Epidemiol 56(12):1150–1156. [DOI] [PubMed] [Google Scholar]
  22. Greasley A, Crook H, Fulford R. (2020) Music listening and hearing aids: perspectives from audiologists and their patients. Int J Audiol 59(9):694–706. [DOI] [PubMed] [Google Scholar]
  23. Guerra-Zúñiga M, Cardemil-Morales F, Albertz-Arévalo N, Rahal-Espejo M. (2014) Explanations for the non-use of hearing aids in a group of older adults. A qualitative study. Acta Otorrinolaringologica 65(1):8–14. [DOI] [PubMed] [Google Scholar]
  24. Hall JA, Holmstrom AJ, Pennington N, Perrault EK, Totzkay D. (2023) Quality conversation can increase daily well-being. Commun Res 2023:009365022211393. [Google Scholar]
  25. Hickson L, Meyer C, Lovelock K, Lampert M, Khan A. (2014) Factors associated with success with hearing aids in older adults. Int J Audiol 53(Suppl 1):S18–S27. [DOI] [PubMed] [Google Scholar]
  26. Holman JA, Drummond A, Hughes SE, Naylor G. (2019) Hearing impairment and daily-life fatigue: a qualitative study. Int J Audiol 58(7):408–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Holman JA, Ali YHK, Naylor G. (2023) A qualitative investigation of the hearing and hearing-aid related emotional states experienced by adults with hearing loss. Int J Audiol 62(10):973–982. [DOI] [PubMed] [Google Scholar]
  28. Hornsby BW. (2013) The effects of hearing aid use on listening effort and mental fatigue associated with sustained speech processing demands. Ear Hear 34(5):523–534. [DOI] [PubMed] [Google Scholar]
  29. Hsu C-C, Sandford BA. (2007a) Minimizing non-response in the Delphi process: how to respond to non-response. Pract Assess Res Eval 12(1):17. [Google Scholar]
  30. Hsu C-C, Sandford BA. (2007b) The Delphi technique: making sense of consensus. Pract Assess Res Eval 12(1):1–8. [Google Scholar]
  31. Humes LE. (1999) Dimensions of hearing aid outcome. J Am Acad Audiol 10(1):26–39. [PubMed] [Google Scholar]
  32. Humes LE, Dhar S, Manchaiah V, et al. (2024) A perspective on auditory wellness: what it is, why it is important, and how it can be managed. Trends Hear 28:23312165241273342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kahneman D, Krueger AB. (2006) Developments in the measurement of subjective well-being. J Econ Perspect 20(1):3–24. [Google Scholar]
  34. Kalluri S, Humes LE. (2012) Hearing technology and cognition. Am J Audiol 21(2):338–343. [DOI] [PubMed] [Google Scholar]
  35. Keeney S, Hasson F, McKenna H. (2011) The Delphi Technique in Nursing and Health Research. West Sussex, UK: Wiley-Blackwell. [Google Scholar]
  36. Kitterick PT, Ferguson MA. (2018) Hearing aids and health-related quality of life in adults with hearing loss. JAMA 319(21):2225–2226. [DOI] [PubMed] [Google Scholar]
  37. Knoetze M, Beukes E, Manchaiah V, Oosthuizen I, Swanepoel DW. (2024) Reasons for hearing aid uptake in the United States: a qualitative analysis of open-text responses from a large-scale survey of user-perspectives. Int J Audiol 63(12):975–986. [DOI] [PubMed] [Google Scholar]
  38. Kochkin S. (2000) MarkeTrak V: consumer satisfaction revisited. Hear J 53(1):38, 40, 42, 45–46, 50, 52, 55. [Google Scholar]
  39. Korkmaz MH, Bayir O, Er S, et al. (2016) Satisfaction and compliance of adult patients using hearing aid and evaluation of factors affecting them. Eur Arch Otorhinolaryngol 273(11):3723–3732. [DOI] [PubMed] [Google Scholar]
  40. Laplante-Lévesque A, Hickson L, Worrall L. (2010) A qualitative study of shared decision making in rehabilitation audiology. J Acad Rehabil Audiol 43:27–43. [Google Scholar]
  41. Laplante-Lévesque A, Jensen LD, Dawes P, Nielsen C. (2013) Optimal hearing aid use: focus groups with hearing aid clients and audiologists. Ear Hear 34(2):193–202. [DOI] [PubMed] [Google Scholar]
  42. Lin FR, Pike JR, Albert MS, et al. ; ACHIEVE Collaborative Research Group . (2023) Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet 402(10404):786–797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lin T-C, Yen M, Liao Y-C. (2019) Hearing loss is a risk factor of disability in older adults: a systematic review. Arch Gerontol Geriatr 85:103907. [DOI] [PubMed] [Google Scholar]
  44. Livingston G, Huntley J, Sommerlad A, et al. (2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396(10248):413–446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Lockey K, Jennings MB, Shaw L. (2010) Exploring hearing aid use in older women through narratives. Int J Audiol 49(8):542–549. [DOI] [PubMed] [Google Scholar]
  46. Maidment DW, Ali YHK, Ferguson MA. (2019) Applying the COM-B model to assess the usability of smartphone-connected listening devices in adults with hearing loss. J Am Acad Audiol 30(5):417–430. [DOI] [PubMed] [Google Scholar]
  47. Manchaiah V, Swanepoel DW, Bailey A, Pennebaker JW, Bennett RJ. (2021) Hearing aid consumer reviews: a linguistic analysis in relation to benefit and satisfaction ratings. Am J Audiol 30(3):761–768. [DOI] [PubMed] [Google Scholar]
  48. Meibos A, Muñoz K, Twohig M. (2019) Counseling competencies in audiology: a modified Delphi study. Am J Audiol 28(2):285–299. [DOI] [PubMed] [Google Scholar]
  49. Mener DJ, Betz J, Genther DJ, Chen D, Lin FR. (2013) Hearing loss and depression in older adults. J Am Geriatr Soc 61(9):1627–1629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Meyer C, Hickson L, Khan A, Walker D. (2014) What is important for hearing aid satisfaction? Application of the expectancy-disconfirmation model. J Am Acad Audiol 25(7):644–655. [DOI] [PubMed] [Google Scholar]
  51. Michie S, Van Stralen MM, West R. (2011) The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci 6(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Mothemela B, Manchaiah V, Mahomed-Asmail F, Knoetze M, Swanepoel DW. (2024) Factors influencing hearing aid use, benefit and satisfaction in adults: a systematic review of the past decade. Int J Audiol 63(9):661–674. [DOI] [PubMed] [Google Scholar]
  53. Okoli C, Pawlowski SD. (2004) The Delphi method as a research tool: an example, design considerations and applications. Inf Manag 42(1):15–29. [Google Scholar]
  54. Oosthuizen I, Manchaiah V, Launer S, Swanepoel DW. (2022) Hearing aid experiences of adult hearing aid owners during and after fitting: a systematic review of qualitative studies. Trends Hear 26:23312165221130584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Oosthuizen I, Kumar LMS, Nisha KV, et al. (2023) Patient-reported outcome measures for hearing aid benefit and satisfaction: content validity and readability. J Speech Lang Hear Res 66(10):4117–4136. [DOI] [PubMed] [Google Scholar]
  56. Oosthuizen I, Swanepoel DW, Boyd RL, Pennebaker JW, Launer S, Manchaiah V. (2024) Exploring adult hearing aid user experiences: meaning extraction methods content patterns and associations with demographic and outcome variables. Int J Audiol 2024:1–11. [DOI] [PubMed] [Google Scholar]
  57. Pennebaker JW. (2011) The secret life of pronouns. New Scientist 211(2828):42–45. [Google Scholar]
  58. Pennebaker JW, Booth RJ, Boyd RL, Francis ME. (2015) Linguistic Inquiry and Word Count: LIWC2015. Austin, TX: Pennebaker Conglomerates. [Google Scholar]
  59. Picou EM. (2020) MarkeTrak 10 (MT10) survey results demonstrate high satisfaction with and benefits from hearing aids. Semin Hear 41(1):21–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Picou EM, Ricketts TA, Hornsby BW. (2013) How hearing aids, background noise, and visual cues influence objective listening effort. Ear Hear 34(5):e52–e64. [DOI] [PubMed] [Google Scholar]
  61. Powell C. (2003) The Delphi technique: myths and realities. J Adv Nurs 41(4):376–382. [DOI] [PubMed] [Google Scholar]
  62. Rayens MK, Hahn EJ. (2000) Building consensus using the policy Delphi method. Policy Polit Nurs Pract 1(4):308–315. [Google Scholar]
  63. Ross F, Wohllebe A, Diez E. (2022) The role of personal assistance in the uptake of smartphone-based Tele-Audiology—an extension of the technology acceptance model. Int J Interact Mobile Technol 16(4):18–31. [Google Scholar]
  64. Sarant JZ, Busby PA, Schembri AJ, Fowler C, Harris DC. (2023) ENHANCE: a comparative prospective longitudinal study of cognitive outcomes after 3 years of hearing aid use in older adults. Front Aging Neurosci 15:1302185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Saunders GH, Vercammen C, Timmer BH, et al. (2021) Changing the narrative for hearing health in the broader context of healthy living: a call to action. Int J Audiol 60(Suppl 2):86–91. [DOI] [PubMed] [Google Scholar]
  66. Sbalchiero S. (2018) Topic detection: a statistical model and a quali-quantitative method. In: Tuzzi A., ed. Tracing the Life Cycle of Ideas in the Humanities and Social Sciences. Cham, Switzerland: Springer International Publishing. [Google Scholar]
  67. Singh G, Lau S-T, Pichora-Fuller MK. (2015) Social support predicts hearing aid satisfaction. Ear Hear 36(6):664–676. [DOI] [PubMed] [Google Scholar]
  68. Solheim J, Kværner KJ, Sandvik L, Falkenberg E-S. (2012) Factors affecting older adults’ hearing-aid use. Scand J Disabil Res 14(4):300–312. [Google Scholar]
  69. Stephens SDG, Meredith R. (1991) Qualitative reports of hearing aid benefit. Clin Rehabil 5(3):225–229. [Google Scholar]
  70. Stone Fish L, Busby DM. (1996) The Delphi method. In: Sprenkle DM, Moon SM., eds. Research Methods in Family Therapy. New York: Guilford. [Google Scholar]
  71. Stull DE, Leidy NK, Parasuraman B, Chassany O. (2009) Optimal recall periods for patient-reported outcomes: challenges and potential solutions. Curr Med Res Opin 25(4):929–942. [DOI] [PubMed] [Google Scholar]
  72. Tausczik YR, Pennebaker JW. (2010) The psychological meaning of words: LIWC and computerized text analysis methods. J Lang Soc Psychol 29(1):24–54. [Google Scholar]
  73. Timmer BHB, Bennett RJ, Montano J, et al. (2024) Social-emotional well-being and adult hearing loss: clinical recommendations. Int J Audiol 63(6):381–392. [DOI] [PubMed] [Google Scholar]
  74. Tyler RS, Baker LJ, Armstrong-Bednall G. (1983) Difficulties experienced by hearing-aid candidates and hearing-aid users. Br J Audiol 17(3):191–201. [DOI] [PubMed] [Google Scholar]
  75. Vázquez-Ramos R, Leahy M, Estrada Hernández N. (2007) The Delphi method in rehabilitation counseling research. Rehabil Couns Bull 50(2):111–118. [Google Scholar]
  76. Vercammen C, Ferguson M, Kramer SE, et al. (2020) Well-hearing is well-being. Hear Rev 27:18–22. [Google Scholar]
  77. Vercammen C, Oosthuizen I, Manchaiah V, Ratinaud P, Launer S, Swanepoel DW. (2023) Real-life and real-time hearing aid experiences: insights from self-initiated ecological momentary assessments and natural language analysis. Front Digit Health 5:1104308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Wang X, Zheng Y, Liu Y, Lu J, Cui Z, Li Z. (2022) Effects of demographic, audiologic, and hearing-aid-related variables on the outcomes of using hearing aids. Eur Arch Otorhinolaryngol 279(8):3857–3865. [DOI] [PubMed] [Google Scholar]
  79. Weinstein BE, Sirow LW, Moser S. (2016) Relating hearing aid use to social and emotional loneliness in older adults. Am J Audiol 25(1):54–61. [DOI] [PubMed] [Google Scholar]
  80. Whitmer WM, Wright-Whyte KF, Holman JA, Akeroyd MA. (2016) Hearing aid validation. In: Popelka GR, Moore BCJ, Fay RR, Popper AN., eds. Hearing Aids. Switzerland: Springer International. [Google Scholar]
  81. World Health Organization. (2021) World Report on Hearing. Geneva, Switzerland: World Health Organization. [Google Scholar]
  82. Woudenberg F. (1991) An evaluation of Delphi. Technol Forecast Soc Change 40(2):131–150. [Google Scholar]
  83. Xue L, Le Bot G, Van Petegem W, van Wieringen A. (2018) Defining interdisciplinary competencies for audiological rehabilitation: findings from a modified Delphi study. Int J Audiol 57(2):81–90. [DOI] [PubMed] [Google Scholar]

Articles from Journal of the American Academy of Audiology are provided here courtesy of American Academy of Audiology

RESOURCES