Preliminary Findings on Self-Reported Voice Disorders in Urban Ghana: A Qualitative Description Study

Qualitative Description Study Design

We used a qualitative description design as an efficient, inexpensive method for gathering knowledge from adults with a self-reported voice disorder living within metropolitan Accra—the capital of Ghana (Sandelowski, 2000). Qualitative description research is a naturalistic approach that uses an inductive process to develop a relativistic, context-specific understanding of a phenomenon from an individual’s subjective knowledge (Bradshaw et al., 2017). Qualitative description studies stay close to individuals’ meaning by providing less interpretation of data than in other qualitative designs (Sandelowski, 2000).

Researcher/Interviewer Characteristics

Structured interviews were administered by male speech-language therapists (SLTs) with more than 5 years of clinical experience (A.Z. and C.A.). One interviewer (A.Z.) from the United States had extensive training and work experience in international settings, was of European descent, and spoke American English. He held a Certificate of Clinical Competence from the American Speech-Language Hearing Association and was licensed to practice as an SLT in the United States. His clinical and research interests were in voice disorders. The other interviewer (C.A.) was from Ghana, of Ga ethnic origin, and spoke English and Ga. He trained as an SLT in the United Kingdom and practiced in the Ear, Nose, and Throat Department at Korle-Bu Teaching Hospital. He had a clinical interest in voice disorders. University of Ghana graduate students in SLT translated for three participants whose languages were not spoken by the primary interviewers (Twi [n = 2] and Ewe [n = 1], both of which are widely spoken in metropolitan Accra). Interviewers transcribed participants’ responses contemporaneously on paper.

Study Context

This study followed the Declaration of Helsinki principles (World Medical Association, 2013). Verbal consent to participate was obtained before administering study procedures. The study was conducted on a single day at Korle-Bu Teaching Hospital’s Speech and Hearing Center in Accra. The interview tool was prepared by the first and last authors (C.A. and A.Z., respectively) and peer reviewed by two University of Ghana SLT faculty.

Sampling Strategy and Recruitment

This qualitative description study used purposive sampling to identify individuals above 17 years of age with a self-reported voice disorder who resided in metropolitan Accra. The research occurred during a free screening at the Speech and Hearing Center that was offered to the public to increase awareness about voice disorders. Radio announcements, paper flyers, and electronic postings occurred the week prior to the event and presented the screening’s purpose, date and time, location, and organizing bodies. Participants enrolled in the study consecutively upon arrival to the site.

Participants were administered the Voice Handicap Index–10 (VHI-10) as a screening tool to identify individuals at risk for a voice disorder (Rosen et al., 2004). The VHI-10 is a valid and reliable patient-reported outcome measure (PROM) that asks about physical, emotional, and functional aspects of a voice disorder. Scoring of the VHI-10 is based on a 5-point Likert-type scale with rating options from 0 = never to 4 = always. The 10 ratings of the VHI-10 items are summed to derive a self-reported vocal handicap score. For this study, a VHI-10 score ≥7.5 was selected as a specificity cut-off score due to its excellent classification abilities in a large endoscopically diagnosed patient sample (Behlau et al., 2016).

Units of Study

Qualitative description studies tend toward smaller sample sizes (Bradshaw et al., 2017; Sandelowski, 2000). This study enrolled a small sample of 10 adult Ghanaians who presented to a screening for voice. Table 1 provides participant characteristics. Reported medical history revealed a relatively healthy set of ambulatory, community-dwelling urban Ghanaians. Four of the 10 individuals who were screened were identified as having a self-reported voice disorder (VHI-10 ≥ 7.5; Behlau et al., 2016). The average age of participants with a self-reported voice disorder was 39.0 ± 12.1 years. All participants with a self-reported voice disorder completed tertiary education and spoke English.

OPEN IN VIEWER

Table 1.

Participant Characteristics by VHI-10 Group.

Participant ID	Age (years)	Sex	Occupation	Education	VHI-10 (out of 40)	Language	Interviewer
High VHI-10 group	39.0 ± 12.1				17.5 ± 9.7
4	51	M	Travel agent	Tertiary	14	English	A.Z.
6	25	F	Student/Sales	Tertiary	11	English	A.Z.
7	47	F	Nurse	Tertiary	13	English	C.A.
9	33	M	Web developer/Sunday school teacher	Tertiary	32	English	A.Z.
Normal VHI-10 group	34.8 ± 8.9				2.3 ± 2.4
1	38	M	Nurse	Tertiary	0	English	A.Z.
2	31	F	Clinical psychologist	Tertiary	3	English	A.Z.
3	29	F	Fashion student	Secondary	4	Twi	C.A.
5	23	F	Reporter	Tertiary	6	Twi	C.A.
8	48	F	Trader/Grocer	Secondary	1	Ga	C.A.
10	40	F	Audiologist	Tertiary	0	English	A.Z.
Overall group, N = 10	36.5 ±9.9				8.4 ± 9.8

Note. VHI-10 = Voice Handicap Index–10 (40 represents maximum handicap); M = male; F = female.

Interview Corpora

During structured interviews, broad information on voice disorders was probed for based on a biopsychosocial classification framework of health and health-related domains, the International Classification of Functioning, Disability and Health (ICF; World Health Organization, 2013, 2017). The ICF considers impairments in physiological body functions and anatomical structures, limitations in carrying out activities, and restrictions in participating in life situations in addition to environmental and personal context factors. The ICF framework offers an opportunity to gain insight into the way people think about a disability by analyzing textual information elicited from interviews on health and health-related domains.

Drawing on the ICF framework, the breadth and depth of questions in this study allowed for rich exploration of an array of topics related to voice disorders (see Figure 1; World Health Organization, 2013, 2017). Information on symptoms (impairments) included voice problem onset and duration; issues with pitch, loudness, and voice quality; as well as any other self-reported symptoms related to voice. Voice use information (activity limitations) included voice habits at work, vocal loudness during typical use, duration of speaking time, and vocal demands. Participation restrictions included in ability to engage in activities requiring voice as well as absences at work, losses of income, and career changes as a result of voice problems.

OPEN IN VIEWER

Figure 1.

Framework for understanding self-reported voice disorders in Ghana.

Lifestyle topics (environmental and personal factors) included tobacco and alcohol use, hydration, eating habits, employment, home and work noise levels, and church worship patterns. Other environmental and personal factors that were queried included medical information (current medications, reflux-related symptoms [heartburn, globus sensation, chronic throat clearing, effort with swallowing], respiratory or stomach problems, head, neck, or chest injury and surgery, prolonged intubation, laryngitis, sinus infections, and post-nasal drainage), information on management (treatments sought and received as well as associated barriers to health care), and demographics (age, gender, educational level, and employment status).

Study Procedures

After establishing rapport and credibility, participants completed the VHI-10 in English (Rosen et al., 2004). English is the official language of Ghana and widely spoken in metropolitan Accra; however, those unable to read or understand English were provided with verbal translations of the 10 items to rate in their native language with the typed English scale presented as a visual aid. All participants rated VHI-10 items related to physical, emotional, and functional aspects of a voice disorder using the 0 to 4 Likert-type scale. Then, an SLT completed a 30-min structured interview in a quiet room with the participant seated across a small table. Interviews were not audio- or video-recorded to improve the participants’ candidness in their responses (Jones et al., 2013; Wylie et al., 2017). The structured interview was preferred over semi- or unstructured interviews for the ease of documenting responses on paper contemporaneously.

The SLT followed an interview tool adapted from published research that consisted of the same series of mostly closed-ended questions (Roy et al., 2005). The initial portion of the interview sought information on symptoms and impact before moving onto treatment-seeking behaviors in a natural progression. Key questions about voice were queried earlier in the interview to reduce the likelihood of respondent fatigue. Demographics were accessed at the end. Appropriate response categories were provided based on current understanding of voice disorders; other was also a response option. Based on a participant’s response, the SLT was guided to the next appropriate question. Interviewers verbally confirmed participant responses at the time of data collection.

Bias and Trustworthiness

To reduce cross-cultural biases, cultural background research on Ghana was conducted when designing the questionnaire followed by editing from University of Ghana faculty in Accra. Trustworthiness was improved by assigning three-digit participant codes to maintain anonymity. Researchers used de-identified data when conducting analyses to remain blinded to participants. Finally, transparency was increased by following recommendations outlined in the document Standards for Reporting Qualitative Research (O’Brien et al., 2014).

Data Processing

Anonymized and transcribed responses were coded and entered into a data management spreadsheet for content analysis by a research assistant who was a US SLT graduate student (A.H.). The last author (A.Z.) blindly verified responses were coded and entered correctly. Codes included 0/1 for dichotomous variables and a series of numeric values for categorical variables with more than two response options. Summary values were generated (frequencies, f; percentages, %) and used in content analysis to identify themes and subthemes. Quantitative data, such as age and VHI-10 scores, were limited to descriptive analyses with reporting of central tendency and variability to appreciate any differences in characteristics among this sample of adult urban Ghanaians.

Qualitative Data Analysis

Content analysis was completed by the last author (A.Z.) performing quantification of the coded interview data. The interview transcripts from the six individuals who scored within normal limits on the VHI-10 (<7.5) also underwent content analysis. The availability of their knowledge permitted an iterative analysis process that yielded a more precise interpretation of themes for adults with self-reported voice disorders (Bradshaw et al., 2017; Sandelowski, 2000).

The iterative process for establishing themes involved a two-step procedure in which thresholds were applied to variables. Variables that met an 80% threshold were identified in participants with high VHI-10 scores (high VHI-10 group) as well as those whose VHI-10 scores were within normal limits (normal VHI-10 group). This threshold was met when the same responses were provided by three out of four participants in the high VHI-10 group and five out of six participants in the normal VHI-10 group. Variables that met the 80% threshold in both groups were determined to be themes shared by the entire sample of adult Ghanaians. A variable that met the 80% threshold in the high VHI-10 group exclusively, that is, responses for that variable were below a 33% threshold for participants in the normal VHI-10 group, was determined to be unique themes for participants identified to have a self-reported voice disorder. Straight descriptions were generated of each high VHI-10 participant’s account, highlighting their distinct personal experience with their self-reported voice disorder (Bradshaw et al., 2017; Sandelowski, 2000).

Themes for Adult Ghanaians With a Voice Disorder

One theme that emerged was the experience of recurring problems with voice. Although multiple episodes of voice problems were common (3/4), their course tended to be short, lasting less than 4 weeks (3/4). Two participants in the high VHI-10 group reported their voice problem occurred off and on, whereas one participant in the high VHI-10 group indicated their voice problem had been continuous. All four participants in the high VHI-10 group endorsed a voice problem at some point over their lifetime.

Another theme that emerged was experiencing a constellation of voice symptoms. Three participants in the high VHI-10 group reported voice quality–related symptoms, of which vocal strain and breathy voice quality were exclusive to this group. Three participants in the high VHI-10 group reported more effort to voice, and two participants reported discomfort with voicing that was exclusive to the high VHI-10 group. All participants in the high VHI-10 group reported loudness-related complaints. Loudness-related symptoms included whisper voice (n = 2), difficulty being loud (n = 2), being too soft (n = 1), or being too loud (n = 1).

The third theme that emerged was the experience of vocal handicap. All participants in the high VHI-10 group reported difficulty with their voice in talking to groups or in noisy settings, and this finding was exclusive to this group. Three participants in the high VHI-10 group did not consider themselves to be talkative. Vocal activity restrictions were common, and only one participant in the high VHI-10 group described that their voice disorder affected their participation in life situations.

The final theme that emerged was the experience of other health-related issues. The entire high VHI-10 group reported a history of stomach problems. Three participants in the high VHI-10 group indicated problems with acid reflux/heartburn, whereas one participant indicated gastritis. Three participants in the high VHI-10 group reported they were prescribed one to two medications. Care for voice disorders was reported by one participant in the high VHI-10 group, which included prescription medication. Three participants in the high VHI-10 group reported barriers to voice care that included limited financial resources (n = 2) and lack of awareness about treating voice disorders using SLT (n = 2).

Straight Descriptions of Each Voice Disorder

Rich descriptive summaries were generated from the interview responses of participants in the high VHI-10 group about their self-reported voice disorder (see Table 3). Although some parallels existed, each descriptive summary showed unique experiences with self-reported voice disorders among this sample of urban Ghanaians. These rich descriptive summaries highlight the importance of obtaining a person’s perception of their voice problem and provide preliminary evidence of the subjective nature of voice disorders.

OPEN IN VIEWER

Table 3.

Straight Descriptions of Self-Reported Voice Disorders.

ID	Straight description
P4	A 51-year-old, non-smoking male with tertiary education who worked as a travel agent stated he had an 8-month voice problem. When his voice problem started, he had ear pain and pain when talking, both of which resolved. He described his current voice problem as a chronic, gradual condition that occurs off and on. His voice symptoms included difficulty being loud, whisper voice, roughness, strain/effort, and discomfort with voicing. He noted his voice deteriorated with use but he had periods with normal voice. He reported his voice problem limited his ability to talk in noisy settings and to groups. P4 reported that he consumed more than 2 L of water and three caffeinated beverages per day. He noted he had five alcoholic beverages per week. He stated that dinner was his heaviest meal of the day but he does not sleep within 3 hr of eating. He described some voice use at work, a moderate loudness during typical voice use, and that he talks on the telephone often. He reported that he attended church once weekly. He described a history of acid reflux, and took two medications. He had sought medical treatment for his voice problem. He reported that financial barriers restricted him from seeking medical attention.
P6	A 25-year-old, non-smoking female with tertiary education who was a student with part-time work as a kiosk vendor selling alcoholic beverages stated she had a voice problem in the past that lasted less than 4 weeks. She described her current voice symptoms as low pitch, too soft, and rough voice quality. She considered herself a talkative person but noted her voice problem limited her ability to talk in noisy settings and to groups. P6 reported that she consumed more than 2 L of water. She stated lunch was her heaviest meal, and she sleeps within 1 to 3 hr of eating dinner. She reported passive exposure to secondhand smoke for 20 years but not within the past 2 years. She described a lot of voice use at work and on weekends (12 hr) and talking for long periods without a break. She described a noisy home environment, a moderate loudness during typical voice use. P6 reported she never attends church but often talks on a telephone. She described a history of acid reflux with pain in the diaphragm at night and sometimes difficulty breathing. She took two medications. She had not sought treatment for her voice problem and reported her lack of awareness of treatment for voice problems was a barrier that restricted her from seeking medical attention.
P7	A 47-year-old, non-smoking female with tertiary education who worked as a nurse stated she had a 3-week problem with her voice. She described her current voice problem as a gradual condition that occurred off and on but that never lasted more than 4 weeks. She reported that using her voice took more effort, and voice symptoms included being too loud. She reported her voice problem limited her ability to talk in noisy settings and to groups. She stated that she missed work due to her voice problem in the past year. P7 reported that she worked 9 hr a day, 6 days a week, and used her voice a lot at her job for long periods without taking a break from speaking. She described soft loudness during typical voice use, and that she talked on the telephone often. She stated that she sung recreationally and attended church once weekly. She described a history of gastritis and took one medication. She reported that she did not drink alcohol or caffeinated beverages, and did not sleep within 3 hr of eating dinner. She stated that she had not sought treatment for her voice problem.
P9	A 33-year-old, non-smoking male with tertiary education who worked as a web developer and Sunday school assistant stated he had an 18-year problem with his voice. He described his current voice problem as chronic and continuous that started gradually and has not changed. Although he had periods of normal voice, he experienced more effort to use his voice, discomfort with voicing, and his voice deteriorated with use. His symptoms included high pitch, loss of high range, difficulty being loud, whisper voice, and a breathy, shaky voice quality with vocal strain. He also reported difficulty breathing with running out of air when talking, noisy inhalation when exposed to air conditioning and smoke, and excessive coughing when exercising. He reported his voice problem limited his ability to talk in noisy settings and to groups. He also described other activity limitations due to his voice including talking on a telephone, talking in a one-to-one conversation, yelling or cheering, and singing. He stated that he wanted to teach but cannot because of his voice problem and stated that he had lost income due to his voice problem. P9 reported 1 to 2 L of water consumption per day as well as one coffee each morning. He stated that dinner was his heaviest meal of the day and he slept within 1 to 3 hr of eating. He reported soft loudness during typical voice use, talked for long periods of time without taking a break from speaking, but had very little voice use at work. He sang recreationally and stated he attended church several days each week. He reported a noisy home environment. He described a history of acid reflux, but was not taking any medications. He described a constellation of symptoms including swallowing difficulties and pain with swallowing, chronic throat clearing, the feeling of something stuck in his throat, and heartburn. He has sought treatment for his voice problem, having been evaluated by an ENT 10 years prior. He reported a lack of awareness of voice therapy for voice problems in addition to finances as barriers that restricted him from seeking medical attention.

Themes for Adult Ghanaians Regardless of Voice Status

Content analysis of interview responses from the entire sample of adult Ghanaians resulted in two themes related to voice. The first theme was high vocal demands. Most participants reported a lot or some voice use at work, whereas two participants reported little occupational voice use. Many participants reported that they used the telephone often (8/10) and talked for long periods without a break (7/9). Most participants (8/10) attended church or worshipped at least weekly.

The other theme was adequate voice care. Seven (78%) participants (five without a voice disorder) reported consumption of greater than 2 L of non-caffeinated fluids per day. Caffeine intake was limited in the seven participants that indicated consumption. All participants denied smoking tobacco (lifetime), and most participants (7/10) did not drink alcohol. Two participants in the normal VHI-10 group (33%) reported less than one drink per week, whereas one participant in the high VHI-10 group (25%) reported five alcoholic drinks per week. Nine participants considered their typical loudness to be soft or moderate, and only one participant who was in the normal VHI-10 group reported being loud or very loud.

Vocal Impairments and Their Restrictions on Vocal Activities and Life Situations

One theme was the experience of recurring problems with voice. Problems with voice were episodic for two participants in the high VHI-10 group, and one participant experienced a continuous issue with their voice. In terms of vocal functioning, the second theme was the experience of a large number of voice-related symptoms. Voice quality–related symptoms and complaints of vocal effort predominated, and loudness-related symptoms were commonly reported. In summary, problems with voice for this sample of urban Ghanaians involved multiple symptoms that occurred mostly episodically.

The third theme that emerged was limitations to vocal activities and restrictions in life situations. Specific activities affected by vocal disability were talking in groups (100% reporting difficulty) and talking in noisy environments (100% reporting difficulty). The finding that participants in the high VHI-10 group felt handicapped by their voice in groups or noisy environments is not surprising given the sound pressure requirements to be heard in those contexts. The consequences of one participant’s voice disorder were missed work, job changes, and a loss of income as a result of not being able to teach. This participant’s restriction with teaching highlights the occupational impact of voice disorders (Roy et al., 2005).

Three participants in the high VHI-10 group indicated they were not talkative. This finding may reflect a negative consequence of voice disorders (Roy et al., 2005). Another explanation is that different patterns of voice use are associated with different voice disorders. Roy and colleagues (2000) observed that extroversion was associated with vocal fold nodules, whereas introversion was associated with functional dysphonia characterized by muscle tension. Taken together, self-reported voice disorders in this sample of urban Ghanaians resulted in non-trivial personal and professional consequences and negatively affected voice-related quality of life.

Environmental and Personal Context Factors

The final theme that emerged was the experience of other health-related issues with barriers to management. The entire high VHI-10 group reported a history of stomach problems, of which acid reflux (n = 3) was common. Three participants in the high VHI-10 group reported pharmaceutical management including one participant who was prescribed medication for acid reflux. Management of voice and other communication disorders has been reported to consist mostly of medical interventions (Fagan & Jacobs, 2009; Wylie et al., 2017).

Financial barriers and lack of awareness of management for voice disorders by SLTs were described by participants in the high VHI-10 group. Few SLT services exist to manage voice and other communication disorders among Ghana’s 10 regions; all are concentrated in urban settings and accessed primarily by patients with sufficient financial resources (Crowley et al., 2013). Perhaps, due to such barriers, Ghanaians with voice and other communication disorders have been reported to seek treatment later in their disease process (Fagan & Jacobs, 2009). Kitcher and colleagues (2006) found in their laryngeal cancer cases the average duration of hoarseness before presentation ranged from 11 to 17 months and was as high as 60 months. These issues emphasize a need for further research on the obstacles faced by Ghanaians in accessing SLT services for voice disorders.

Shared Experiences With Voice in Urban Ghana

Two themes emerged for the entire sample of urban Ghanaian adults. The first theme related to high vocal demands. The group described a large number of daily activities that required voice. Many participants reported frequent telephone use (8/10), worshipped at least weekly (8/10), and talked for long periods without a break (7/9). Most participants reported a lot or some voice use at work, whereas only two participants reported little occupational voice use. Given the increased risk for voice disorders with heavy vocal demands (Roy et al., 2005), potential risk factors for voice disorders in Ghana need to be examined further.

The second theme for this sample of adult urban Ghanaians related to adequate voice care. Aside from the heavy vocal demands, the group generally described behaviors that were associated with maintaining vocal health. Participants mostly reported a soft to moderate level of vocal loudness when speaking (90%). They indicated a high level of hydration, and alcohol consumption was reported to be minimal. None of the participants in this sample reported any prior or current tobacco use, a finding that is in line with data showing only 3.8% of Ghanaians are reportedly current smokers (8.9% male, 0.3% female) and less than 10% reported having ever smoked (22.0% male, 1.2% female; Owusu-Dabo et al., 2009). These preliminary findings suggest that protective factors helped to maintain the vocal health and wellness of many from this sample of adult Ghanaians.

Implications for Voice Disorder Management in Ghana

Management of voice disorders in Ghana is limited by few university-trained SLTs, and SLT training programs in Ghana are only commencing (Crowley et al., 2013; Wylie et al., 2014, 2016). Increasing the rehabilitation workforce in Ghana to provide individualized services for voice disorders that meets population needs will take considerable time. Efforts in Ghana likely require a broader, public health approach of community-level SLT services for those at risk for a voice disorder (McAllister et al., 2013).

Community-level SLTs can have a major impact in a public health role. The public health SLT can perform outreach to develop positive attitudinal responses to voice disorders, provide education on voice disorders that includes early detection, and offer vocal injury prevention to at-risk groups of Ghanaians such as teachers and preachers (Wylie et al., 2018; Ziegler, 2020). The findings of this study indicate the need for a multi-dimensional voice evaluation in individuals who access care for voice problems (i.e., tertiary prevention/treatment). Finally, study findings suggest PROMs (e.g., VHI-10) can yield information about the symptoms and impact of voice problems that complements clinical signs of voice disorders obtained from auditory-perceptual evaluation of the voice and visual inspection of the larynx. PROMs can be re-administered post-intervention to appraise the patient’s perception of the response to treatment.

Limitations

Several limitations deserve mention. Structured interviews involved closed-ended questions that may have limited information sharing. Participants were not contacted to confirm interpretation of study findings and interviews were not recorded, both of which have the potential to affect the accuracy of reporting on participants’ experiences. Using the VHI-10, which has never been validated for administration in Ghana, may have introduced cross-cultural bias. Finally, this sample of participants primarily consisted of educated, young, and middle-aged adult females who resided in or near Accra, which does not fully reflect the diversity of Ghanaian society in terms of educational level, age, gender, and geographical region.

Future Directions

Large-scale epidemiological research is needed to examine the risk factors for voice disorders in Ghana and their burden, including in occupational voice users such as teachers and preachers. Such research will provide targets for cost-effective prevention programs for voice that can be implemented by public health SLTs and tested for effectiveness on a national scale.