Materials and Methods
In this cross-sectional qualitative study, the multidisciplinary study team conducted in-depth interviews with self-identified African-American men and women over the age of 65. The study used the grounded theory method with two basic principles: (a) questioning rather than measuring and (b) generating hypotheses using theoretical coding. 22
Sample Selection
Participants were recruited using multiple methods, including distributing recruitment fliers at community health fairs, churches, senior centers, community centers, primary care clinics, dental offices, and senior residential facilities in the Portland metro area. Eligible participants were African-American, 65 years or older at the time of the interview, capable of making cognitive also used a snowball sampling technique, 23 which asked interview participants to recommend other African-American seniors who met study inclusion criteria. Participants were asked to subjectively select their social status in the community using the MacArthur Subjective Social Status Scale 24 and to complete the Mini-Cog test, a brief test of cognition. 25 Participants that met inclusion criteria were invited to take part in a one-on-one, one-hour semistructured interview at a central community location or at the researcher’s office.
Data Collection
From , a research assistant trained in qualitative methods (SH) obtained informed consent and conducted interviews using a semistructured interview guide. The interview guide was first pilot-tested and refined with the sample of African-American seniors at the 24th Annual Event of African-American Wellness Village organized by African-American Health Coalition in Portland, OR. The interview guide included a definition of oral health and asked questions about participant’s oral health history, personal history, current oral health, and where they sought dental care. Interviews were performed at a mutually agreed site by the interviewer and interviewee in and around Portland, OR. Each interview was audio recorded (average length 35 min), professionally transcribed, de-identified, and transferred to ATLAS.ti (Version 7.5.18) for data analysis and retrieval. Participants received a $25 gift card once the interview was completed. We enrolled 16 African-American seniors, out of which 10 participants (63%) were recruited using snowball sampling.
Data Analysis
We conducted interviews and analyses concurrently, monitoring for data saturation, the point at which no new findings emerge. 26 Analysis began after three months of data collection and assisted the study team in determining if saturation had been met. We used the six phases of thematic analysis to identify emergent themes: familiarizing with data; generating initial codes; searching for, reviewing, then defining, and naming themes; and producing a scholarly report. 27 The research team read early transcripts and defined a preliminary coding scheme. Transcripts were coded independently or by groups of two study team members. Over the course of multiple Se her nГҐ analysis meetings, the full study team met to discuss emergent themes and to resolve discrepancies through consensus. 28 During these meetings we organized emerging themes according to levels of the social–ecological framework: individual, relationship, community, and societal factors. 29 We then iteratively refined themes and produced a summary report.
Results
Demographic characteristics of the 16 participants are summarized in Table 1 . Mean participant age was 70.8 years (range: 66–78); 75% (n = 12) were females. All participants had health insurance through Medicare (n = 16, 100%; dental insurance is generally not included with Medicare and, therefore, it is an additional product that must be purchased and managed by the recipient). Nine participants (56%) had no dental insurance; the seven participants with dental insurance had it through Medicaid (n = 4, 25%) or through commercial plans (n = 3, 19%). Mean annual income was $32,000 across the participants; eight participants (50%) made $10,000 to $20,000 per year.