Understanding the Meaning of Food in People With Type 2 Diabetes Living in Northern Appalachia

Focused ethnographic methods were used in this study to describe the meaning of food in adults with type 2 diabetes living in Northern Appalachia. Focused ethnography is a means to accessing beliefs and practices in the context in which they occur, thereby improving our understanding of factors surrounding health and disease (49). As in other forms of qualitative research, the data collected are rich in their descriptions of beliefs, experiences, and behaviors of a population. In focused ethnography, informants, or participants, may not be connected by the same culture in the broadest sense; however, they typically share behavioral norms, local knowledge, and a common language from living through a common experience (e.g., living with type 2 diabetes). For the purposes of this study, Northern Appalachia, and specifically Central Pennsylvania, was the sociocultural context of the informants. The diagnosis of type 2 diabetes was an additional sociocultural context such that participants shared the diagnosis of type 2 diabetes, diabetes education practices, diabetes beliefs (e.g., high blood glucose is damaging), diabetes language (e.g., diabetes terminology such as A1C, insulin, and hypoglycemia), and recommended diabetes management behaviors (e.g., following a healthful diet, getting regular physical activity, checking blood glucose levels, and practicing preventive foot care).

To maximize the phenomenon of interest, we employed judgmental sampling, a form of purposive sampling (50), to recruit community-dwelling adults who were English-speaking and mentally alert and had been diagnosed with type 2 diabetes for at least 1 year. Consistent with the philosophical underpinnings of qualitative research, we used judgmental sampling to identify participants who were able to richly describe the experience of living with type 2 diabetes. Judgmental sampling techniques are useful in learning more about how a group (i.e., people with type 2 diabetes living in Northern Appalachia) thinks and experiences a given sociocultural phenomenon.

We screened potential participants for eligibility and collected sociodemographic information by telephone. Participants were excluded if they reported being diagnosed with Alzheimer’s disease, other dementia, stroke, or cancer in the past year; having severe psychopathology (i.e., schizophrenia or bipolar disorder); alcohol or drug abuse; or impaired activities of daily living (e.g., difficulties with bathing, dressing, personal hygiene, or walking). We recruited participants via the university diabetes database and through direct mailings and flyers in the communities surrounding Pennsylvania State University in University Park. The university’s institutional review board approved the study. All participants provided written informed consent before participating and received compensation for their time.

A multidisciplinary research team developed a semi-structured interview guide and field-tested it with two people. An experienced qualitative researcher conducted all interviews, asking participants broad, open-ended questions about the role of food in type 2 diabetes self-management and the personal, social, and cultural meanings of food in their lives. Participants were also asked how the diagnosis of diabetes has changed the meaning of food in their lives. The interviewer used directive probes to elicit additional information and clarify questions. Interviews were conducted at community sites (recreation centers and churches) and university conference rooms and lasted 30–120 minutes each. All interviews were digitally audio-recorded and transcribed verbatim. The researchers performed quality checks of the transcribed files while listening to the interview recordings to validate the transcriptions. Participants’ names and identifiers were removed to ensure confidentiality.

Additionally, a reflexive journal (written by the interviewer) was maintained during data collection. The purpose of the reflexive journal was to record the interviewer’s thoughts, beliefs, and experiences during the research process. The personal notes were used to evaluate the investigator’s response to specific interviews. The value of the reflexive journal was to reduce personal bias and maintain objectivity (51). In addition, memoing was incorporated to record ideas and insights regarding the data. Memoing is a form of data coding completed during data collection. Memos were written to connect notations with shared meaning. Such reflections helped the investigator recognize the need for additional interviewing of the informants (52). Salient information from the reflexive journal and memos was integrated into the data analysis.

To analyze the qualitative data, we employed Spradley’s Developmental Research Sequence (DRS) of domain analysis, taxonomic analysis, componential analysis, and cultural themes (53). Spradley defines ethnographic analysis as a systematic examination of something to determine its parts, the relationships between the parts, and the relationships of the parts to the whole. The DRS method is an iterative process through which cultural ideas that arise in the field are transformed, translated, and represented during data analysis. This process involves sifting and sorting through pieces of data to detect and interpret thematic categorizations, searching for inconsistencies, and generating conclusions about what is happening and why. All levels of data analysis occur concurrently, informing each level throughout the analysis process. This process continues until data saturation is reached (i.e., until no new codes emerge).

For domain analysis, we identified categories evident in the transcripts with the intention of revealing participants’ role and personal, social, and cultural meanings of food. This analysis grouped words, phrases, and experiences into a collection of categories labeled by cover terms. For taxonomic analysis, we categorized the items in each domain to identify subsets of cultural knowledge. For semantic relationships, we noted links between the generic and specified concepts within and between categories. With componential analysis, we searched for attributes or components of meaning associated with the cultural categories. The objective of the componential analysis was to specify and understand the conditions under which the participants assigned something a specific name. Finally, we conducted thematic analysis to thread cultural themes via the domain, taxonomic, and componential analyses. The cultural themes represented propositions that guided the meaning of food and diet-related decision-making and, in turn, the optimal management of diabetes. Specifically, the research team selected themes that characterized the participants’ perceptions and experiences from data that occurred multiple times, both within and across transcripts. Each theme included words expressed by the participants to demonstrate that the concepts were grounded in the data. We used NVivo 10 software (QSR International, Victoria, Australia) to organize, sort, and structure the large amounts of text generated from the in-depth interviews.

To support credibility (i.e., validity), we triangulated the investigators with experts from different disciplines (i.e., behavioral medicine, clinical psychology, and gerontology); this analyst triangulation provided a check on selective perceptions, as well as a means of identifying blind spots in the analysis (54). Furthermore, three participants reviewed the findings to achieve participant corroboration (55). Transferability (i.e., external validity) was supported via rich descriptions and verbatim quotations included in the data. To support dependability (i.e., reliability) of the data, a researcher not involved with the study conducted an external audit both to examine the process of the research study and to evaluate whether the findings, interpretations, and conclusions were supported by the data (56). Confirmability (i.e., objectivity) of the data was supported by tracking the decision-making process with an audit trail, which is a detailed description of the research steps conducted from the development of the project to the presentation of its findings (54,57).

Nineteen adults with type 2 diabetes (mean age 68.7 ± 10.6 years, mean A1C 7.4 ± 1.4%, mean diabetes duration 10.9 ± 11.9 years, mean BMI 28.6 ± 5.2 kg/m², mean education level 15.7 ± 2.6 years, 78.9% prescribed oral hypoglycemic agents, 31.6% prescribed insulin injections, 52.6% female, 100% white, 84.2% married, and 68.4% retired) participated in in-depth, face-to-face interviews lasting between 30 and 120 minutes (Table 1). All but one participant were born and raised in Northern Appalachia; the one participant not born in the region had lived there for >10 years.

Below are descriptions of the themes that emerged from the data analysis. Transcript identifiers are used with quotations to indicate participant number, age, and sex.

All of the participants agreed that food was an important component of diabetes management. Participants described the role of food in self-management as operational:

Although most participants noted that the role of food was necessary to sustain life, many said diabetes changed how they approached food. Participants described making food-related decisions based on carbohydrate content and nutrition quality. The most commonly discussed barrier to following a healthful diet was difficulty with counting carbohydrates or following a low-carbohydrate meal plan:

Participants who were prescribed insulin reported challenges in the timing of meals and carbohydrate counting. As described by this participant, insulin demanded that her eating be dictated by time rather than by hunger or pleasure:

Others noted the challenge of insulin administration and how it requires a balance between what one eats and one’s subsequent self-care behaviors:

Interestingly, participants did not discuss many of the common barriers to healthful eating cited in the literature. For example, participants did not mention a lack of knowledge about dietary plans, meal planning and preparation, food shopping, cost, or age-related physical changes (e.g., dentation or altered taste or smell perception) affecting their ability to follow a healthful diet.

When asked to reflect on the personal and social meaning of food before they were diagnosed with diabetes, participants described it as communal and celebratory. To these participants, food was associated with family, religion, happiness, comfort, fond memories, and togetherness. Participants expressed joy with smiles, widened eyes, and looks upward while describing their reflections on food:

Participants expressed that they gained more awareness about their food choices after their diabetes diagnosis. Participants said they had to think more about portion sizes and carbohydrate content, which often increased their caution and restraint when eating:

Because they needed to pay more attention to nutritional content, participants said they no longer associated food with celebration or pleasure. In fact, most participants described a negative or ambivalent relationship with food, with many likening it to an enemy or poison. This negative relationship may have contributed to participants’ expressed sense of guilt or personal failure when they indulged in “bad” foods:

Increased awareness, consciousness, and concern about food may have influenced participants’ views of food as less enjoyable. One participant conveyed visible frustration, with restless hands and a raised voice while saying:

Another participant echoed this sentiment:

Pressure to eat at social events was prominent and may have contributed to participants’ difficulty following a healthful diet. Some individuals chose to avoid social events and holiday get-togethers so as to not be tempted by food or offend family members and friends who prepared meals. Participants also expressed being annoyed and discouraged, with their hands balled into fists, because they had to think about blood glucose levels and complications, whereas other family members and friends did not:

All 19 participants referenced fast food when describing the cultural meaning of food in the United States. Participants described how abundant and readily available food is in American culture. This abundance and availability was widespread in Northern Appalachia:

Participants felt that the social and celebratory aspect of food, coupled with the perceived fast-food American culture, was in direct conflict with the demands of diabetes. All participants referred to this socio-cultural-personal conflict and the challenge of following a healthful diet in a culture that advertises the opposite in many venues:

As people living with diabetes, participants noted that they had to exert more willpower toward food compared to the average American:

Several participants felt they were losing this battle to the media and society. These participants did not believe that a change in eating habits was possible for them; however, they wished for change in the younger generations and in people with prediabetes:

This qualitative study explored the meaning of food among adults with type 2 diabetes living in Northern Appalachia. All participants recognized the importance of following a healthful diet in diabetes self-care. They knew that healthful eating was essential for effective management. However, the diagnosis of diabetes changed not only how they approached food, but also what food meant to them. For many of the participants, eating was dictated by time rather than by hunger and pleasure. Furthermore, participants described feeling more aware of food choices, carbohydrate content, and portion sizes after their diabetes diagnosis. Participants also noted that, before their diagnosis of diabetes, the meaning of food was social and celebratory. After their diagnosis, participants felt that food was associated with negative emotions and, for some, a sense of personal failure. Several participants chose to avoid social events and holiday get-togethers so as to not be tempted by food. Finally, they viewed the perceived fast-food American culture as being in direct conflict with the demands of diabetes management. Participants reported struggling to follow a healthful diet in a culture that often advertised the opposite.

For these reasons, patients may benefit from having their health care providers (HCPs) identify and help to shift the existing cultural norms toward health promotion and disease prevention. Liburd (25) notes how this must occur in African-American culture, in which certain social settings contribute to unhealthful food choices for women, and the rituals of eating need to be recreated to emphasize health. HCPs must begin to address both the personal and sociocultural factors that guide the food choices and behaviors of patients with type 2 diabetes.

Our research builds on earlier work exploring perspectives of healthful eating among people in Appalachia (36). Consistent with this previous work, we found that church, family, and the media influenced participants’ meaning of food and perspectives of healthful eating. However, our sample did not discuss economic constraints or social policies (e.g., welfare), as was found in the study by Schoenberg et al. (36). Volunteer bias may partially explain why these factors were not discussed during the interviews; adults with major financial barriers may not have been able to participate because of transportation costs or reluctance to disclose their financial constraints.

Participants in this study reported changes in how and what they are eating; however, these changes were perceived as losses to their social environment. Participants described a sense of deprivation and unfairness, especially as they described not enjoying food as much and having lost a sense of celebration during events where food is served. HCPs are well positioned to give patients the message that having diabetes offers them an opportunity to consider whether they want to start eating more healthfully. This message may provide a teachable moment to educate patients that the average person in the United States is eating unhealthfully and that unhealthful eating patterns are contributing to obesity, heart disease, diabetes, some types of cancer, and other illnesses. Thus, the patient-provider discussion of diabetes can be the impetus for patients to move toward a more healthful eating style and more healthful behaviors.

Patient-provider discussions about behavior change can be reinforced through diabetes education. Diabetes education is an important component of diabetes care for all people with diabetes (58), and ongoing and repeated education is necessary to help people maintain what they learn during initial education (59). The American Diabetes Association (ADA) recommends individualized medical nutrition therapy (MNT) to help patients with diabetes to achieve treatment goals (60). The goals of diabetes nutrition education are to promote healthful eating and achieve glycemic, lipid, and blood pressure targets (60–63). For people with type 2 diabetes, nutrition therapy focuses on reduced caloric intake (63). Unfortunately, only half of diabetes patients receive diabetes education, and even fewer receive education from a dietitian or nutritionist (64).

In this study, none of the participants reported participating in formal diabetes self-management education (DSME) or MNT delivered by a registered dietitian; however, we did not review medical chart data to confirm the reported lack of DSME or MNT.

Previous research shows that diabetes nutrition education interventions result in average A1C reductions of 1–2% (range –0.23 to –2.6%, with reductions dependent on type and duration of diabetes) (60–63). A variety of dietary patterns, including low-carbohydrate, low–glycemic index, Mediterranean, and high-protein diets, have been shown to improve glycemia and reduce cardiovascular risk factors (e.g., lipid profiles, blood pressure, and weight) (65). Thus, diabetes patients can choose from various effective options to take their differing personal, social, and cultural food preferences into account (66).

Although having a variety of dietary options allows for consideration of patients’ food preferences, it also may create a mismatch between nutrition goals and the sociocultural environment (27,67). In the United States, the food industry has created a surplus of accessible, inexpensive, and calorie-dense foods. Peer group pressure and fast-food consumption further compound the dietary challenges faced by diabetes patients (68–70).

For this reason, diabetes patients may require education focusing on problem-solving skills, coping, and risk prevention (71). The ADA Standards of Medical Care in Diabetes recommend culturally relevant content and delivery of DSME (72,73). Current evidence-based MNT recommendations include cultural assessment for an individualized approach to eating patterns (61). Findings from our study may assist DSME and MNT providers in tailoring meal plans for adults with type 2 diabetes living in Northern Appalachia.

When education is provided in a culturally relevant context, the information about diabetes and self-care behaviors becomes more meaningful to individuals (74–77). In addition, education that incorporates scaffolding techniques and brief cognitive behavioral strategies may help participants identify and develop solutions to personal and sociocultural barriers to following a healthful diet (78). Strategies for scaffolding include 1) incorporating participants’ existing knowledge about diabetes and nutrition, 2) offering a motivational context to pique participants’ interest, 3) breaking complex behaviors/tasks into easier and more attainable steps, 4) modeling a behavior/task to participants before asking them to complete the same behavior/task on their own, 5) facilitating active engagement and participation, 6) using graphic organizers to provide a visual framework for assimilating new information, and 7) providing hints or partial solutions to problems or barriers. Finally, numerous psychosocial factors affect how well diabetes patients are able to follow a healthful diet (79–82). Thus, education that acknowledges and addresses the impact of social relationships, as well as depression, distress, and anxiety, on eating may assist patients in integrating effective self-care behaviors.

Limitations of this study include homogeneity of the study sample from one city in the Northern Appalachian region of the United States with regard to race/ethnicity, small sample size, participant self-selection, and the use of self-reported data. The all-white study sample is reflective of the racial and ethnic distribution in Northern Appalachia; however, our participants were more educated than the general Northern Appalachia population. The cross-sectional study design limited our ability to detect any causal associations between diabetes diagnosis, changes in the meaning of food, or changes in dietary self-care behaviors. The small sample size and participant self-selection also limited the generalizability of the findings. Self-reported data are vulnerable to social desirability bias. To minimize bias, the researchers informed participants that their responses were confidential and could not be linked back to their personal identity. The researchers also emphasized the voluntary nature of participation and explicitly informed the participants that their responses had no bearing on their diabetes care.

Future research with a larger, more heterogeneous sample should involve the collection of mixed-method data from adults with type 2 diabetes to assess objective measures of diabetes self-care behaviors, glycemic control, and physical and psychosocial changes. Additionally, research is needed to assess how the meaning of food changes over time and at different stages in the life span as well as its impact on self-care, glycemia, and well-being. Finally, the relative importance of these findings depends on a variety of factors including how recently individuals immigrated to the United States, their generational cohort, their degree of acculturation in the United States, and their self-identification as American or as a member of another culture.

For people with type 2 diabetes, healthful food choices and dietary modifications are of great importance to self-management. However, many patients struggle to manage food choices and to change the amount and type of carbohydrates they consume. The demands of diabetes may counter the social and celebratory aspects of eating and contribute to social isolation or promote a kind of anhedonia (reduced ability to experience pleasure) (83). The findings reported here expand the understanding of the meaning of food among adults with type 2 diabetes living in Northern Appalachia. Exploring the meaning of food among type 2 diabetes patients prompts HCPs to consider how patients’ personal and sociocultural beliefs toward food need to be embraced in diabetes education. Diabetes patients need strategies to help balance the demands of diabetes with eating and interpersonal relationships for the rest of their lives.

The authors thank the participants who shared their experiences and perceptions for this project.

This work was supported by the Heritage College of Osteopathic Medicine, National Institutes of Health training grant T32 DK007260, and a Pennsylvania State University College of Health & Human Development Alumni Association research grant.

No potential conflicts of interest relevant to this article were reported.

E.A.B. and M.D.R. researched data, contributed to the discussion, and wrote, reviewed, and edited the manuscript. L.A.W. and C-J.C. contributed to the discussion and reviewed and edited the manuscript. E.S. reviewed and edited the manuscript. E.A.B. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.