Facilitators and barriers to performing dietary behaviors among chronic community-dwelling stroke survivors: A qualitative secondary analysis

Abstract

Background:

Following a healthy dietary pattern is recommended for secondary stroke prevention, but stroke-related impairments may hinder the performance of dietary behaviors (i.e., accessing, selecting, and preparing food).

Objective:

The purpose of this study was to characterize facilitators and barriers to performing dietary behaviors in stroke survivors.

Methods:

We completed a secondary analysis of focus group data using a qualitative descriptive approach and content analysis to examine how 15 chronic (≥6 months) community-dwelling stroke survivors and 10 care-partners perceived dietary behavior facilitators and barriers.

Results:

We identified three key themes. First, changes in body functions/structures (e.g., hemiparesis, balance) result in dietary behavior activity limitations (e.g., difficulty grocery shopping, meal preparation). Second, environmental supports (e.g., care-partners, adaptive equipment) and activity modification (e.g., using pre-cut foods, dining out) are used to overcome dietary behavior limitations. Third, negative affect (e.g., dissatisfaction, frustration) and activity limitations lead to participation limitations (e.g., not being able to perform dietary behaviors independently, not being able to socialize when dining out).

Conclusions:

Dietary behaviors are negatively impacted following stroke, but environmental supports and compensatory strategies may be implemented to overcome activity limitations. More research is needed to develop interventions to facilitate dietary behaviors and participation following stroke.

Introduction

An overall healthy lifestyle is important to reduce stroke recurrence.¹ Recently updated guidelines from the American Heart Association recommend following a Mediterranean-type diet, with an emphasis on monosaturated fat, plant-based foods, and fish consumption to reduce risk of recurrent stroke.² To consume a nutrient-rich diet, stroke survivors require the skills, abilities, and resources to access and prepare healthy food. However, many survivors experience barriers to following a healthy dietary pattern due to stroke-related impairments, including fatigue, hemiparesis, and perceptual deficits that make it difficult to engage in necessary dietary behaviors (i.e. behaviors necessary for accessing, selecting, and preparing food), such as grocery shopping, meal preparation, and cooking.^3,4

Ample research has been conducted regarding eating difficulties experienced by acute stroke survivors and survivors with swallowing difficulties (e.g. dysphagia),^5–7 but much less is known about how chronic (≥6 months) community-dwelling stroke survivors without swallowing difficulties perform dietary behaviors. Among chronic stroke survivors, quantitative studies have reported dietary behavior-related difficulties with decreased appetite, depression, reduced energy intake, dissatisfaction with meal preparation ability, and difficulty cooking and buying groceries.^3,6,8 Medin et al.^9,10 have published qualitative findings from community-dwelling stroke survivors at 3- and 6-months post-stroke, reporting dietary behavior-related difficulties in meal preparation, socializing during meals, decreased appetite, and requiring assistance from others for grocery shopping and cooking. These studies provide evidence of diminished dietary behavior performance in chronic stroke survivors, but a knowledge gap exists regarding facilitators and barriers to performing dietary behaviors in this population.

Previously, we conducted a qualitative study to explore how the Diabetes Prevention Program (DPP) could be adapted to accommodate stroke-related disability.¹¹ The DPP is a lifestyle intervention that helps individuals prevent or delay type 2 diabetes through weight loss, physical activity, and dietary behavior modification.¹² We conducted focus groups with stroke survivors and their care-partners to identify specific suggestions for how to adapt DPP curricular content related to physical activity and dietary behaviors to accommodate stroke-related physical, cognitive, and psychosocial impairment. Example suggestions included simplifying session content to reduce cognitive burden, and providing meal preparation strategies and adaptive equipment to compensate for upper-limb hemiparesis.¹¹ During focus groups, participants discussed barriers and facilitators to performing dietary behaviors, but that information was not published because it was beyond the scope of the original study’s purpose.

Understanding how stroke survivors perceive facilitators and barriers to performing dietary behaviors could inform development of interventions to compensate for stroke-related impairments and facilitate performance of healthy dietary behaviors. Therefore, the purpose of this study was to conduct a secondary analysis of the original study’s qualitative data to explore how chronic community-dwelling stroke survivors and their care-partners perceive facilitators and barriers to performing dietary behaviors. Because this was an exploratory secondary analysis, no a-priori hypotheses were identified.

Methods

This qualitative descriptive study is a secondary analysis of previously collected data that examined how to modify the DPP to accommodate stroke-related disability. A qualitative descriptive approach was chosen to guide data analysis. This approach is grounded in naturalistic inquiry and provides a comprehensive summary of an event or phenomenon while limiting researcher inference and data abstraction.¹³ Hallmarks of this approach include low levels of researcher interpretation and emphasis on a comprehensive summary of collected data.¹³ To further facilitate qualitative description and limit researcher inference, we used the International Classification of Functioning, Disability, and Health (ICF)¹⁴ framework, a valid, evidence-based framework¹⁵ that conceptualizes health and function as an interaction among health conditions, environmental factors, and personal factors to guide data analysis.

The original study did not publish findings regarding dietary behaviors post-stroke; rather, findings identified specific recommendations for adapting the DPP.¹¹ Therefore, this secondary analysis reports previously unpublished findings regarding dietary behaviors in community-dwelling stroke survivors. Because de-identified data were used, this study was exempt from IRB approval and written, informed consent was not required.

Participants

Fifteen stroke survivors and 10 care-partners were recruited from community-based stroke support groups in St. Louis, Missouri, USA, using purposive sampling. Individuals were eligible for recruitment if they were aged 30–80 years, ≥6 months post-stroke (stroke survivors only), able to walk ≥150 feet with or without an assistive device, and were community-dwelling.

Procedure (Original Study)

Eligible individuals attended one of four focus groups between November 2018-February 2019. After providing written, informed consent, participants provided demographic and health information (e.g., time since stroke, Body Mass Index [BMI]) and level of functional disability as measured by the Modified Rankin scale.¹⁶ Focus groups lasted 1.5–2 hours and were audio-recorded. Focus group participant composition varied across sessions: Group 1 (4 stroke survivors, 2 care-partners), Group 2 (6 stroke survivors, 5 care-partners), Group 3 (2 stroke survivors, 2 care-partners), and Group 4 (3 stroke survivors, 1 care-partner). We developed an interview guide using the ICF framework¹⁴ to generate questions for structuring focus group discussion in identifying facilitators and barriers to physical activity and dietary behaviors since these activities are extensively addressed in the DPP curriculum. Example interview questions included “What challenges do you experience currently or anticipate as part of the Diabetes Prevention Program to changing eating behaviors?”, “What challenges do you experience related to cooking at home, preparing meals, or shopping?”, and “What kitchen adaptations or adaptive equipment do you use?” Additional methodological details can be found in the original study publication.¹¹

Data Analysis

In the original study, audio recordings of all focus groups were transcribed. For this secondary analysis, analyses were conducted by the study investigator, RRB, and an occupational therapy graduate student, MBI. Nvivo 12 software (QSR International, Melbourne, Australia) was used by the researchers to independently code all focus group transcripts. Initially, deductive content analysis^17,18 was used to code transcripts according to the ICF constructs body functions/structures, activities (i.e., task or action execution), participation (i.e. participation in a life situation), environmental factors, and personal factors. This approach was chosen to categorize qualitative data using validated ICF constructs as a means of limiting researcher inference in the coding process. After this initial coding process, the researchers met to discuss the coded material and resolve discrepancies.

Next, the researchers used inductive content analysis¹⁸ to independently code data across ICF constructs to identify facilitators and barriers for performing dietary behaviors. The researchers again met to discuss the coded material, resolve discrepancies, and identify key themes from the coded data. Inductive content analysis was chosen to summarize qualitative data, consistent with a qualitative descriptive approach.

Study Rigor.

The rigor of this secondary analysis was enhanced using several approaches.¹⁹ First, deductive content analysis was conducted using the ICF framework, which is a valid evidence-based framework¹⁵ that is appropriate for examining dietary behaviors in the context of stroke and stroke-related disability from a biopsychosocial viewpoint. Second, during both deductive and inductive content analyses, participant responses were triangulated within and across the 4 focus groups, which promotes trustworthiness (i.e., validity) of study findings.¹⁹ Third, both researchers developed code sets and coded all transcripts and independently; thus, data were triangulated across researchers. The process of using deductive content analysis followed by inductive content analysis allowed the researchers to categorize the qualitative data into validated biopsychosocial ICF constructs prior to identifying relationships within and across constructs, which resulted in a comprehensive qualitative description of facilitators and barriers to the performance of dietary behaviors among a sample of individuals with stroke and their care-partners.

Reflexivity.

RRB has a background in sociocultural anthropology and has been involved in research for 10 years; he is also an occupational therapist and researcher, with expertise in secondary stroke prevention and stroke rehabilitation. MBI has undergraduate training in exercise science and psychology and is currently a student in an occupational therapy master’s degree program; she has professional experience working with adults with developmental disabilities. Both researchers are trained in theories and frameworks commonplace in the occupational therapy literature. Neither researcher has experienced a stroke nor self-identifies as having a disability. The researchers acknowledge that they can only approach the data from an etic perspective. To avoid over-interpretation of the data, the researchers chose to use a qualitative descriptive approach,¹³ including deductive content analysis using the ICF Framework,¹⁴ to facilitate data analysis.

Results

Demographic and health information of the 15 stroke survivors and 10 care-partners are described in the original study.¹¹ Briefly, mean age was 61 (SD: 9) years; 60% were male; race was 68% White, 24% Black, and 8% Asian; time since stroke (survivors only) was 4 (SD: 3) years; BMI was 29 (SD: 5) kg/m², indicative of overweight (i.e., BMI of 25 to <30 kg/m² is overweight²⁰); and 100% of survivors reported slight or moderate disability as measured by the Modified Rankin scale whereas only 20% of care-partners reported slight disability.

Deductive Content Analysis

Facilitators and barriers to performing dietary behaviors categorized according to ICF constructs are displayed in figure 1. Body functions and structures that hinder dietary behaviors included stroke-related changes in neuromuscular, vestibular, and sensory functions and appetite. Dietary behavior activity and performance limitations resulting from stroke included grocery shopping and shopping-related transportation and mobility, meal preparation and cooking, eating, dining with others, and meal routines. Environmental factors that facilitate dietary behaviors included care-partner assistance and adaptive equipment. In contrast, the personal factor negative affect (e.g., dissatisfaction, fear, frustration) negatively affected dietary behaviors.

Fig. 1.

Facilitators and barrirers to performing dietary behaviors categorized according to the the International Classification of Functioning, Disability and Health framework.

Inductive Content Analysis

When analyzing data to identify facilitators and barriers within and across ICF constructs, we identified three key themes: 1) changes in body functions/structures result in dietary behavior activity limitations, 2) environmental supports and activity modification are used to overcome activity limitations, and 3) negative affect and activity limitations lead to participation limitations. Generalized examples and specific example quotations supporting each theme are included below. Additional example quotations are provided in table 1.

Table 1.

Additional example quotations from participants illustrating each key theme.

Theme 1: Changes in body functions and structures result in dietary behavior activity limitations.	Theme 2: Environmental supports and activity modification are used to overcome dietary behavior limitations.	Theme 3: Negative affect and activity limitations lead to participation limitations
“I’ve gotten behind the wheel and there’s just something about driving with that left foot, I don’t feel comfortable. I feel like it wouldn’t, my mind wouldn’t tell me to move fast enough to stop in an emergency or something like that, I’m just scared.” (Stroke Survivor) “And I can do that because all I need for my balance is to hold on. Same way when I go to the grocery store, all I need is a cart, it’s kind of like a walker for me.” (Stroke Survivor)	“I think ours [splitting up tasks] would be 50–50, to be honest. Like while he is cooking… he can open up the package of bacon, I found scissors he can use, I try to find anything that lets him do it. Then, I might set the table and get the butter and stuff out… while he’s cooking I’m doing the other part.” (Care-Partner) “We have bought like a finger-touch, one-handed can opener. We have bought specific, bigger knives like a rocker-type.” (Care-Partner)	“Yeah, but something sweet up on the high shelf, I won’t even bother with it, because he’ll make a face at me or something so I don’t bother with it anymore.” (Stroke Survivor) “And the people feel so sorry for you, and you hate that. You don’t have to feel sorry for me, just see me as the same person. They look at you as that special person.” (Stroke Survivor)

Theme 1: Changes in body functions and structures result in dietary behavior activity limitations.

Participants reported experiencing difficulty grocery shopping (e.g., walking, pushing a grocery cart), driving to the grocery store, meal preparation tasks (e.g., cutting, chopping, measuring, pouring), cooking (e.g., holding a pan handle while stirring, lifting a heavy skillet), and eating (e.g. cutting food on a plate, difficulty bringing food to the mouth with the non-dominant upper limb because of a hemiparetic dominant upper limb) due to changes in body functions and structures. Hemiparesis of the upper limb made any activity requiring the use of two hands difficult to complete, and hemiparesis of the lower limb and balance problems affected transportation and mobility activities necessary for accessing foods at a grocery store. For example, one stroke survivor explained that eating was difficult due to hemiparesis affecting their doming upper extremity, “I’m left-handed and [since the stroke] I use my right hand to eat, and it’s harder to use my right hand and so I get tired [physically] of eating quicker, so I don’t eat as much as I used to, because of having to use the other hand.”

Changes in vision were reported by one participant, who reported difficulty reading nutrition labels. Decreased appetite following stroke was also reported, resulting in decreased frequency of meal consumption (i.e., two meals per day); some participants explained they were less hungry since their stroke, others explained that they were not as active and didn’t need the energy intake, and others explained that they no longer enjoyed eating. Related to eating less, some participants reported changes in taste such that food did not taste the same, resulting in a lessened desire to eat. As explained by one stroke survivor, “[After the stroke] everything tasted horrible; I didn’t eat nothing… it just don’t taste the same as it used to. I get a craving for something… and then it just tastes like crap.”

Theme 2: Environmental supports and activity modification are used to overcome dietary behavior activity limitations.

The most common environmental support reported by participants to facilitate performance of dietary behaviors was care-partner assistance. Care-partner assistance was exemplified through providing transportation to the grocery store; helping survivors reach for items above shoulder height, which often occurred for individuals seated in a wheelchair or a motorized shopping cart; measuring, cutting, and chopping foods; and managing pots and pans while cooking. Although survivors appreciated the assistance provided by care-providers, they also reported that such assistance resulted in feelings of dependency, which was a source of frustration and discouragement. As an example, one stroke survivor summarized her frustration when grocery shopping: “It’s horrible cause I have to rely on him [care-partner] for everything. He has to go get the [motorized] scooter and bring it to the car for me to get in.”

The second most common environmental support reported was assistive technology. Participants identified walkers and motorized grocery carts as useful for facilitating grocery shopping. To assist with meal preparation and eating, participants reported using rocker knives and single-handed cutting boards (i.e., cutting board with spikes) for stabilizing food while cutting, push-button electric can openers for opening cans without needing to use two hands, and Dycem for opening jars and preventing dinner plates from slipping while eating.

Activity modification was a compensatory strategy for facilitating dietary behaviors also reported by participants. The first type of activity modification was using aspects of the physical environment. For example, instead of opening a jar with two hands, several participants reporting stabilizing the jar between their legs and removing the lid with their non-paretic limb. For example, one stroke survivor explained how she cut food, “I’m left handed but I can’t hold [food] with my left hand, so I have to put it up against the refrigerator and slice it like this [miming a slicing motion with her right hand].” The second type of activity modification was limiting the amount of meal preparation necessary for cooking or eating. Example modifications included using pre-cut frozen foods, heating microwaveable meals, or purchasing fast food or take-out meals from a restaurant to eliminate meal preparation completely. One stroke survivor shared how they used frozen meals to reduce their care-partner’s responsibility for meal preparation, “I like the frozen stuff. That’s only to give him [care-partner] a rest sometimes because I know he gets tired of cooking ALL the time… So yeah, we buy a lot of frozen stuff.”

Theme 3: Negative affect and activity limitations lead to participation limitations.

Survivors reported negative feelings associated with activity limitations caused by stroke. These feelings included fear of driving due to safety concerns, fear and hesitancy to cook while home alone in case they dropped a pan or started a fire, and dissatisfaction and frustration with how meals tasted because one’s care-partner didn’t prepare meals the same way that the stroke survivor did prior to their stroke. As explained by one stroke survivor, “Now I can’t cook. He [care-partner] does all the cooking. It’s hard because he does what he can but the food ain’t the way I’d have made it, not the way I want it… A lot of it is me being picky, but it’s just that anything, ANYTHING that you eat, it’s just not the same as if you were doing it yourself.”

These negative feelings in combination with activity limitations resulted in participation limitations, such that survivors experienced diminished satisfaction with previously meaningful activities. Two stroke survivors reported that they no longer enjoyed grocery shopping because they could no longer shop alone and their care-partners acted like a “gate-keeper” to prevent them from purchasing unhealthy snacks. Several survivors reported that cooking was no longer a source of enjoyment because they couldn’t cook the same way they used to or had to change the ingredients they used to cook to follow a heart-healthy diet (e.g., eliminating salt from cooking but not liking salt-free seasonings).

Some survivors with moderate upper-limb motor impairment also reported that they no longer dined with others because the motor control and effort required for eating made it too difficult to socialize with others simultaneously. One stroke survivor explained, “I noticed that I don’t get to talk when there’s a group of us eating because it takes me so much longer to eat. I have to just constantly try to eat to get that fork up to my mouth without spilling it… I have to concentrate, I can’t be in the conversation.” As a result, they no longer enjoyed dining with others. Furthermore, some participants indicated that they felt pitied by others when dining out, which was another reason to avoid social interaction.

Discussion

To our knowledge, this is the first study to use the ICF framework to examine facilitators and barriers to dietary behaviors among chronic community-dwelling stroke-survivors. Our results provide evidence of the complex interplay among body functions and structures, activity, participation, environmental factors, and personal factors experienced by stroke survivors for performing dietary behaviors. Changes in body functions and structures may create barriers to performing dietary behaviors that result in activity and participation limitations (i.e. Themes 1 & 3), but survivors can use a variety of environmental supports and compensatory strategies to facilitate performing dietary behaviors and minimize activity and participation limitations (Theme 2).

Similar to the changes in body functions and structures identified by our participants, other studies have reported how stroke-related impairments affect dietary behaviors. Klinke et al.⁴ conducted a concept analysis across 33 studies of eating difficulties following stroke and identified many neurological deficits that create functional barriers to dietary behaviors, including impaired movement or motor control, changes in smell and taste, depression, and visual deficits. Given these deficits, it is unsurprising that studies of stroke survivors report difficulty with grocery shopping, preparing meals, and cooking among stroke survivors.^3,10

Utilizing environmental supports and compensatory strategies to improve activity performance is a hallmark of stroke rehabilitation; therefore, it is not surprising that participants implemented supports and strategies to enhance their performance of dietary behaviors. The role of care-partners in providing assistance for meal preparation, shopping, and transportation has been previously described, with studies reporting that up to 52% of survivors require assistance for meal preparation,²⁴ 56%−80% require assistance for shopping,^3,24 and 52%−57% require assistance for transportation.^24,25 While study participants viewed care-partner assistance as essential for facilitating these dietary behaviors, participants did not like feeling dependent upon their care-partners help, which is a justified concern because care-partner dependency is associated with decreased quality of life in stroke survivors.²⁶ It seems apparent that there is a need to balance the perceived benefits and drawbacks of care-partner assistance for performing dietary behaviors.

The use of adaptive equipment and activity modification for overcoming deficits in dietary behaviors is common in rehabilitation programs, where interventions for addressing dietary behavior deficits might include modification of the physical environment (e.g. kitchen) to accommodate mobility and motor control deficits, provision of adaptive equipment to accommodate hemiparesis, or simplifying recipes and cooking needs to be less cognitive demanding.^21,27 We did not ask participants where they learned about the compensatory strategies they identified, so we cannot comment on the source of their learning. Teaching such strategies is not unique to rehabilitation settings, however. At least one randomized controlled trial designed, in part, to improve dietary behaviors in individuals with mobility impairment utilized many of the strategies identified by our participants, including adaptive equipment and accessible grocery shopping,²⁸ suggesting that opportunities for learning to use compensatory strategies do not end when rehabilitation services are discontinued.

Emotional distress and changes in mental health and quality of life occur following stroke.²⁹ Our participants reported feelings of dissatisfaction, fear, frustration, and hesitancy with performing dietary behaviors—emotions that are consistent with emotional distress. All negative emotions were expressed in relation to activities that participants found meaningful, which is congruent with the ICF definition of “participation” (i.e. engaging in a life situation).¹⁴ Some activities were solitary (e.g. meal preparation and cooking) while others involved other people (e.g. grocery shopping, dining with others), but regardless of the activity, participants ceased to derive enjoyment. Our findings are similar to other studies that report negative affect related to participation limitations in driving,³² shopping,¹⁰ meal preparation,¹⁰ and dining with others⁹ following stroke.

Recommendations for Future Research

Given the difficulties surrounding dietary behaviors post-stroke, it is surprising that little research has been conducted to develop evidence-based interventions for enhancing the performance of dietary behaviors in stroke survivors. Multimodal (i.e. consisting of ≥2 intervention components) interventions for secondary stroke prevention and health promotion have been trialed that provide dietary counseling and advice,³³ but few studies provide hands-on training to develop functional skills for managing stroke-related changes in body functions and structures. Rimmer et al.’s (2000)³⁴ intervention included a weekly nutrition and cooking class for stroke survivors, but their relevant dietary behavior outcome dietary fat intake showed no change. Towfighi et al.’s (2020)³⁵ intervention included hands-on training in preparing nutritious drinks and meals and grocery shopping, but their dietary behavior outcome fruit and vegetable intake showed no change. The limited number of studies published that provide training for improving dietary behaviors suggests that additional intervention and outcomes research is needed in this area.

Additionally, the relationship of low appetite, decreased food consumption, and depression deserves further exploration. Several participants in our study expressly mentioned that they consumed food less frequently due to decreased appetite, but no participant mentioned depression as an explanation for decreased appetite or frequency of eating. Low appetite is commonly reported following stroke, and low appetite is associated with depression.²² A better understanding of depression, low appetite, and decreased food consumption could inform the development of future dietary behavior interventions.

Limitations

There are limitations inherent to this study. First, this a secondary analysis. The parent study on how to modify the DPP for stroke survivors was not specifically designed to explore dietary behaviors;¹¹ however, we heavily investigated dietary behaviors in our parent study because of its importance within the DPP. Sampling saturation is also a concern when conducting a secondary analysis. Although saturation was achieved in the parent study,¹¹ saturation cannot be directly addressed in this secondary analysis because focus groups did not explicitly explore dietary behaviors.

Second, characteristics of the study sample must be considered. Stroke survivors and care-partners were recruited and interviewed together during focus groups in the parent study. As a result, stroke survivors and care-partners may not have shared their thoughts as thoroughly as they might have if they had been interviewed separately. Additionally, the study sample was relatively homogenous and lacked diversity (i.e., a majority of White males), so data may not generalize to other demographic groups. It is important to note, however, that females and non-White participants were represented within our sample, and participants varied in age, time since stroke, BMI, and level of disability.

Third, the effect of age-related disability distinct from stroke-related disability was not accounted for in this study. The prevalence of functional disability is higher among older adults than among young and middle-aged adults,³⁶ and likely affects dietary behaviors. Future research should examine the independent and joint effects of aging- and stroke-related disability on dietary behavior performance.

Finally, our choice to use a qualitative description approach, and our use of deductive content analysis followed by inductive content analysis, resulted in an interpretation consistent with ICF framework constructs. Had no initial deductive analysis step been conducted, or had a different framework been used, the overall interpretation likely would be different. However, our use of the ICF framework was intentional to contextualize findings within a valid framework, to limit researcher interpretation, and to prevent over-interpretation of data, which may have occurred if a different approach had been used.

Conclusion

Dietary behaviors are negatively impacted by stroke-related impairments, but survivors and care-partners may use strategies for overcoming activity limitations. Furthermore, negative affect and activity limitations lead to diminished dietary behavior participation and satisfaction. The research literature is sparse regarding effective interventions for improving the full spectrum of dietary behaviors, including accessing and preparing healthy foods, with most interventions focusing on healthy dietary consumption. Future research is needed to develop and test effective interventions to facilitate dietary behaviors in stroke survivors.