Cardiac rehabilitation (CR) is an evidence-based, standard of care for patients with CVD(Reference Piepoli, Hoes and Agewall1). Comprehensive CR should include exercise training, risk factor modification, psychological support, patient education and dietary recommendations(Reference Piepoli, Hoes and Agewall1). Dietary recommendations are considered a quality indicator for these programmes(Reference Butscher, Keisha and Richard2,Reference Moghei, Oh and Chessex3) , as diet has been identified not only as a primary modifiable risk factor but also as a determinant of secondary CVD events in patients with existing CVD(Reference Anand, Hawkes and de Souza4,Reference Trichopoulou, Costacou and Bamia5) . Despite this, dietary recommendations and education are inconsistent within CR Guidelines(Reference Sofi, Abbate and Gensini6).
Generally, a healthy dietary pattern that reduces cardiovascular risk includes higher consumptions of whole grains, vegetables, fruits, legumes, nuts and olive oil, moderate consumption of fish and dairy products, and a low intake of sugars, sweets, and red or processed meats(Reference Anand, Hawkes and de Souza4). Foods that are naturally low in Na, sugar and saturated fat, which contain important sources of vitamins, minerals, antioxidants, mono-and-poly unsaturated fat acids and fibre, ultimately contribute to a wide range of health benefits(Reference Trichopoulou, Costacou and Bamia5,Reference Martinez-Gonzalez and Bes-Rastrollo7–Reference Vitale, Masulli and Calabrese9) . Most recently, dietary recommendations have shifted from single nutrients to adopting food-based recommendations emphasised in the Mediterranean Diet, the Dietary Approaches to Stop Hypertension Diet, Vegetarian or Vegan diets and the Portfolio Diet(Reference Trautwein and McKay10).
The Mediterranean Diet, which consists of fish, unsaturated fat, whole grains, fruits and vegetables, nuts and legumes, has promoted favourable benefits in cardiovascular risk profiles, glucose control and BMI(Reference Appel, Moore and Obarzanek11). The Dietary Approaches to Stop Hypertension Diet emphasises fruits and vegetables, low-fat dairy foods and whole grains and has also attracted attention due to its beneficial effects on blood pressure and its potential to reduce risk of left ventricular dysfunction(Reference Appel, Moore and Obarzanek11–Reference Chiavaroli, Viguiliouk and Nishi13). The Vegetarian Diet is based on plant foods such as cereals, legumes, fruits, leafy vegetables, nuts, seeds and sea vegetables. The consumption of vegetarian foods is also associated with reductions in the risk of left ventricular dysfunction and in blood pressure and increases in antioxidant levels in the body, which contributes to reductions in levels of LDL and total cholesterol, and improvements in levels of HDL-cholesterol(Reference Pilis, Stec and Zych14). The Portfolio Diet, also a plant-based diet, is composed of four core components including nuts, plant protein, soya products or dietary pulses, viscous soluble fibre and plant sterols(Reference Jenkins, Kendall and Faulkner15–Reference Chiavaroli, Nishi and Khan17). It leads to reductions in the primary therapeutic lipid target for CVD prevention and other cardiometabolic risk factors(Reference Chiavaroli, Nishi and Khan17).
The effectiveness of nutrition counselling within a comprehensive CR programme has been extensively evaluated in the literature, and studies have shown that it is associated with improvements in abnormal blood lipid profiles, high blood pressure, glucose intolerance and overall cardiometabolic risk factors(Reference Valentino, Galgani and Álamos18–Reference Frediani, Reilly and Higgins24). Despite well-established benefits, only a low proportion of CVD patients attend CR programmes(Reference Ruano-Ravina, Pena-Gil and Abu-Assi25), and of those who complete the programme, only 34–49 % of them adhere to a healthy diet upon the completion of their comprehensive programme(Reference Song26,Reference Twardella, Merx and Hahmann27) .
Given the importance of a healthy diet for CR patients, it is important to identify the barriers and facilitators impacting patients’ adherence to the dietary recommendations offered in comprehensive CR programmes. Identifying these factors will help CR staff to address patients’ concerns and improve the educational strategies to ensure greater adoption. On summary, this systematic review was conducted with the aim to identify the individual-, provider- and organisational/environmental-level barriers affecting participant adherence of dietary recommendations within comprehensive CR programmes.
Methods
This systematic review was registered prospectively with the Open Science Framework. Data are reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses(Reference Moher, Liberati and Tetzlaff28).
Eligibility criteria
Studies from peer-reviewed literature identifying barriers and facilitators affecting adherence to dietary recommendations in patients participating in comprehensive CR programmes were considered for inclusion. We included studies reporting data from patients at risk or diagnosed with CVD receiving any dietary recommendations while participating in a comprehensive CR programme. All barriers to dietary intake described in individual-, provider- and system/environmental-levels were considered and reported. Barriers reported by family members or healthcare providers were not considered. The most common dietary patterns included in the search strategy were the following: the Mediterranean Diet, the Dietary Approaches to Stop Hypertension Diet, Vegetarian or Vegan diets, the Portfolio Diet, and the Carbohydrate-restricted Diet. However, any type of diet recommendation was included in our search and stated in our findings.
Studies of any methodological design were considered for inclusion in this study (i.e. quantitative, qualitative and mixed methods). Narrative, scoping and systematic reviews were considered as a source of additional primary studies. Pilot and case report studies, non-peer-reviewed literature and studies published in a language other than English, Portuguese or Spanish were excluded.
Information sources and search strategy
Six databases were searched from inception to 15 March 15 2021: APA PsycInfo (Ovid), Cumulative Index to Nursing and Allied Health Literature, Embase (Ovid), Emcare (Ovid), Medline (Ovid) and PubMed (non-Medline). The search strategies were developed by an Information Specialist utilising the PICO framework, subject headings as appropriate for each database and free-text terms relevant to the topical concepts. Relevant studies identified by reviewing the references from included studies were also added (snowballing).
The search strategies were comprised of three main concepts: (1) comprehensive CR: defined as an outpatient CR programme that includes a range of approaches including report risk factor modification support and/or educational strategies associated with dietary recommendations(Reference Piepoli, Hoes and Agewall1); (2) dietary recommendations: defined as any type of foods and/or dietary patterns recommended for patients while participating in CR programmes; and (3) patient compliance: defined as the act of an individual conforming to professional recommendations with regard to prescribed dosage, timing and frequency of an intervention(Reference Jay, Litt and Durant29). The Medline search strategy is included as an online supplementary material, Supplemental Table 1).
Study selection
Two reviewers (LMV and GLMG) independently conducted an initial screening of all records identified by the search strategy (title and abstract). To be selected for secondary screening, abstracts had to clearly mention all three concepts. Full text of the relevant screened articles was then obtained and assessed independently for eligibility by the two reviewers, based on defined inclusion and exclusion criteria. Disagreements in any part of the screening process were discussed and used to refine results.
Quality appraisal
The Critical Appraisal Skills Programme was used to assess the quality of qualitative studies included in this systematic review. Critical Appraisal Skills Programme includes two screening questions and eight detailed questions regarding study design, sampling, data collection, reflexivity, ethical issues, data analysis, findings and values of the research(Reference Tong, Flemming and McInnes30).
The Downs and Black Modified Checklist was used to assess the quality of quantitative studies. These studies were classified as ‘good’, ‘fair’ and ‘poor’ according to the US Preventive Services Task Force Approach(Reference Downs and Black47). Qualitative appraisal was performed by both reviewers (LMV and GLMG). No disagreements between reviewers were identified after this classification.
Data extraction, synthesis and analysis
The COVIDENCE online systematic review software programme was used to eliminate duplicate records and conduct initial and secondary screening of the database results. Data from included studies were extracted to a table by the first author and verified by the last. These data included the following: author(s), year of publication, country, study design, method of data extraction, sample size, characteristic of participants, dietary recommendations, components of dietary recommendations within CR, and barriers and facilitators to participant adherence of dietary recommendations. Components of dietary recommendations within CR were reported according to the Workgroup for Intervention Development and Evaluation Research reporting guideline and included the following: characteristics of those delivering the intervention, detailed description of the intervention content, intensity, mode of delivery, duration, the setting for outpatient CR and adherence to delivery protocol(48,Reference Albrecht, Archibald and Arseneau49) .
Thematic analysis was considered for analysing data from the qualitative studies. Line-by-line coding and free coding were organised and interpreted to generate analytical themes that offered new insights and interpretation(Reference Tong, Flemming and McInnes30,Reference Barnett-Page and Thomas50) . Data from quantitative studies were presented according to the Synthesis Without Meta-analysis guideline(Reference Campbell, McKenzie and Sowden51). Barriers extracted from the quantitative and qualitative studies were grouped and reported on the basis of individual-, provider- and system/environmental-levels, according to the socio-ecological theory(Reference Richard, Potvin and Kishchuk52).
Results
Characteristics of included studies
The initial database search yielded 2083 records, and two additional records were identified through a snowball hand-search. After removing duplicates and conducting an initial screening of titles and abstracts, sixty-one full articles were assessed for eligibility; all were written in English. Overall, sixteen (26 %) articles were included in this study(Reference Astin, Atkin and Darr31–Reference Leong, Molassiotis and Marsh46). The Preferred Reporting Items for Systematic Review and Meta-Analyses flow diagram, depicting the search results, reasons for exclusion and the study selection totals are shown in Fig. 1.
Summary of the study selection process
The characteristics of the included studies are shown in Table 1. Most of the studies were qualitative in design (n 9; 56 %)(Reference Astin, Atkin and Darr31–Reference Kärner, Tingström and Abrandt-Dahlgren39), with semi-structured interviews used in seven (43 %)(Reference Astin, Atkin and Darr31–Reference Rowland, Schumacher and Leinen37) and focus group sessions used in two (12 %)(Reference Fletcher, Burley and Thomas38,Reference Kärner, Tingström and Abrandt-Dahlgren39) . Seven studies (46 %) were observational(Reference Sharp and Salyer40–Reference Leong, Molassiotis and Marsh46), in which four (25 %) were cross-sectional and three (18 %) were prospective cohort in design.
Characteristics of the included studies
ND, not described.
* Same name used in the publication; quality assessment score (classification).
† Critical Appraisal Skills Programme, scores ranged from 0 to 10 (poor: 0–4; fair: 5–7; good: 8–10).
‡ Downs and Black Modified Checklist; scores ranged from 0 to 18 (poor: 0–6; fair 7–12; good 13–18).
In total, 742 patients were included in these studies. Patients were geographical comprised of people from the United Kingdom (n 117 patients in two studies; 12 %)(Reference Astin, Atkin and Darr31,Reference Leong, Molassiotis and Marsh46) ; Canada (n 42 patients in two (12 %) studies)(Reference Galdas and Kang32,Reference Galdas, Oliffe and Wong33) ; Finland (n 260 patients in three (18 %) studies)(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Lappalainen, Koikkalainen and Julkunen45) ; Australia (n 45 patients in two (12 %) studies)(Reference Meyer, Coveney and Ward35,Reference Fletcher, Burley and Thomas38) ; Ireland (n 87 patients in two (12 %) studies)(Reference Reid, Graham and Hickey36,Reference Cannon42) ; the USA (n 126 patients in four (25 %) studies)(Reference Rowland, Schumacher and Leinen37,Reference Sharp and Salyer40,Reference Franklin, Kolasa and Griffin43,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) ; and Sweden (n 113 patients in one (6 %) study). Figure 2 illustrates these countries and notes other countries with comprehensive CR programmes, but with undefined barriers and facilitators to participant adherence of dietary recommendations(Reference Turk-Adawi, Oldridge and Tarima53).
Countries in which cardiac rehabilitation programmes are reported. Data from availability of CR programs around the world were extracted from study published by Turk-Adawi et al. (Reference Turk-Adawi, Oldridge and Tarima53)
Table 1 also shows the quality appraisal of the included studies. Nine studies were classified as ‘good’ (56 %)(Reference Astin, Atkin and Darr31–Reference Meyer, Coveney and Ward35,Reference Rowland, Schumacher and Leinen37–Reference Kärner, Tingström and Abrandt-Dahlgren39,Reference Leong, Molassiotis and Marsh46) , four (25 %) as ‘fair’(Reference Reid, Graham and Hickey36,Reference Sharp and Salyer40,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) and three (18 %) as ‘poor’ quality(Reference Cannon42,Reference Franklin, Kolasa and Griffin43,Reference Lappalainen, Koikkalainen and Julkunen45) . Online supplementary material, Supplemental Tables 2 and 3 show detailed information regarding the quality appraisal of included studies.
Dietary recommendations
Of the sixteen studies, thirteen (81 %) studies(Reference Astin, Atkin and Darr31–Reference Meyer, Coveney and Ward35,Reference Rowland, Schumacher and Leinen37–Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44–Reference Leong, Molassiotis and Marsh46) recommended CR patients to develop healthy eating habits, including a prudent diet, and a cholesterol-lowering diet. One (6 %) study(Reference Reid, Graham and Hickey36) recommended a weight-reducing diet. One (6 %)(Reference Cannon42) study recommended the increase of fish oil and n-3 intake. Finally, one (6 %)(Reference Franklin, Kolasa and Griffin43) study recommended CR patients to transition to vegetarian-rich diets.
Characteristics of dietary recommendations described according to Workgroup for Intervention Development and Evaluation Research were reported by the following number of studies: delivering the intervention by eight8 (50 %) studies(Reference Galdas and Kang32,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Sharp and Salyer40,Reference Cannon42–Reference Lappalainen, Koikkalainen and Julkunen45) , detailed description of the intervention content by four (25 %) studies(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) , intensity by five (31 %) studies(Reference Reid, Graham and Hickey36,Reference Sharp and Salyer40,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Franklin, Kolasa and Griffin43,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) , mode of delivery by ten (62 %) studies(Reference Galdas and Kang32–Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Sharp and Salyer40–Reference Lappalainen, Koikkalainen and Julkunen45) , duration by five (31 %) studies(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Sharp and Salyer40,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) , setting by 16 (100 %) studies(Reference Astin, Atkin and Darr31–Reference Leong, Molassiotis and Marsh46) and adherence to delivery protocol by 6 (37 %) studies. Overall, one (6 %)(Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44) study reported all seven characteristics of the nutrition education, and two (12 %)(Reference Reid, Graham and Hickey36,Reference Koikkalainen, Mykkānen and Erkkilā41) reported six of them. The other studies included in this systematic review described an average of 2·6 characteristics (range 0–5)(Reference Astin, Atkin and Darr31–Reference Meyer, Coveney and Ward35,Reference Rowland, Schumacher and Leinen37–Reference Sharp and Salyer40,Reference Cannon42,Reference Franklin, Kolasa and Griffin43,Reference Lappalainen, Koikkalainen and Julkunen45,Reference Leong, Molassiotis and Marsh46) .
Barriers affecting adherence to dietary recommendations
Barriers affecting adherence to dietary recommendations were identified by CR participants in fourteen studies. These barriers cover individual- and system/environmental-levels. No barriers at a provider-level were identified by any of the included studies (refer to Table 1 and Fig. 3).
Diagram of barriers and facilitators affecting adherence to dietary recommendations in CR programmes in an individual-, provider- and system/environmental-level. CR, cardiac rehabilitation
The identified individual-level barriers included were the following: food habits, financial situation, personal reasons, time constraints, psychological and cultural aspects. Food habits were the most common barrier, being reported in eight studies (50 %). These studies recommended CR participants to develop healthy eating habits and transition to vegetarian-rich diets(Reference Astin, Atkin and Darr31,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Kärner, Tingström and Abrandt-Dahlgren39–Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Franklin, Kolasa and Griffin43–Reference Lappalainen, Koikkalainen and Julkunen45) . CR participants reported that the taste of healthy foods and difficulties in not eating their favourite foods were important aspects contributing to low adherence to dietary recommendations.
Financial situation(Reference Astin, Atkin and Darr31,Reference Galdas, Oliffe and Wong33,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Cannon42,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44,Reference Lappalainen, Koikkalainen and Julkunen45) and personal reasons(Reference Galdas, Oliffe and Wong33,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Cannon42,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44,Reference Lappalainen, Koikkalainen and Julkunen45) were reported by seven studies each (37 %), as barriers to adopting healthy eating habits and increasing fish oil and n-3 intake. The financial cost of food emerged as a strong barrier to adopting the nutritional changes recommended. Participants reported that ‘the instructed diet was expensive’. Personal reasons included insecurity about how the diet would affect their health or body, a lack of interest or motivation to change their current eating habits, having their foods prepared by others, and difficulties in controlling their appetites and food choices when seeing food advertisements or presented with the greater availability of different or cheaper foods in local stores.
Four studies (25 %) reported time constraints(Reference Astin, Atkin and Darr31,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Lappalainen, Koikkalainen and Julkunen45) as a barrier affecting adherence to dietary recommendations. CR participants identified irregular work hours and lack of time to prepare healthy foods as a barrier. Making dietary changes was described as ‘time consuming’ and ‘factors that create unnecessary work’. Psychological aspects(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Kärner, Tingström and Abrandt-Dahlgren39) and cultural aspects(Reference Astin, Atkin and Darr31,Reference Galdas, Oliffe and Wong33) were also reported as barriers affecting adherence to dietary recommendations in two studies each (12 %). Patients indicated they did not have enough energy or a psychological state that motivate them to engage in healthy eating habits. Cultural aspects were raised by the South Asian population attending CR programmes. In their culture (Punjabi Sikh), meals are usually prepared by women for the whole family, and men have little control on what will be prepared. According to this cultural group, it was not appropriate to change dietary habits for the whole family because of healthy eating requirements necessary for one family member.
Barriers at the system/environmental-level were listed as lack of family support and a poor availability of recommended foods. Lack of family support was reported in six (37 %) studies(Reference Galdas, Oliffe and Wong33,Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Reid, Graham and Hickey36,Reference Rowland, Schumacher and Leinen37,Reference Kärner, Tingström and Abrandt-Dahlgren39,Reference Koikkalainen, Mykkānen and Erkkilā41) . Since meals are often prepared by one family member, it was often regarded as ‘too much work’ to prepare a different meal for only one person within the family.
The availability of certain foods was reported as a barrier in six (37 %) studies that recommended adopting healthy eating habits, increasing fish oil and n-3 intake, and transitioning to vegetarian-rich diets(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Sharp and Salyer40–Reference Franklin, Kolasa and Griffin43,Reference Lappalainen, Koikkalainen and Julkunen45) . Lack of healthy choices in restaurants, limited healthy ingredients in local grocery stores, the absence of fish shops close to home and difficulty in finding fresh vegetables during certain seasons of the year (such as winter) were described by patients.
Multi-level barriers included a lack of knowledge, language difficulties and social aspects. Lack of knowledge was the most common barrier reported by studies that recommended healthy eating habits (n 9; 56 %)(Reference Astin, Atkin and Darr31,Reference Galdas and Kang32,Reference Koikkalainen, Lappalainen and Mykkänen34–Reference Reid, Graham and Hickey36,Reference Kärner, Tingström and Abrandt-Dahlgren39–Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Lappalainen, Koikkalainen and Julkunen45) , and it was characterised as both an individual- and provider-level barrier. Some participants indicated they did not know how to choose healthy and acceptable foods or expressed challenges in understanding the association between diet and their cardiovascular condition. Language was also characterised as both an individual- and provider-level barrier. One study(Reference Galdas and Kang32) recommended establishing healthy eating habits to Punjabi-speaking CR patients noted that they were not able to engage in private discussions about their dietary choices because translators were not available. Lastly, social factors were reported as both an individual- and system-level barrier to dietary adherence by four (25 %) studies(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Franklin, Kolasa and Griffin43,Reference Lappalainen, Koikkalainen and Julkunen45) . CR participants identified challenges of choosing healthy foods on workdays or while eating with others. While associating with other people, patients prefer to eat what others are eating because they ‘do not want to be different’ from their friends and colleagues.
Facilitators affecting adherence to dietary recommendations
Seven (43 %) studies reported specific facilitators for improving adherence to dietary recommendations in comprehensive CR programmes(Reference Astin, Atkin and Darr31–Reference Galdas, Oliffe and Wong33,Reference Fletcher, Burley and Thomas38,Reference Kärner, Tingström and Abrandt-Dahlgren39,Reference Franklin, Kolasa and Griffin43,Reference Leong, Molassiotis and Marsh46) . These factors, also characterised on an individual-, provider- and system/environmental-level, are described in Table 1 and illustrated in Fig. 3.
Two individual-level facilitators for developing healthier eating habits included the participants’ financial situation and an older age. Financial circumstances were reported by one (6 %) study(Reference Astin, Atkin and Darr31) and impacted the nature of meals since as dependent on the food purchased. Participants were more likely to purchase healthy foods when shopping with family members. A positive correlation between adhering to healthy eating habits and older age was reported by one study(Reference Leong, Molassiotis and Marsh46). According to this study, older people are more likely to follow healthy eating recommendations.
At the provider-level, support from CR programmes was reported by four studies (25 %) as a facilitator to developing healthy eating habits or adopting a vegetarian-rich diet adherence(Reference Galdas and Kang32,Reference Galdas, Oliffe and Wong33,Reference Fletcher, Burley and Thomas38,Reference Franklin, Kolasa and Griffin43) . Participants indicated receiving ongoing support and reassurance from CR providers was a critical element for their adhering to healthier eating habits.
At the system/environmental-level, family support was identified as a key facilitator to developing healthier eating habits in four studies (25 %)(Reference Astin, Atkin and Darr31,Reference Galdas, Oliffe and Wong33,Reference Kärner, Tingström and Abrandt-Dahlgren39,Reference Leong, Molassiotis and Marsh46) . There are ‘several ways in which families supported patients in adopting healthier dietary habits’, including partners deciding to change their diet habits as well or family members helping with healthy food preparation and encouraging CR participants to follow dietary recommendations.
Knowledge was identified as a multi-level (provider- and individual-level) facilitator for establishing healthy eating habits as reported by three studies (18 %)(Reference Galdas and Kang32,Reference Galdas, Oliffe and Wong33,Reference Fletcher, Burley and Thomas38,Reference Franklin, Kolasa and Griffin43) . Participants who understood the association between unhealthy eating habits and their cardiovascular conditions were more likely to change their eating habits.
Nine studies (56 %) did not identify specific facilitators for establishing a healthy diet in CR participants(Reference Koikkalainen, Lappalainen and Mykkänen34–Reference Rowland, Schumacher and Leinen37,Reference Sharp and Salyer40–Reference Cannon42,Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44,Reference Lappalainen, Koikkalainen and Julkunen45) . Unfortunately, none of the studies identified facilitators for increasing fish oil and n-3 intake.
Synthesis Without Meta-analysis of quantitative studies
Seven quantitative studies included in this systematic review could not be meta-analysed as the relevant information required for statistical analysis was not reported(Reference Sharp and Salyer40–Reference Leong, Molassiotis and Marsh46). These seven studies used different strategies to report barriers and facilitators, including average(Reference Sharp and Salyer40), total absolute number(Reference Hämäläinen, Paalosmaa-Puusa and Seppänen44), total percentage of participants that reported barriers and facilitators affecting dietary intake(Reference Cannon42), the mean number of barriers and facilitators reported by CR participants(Reference Koikkalainen, Mykkānen and Erkkilā41,Reference Lappalainen, Koikkalainen and Julkunen45) or correlation between facilitators that affects adherence to dietary intake(Reference Leong, Molassiotis and Marsh46). Furthermore, although classified as quantitative, one study reported barriers based on qualitative clinical interviews(Reference Franklin, Kolasa and Griffin43). Data from these results are shown in Table 1, and barriers and facilitators were incorporated into the model described above.
Discussion
To the best of our knowledge, this is the first systematic review that investigates barriers and facilitators affecting adherence to dietary recommendations in patients participating in comprehensive CR programmes. Although a limited number of published studies were identified, the existing data suggested that CR patients experienced barriers affecting dietary recommendations at an individual-, provider- and system/environmental-level. Dietary recommendations included in these studies were listed as developing healthy eating habits, transitioning to vegetarian-rich diets, and increasing fish oil and n-3 supplements. Barriers affecting adherence to dietary recommendations, in descending order of appearance, were lack of knowledge, financial situation, personal reasons, lack of family support, availability of food, food habits, time constraints, social aspects, food habits, cultural aspects, psychological aspects and language. Facilitators affecting adherence to dietary recommendations were identified and, in descending order, included: support from the CR programme, family support, knowledge and older age.
Although barriers and facilitators to participant adherence of dietary recommendations were identified, those are only representative of recommendations to developing healthy eating habits, transitioning to vegetarian-rich diets and increasing fish oil and n-3 supplements. The dietary patterns noted in the Mediterranean Diet, the Dietary Approaches to Stop Hypertension Diet, the Portfolio Diet, and Vegetarian and Vegan diets are highly recommended for patients participating in CR programmes. However, the perceived barriers and facilitators experienced by patients trying to follow these dietary patterns are often not reported in the literature(Reference Appel, Moore and Obarzanek11,Reference Sacks, Svetkey and Vollmer12,Reference Pilis, Stec and Zych14,Reference Jenkins, Kendall and Faulkner15,Reference Chiavaroli, Nishi and Khan17,Reference Valentino, Galgani and Álamos18,Reference Appel, Champagne and Harsha54) . This is an important literature gap, identified by this systematic review, that should be addressed in future studies.
Furthermore, this review was not able to identify a standardised intervention for delivery of dietary recommendations in CR as most of the studies did not reported main characteristics following the Workgroup for Intervention Development and Evaluation Research recommendations(48). This lack of existing data to guide healthcare providers on the optimal setting, mode of delivery, intensity, duration and content of dietary recommendations for patients participating in CR is concerning. Cardiac College – a comprehensive and evidence-based virtual patient education programme(Reference Ghisi, Grace and Anchique55,Reference Ghisi, Rouleau and Ross56) – is available in eight languages for CR programmes to use and address nutritional education in a standardised way. This tool could be of great use to programmes around the globe in addressing some of the educational gaps related to dietary recommendations.
Availability of food was reported as a barrier affecting adherence to healthy eating habits, increasing fish oil and n-3 supplement intake and transitioning to vegetarian-rich diets, as patients reported difficulty in finding healthy food options close to their homes(Reference Koikkalainen, Lappalainen and Mykkänen34,Reference Sharp and Salyer40–Reference Franklin, Kolasa and Griffin43) . Previous studies have identified an association between availability of healthy food and race/ethnicity, financial situation, and access to grocery stores and supermarkets(Reference Morland, Wing and Diez Roux57–Reference Whelan, Wrigley and Warm59). In the American city of Baltimore, Morland et al. (Reference Morland, Wing and Diez Roux57) suggested that high availability of healthy foods was present in only 19 % of predominately black neighbourhoods compared with 68 % of white neighbourhoods. People living in areas with long distances to supermarkets, low access to fresh foods and with an abundance of unhealthy processed fast food outlets were less likely to adhere to dietary recommendations(Reference Morland, Wing and Diez Roux57–Reference Whelan, Wrigley and Warm59). Patients’ income also has an important impact on their adherence to dietary recommendations(Reference Larsen and Gilliland60). A comprehensive analysis of food purchase patterns from a representative sample of 4826 American households showed that low-income households purchased lower amounts of healthy food than recommended(Reference Morland, Wing and Diez Roux57). Overall, low-income households were less likely to purchase fruits and vegetables and more likely to purchase sugar-sweetened beverages. The higher food prices for higher quality foods may account for this behaviour as healthy food purchases required spending more money, which might force patients to choose unhealthy eating options instead(Reference Larsen and Gilliland60).
The global burden of CVD shifted to low-and-middle income regions such as Eastern Europe and Central Asian countries as a result of a combined effect of population growth, ageing, high intake of processed foods and alcoholic drinks, high prevalence of smokers and inadequate strategies to reduce cardiovascular risk factors(61–Reference Finegold, Asaria and Francis63). However, barriers and facilitators affecting adherence to dietary recommendations reported by patients participating in CR programmes of these countries remain unknown. This systematic review included studies conducted in high-income countries and, therefore, conclusions cannot be extrapolated to middle- and low-income countries where socio-economic disparities are more evident. In high-income countries such as Canada and the USA, immigration status can also impact food purchases as immigrants usually incorporate unhealthy and fast foods into their traditional diet(Reference Finegold, Asaria and Francis63). In this context, studies have shown that culture creates challenges to immigrant adherence to dietary recommendations in CR programmes(Reference Popovic-Lipovac and Strasser64,Reference Jin, Neubeck and Koo65) . The main reasons for this fact include limited access to familiar foods or ingredients, such as types of vegetables or spices, uncertainty or unfamiliarity with new foods and cultural preparation practices, and digestion problems related to the consumption of unfamiliar products(Reference Visram, Crosland and Unsworth66–Reference Siddiqui, Winther and Kurbasic68).
Food habits are also important barriers to healthy dietary intake, mainly associated with the taste of the food(Reference Drewnowski69). The human attraction to sweet and savoury foods is known, so food companies add a large amount of sugar and Na to most processed products, which may contribute to an increase of low-quality food intake, obesity and cardiovascular risk in overall populations(Reference Drewnowski69–Reference Hu72). Although not reported by the included studies, food taste and choice can also be influenced by branding and marketing, as advertisements that promote fun, happiness and excitement are linked to positive sensory thoughts with the target food, and contribute to the specific consumption of that product(Reference Elder and Krishna73).
Social determinants, including family and social support, have been recognised as a powerful influence in food choices and eating patterns(Reference Shepherd74–Reference Kettunen, Solovieva and Laamanen77). Family support was identified in this systematic review as both a barrier and a facilitator to adherence of dietary recommendations. Overall, conflicts with home habits and social environments result in patients being poor compliers, whereas when family and friends build nutrition changes together it influences patients to adopt healthier dietary choices(Reference Shawon, Hossain and Adhikary76,Reference Kettunen, Solovieva and Laamanen77) . Lack of family support may also affect patients by producing negative emotions such as anxiety, depression, uncertainty, fatigue, irritability, poor concentration and insomnia(Reference O’Farrell, Murray and Hotz78–Reference Webster, Thompson and Mayou80). Some reasons affecting family support include beliefs and attitudes towards CVD, motivations to adopt dietary changes with their family members diagnosed with CVD and knowledge regarding the impact of dietary habits on patients’ cardiovascular health(Reference O’Farrell, Murray and Hotz78–Reference Webster, Thompson and Mayou80). Language barriers, identified in this systematic review as a barrier affecting adherence to dietary recommendations, might be associated with patients and family members’ limited information and misconceptions(Reference Moser, Dracup and Marsden79). Support provided by the CR team, identified in this study as a facilitator affecting dietary recommendations, might contribute to overcoming lack of family support and language barriers(Reference Galdas and Kang32,Reference Galdas, Oliffe and Wong33,Reference Fletcher, Burley and Thomas38) . Knowledge improvements and translation services can help patients and family members understand their cardiovascular conditions and impact their potential risk of having future events, improving the engagement of family members.
Finally, our review identified that elderly populations are more likely to follow dietary recommendations than young individuals(Reference Leong, Molassiotis and Marsh46). This finding may be potentially associated with time constraints commonly reported by young productive populations, such as working long hours and family commitments to children and senior family members(Reference Welch, McNaughton and Hunter81–Reference Ziol-Guest, Deleire and Kalil84). Studies with young individuals identified difficulties committing time to exercise and adopt healthy eating behaviours due to competing priorities or poor time-management skills(Reference Cluskey and Grobe85,Reference Strong, Parks and Anderson86) .
Limitations
Results from this study should be interpreted with caution. We acknowledge that most of the studies included in this systematic review discuss barriers and facilitators to improving healthy eating habits. Only a small number of studies included detailed descriptions of healthy eating strategies recommended to patients participating in CR programmes. This makes it difficult to understand whether specific dietary characteristics address barriers and facilitators for healthy dietary intake. Furthermore, our results report data from CR participants only in high-income countries. Barriers and facilitators identified by people participating in CR programmes in other countries may be different. In addition, results extracted from these studies are only generalisable to patients who are referred and attend CR programmes, which are a low proportion of total cardiac outpatients.
Conclusions
Some recommendations described in the literature would help to overcome barriers related to patients’ financial situation, lack of knowledge, lack of family member support and the cultural aspects reported by CR participants. Overall, the literature suggests financial incentives as important modulators of behaviour change(Reference Strong, Parks and Anderson86,Reference Gittelsohn, Trude and Kim87) . Restrictions on advertisements of unhealthy food may also help to improve diet quality in this population(Reference Gittelsohn, Trude and Kim87,Reference Afshin, Peñalvo and Del Gobbo88) . CR programmes should also consider improving knowledge and providing culturally sensitive resources for patients and family members on culturally relevant dietary recommendations and on the importance of healthy food consumption and its impact on cardiovascular health. Furthermore, the use of technology (such as text messages)(Reference Mozaffarian, Angell and Lang89) may help CR providers to disseminate nutritional information to patients in need.
Results from this review highlighted the complex interaction between individuals, providers and systems in the adherence of dietary recommendations. Efforts to change eating behaviours, especially in secondary prevention programmes like CR, must carefully consider all described factors in this review to be successful. In addition, these results also suggest avenues for future research. First, researchers need to better understand barriers and facilitators from highly recommended diets to patients participating in CR programmes to help with policy changes. Second, healthcare providers need to counsel patients, if relevant, about culturally appropriate ways to engage in a healthy diet and to increase their knowledge and motivation, which can help with the provision of direct assistance in ways to access education resources. Finally, intervention strategies tailored to the unique life contexts of patients should be implemented at multiple levels to encourage healthy dietary behaviours, including all three levels addressed in this review.
Acknowledgements
Acknowledgements: The authors would like to acknowledge the Coordination of Improvement of Higher Education Personnel – CAPES/Print Programme, Brazil to support LaisVanzella’s work. Financial support: This work was supported by the Coordination of Improvement of Higher Education Personnel – CAPES. Conflict of interest: There are no conflicts of interest. Authorship: L.V. contributed to the conception, design, acquisition, interpretation of data and written of the manuscript. V.R., F.A. and N.D. contributed to the interpretation and revising the version to be published. M.P. contributed to the conception, design and revising the version to be published. P.O. contributed to the conception and revising the version to be published. G.G. contributed to the conception, design, interpretation of data and written of the manuscript. Ethics of human subject participation: Not applicable.
Supplementary material
For supplementary material accompanying this paper visit https://doi.org/10.1017/S1368980021002962
