Diabetes self-management arrangements in Europe: a realist review to facilitate a project implemented in six countries

The EU-WISE collaborative project (http:http://​www.​eu-wise.​com) involves 6 countries including The United Kingdom, The Netherlands, Norway, Spain, Bulgaria, and Greece with the aim to produce a theoretical and methodological background of systems of support, summarise the structure and governance of health and welfare systems of the six participating countries and to describe, explain and profile the structure of personal networks of people with diabetes in different national contexts. Moreover, this project aims to describe and analyse the support mechanisms that operate on the meso-level, to survey the systems of support of people with diabetes and the networks of the organizations. Finally, the EU-WISE collaborative project aims to identify existing mechanisms involved in diabetes’ management with the aim to develop and implement new strategies of engagement and support self-care.

The realist review reported here will inform the second work package of this EU project through the identification of quantitative published peer-reviewed literature with regard to diabetes self-management arrangements currently in place (within the health care systems of the different partner countries) and how these are integrated into the broader welfare systems. Within the EUWISE framework, it was agreed to explore the existing different welfare policies, systems provision, structure and reforms of partner healthcare systems and association through direct and indirect impact on the configuration of support for people with chronic illness, with a focus on those with type 2 diabetes.

This realist review aims to assess the key government objectives and direction of change in the provision of self-care and how the role of health care professionals is changing within this environment. Whilst for the purpose of this online search a number of research questions were identified, and they were in line with those raised during the formation of the research protocol, as the methodology of the realist review suggests, the purposes of the review were refined during the process of the literature search. Eventually (and primarily after the background search that gave a feel for the literature), the key themes to explore were guided by the following questions:

A realist review is a systematic method which has been used for assessing complex interventions, focused on identifying determinants of their impact, and evaluating or developing theories [10]. The aim of the realist review is to develop concepts and knowledge about variations in a phenomenon rather than seeking one comprehensive effect size. Furthermore, it can be used with quantitative, qualitative and multiple methods [11].

We drew on the principles of the realist review as this strategy for synthesising research aims to unpack the mechanisms of how complex programmes work in particular contexts and settings (what works for whom and in what circumstances). As first step in the design of the review, the underlying assumptions about how an intervention is meant to work and what impacts it is expected to have were identified and the relevant theoretical framework was put together [11]. To that end, a literature review to summarise the peer-reviewed published scientific literature in the field (stage 1) was followed. Relevant papers were then synthesised narratively (stage 2).

Stage 1 included identification of all material which had been published following a peer-reviewed process (original or policy articles). A search diary was maintained detailing the names of the databases that were searched, the keywords used and the search results.

For the literature search, a systematic approach was undertaken, utilising the specific keywords and criteria, as indicated below. The methodology of the realist review, that combines theoretical understanding with empirical evidence, was followed throughout [11]. The studies identified were grouped and presented in summary tables featuring the key messages of each one.

Publications of interest were identified by a search of the following information resources since 2000: PubMed, Scopus, Web of Science, PsycINFO, European Observatory on Health Systems and Policies, Centre for Reviews and Dissemination - Database of Abstracts and Reviews of Effects (DARE).

Various search terms were used to include both MeSH terms and other glossary databases. Guided by entry terms in MeSH and mapped terms in Emtree thesaurus, the main ones used were: Chronic Disease, Diabetes Mellitus, Self Care Support, Self Care Management, Health Care System, Social Welfare, Chronic Illness Management, Chronic Illness Policy, Self Management Systems, Health Resources, Health Care Providers, Health Plan Implementation, Government Programmes, Skill-Mix. Within this context, indicatively, the electronic search strategy that was followed was constructed on the following search algorithm for Medline bibliographical database: (diabetes [Title/Abstract]) AND (self care support OR self care management) AND (health care system OR social welfare OR policy OR providers OR government OR skill-mix). Advanced search options have been used to combine the terms (and their truncations) in every database.Thereafter, reference lists were systematically searched for further relevant articles by two reviewers; finally, corresponding authors were contacted with regard to missing data. A relevant flow chart was constructed to detail the number of papers retrieved and the steps undertaken (Figure 1).

The specific inclusion and exclusion criteria that have been proposed in the review protocol are illustrated in Table 1.

A standard ‘search diary’ was maintained detailing the names of the databases that were searched, the keywords used and the search results. Titles and abstracts of studies to be considered for retrieval were recorded on a form, along with details of where the reference had been found. Inclusion and exclusion decisions were recorded on a database. Retrieved studies were filed according to inclusion and exclusion decisions. Studies were selected for retrieval after two independent reviewers had appraised titles and abstracts identified in electronic searches. All references provided by expert contacts were also retrieved.

Two reviewers (AK, EP) independently abstracted data and all relevant information, which was placed onto a form and summarised in order to identify what was considered to be the most important results from each study. These summaries were compared and any differences of opinion were resolved by discussion and consultation with the original study.

For the narrative synthesis, the studies were grouped and presented in Summary Tables featuring key points of each study (Tables 2 and 3). A standard form was used to assist the sorting of primary source material, but in the context of the realist review a bespoke form was developed to serve a different purpose (see Table 3). Preliminary primary sources (as in the paper by Bower et al. for example [12]) were included in the results as it is important in a realist review to identify possible relevant concepts and theories [11]. Furthermore, in terms of quality appraisal and to identify fatally flawed empirical papers (which in a realist review differs from the methodological approach of a systematic review), we decided to assess the five prompts suggested by Dixon-Woods et al for all studies: (i) Are the aims and objectives of the research clearly stated? (ii) Is the research design clearly specified and appropriate for the aims and objectives of the research? (iii) Do the researchers provide a clear account of the process by which their findings were reproduced? (iv) Do the researchers display enough data to support their interpretations and conclusions? (v) Is the method of analysis appropriate and adequately explicated? [13]. The quality appraisal of the evidence occurred in parallel with the synthesis and narration of results, and judgment had to be used to supplement the formal critical appraisal checklist and consider all included articles in terms of relevance and rigour [11].

In this section the raw data results from the literature search are presented. A more elaborated presentation of the results, in the context of the realist study and its conclusions in terms of factors potentially relating to the variation in the phenomena of interest, is detailed further in the Discussion section.

The multi-step search strategy yielded 880 documents from six bibliographical databases. As a result of the evaluation, 535 articles were deemed irrelevant, 101 matched the exclusion factors, 83 did not fit the inclusion criteria as outlined in Table 1, and 106 did not constitute original research. All in all, 56 studies were included in the final review reporting design methods and findings on 21 interventions and programmes for diabetes and chronic disease self-management, since 2000 (Figure 1). Twenty-four out of the 56 studies were randomised controlled trials, 18 used a cohort design, and 14 followed other designs including cost-effectiveness analyses, complex intervention framework and presentations of programmes. The themes discussed by all were revolving around the same basic axes. Secondary research and relevant systematic reviews were only later integrated to further enhance the Discussion. Thirteen of the studies (61.9%) came from the The United Kingdom, five (23.8%) from The Netherlands, one (4.76%) from Norway, one (4.76%) from Spain, and one (4.76%) from Bulgaria; only initial discussions and approaches have taken place in Greece [68]. Characteristics and detailed information of the included studies are listed in Table 2. Relevant productive information on the background to the introduction of the selected initiatives is included in Table 3.

Initiatives that only provided information in a didactic format or manipulated delivery of information (e.g. assessment of telephone consultations) were not included because provision of information alone has been recognised as insufficient for improved self-management [69]. Nevertheless, information and contextual aspects of these papers have been discussed. Many (11/21, 52%) of the interventions were designed to fit within the context of general practice as the primary setting. Fifty-two per cent (11/21) clearly stated behavioural change as an important goal, whereas the concepts of empowering patients and addressing individual responsibility were stated in the remaining 48% (10/21). Five interventions (24%) referred to Internet-based tools.

The professional role of diabetes specialist nurses, the need for multidisciplinary approaches and the focus on patient education emerge as fundamental principles in the design of relevant programmes from the majority of the papers (16/21, 76%). The results of the current review, indicating a shift towards patient-centred self-care of diabetes, derive from a total of at least 5,500 individuals residing in the six participating countries.

As evident by the review, success of self-management as a policy solution will be affected by interacting influences at three levels: [a] at micro-level by individuals’ dispositions and capabilities; [b] at meso-level by roles, relationships and material conditions within the family and in the workplace, school and healthcare organisation; and [c] at macro-level by prevailing economic conditions, cultural norms and expectations, and the underpinning logic of the healthcare system [8]. Even though interventions are usually developed for specific chronic diseases, the themes and behavioural strategies are comparable across various disease-homogeneous groups of patients. All chronically ill patients face psychological demands, such as maintaining an optimal quality of life, preserving a reasonable emotional balance and sustaining relationships with family and friends [54]. Social networks, especially, seem to offer an opportunity to redress the balance of an exclusively individual focus on self-management [70], because they address the broader set of contributions and resources available to people in need of chronic illness management and support [5].

In an overview of its key findings, the literature review identifies the need for development of accessible and relevant educational material; improved communication of disease-specific information between patients and providers, as well as providers and community resources; and strategies to improve the convenience and cost of monitoring devices. However, basic diabetes education (including comprehensive programmes and training to enhance personal understanding of the disease and its implications and advance the practice of self-management) should be regarded as having broad patient-based positive outcomes, but should not be expected to have lasting benefits on glycaemic control [25]. In general, the literature indicates reservations in effects of different measures. For example, exercise consultation is more effective in stimulating exercise behaviour change in the short term than a standard exercise leaflet [71]. Educational interventions should have multiple components. They should aim to improve patients’ sense of self-efficacy and empowerment [72], and develop attitudes towards diabetes that will support the lifestyle changes needed for successful self-management [73]. Interestingly, locally developed education programmes seem to have less effect than interventions developed for the purposes of research studies. This is most likely attributed to the fact that study interventions are applied in a controlled environment, strictly following guidelines and not subject to the limitations of routine practice, but long term impact should also be taken into account [58].

It has become explicit in this review that self-management of diabetes mellitus is physically, intellectually, emotionally and socially demanding. Thus, it will be expected for some patients not to engage with self-management due to personal resources (e.g. health literacy, resilience) or overwhelming personal, family and social circumstances.

Self-management of type 2 diabetes is referred to in the literature as being strongly interrelated with contextual factors [74]. Research indicates that immigrants might perceive the causes of type 2 diabetes differently compared to people of European origin. An English study [75] found that patients of European origin tend to blame themselves for developing diabetes. In contrast, patients with Pakistani and Indian backgrounds have more varied accounts to explain their diabetes onset, emerging from a more fatalistic sentiment explained as ‘God’s will’ or genetic tendencies, or as “a thread in the fabric of life” often with a strong focus on external stress factors. An opposite perspective to self-blame is seen in accounts from white diabetes patients who externalize the responsibility for the disease. These are primarily cultural issues, but tend to be provoked by economic influences and most notably, poverty [76]. Thus, they are key issues quite visible and of relevance to the capacity to self manage in countries that have been affected by the economic crisis, including Greece, Spain and Bulgaria [77].

Socioeconomic factors impact on the experience of tension, stress and moral dilemmas in patients struggling with their self-management experience in their encounter with health professionals, even to the extent that health care workers in some cases are perceived as adding to the burden of living with a chronic disease [78]. Whilst, the literature identified few studies explicitly seeking data to illuminate the experiences of patients belonging to either socially or economically deprived groups (the concept of deprivation in this review was considered following the WHO guidelines for Europe [79, 80]) both patients and health professionals indicate that lack of social, personal and economical resources are tightly interconnected and affect patients’ self-management. In cases where there is a large social distance between patient and GP’s and other health care workers this might hinder the health professional’s capacity to fully engage with patients’ problems [78]. Recent evidence suggests that lower social class and presumptions of being able to participate are also implicated in chronic conditions overall though this has not been looked at in relation to type 2 Diabetes [81].

These socioeconomic facts might impact further on the tensions and moral dilemmas that patients experience in their efforts to self-manage, and thus add to the burden of having a chronic disease. The literature illuminates patients’ vast experience of the magnitude of economic strain experienced when living with chronic illnesses. There are certain reports from Greece where the impact of economic crisis on the quality and provision of primary care and health care services is high [82]. The economic burden of living with a chronic illness is described by patients as considerable. It affects not only ability to afford essential treatment and medication but also the ability to maintain a healthy lifestyle and quality of life. This burden does not seem to be fully acknowledged by health professionals nor by health politicians. Descriptions of economic burdens will be even more seriously impacted by the global economic downturn.

Not unexpectedly, there is a variation of health care systems across the participating countries, different funding models and payment systems with General Practitioners (GP’s) as gate keepers in some of them (England, The Netherlands, Bulgaria, Norway), and various governance schemes (primary care trusts in UK, NHS in Bulgaria and Spain, Municipalities in Norway); all seem to have an impact on the management of long term conditions.

Table 3 provides the background for the governance of self-care programmes in the participating countries, as well as the resources that have been allocated to this section. Political commitment and government initiatives have supported many of these approaches. A few of them have been designed for nationwide implementation with regional organisation.

However, certain barriers to change, most probably originating from the health care system structure, seem to exist in a number of countries (vaguely illustrated in the countries that yielded the fewest results in this review: Greece, Bulgaria, Spain and Norway) when discussing how governance and health system organisation promote or inhibit the provision of chronic illness self-management. It should be expected that national health care systems, especially at times when limited resources are available, operate targeted to cost-cutting and identification of effective alternatives.

Within this context, out of the pool of the six participating countries, the majority of research comes from The United Kingdom. The emphasis of care in the UK has now changed from being predominantly reactive to a more preventive management approach, aiming to reduce emergency hospital admissions and encourage patients to make healthier choices about diet, physical activity and lifestyle through self-management of their condition [2]. Despite the long-accepted importance of patient education in diabetes mellitus, an authoritative UK report in 2000 exposed severe deficits in patient education services, describing them as incomplete and not based on current research evidence [83]. As such, self-management interventions have been recognized as a key part of care in the National Service Framework (NSF) in the UK since 2002, spanning the decade studied in this report [84]. Until recently, however, few centres in the UK have offered self-management programmes and rarely were these evidence-based or replicable [29]. In spite of diabetes education offered to all newly diagnosed patients, no universal programme is used. However, the Diabetes Education and Self Management for Ongoing and Newly diagnosed (DESMOND) and Dose Adjustment for Normal Eating (DAFNE) programmes are fairly widely available [25, 85].

In The Netherlands, many Internet-based programmes have been designed, since the most frequently reported self-tests are those for diabetes [86]. In Norway, research on the use of technology, in order to enhance patients’ experience, has been undertaken [87, 88]. Further, there are programmes for patient education located in hospitals [89]. Even though it seems that usage of IT is also the trend in Spain and Greece [60, 68], a literature review has suggested that, despite its potential effect, the use of technology to enhance diabetes self-management is still early in its evolution [90]. The Eurozone economic crisis comes to pose further barriers in the countries that face these limitations, as studies have highlighted the impact of low income status on an effective diabetes self-management [91].

In Bulgaria, a Diabetes Education Program was designed in 1997 to introduce a large-scale unified structured educational program for insulin-treated diabetic patients. Fifty-six in-patient training centers for people with diabetes were established which are still active in all larger cities of the country, while within the required visits to an endocrinologist (which is twice yearly mandatory for patients with diabetes) 25 minutes are devoted to patient education in self-management, under the National Regulation and Instruction [7]. Informational and screening programmes for high-risk population groups are also organized for people with diabetes in Bulgaria [92].

Since 2008, there has been an ongoing, global economic recession and a changing European shift towards a demand for cost-effective health policy. It has been an objective to explore the potential impact on the nature and direction of change in health care systems, particularly in relation to chronic illness management. Within this perspective, UK consensus guidelines recommend limited use of self-monitoring of blood glucose (SMBG) in patients with type 2 diabetes using diet and exercise, metformin and/or a glitazone. The study estimates that the potential savings of up to 17 million pound sterling could be made each year if guidelines were followed more closely. There is a need for further research into SMBG use in patients with type 2 diabetes [93]. Moreover, despite effective self-management of diabetes requiring considerable behavioural change and continuous support from health professionals which can be expensive, information technology has the potential to offer cost-effective patient support, but Internet use mostly relies on the active seeking of information [94]. Several implemented initiatives have proven to be cost-effective in their initial operation [66]. However, programmes using the cognitive behavioural approach may be relatively costly when delivered by specialised nurses who have to be trained by behavioural therapists [54].

Expanding on the above concepts, an apparent trend is the delivery of group self-management education. These group-based programmes, delivered in multiple sessions, can assist in the achievement of meaningful changes. Undeniably, as the aforementioned evidence mainly from The Netherlands and the UK indicate, both governmental and various organisations’ initiatives target modern IT services and web-based applications [28, 29, 51, 54]. IT is a way forward to achieve long-term continuous long term self-management support and this trend should definitely be seriously taken into account for the empowerment of new generations of patients, while exploring how to promote community participatory medicine by public institutions and build social capital engaging inter-organisational interactions.

The literature review clearly illustrates a shift in the management of chronic diseases from secondary to primary care that poses stakeholders with new challenges, while the fundamental role of primary care in the management of chronic diseases has been focused on policymakers and politicians in the European Union.

Structural, organisational and socio-economic influences are relevant both to the capacity of patients to self manage and policy and governments and other agencies to support progress in this field. The identification of the importance of socio-economic influences suggest that the gains to date and further progress are likely to be hard to maintain in countries impacted upon most by economic austerity and high levels of poverty and low incomes. Revision of professional roles in multi-disciplinary clinical teamwork in primary care is advocated by European governments. Nurses, GPs, dieticians, psychologists, market analysts and technology experts should collaborate to design or implement specific self-management interventions [12, 45, 56]. However, patients still express fragmentation and lack of coordination and often receive inconsistent and contradictory advice [95]. Certainly, not only patients, but also health care workers in primary care describe lack of communication and collaboration between healthcare professionals. Health professionals relate this lack of cooperation to general challenges of multidisciplinary work, lack of time and burden of administrative work [96]. Both GPs and nurses with a specialist education (diabetic nurse) are described as having a key role in the management of patient with long term conditions, but patients as well as professionals describe issues in inter-professional co-operations between them [96, 97].

Practice nurses play an important role in the nurse-led, shared care which aims to encourage chronically ill people to participate actively in the management of their disease in selecting the organisations and interventions for care [98]. In addition, those that are trained as Diabetes Specialist Nurses not only teach and advise people with diabetes, but also modify and monitor self-management strategies and help develop shared goals, action plans, and skills [99]. Indicatively, in the UK and The Netherlands nurses have enhanced roles and some are trained in empowerment, counselling and communication skills [28, 98], in Greece, it is known that nursing staff in public health care operate within a restricted and task-oriented framework and their educational preparation has little effect in practice role variations and professional needs and this may also be the case for other countries [100]. In the cultural context of these countries, where family coherence is important and can become the thorax to resist challenging situations, family caregivers can play a significant role in sharing and supporting the self-management activities of people with Type 2 diabetes [99].

GP’s should perhaps take responsibility for coordinating these initiatives, reviewing guidelines and safeguarding participation [14, 55]. In addition, findings suggest that although GP’s express the view that they value the increasing patient involvement and the focus on self-management, it is not always prioritized as it clashes with time constraint [101]. GP’s have been found to report that their frustration with patients’ non adherence has resorted to them implementing tactics of “shocking” their patients by using threats as they feel they have no other therapeutic options with which to engage patients [102].

In order for the much desirable behavioural change to be achieved, consistent and long-term education and constant access to relevant information is needed [103, 104]. For example, recent meta-analysis identified that behavioural interventions increased free-living exercise and produced clinically significant improvements in long-term glucose control [105]. These features can be better sustained in the setting of primary care.

Interestingly, over half of the studied initiatives (52%, N = 11/21) have been designed to operate within a primary care framework. Genuine and enduring change requires a multilevel approach, ideally integrated into routine delivery within primary care [12]. Thus, health care services are needed that are better aligned to patient practices of self-management. The above, not unexpectedly, raises the issue of the health care professional-patient relationship and underlines that both university education and continuous medical education can be enhanced with this concept. Various points within the articles included in this review highlight the influence of the setting (especially primary care) as a key determinant in the adoption of self-management behaviours. As the international discussion on health care continues, especially intertwining among the herein participating countries [106‐109], each government, along with local organisations, can both learn and extrapolate meaningful lessons to this end. This interaction between health care professionals and patients renders a proper undergraduate education that is not the case for all the European countries.

In general, the review revealed that it is still unknown how the gender and education of patients interrelate and affect self-management and the encounter with health care practitioners. In addition, individuals’ perception of social support has received less attention from researchers. In other words; there is a lack of research adapting what might be described as a biopsychosocial approach [110], or a contextual approach [76] or a partnership approach [111] to explore the way people cope with conflicting demands and economic hardship in their striving to achieve balance between managing chronic illness and living a normal life. For example, patients with long term conditions all describe their immediate and extended family as sources of not only emotional, but also practical support [112]. Family support is especially important [113], as the need for practical support related to everyday challenges is described as frequently being ignored by healthcare practitioners [114].

Within the six participating countries, there is evidence that certain actions either as laws, initiatives or programmes have been undertaken. Thus, different management programmes serve actually the same end; specific support for patients with DM in the UK (with Diabetes Education, self-monitoring for Ongoing and Newly Diagnosed, and Dose Adjustment for Normal Eating), the National Framework and contract for patients with DM in Bulgaria, and the telemedicine and technology communication in Norway, The Netherlands and Spain.

Intensive efforts were made to elaborate on the different aspects explored in the review that primarily conveys messages to the countries that have a benefit from this FP7 project. Nevertheless, limitations of this paper should be declared. Realist reviews embrace complexity and the amount of information is sometimes hard to follow. We have attempted to make the extrapolated information more explicit by presenting the findings in different contexts in the Results and Discussion sections. Moreover, a realist review (that incorporates judgment in the interpretation and presentation) is not necessarily standardisable or reproducible in the same sense as a conventional Cochrane review [11]. Of note, while aiming at producing meaningful results at a multinational level, limited availability of studies from certain countries led to the overrepresentation of results from others, especially, The United Kingdom and The Netherlands. Based on the strict systematic method and quality appraisal of included articles, every effort was made not to omit published papers as well as to eliminate bias. Finally, recognising that social interventions are complex and that reproduction of their directions may be difficult is important. Therefore, a meaningful objective is to gather experience and identify what types of interventions work for what kinds of subjects and in which situations.