Background

Over the last few decades, the global increase in type 2 diabetes mellitus (T2DM) prevalence has become a significant economic burden to society, health systems, and patients, most directly affecting low and middle-income countries [1,2,3]. T2DM is associated with a reduced life expectancy, significant morbidity, and diminished quality of life [2, 4].

Self-management interventions (SMIs) represent a component of high-quality care for patients with T2DM [2]. SMIs are complex, multifaceted interventions, with increasing evidence of their beneficial effects, such as improving personal skills, knowledge, and self-efficacy, and in T2DM, a positive impact on glucose control [5,6,7,8,9,10]. However, it is still necessary to identify the most important SMI components, for each type of outcome, and under what circumstances they can be recommended and implemented.

Developing recommendations in healthcare requires a transparent step-by-step process, considering not only the effectiveness of interventions but also how patients value the importance of outcomes [11]. Patients’ preferences regarding outcomes can be reported as utility and non-utility measures [12,13,14]. Utilities are used to represent the strength of individuals’ preferences for different health states. Conventionally the valuations fall between 0 and 1, with 1 representing the valuation of a state of perfect health and 0 representing the valuation of death (non-existence) [15, 16]. Non-utility measures include measures explaining patients’ preferences through qualitative or quantitative methods [13]. Qualitative evidence is particularly informative regarding complex scenarios, interventions, and experiences [17, 18].

Systematic reviews (SRs) represent the most trustworthy source of evidence of patients’ values and preferences on outcomes. Over the last decade, there has been year-on-year growth of newly published articles regarding experiences or preferences in T2DM, with a corresponding increase in published SRs [19]. SRs can explore patients’ experiences with T2DM using quantitative, qualitative, or mixed-methods approaches [18, 20, 21].

A technique to summarise a large body of evidence with many SRs is to conduct a meta-review or an overview of reviews. Overviews of reviews provide broad perspectives and synthesise research fields, ideal for informing policymakers, commissioners, and providers of healthcare services [22]. They also provide insight into areas already extensively researched versus those underresearched. We considered that an overview of reviews using a mixed-methods approach, including quantitative utility-based measures and qualitative evidence, would provide a comprehensive and valid source of evidence to inform decision-making in SMIs for T2DM. Therefore, we aimed to identify and summarise available SRs exploring the perspectives of patients with T2DM and their informal caregivers, on outcomes of SMIs. This paper presents the qualitative branch of evidence and the integration of quantitative and qualitative findings.

Methods

This study was conducted in the context of COMPAR-EU, a European project aimed to identify the most cost-effective SMIs for T2DM, among other chronic diseases [23]. We conducted a mixed-methods overview of reviews, with a three-step approach and convergent parallel analysis [21, 24,25,26,27,28,29]. The first step consisted of summarising the evidence derived from SRs informing about patients’ preferences on outcomes, using utility-based measures. In parallel, the second step consisted of summarising SRs exploring patients’ perspectives, and non-utility measures, including qualitative findings. The third step involved the integration of both sources of evidence. We have published the quantitative utility SR results elsewhere [30]. In this paper, we present results from the second and third steps. The methodological details are summarised below, and further details are available in the protocol (PROSPERO CRD42019117867) [31]. We adopted the PRISMA statement for reporting [32].

Eligibility criteria

Type of reviews. SRs were defined as so if reporting: (1) a systematic search (at least in one database), (2) a list of primary studies, and (3) a description of the method of analysis. SRs could be (1) quantitative SRs, including studies with a quantitative design (e.g., surveys, cross-sectional studies); (2) qualitative evidence syntheses, including studies that applied qualitative methods (e.g., focus groups, interviews); or (3) mixed-methods reviews, including quantitative and qualitative studies that applied qualitative synthesis.

Phenomenon of interest. Perceptions and experiences of patients with T2DM and their informal caregivers with outcomes of T2DM either in relation to SMIs, self-management (SM), or the experience with the disease. The outcomes of interest were the 23 core outcomes set for T2DM SMIs of the COMPAR-EU project [23]; definitions are available in the Additional file 1.

Population. Adult patients with T2DM and their informal caregivers. The informal caregivers were considered family or friends who help patients with disease management and daily activities without payment. We included SRs that covered more than one type of population or disease if primary studies had at least 80% of adults with T2DM or if results were reported disaggregated. We excluded SRs focused only on Type 1 diabetes mellitus (T1DM), children, gestational diabetes, or healthcare professionals’ perspectives.

Setting and language. We did not establish geographical or setting restrictions, except those confined to inpatient care. We included studies published in English only.

Search strategy

We searched MEDLINE (accessed through PubMed), the Cumulative Index of Nursing and Allied Health Literature (CINAHL), and PsycINFO from inception to November 2020. The search was updated monthly using each database’s alert system until June 2021. We applied a sensitive content search strategy previously published [33] and specific terms for T2DM [Additional file 2]. We limited our search to SRs by using methodological filters in each database. Other sources were the reference list of overviews identified through our search strategy and a forward citation search of selected SRs in Scopus.

Study selection

After achieving at least 80% agreement with an initial calibration exercise (with 10% of the references), a pair of authors (JB, CRC) screened titles and abstracts, with a subsequent independent full-text assessment. Disagreements were solved by discussion or with the help of a third author (ENDG). We managed references using Endnote X9 and Rayyan.

Data collection

We used Nvivo 12 PRO software and Excel spreadsheets to collect and analyse qualitative data. After pilot-testing the data extraction table and the coding process, two authors (ENDG, JB) extracted the predefined data, including general characteristics, methodological characteristics, settings, participants and intervention characteristics, and themes or findings. The latter was the first stage of thematic synthesis.

Assessment of methodological quality

We applied the 11-item Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Systematic Reviews and Research Syntheses [24]. One item regarding the likelihood of publication bias was removed since it only applies to SRs conducting metanalysis. After initial calibration, one author (JB) applied this checklist, and a second author validated the responses (CRC). Disagreements were solved by consensus or, if necessary, with the help of a third author (ENDG). We calculated the percentage of positive responses and classified the quality as low (0–33%), medium (34–66%), or high quality (≥ 67%) [34].

Assessment of overlap

We measured the extent of overlap of primary studies using a citation matrix and calculated the “corrected covered area” (CCA) [35, 36]. Overlap was classified as slight if the CCA is < 5%, moderate if it is ≥ 5% and < 10%, high if it is ≥ 10% and < 15%, and very high if CCA is ≥ 15% [36].

Data synthesis and analysis of qualitative data

We applied a thematic synthesis approach for developing fourth level descriptive and analytical themes [37]. There are four levels of interpretation in a qualitative meta-review (overview of reviews). The first-level is the participant’s interpretation of their experiences in the primary research; the second-level is the researcher’s reflections and report on the primary study; the third-level involves the synthesis of findings from studies included in a SR, and the last is the meta-review level (fourth level). In this study, we focused on the second (as reported in SRs) and third-levels of interpretation (results and discussion sections of SRs) to derive the fourth level themes. We did not return to the original studies [38].

The process consisted of (1) line-by-line coding of results and discussion sections; (2) classifying codes according to the 23 core outcomes set of COMPAR-EU (including an active search of terms using Nvivo); (3) developing fourth level descriptive themes based on semantic correspondence; (4) developing fourth level analytical themes to establish broader patterns or relationships among findings. We analysed and reported fourth level themes with different levels of detail: (1) analytical themes per outcome, including descriptive themes providing narrative accounts per each, (2) overarching analytical themes, and (3) summary, proposing inductively a plausible explanation of how patients with T2DM (or informal caregivers) perceive the importance of outcomes of SMIs. We conducted a subanalysis for two populations, informal caregivers and patients from ethnic minorities living in Western countries.

Two reviewers (ENDG, JB) developed the synthesis of descriptive and analytical themes; one independently proposed themes, and a second reviewed them independently, followed by an iterative collaborative analysis. Both reviewers discussed alternative interpretations and ensured that fourth-order levels of interpretation remained grounded in the third and second levels (as reported in SRs). A third reviewer (PAC) independently reviewed fourth level themes and coherence of the narrative accounts.

Integrating quantitative and qualitative evidence

We used a side-by-side comparison conjoint display [25, 39] to integrate synthesized findings from SRs reporting utility-based measures (published elsewhere) with findings from SRs reporting non-utility measures (including qualitative findings), presented in this overview. The integration process included assessing whether datasets were in discordance, confirmation, or expansion; or if there were missing or less explored areas [40]. Discordance was defined as when results were inconsistent or contradictory. Confirmation occurred when findings reinforced each other. When findings from one type of data were expanded upon insights from another, this was considered expansion [40]. We synthesized and reported narratively. One reviewer independently conducted this step (ENDG), and a second reviewer cross-checked the analysis and synthesis (PAC) [21].

Results

Study selection

We included 54 SRs, selected as described in the PRISMA flowchart (Fig. 1) (PRISMA 2020). Reasons for excluding studies reviewed in full-text are available in Additional file 3.

Fig. 1
figure 1

PRISMA flow diagram of study selection

Full size image

Study characteristics

The main characteristics of included SRs are described in Table 1 and Additional file 4. We included 25 qualitative evidence syntheses (25/54; 46.3%), 22 mixed-methods reviews (22/54; 40.7%), and seven quantitative SRs (7/54; 13.0%), corresponding to 939 primary studies. These were conducted in 19 different countries; the most frequent were UK (14/54; 25.9%), USA (8/54; 14.8%) and Australia (6/54; 11.1%). The majority were published between 2016 and 2020 (32/54; 59.3%) and included a range from 5 to 120 primary studies. Most SRs included patients with T2DM only (38/54; 70.4%), and some focused on patients from ethnic minorities living in Western countries (8/54; 14.8%), and others included informal caregivers (7/54; 11.1%). The majority addressed features of SM (39/54; 72.2%) (Fig. 2, Additional file 4).

Table 1 General characteristics of included reviews
Full size table
Fig. 2
figure 2

Phenomena of interest across systematic reviews

Numbers represent the number of SRs in each category. Diabetes: SRs exploring features of the disease trajectory with broad lenses, referring to the lived experience of dealing with diabetes. It included the views of specific subpopulations, including older patients and patients from ethnic minorities. Diabetes SM: SRs focused on themes regarding SM behaviours, and abilities patients acquire in diabetes care. The subcategory Barriers represents SRs exploring challenges and facilitators to SM, and Caregivers, refer to SRs exploring the role of social support in SM. Diabetes SMI: SRs exploring perspectives and experiences with components of SMI, such as telemedicine, educational programmes, and culturally tailored SMI (in ethnic minorities)

Full size image

Studies’ quality and overlapping

Most SRs evaluated the methodological quality of included primary studies (42/54; 77.8%) using different tools. The most frequent was the Critical Appraisal Skill Program checklist (18/54; 33.3%). One study applied CERQUAL to assess the certainty of evidence [41]. When evaluating the SRs’ quality with the JBI Critical Appraisal Checklist, most were considered high quality, with ≥ 67% (47/54; 87%). The most frequent concern was the absence or unclear criteria to appraise studies (23/54; 42.6%) (Additional file 5). The overlap analysis of the primary studies showed a corrected covered area of 0.75%, meaning there was a slight overlap. When comparing SRs, we identified that 126 studies were included in at least 2 SRs, 40 in 3 SRs, and 20 in 4 SRs (Additional file 6).

Fourth level themes

We developed summaries, including descriptive and analytical themes with narrative accounts for all the outcomes of SMI for T2DM from the core outcome set of COMPAR-EU, except for unscheduled care. Summaries per outcome are available online. See: https://osf.io/dj3wy/?view_only=f0f3a82ea97747b59beb4de2a11c05f8. We identified 101 descriptive themes (Table 2) and six overarching analytical themes; their narrative accounts are described below.

Table 2 Fourth level descriptive themes per outcomes of Self-Management interventions
Full size table

Diabetic identity

Patients with T2DM need to adopt a “diabetic identity” to confront the diagnosis and have a sense of ownership in disease management. However, not all patients are willing or able to SM [42,43,44]. Adopting a diabetic identity is fundamental to effective SM. The ability to engage in diabetes care (“sense of agency”) is determined by several contextual factors, of which financial ability is central [42, 45,46,47,48,49,50]. Self-care ability evolves in the tension between disease taking control of life, and patients taking control of the disease and living in the present, and for the future [44, 51]. Patients can start their T2DM journey with a medication prescription, associated with negative emotions. Their social environment, beliefs and experiences can reinforce their perceptions [52]. Talking to peers and sharing experiences is an important source of emotional support [53, 54]. These interactions provide opportunities to discuss challenges and fears openly and a sense of belonging to a group, which enhances confidence to overcome the barriers to initiate and maintain SM [55].

Beliefs about the seriousness of diabetes, treatment effectiveness and a passive role in interacting with healthcare providers (HCPs) predict the level of SM in diet, exercise, and glucose testing [42, 44, 47, 48, 51, 52, 54, 56,57,58,59]. Good communication and effective collaboration between patients and HCPs are crucial to generate opportunities for self-care knowledge, facilitated by regular, repeated and timely contact [42,43,44,45, 48, 60].

Accessing healthcare

Patients need access to and interaction with healthcare services and HCPs, to make informed decisions. Low numeracy, literacy skills or health literacy impede patients from understanding health-related information and receiving or accessing diabetes care services [47, 50, 57]. Challenges to gaining knowledge and finding information are higher in patients who do not speak or read the dominant language fluently [50, 61, 62]. Patients with inaccurate cognitive disease representations have alternative interpretations or biased perceptions of their risk of complications [45, 53, 63,64,65]. Preferences for the amount and quality of the information provided by HCPs are mixed. Some patients are satisfied with the information, others ask for more pragmatic advice, and others feel reluctant or unable to articulate their concerns [45, 60]. While HCPs of patients with low health literacy seek to improve knowledge by repeatedly checking for comprehension, patients value being heard and asked about their circumstances, making the information more relevant [65].

Experience of care

The experience of care depends on the degree of individualised care, the nature and characteristics of the patient-HCP relationship, and cultural appropriateness. Most patients value an individualised provision of care, feeling heard and being given recommendations, tailored to their perceived needs [43, 45, 49, 60, 63, 66]. Patients identify that relationships with HCPs can be collaborative, sharing control and responsibility, or paternalistic, where HCPs are an authority [46, 51, 53, 60, 66, 67]. Cultural appropriateness of care includes linguistic appropriateness and awareness of the mismatch between recommendations and patients’ beliefs [46, 50, 54, 67,68,69,70].

The divergent agenda between patients and HCPs, the quality of communication, the cultural context, and the lack of time, influence the decision-making process in diabetes care. Patients value emphatic interactions. However, they usually face criticism and blame, which undermines communication and autonomy [44, 53, 54, 60, 67]. High-quality communication defined as the effective and meaningful exchange of information that considers patients’ unique requirements, values, and preferences, is associated with improved SM and enhanced well-being [53, 60, 67, 71, 72]. Patients perceive a lack of time as inadequate support [42, 53, 54, 60, 67, 70]. Patients who feel unheard do not attend appointments, avoid asking questions, or discuss personal issues [60]. Patients perceive HCPs as disease-oriented, leaving patients feeling that their experiences are devalued, discouraging SM efforts [42, 56, 60]. Trustful and friendly communication with HCPs facilitates participation, responsibility, and safety in self-care [66].

Screening for complications

Receiving a recommendation from primary care HCPs facilitates retinopathy screening [64, 73]; this decision is influenced by the perception of how competent or qualified the HCPs are [73]. Patients often report problems with referrals, scheduling appointments, long waiting times and lengthy appointments, which can be problematic because of food abstinence [64, 73]. Barriers to primary care or retinopathy screening clinics include transportation, language and cultural barriers, and work commitments [42, 49, 70, 73]. Although most patients prefer flexibility when setting appointments, some value fixed ones [73]. Most patients are willing to use and are generally satisfied with technological health interventions [74, 75]. The willingness to adopt them is influenced by the sense of independence, the perceived improvement and ease of use [75, 76]; however, some perceive it affects trust and confidentiality with HCPs [76].

Barriers to attending retinopathy screening and foot care are multi-dimensional [53, 64, 73]. Patients usually have limited comprehension of diabetic foot ulceration (DFU) and amputation, which have significant and enduring effects on patients’ quality of life. Overall, most patients ignore foot care advice to maintain a normal life [53]. Telehealth interventions for foot care could improve patient-HCP interactions [76, 77].

Engagement with self-management

Self-efficacy

Patients’ reflections on SM are connected to everyday life habits, traditions, cultural beliefs, preferences, attitudes, and the patient’s social network [42, 44, 46, 50, 58, 61, 70, 78,79,80,81]. Developing self-efficacy skills can be challenging, especially when patients perceive having full responsibility, limited control, a sense of hopelessness and resignation [42,43,44,45, 53, 55]. Patients become more proactive when they start achieving goals, feel in control, understand their responsibility, and gain confidence [44, 53]. Patients consider SM support adequate when they receive timely information and advice, and perceive HCPs have considered their circumstances, the psychological and emotional burden of T2DM, disease chronicity, and treatment complexity [45, 51, 68, 82]. The burden of prescribed self-care and unrealistic expectations of HCPs are barriers to SM compliance [44].

Healthy diet

Most patients and informal caregivers recognise having limited knowledge regarding nutritional concepts and how to implement them [42, 46, 57, 61, 70, 83, 84], and perceptions of a healthy diet are culturally influenced [46, 61, 69, 70, 79]. Adopting a healthy diet requires self-discipline strategies, such as portion control, avoiding tempting food and being proactive. However, healthy options are often reported as unreachable due to their high costs [41, 46, 49, 58, 61, 66, 68, 70, 71, 85]. Social situations and family support influence adherence to a healthy diet; some patients may need to adjust and subordinate their diet due to social factors [41, 43, 46, 49, 58, 66, 68, 70, 71, 78, 82].

Physical activity

Knowledge is insufficient to initiate and engage in physical activity; patients need physical and non-physical skills [55]. Expected and experienced benefits influence these decisions [49, 55, 70]. Barriers to physical activity include low self-efficacy, negative attitudes, physical weakness, symptoms persistence, comorbidities, physical limitations, financial constraints, unsafe neighbourhoods, competing demands, perceived social support, travel distance, lack of culturally sensitive options and transport [44, 46, 48, 50, 51, 55, 58, 70, 73, 83, 86].

- Treatment and medicine-taking behaviour.

Adherence to treatment and lifestyle behaviours requires self-regulation, and a deliberate effort to live as normally as possible [48, 52, 85, 87]. Patients can present five patterns of medicine-taking behaviour: (1) strict adherents who strongly believe in treatment benefits; (2) those who accidentally miss a dose, who usually feel guilty; (3) unintentional non-adherents who do not feel guilty, possibly due to lack of symptoms and a belief that diabetes is not serious; (4) intentional non-adherents who delay, skip, or adjust doses, often manipulating blood glucose and diet, without feeling guilty, and (5) intentional non-adherents who feel guilty and usually had negative experiences with treatment [52]. Most patients wish to minimise daily medications or simplify their regimens [47, 52, 71]. Factors influencing treatment adherence are multi-dimensional [46, 47, 50, 56, 58, 59, 61, 70, 71, 78, 83, 85, 87, 88]. Patients’ worries include the treatment’s effectiveness, side effects (hypoglycaemia or weight gain), interactions between different medication regimens that could have negative long-term effects, needle anxiety, fear of injection, pain, stigma and discrimination [52, 70, 71]. Insulin is perceived as an indicator of the worse type of diabetes and is associated with more side effects than oral glucose-lowering agents. Insulin therapy is considered inaccessible, impractical and unacceptable, restricting patients’ lives, including daily and social activities [46, 52, 56, 59, 85, 88, 89]. The complexities of managing insulin include injection difficulties and regimen inflexibility, forgetting doses, and the titration of the insulin dose [46, 52, 56, 59, 85, 90]. Fear of complications, experiencing comorbidities, awareness of treatment necessity, receiving support and temporary trials, reduce barriers to treatment progression to insulin [44, 49, 52, 59, 71, 88, 90]. Patients prefer non-judgmental guidance on alleviating negative experiences with insulin, having some control over changes and real-life advice [45, 52, 56].

Self-monitoring

Self-monitoring of blood glucose (SMBG) helps patients understand the relationship between blood glucose levels and disease progression, treatment and prognosis. However, despite knowing what to do, patients might not always have the time or energy to respond [43, 46, 47, 59, 63, 66]. Patients prefer when HCPs customise blood glucose plans based on their condition; however, inadequate HCPs interactions decrease the willingness and efficacy of SMBG [45, 63]. Barriers to SMBG include problems with monitoring devices, lack of confidence, misconceptions, fatalistic beliefs, and the perceived burden [44, 58, 63, 66, 78, 84]. Patients not using insulin focus on regulating daily food intake and lifestyle, while patients who use insulin focus on insulin dosage. SMBG has become a helpful habit for preventing and detecting hypo- and hyperglycaemic symptoms [56, 63]. Some patients who perform SMBG refuse to assume an active role, while others are willing to accept the responsibility actively [56, 63, 75, 86]. Foot self-care is often considered a lower priority than immediate demands [53, 72].

Outcomes awareness

Poor glycaemic control is frequently associated with negative behaviours, impatience, and fatalistic beliefs, or being just aware of their HbA1c values but not feeling confident and motivated to improve their diabetes [47, 50, 81]. Some patients develop incorrect causal relationships between symptoms and complications, considering symptoms of poor SM and complications inevitable [46]. Patients often deal with psychological and emotional distress or a sense of failure due to unacceptable blood sugar levels and the constant threat of hypo- and hyperglycaemia [47, 54, 66, 83].

Most patients do not experience warning signs of hypoglycaemia. Its ocurrence causes fear and reduced treatment adherence, which explains why some patients prefer maintaining high blood glucose levels [46, 52, 54, 56, 57, 59, 66, 85, 86, 88]. Insulin-related nocturnal hypoglycaemia disrupts diabetes SM and quality of life [43, 56, 78]. Weight and lipid control are SM goals. Fear of weight gain or the inability to lose weight is a common worry for patients using insulin; however, for others, weight loss can be an expected benefit [52, 56, 90].

Social support and interactions with HCPs influence adherence to weight loss educational programmes [47, 52, 60]. Patients often do not perceive the benefit of lipid control [71], probably due to the absence of symptoms and the lack of awareness of the increased cardiovascular risk [65].

Challenges adhering to self-management

Barriers to SM include insufficient knowledge about the seriousness of the disease, the risk of complications or the importance of preventive care [44, 48, 51]. Patients’ beliefs regarding diabetes causes and the perceived sense of control influence SM behaviours [42, 46, 50, 61]. Even though knowledge can improve well-being, it is not enough to motivate patients to engage in healthy behaviours. The lack of motivation is the main obstacle to seeking information. Some patients feel it is pointless to manage the disease since complications would manifest regardless of any action taken [47, 49, 59, 74].

Patients often feel overwhelming negative emotions related to diabetes diagnosis, complications, and SM [46, 56, 57, 73, 80, 86]; they experience a loss of confidence, the disruption of usual roles and their sense of independence [43, 45, 46, 53, 56, 57, 73, 80, 86, 89]. The stigma associated with T2DM, defined as patients’ expressions of embarrassment and moral failure, can be related to diagnosis, complications, and medication use. Stigma may prevent patients from disclosing their diagnosis, leading to an impaired ability to SM [41, 46, 53, 56, 71, 73, 85, 89, 91]. Simultaneous demands of competing comorbidities represent an extra burden [44, 47, 49, 85, 88].

Patients usually feel worried or anxious about premature death, despite underestimating the likelihood of fatal events [65, 86]. Some patients express hopelessness, powerlessness, a sense of inevitability, and fatalism regarding diabetes, irrespective of treatment. They consider decisions in the past cannot be redressed [41, 50, 53, 56, 71]. The sense of fatalism leads to low patient motivation to partner HCPs, which could be influenced by culture.

Patients’ finances have an impact on disease management. Patients experience a continuous trade-off between health-related costs and other concerns [41, 42, 44, 47,48,49, 61, 70, 72, 76, 77, 83, 88]. Patients with acute or chronic complications report more significant healthcare expenditures. Diabetes can considerably impact patients’ and informal caregivers’ finances [56, 84]. Insurance schemes with reduced or non-existent co-payments improve treatment adherence. Some patients perceive the limited availability of personnel and resources due to the need for more healthcare system funding [41, 42, 47, 53, 62, 71, 85]. Patients and informal caregivers perceive telehealth care as alleviating financial burdens due to reduced healthcare utilisation and lower treatment and travel costs [42, 76, 77]. The economic impact of medical expenses on patients and informal caregivers was most frequently reported in high-income countries without public health system funding. There is a significant disparity in the financial burden for vulnerable subpopulations, such as immigrants, older people, and patients with lower levels of education; in some cases, cultural factors may also play a role.

Subgroup analysis

Family informal caregivers

Family informal caregivers face the dilemma of protection versus enabling autonomy [89]. Informal caregivers often support patients with depressive and behavioural problems. Nevertheless, they report lacking support when dealing with events that could make it difficult for patients to follow SM behaviours [84]. Patients often need family support for privacy and shared responsibility when checking blood glucose [58, 78]. Fluctuating blood glucose levels are a serious concern to informal caregivers; for instance, hypoglycaemia is challenging since they may need to take control in acute situations [58, 78, 84]. Family informal caregivers can promote dietary adherence and physical activity. Encouragement with a gentle, positive approach is well-received and effective [82]. Family is usually perceived as a source of motivation and confidence [55, 68]. However, for some patients, it is perceived as interference. Supportive interactions occur when the goal is to adhere to the recommended lifestyle, to achieve physical and psychological well-being for both of them. Non-supportive actions include sabotaging diet and promoting unhealthy family habits or routines [44, 46,47,48,49, 54, 55, 58, 70, 73, 78, 88]. Some informal caregivers fear being affected in their job and financial status due to diabetes demands [56, 84].

Ethnic minorities living in Western countries

Patients from some ethnic minorities living in Western countries often report inevitability and fatalism in their diabetes perception, considering disease management beyond their control or a consequence of their past decisions. However, some perceive faith has a direct influence on the outcome but does not absolve them from taking responsibility [41, 46, 50, 61]. A sense of fatalism leads to low motivation to partner with HCPs to address diabetes management [41, 50, 70, 71]. The cultural significance of food usually conflicts with the Western concept of a healthy diet, which is portrayed as a form of self-denial to improve health. These patients often describe strong food traditions, despite awareness of the detrimental effects of some types of food or behaviours on health, and the importance of following a healthy diet. When HCPs advise against some traditional foods and discount them as harmful, it can be difficult to follow dietary recommendations, preventing change and SM [41, 43, 46, 50, 61, 70]. Patients are often unaware of the risks of experiencing complications, especially macrovascular events [65].

Most South Asians in Western countries prefer traditional therapies and lay sources of knowledge. They perceive that exercise depletes energy and prioritises the family’s diet over their needs; they also perceive that diabetes care burdens their family [61, 71]. The main factors affecting the experience of care are linguistic appropriateness, and the mismatch between HCPs’ recommendations and patients’ beliefs [46, 50, 54, 67, 70].

Middle Eastern or South Asia cultures rely more on tradition and authority, making it challenging to engage in shared decision-making [54, 67, 70]. Most patients consider physicians the primary information source and acknowledge low self-efficacy [42, 46, 50, 61, 70]. Patients value linguistically concordant support highly [50, 69]; however, some feel reluctant to communicate with HCPs through interpreters [46, 50, 70]. Most patients prefer professional interpreters to relatives or friends [50, 70]. Participants of culturally tailored interventions consider that these interventions facilitate healthcare access [68].

Integrating utility and non-utility measures for outcomes of T2DM-SMIs

We developed a conjoint display integrating quantitative (utility-based measures) and qualitative findings (non-utilities measures) (Table 3). For glycaemic control, results from both sources of evidence expanded each other; however, some discordance was identified in poor control that mismatched qualitative findings. Evidence from utility measures was informed by a single study, and would require additional research to confirm or not this value.

Table 3 Conjoint display utility and non-utility measures of outcomes importance
Full size table

In weight change, findings also expanded each other. Utility measures referred to a higher importance of extreme obesity than being overweight, while non-utility measures referred to the fear of weight changes concerning treatment consequences. Long-term complications were more detailed by utility-based measures. The most important outcomes for patients were diabetic neuropathic pain, blindness and amputation; in qualitative findings, we also found evidence for diabetic foot ulceration, amputation and cardiovascular risk. The outcome burden of SM included findings from adherence to a healthy diet, physical activity, and treatment, quality of life, scheduled care, and experience of care. Quantitative utility-based findings were discordant. This difference can be explained by the indirectness in the population and measures considered. Hypoglycaemia findings were complementary and confirmatory; on the one hand, a body of evidence expressed the outcomes’ importance in terms of severity, and on the other hand, qualitative evidence informed the experience and the difficulties in detecting, treating and preventing this event. Non-utility measures informed lipid control and quality of life.

Summary

The identified research evidence shows there is important variability in how patients with T2DM value critical outcomes, mainly influenced by contextual factors and the degree of disease progression. SM is possible when patients can adjust and accept their diagnosis and treatment; furthermore, health literacy is critical since it unlocks healthcare access. Knowledge provision is better received in a positive patient-HCP relationship within a culturally sensitive approach. In this sense, the decision-making process with HCPs enables patients’ engagement in SM. SMIs with adequate support enhance patients’ self-efficacy; however, it requires building capabilities, behavioural skills, social support, and scheduled care.

It is difficult for most patients and their informal caregivers to perceive the risk of long-term complications. Being able to perform self-monitoring facilitates awareness of glycaemic complications and glucose control; however, fear of hypoglycaemia and weight change may hinder adherence to treatment. The loss of quality of life, physical and psychological constraints, associated costs, comorbidities and misleading life expectancy beliefs represent barriers to SMIs. Thus, to integrate SM into everyday life, patients and informal caregivers require tailored, contextualized, self-paced training.

Discussion

Main findings

We synthesised 54 SRs of diverse nature on how patients perceive or experience outcomes when dealing with SM, corresponding to 939 studies across 19 countries. We found important variability in how patients with T2DM and their informal caregivers value critical outcomes of SMIs, mainly influenced by contextual factors and disease progression. SM is possible when patients can adjust and accept their diagnosis and treatment. It requires building capabilities, behavioural skills, social support, and attending to scheduled care. Patients and informal caregivers need contextualized self-paced training to integrate SM into everyday life.

The themes frequently overlapped since qualitative research findings are not necessarily neat and discrete. Family informal caregivers can also be affected by the SM process and financial requirements of diabetes care. They are involved when patients use SMIs and recognise having limited knowledge to engage with SM successfully. The family acts as a facilitator of SM, especially when they need to assume responsibility for the patient’s health in acute events. However, in some cases, non-supportive family interactions limit SMIs’ effectiveness, adding extra burden for patients. Most patients from ethnic minorities living in Western countries prefer to receive support that considers their cultural beliefs and traditions. Culturally tailored interventions are generally perceived as facilitators.

When comparing utility and qualitative findings, the former informed the direction of patients’ preferences and the typology and severity of outcomes of SMIs (e.g., severe hypoglycaemia vs. mild one or extreme obesity vs. overweight). The latter provided information regarding the burden associated with SM (e.g., barriers to adhering to a healthy diet, or physical activity).

Our findings in the context of previous research

Our results are consistent with findings from previous metareviews, exploring the perspectives of patients with other chronic diseases [19, 92,93,94]. Overall, patients’ perspectives on SMIs vary according to the disease stage and the specific SM process, which requires time and deliberate effort to integrate into daily life [19]. These perspectives are influenced by contextual factors, including the perceived benefit and usability of the intervention, the sense of community, and the level of individualised care [19, 92]. Patients with hypertension also perceived adherence to SM as challenging [94]. Barriers to SM included familial (lack of support, need for separate meals), environmental (sense of security, local amenities, healthy food availability), financial status and logistical (frequency of appointments, work schedules, accessibility) reasons. The reviews also identified a degree of deliberateness in non-adherent behaviours, which was influenced by the perception of symptoms, the disease severity, stress or fear of dependency. A common finding across reviews, is that having a collaborative, supportive relationship between patients and HCPs is crucial for effective SM [93, 94].

Previous reviews that included HCPs’ perspectives found that their views complement patients’ and informal caregivers’. HCPs in diabetes care considered the main barriers to implementing SM were the limited resources, heavy workloads and environmental constraints [19, 95]. Consultation time was the most frequently reported factor [56, 60, 96, 97]. HCPs often reported being unable to address patients’ broader psychosocial aspects of diabetes care, because of a lack of experience or training in effective communication, counselling, goal setting and shared decision-making [47, 60]. Some described experiencing frustration around patient compliance; and inadequacy and helplessness at being unable to address psychosocial concerns [58, 95], and others considered non-adherence primarily as patients’ own failure [60].

Our findings resonate with theoretical frameworks that explain attitudes and behaviours in patients with chronic conditions. These include the “Common sense model of self-regulation of health and illness” [98], “The shifting perspectives model of chronic illness [99], “the integrative model of behaviour prediction” [100] and the “Health Belief Model” [101]. Consistent with these frameworks and their different but complementary components, our findings described that a crucial step to start with SM is adopting the diabetic identity. The existence of barriers such as limited knowledge, weight change beliefs, the fear of insulin injections, fatalistic beliefs, costs or quality of life constraints can limit the chances of seeking treatment. However, being aware of the importance of outcomes and starting self-monitoring can help to gain insight into their disease, and learn the influence of lifestyle behaviours and treatment. Shifting processes, such as changes in treatment requirements and acute events, can affect adherence to SM. The experience of an acute event or long-term complications could represent cues to action where patients can find SMIs valuable, especially when they start perceiving benefits and self-efficacy.

Strengths and Limitations

Our study has several strengths, including using mixed-methods, selecting a broad scope based on the principles of SM, and considering SM as a continuous process transversal to the diabetes journey instead of limiting to the term “self-management” or focusing only on SMIs. We also applied a sensitive search using a validated content search strategy. We also applied rigorous iterative methods to collect and analyse data, and ensure a balanced interpretation. We provided detailed information on the outcomes of T2DM SMI and proposed a logical sequence of findings to explain how outcomes interact and their relevance in the SM process. The high level of data saturation for some outcomes, with only a slight overlap across reviews (CCA = 0.73%), increases the credibility of the findings.

Our study also has some limitations. Since we analysed data based on second-order findings, we acknowledge that individual SRs authors’ analyses could influence our results. However, data saturation for some outcomes makes this potential source of bias unlikely. Moreover, obtaining and reanalysing first-order findings (primary data) was beyond the scope of this overview. Despite most included SRs being high quality and applying rigorous and transparent synthesis methods, the high variability of data reporting has limited us from providing complete information regarding the quality of primary studies and population characteristics. Since we included only English publications, sample origins included patients mostly from high-income countries. Our findings might lack generalisability to patients from low or middle-income countries. Finally, given the inconsistent reporting of information, we could not evaluate the certainty of evidence.

Implications for practice and research

SMIs are complex and context-specific interventions that require an in-depth understanding of the unique experiences of specific T2DM subpopulations. We propose recommendations for policymakers and HCPs when developing SMIs For example, the need to develop culturally sensitive programmes to facilitate patients’ adaptation to T2DM diagnosis, treatment, and decision-making with HCPs (Table 4). The main areas of improvement include providing psychological support, tailoring messages to health literacy and ethnic considerations, detecting people at high risk early and supporting patients based on their specific needs at different stages.

Table 4 Recommendations for Policy, Practice and Future Research
Full size table

We identified some quantitative and qualitative research gaps, reflecting either the lack of SRs or primary studies in these areas. The underresearched outcome in both types of evidence was unscheduled care. In the qualitative branch, we found scarce evidence for comorbidities, hyperglycaemia, lipid control, and complications such as cardiovascular diseases, neuropathic pain, renal disease, and stroke. On the other hand, we consider some themes close to saturation, such as barriers to SM, adherence to treatment/insulin, and SM in ethnic minorities. In the quantitative utility-based research, we did not find evidence for psychological distress and lipid control, and scarce findings for the burden of SM and the importance of glycemic control.

Summaries can vary in extension or detail according to the outcome definition or the scope. Our findings can inform the formulation of recommendations for other healthcare decisions in diabetes care. These findings can also inform the development of educational materials and decision-making tools.

Conclusion

Our results represent what patients with T2DM and their informal caregivers perceive as most important when dealing with SMIs. Their perspectives on SMI outcomes are variable since they are affected by the progression of the disease and several contextual factors. We found quantitative or qualitative evidence for almost all outcomes of the core outcome set of T2DM SMIs. Considering the availability of different types of SMIs, decision-making must incorporate patients’ and informal caregivers’ values and preferences on the importance of SMIs outcomes. We have summarised available SRs on this topic and identified some research gaps. Our results can facilitate the development and evaluation of SMIs, and guide decision-making in diabetes care, including the formulation of decisions and recommendations.