Contribution of family social support to the metabolic control of people with diabetes mellitus: A randomized controlled clinical trial

Applied Nursing Research 36 (2017) 68–76

Contents lists available at ScienceDirect

Applied Nursing Research journal homepage: www.elsevier.com/locate/apnr

Contribution of family social support to the metabolic control of people with diabetes mellitus: A randomized controlled clinical trial Lilian Cristiane Gomes a,⁎, Anna Claudia Martins Coelho a, Danielle dos Santos Gomides a, Maria Cristina Foss-Freitas b, Milton César Foss b, Ana Emilia Pace c a b c

Fundamental Nursing Graduate Program, EERP/USP, Ribeirão Preto, São Paulo, Brazil Department of Medical Clinics of the School of Medicine, Ribeirão Preto/USP, Ribeirão Preto, São Paulo, Brazil Department of General Nursing and Specialist at EERP/USP, Ribeirão Preto, São Paulo, Brazil

a r t i c l e

i n f o

Article history: Received 4 December 2016 Revised 11 April 2017 Accepted 25 May 2017 Available online xxxx Keywords: Nursing care Diabetes mellitus Health education Social support Clinical trial

a b s t r a c t Aim: This randomized controlled clinical trial aimed to evaluate the contribution of family social support to the clinical/metabolic control of people with type 2 diabetes mellitus. Background: Diabetes mellitus is a chronic disease that requires continuous care in order for individuals to reach glycemic control, the primordial goal of treatment. Family social support is essential to the development of care skills and their maintenance. However, there are few studies that investigate the contribution of family social support to diabetes control. Methods: The study was developed between June 2011 and May 2013, and included 164 people who were randomized using simple randomization. The intervention group differed from the control group in that it included a family caregiver, who was recognized by the patient as a source of social support. The educational interventions received by people with diabetes mellitus were used as the basis of the education provided through telephone calls to patients' family members and caregivers, and their purpose was to encourage dialogue between the patients and their relatives about the topics related to diabetes. Results: Regarding the clinical impact, the results showed that there was a greater reduction in blood pressure and glycated hemoglobin in the intervention group than in the control group, showing a positive effect on the control of the disease. Conclusions: Families should be incorporated into the care of people with diabetes mellitus and especially in health care programs, in particular those that can promote different forms of social support to strengthen the bond between family members. © 2017 Published by Elsevier Inc.

1. Introduction Type 2 diabetes mellitus (T2DM) is a chronic disease of particular importance not only because of the increase in its prevalence but also Abbreviations: T2DM, Type 2 diabetes mellitus; HbA1c, glycated hemoglobin; BMI, body mass index; WC, waist circumference; DM, diabetes mellitus; SS, social support; H0, null hypothesis; AH, alternate hypothesis; T0, before the start of the intervention (baseline); T6, six months after the start of the intervention; T12, three months after the end of the intervention; ICF, Informed Consent Form; SCT, Social Cognitive Theory; IG, intervention group; CG, control group; SD, standard deviation; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose; CI, confidence interval; LLCI, lower limit of the confidence interval; ULCI, upper limit of the confidence interval; UKPDS, United Kingdom Prospective Diabetes Study; T1DM, Type 1 diabetes mellitus. ⁎ Corresponding author at: Rua João Carlos Farah, 340, Jardim Agenor de Lima, CEP: 37.800-000, Guaxupé, MG, Brazil. E-mail addresses: [email protected] (L.C. Gomes), [email protected] (A.C.M. Coelho), [email protected] (D.S. Gomides), [email protected] (M.C. Foss-Freitas), [email protected] (M.C. Foss), [email protected] (A.E. Pace).

http://dx.doi.org/10.1016/j.apnr.2017.05.009 0897-1897/© 2017 Published by Elsevier Inc.

because when poorly controlled, it is associated with other consequent morbidities, including cardiovascular, renal, ophthalmic, and neurological diseases; these comorbidities can result in physical disability and premature death (Francisco et al., 2010; Lessa, 2004). Brazil, which was ranked eighth among the top ten countries with the highest number of T2DM cases in the world (4.6 million) in 2000, is expected to be ranked sixth in 2030, with an estimated 8.9 million people diagnosed with the disease (Wild, Roglic, Green, Sicree, & King, 2004). Due to the chronic nature of the disease, persons with T2DM require long-term monitoring to maintain their care and glycemic control; this control is the main focus of treatment, which primarily aims to prevent and/or delay the serious chronic complications of the disease (Sociedade Brasileira de Diabetes, 2016). The need for behavioral changes and the incorporation of new technologies into treatment can be challenging for persons with T2DM and their family and may require new skills for maintaining care. Accordingly, health education is considered essential to the treatment of T2DM (Francisco et al., 2010) because of its contribution to preventing acute events and to reducing the risk of

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

chronic morbidities during the disease course (American Diabetes Association, 2016). Health education comprises a set of planned activities that aim to improve people's knowledge and skills to maintain health, improve engagement in healthy behaviors, and promote changes in attitudes and perceptions about disease to facilitate adjustment to new situations of daily living (Chrvala, Sherr, & Lipman, 2016; Hoving, Visser, Mullen, & van den Borne, 2010). Health education is also considered a strategy that can be used to help persons take responsibility for self-care, function autonomously and become motivated to use their own knowledge and skills in problem solving (Bagnasco et al., 2014; Haas et al., 2012). Previous studies have demonstrated the benefits that educational interventions can have on glycemic control in people with T2DM, especially interventions that combine behavioral, cognitive and affective aspects in a collaborative approach and include regular reinforcements for participants (Chrvala et al., 2016; Cooper, Booth, Fear, & Gill, 2001; Eakin et al., 2014; Klein, Jackson, Street, Whitacre, & Klein, 2013; Lynch, Liebman, Ventrelle, Avery, & Richardson, 2014; Minet, Møller, Vach, Wagner, & Henriksen, 2010; Norris, Engelgau, & Narayan, 2001; Norris, Lau, Smith, Schmid, & Engelgau, 2002a; Norris et al., 2002b). However, these same studies have differed regarding the frequency of contact with the educator, the follow-up time (Chrvala et al., 2016; Cooper et al., 2001; Eakin et al., 2014; Klein et al., 2013; Lynch et al., 2014; Minet et al., 2010), the sample size (Eakin et al., 2014; Lynch et al., 2014; Minet et al., 2010) and the effects of the intervention on other clinical parameters, such as blood pressure (Cooper et al., 2001; Lynch et al., 2014; Norris et al., 2002b), lipid profile and body weight (Eakin et al., 2014; Lynch et al., 2014; Norris et al., 2001; Norris et al., 2002a). Other studies have noted that the most effective interventions for reducing glycated hemoglobin (HbA1c) were those that focused on the patients (Chrvala et al., 2016; Windrum, García-Goñi, & Coad, 2016) and were based on the empowerment of individuals (providing motivation for self-awareness, assessing problems, establishing personal goals and monitoring the achievement of goals) (Chen, Wang, Lin, Hsu, & Chen, 2015), as well as those that included typical elements of the subjects' culture (culturally adapted education) (Bhurji, Javer, Gasevic, & Khan, 2016). Culturally tailored educational interventions have also resulted in a significant reduction in body mass index (BMI) and waist circumference (WC) and a slight improvement in blood pressure and lipid profile (Bhurji et al., 2016). Furthermore, it should be noted that chronic disease management occurs in a context that includes health professionals, social network members and the physical environment. Understanding the social context has important implications for the planning of interventions that aim to improve people's health and well-being (Gallant, 2003). In particular, social support (SS) is considered a facilitator of diabetes mellitus (DM) self-care (Cardoso, Queirós, & Ribeiro, 2015) because it is an interpersonal process that is centered on the exchange of information (Finfgeld-Connett, 2005) in which one has the perception or belief of being connected to and feeling loved and esteemed by others (Martins, 2005). Systematic reviews have shown that families are one of the main sources of SS for adults with DM (Rintala, Jaatinen, Paavilainen, & Astedt-Kurki, 2013; Strom & Egede, 2012) and that families actively participate in the health care of adults and elders (Gallant, 2003). Care is most often provided by a family member when patients do have blood relatives, not only because of their existing relationship but also because this responsibility is culturally assigned (Cattani & GirardonPerlini, 2004). SS can be considered a personal dimension of family relationships, i.e., to occur as a result of these relationships (Pedro, Rocha, & Nascimento, 2008), regardless of family structure. Although the literature shows that educational interventions developed together with family carers of people with DM may contribute to the management of the disease, few studies have examined the effects of family SS on metabolic control in adults with DM (Rintala et al.,

69

2013). In addition, the gaps and inconsistencies between studies, which have been conducted in different population samples, require further research to better understand the true influence of SS on disease-related outcomes (Strom & Egede, 2012), preferably using randomized controlled trials. Based on the background presented, this study proposed to investigate whether the inclusion of family SS in an education program for DM would help improve clinical/metabolic parameters of outpatients with DM in a tertiary health care unit. The hypotheses of the study were as follows: Null hypothesis (H0): including family SS in the educational process does not result in better clinical/metabolic control among people with T2DM (there are no differences between groups). Alternate hypothesis (AH): including family SS in the educational process results in better clinical/metabolic control among people with T2DM (the intervention group shows better clinical/metabolic control than the control group). 2. Method This single-blind randomized controlled clinical trial focused on educational interventions for people with T2DM and a family member, who was referred to as a source of SS in caring for the disease. The participants were outpatients in a tertiary health care unit from June 2011 to May 2013 (Fig. 1). 2.1. Study sample The sample size was calculated based on the expected results for the primary outcome variable (HbA1c) in accordance with the literature, using the following formula (Pocock, 1983):

n¼

p1 ð1‐p1Þp2 ð1‐p2 Þ ðp1 −p2 Þ2

 f ðα; βÞ;

where: p1 = 0.22; p2 = 0.10; p1- p2 = 0.12; α = 0.05; β = 0.10; 1-β = 0.90. Therefore, the sample size was estimated to be 190 people or 95 per group. Potential participants were identified weekly by reviewing the medical records of subjects scheduled for that date and assessing the inclusion/exclusion criteria. 2.1.1. Inclusion criteria The inclusion criteria included a T2DM diagnosis, a minimum age of 40 years and a lack of complications in an advanced stage. The selection of 40 years as the minimum age was justified by the fact that T2DM is commonly diagnosed in this age group (American Diabetes Association, 2016). 2.1.2. Exclusion criteria The exclusion criteria included the presence of complications in an advanced stage, such as hemodialysis treatment, amaurosis, cerebrovascular accident sequelae, heart failure sequelae, previous amputations and/or active ulcers in the lower limbs; use of a wheelchair or stretcher; inability to maintain dialogue; and other serious diseases whose treatment could prevent participation in an educational intervention (e.g., chemotherapy, radiotherapy). 2.2. Randomization Randomization was performed every 40 recruited subjects through simple randomization and occurred after the first data collection time point and before the start of the educational intervention. To minimize bias, randomization was performed by statistical software R.

70

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

Fig. 1. Study design.

2.3. Data collection Data collection occurred through individual interviews with subjects with T2DM and a family member telephone interview conducted on two occasions, T0 (before the start of the intervention) and T12 (three months after the end of the intervention); in addition to these time points, patients also received laboratory tests at T6 (six months after the start of the intervention). An Informed Consent Form (ICF) was signed personally by the subject, and an ICF was also sent to the family member and returned by the patient after being signed.

2.4. Interventions The Diabetes Mellitus Education Program for patients was developed in groups of up to 10 people, in four sessions with an average duration of 2 h each, through interactive tools, illustrated with pictures and brief texts, based on the Social Cognitive Theory (SCT) (Bandura, 1977; Reaney, Eichorst, & Gorman, 2012). These educational tools, called Diabetes Conversation Maps, were specially designed for people with T2DM by Healthy company and translated to Brazil by Lilly company. These maps include four main themes that guided both the development of the educational program to the patients, as well as to the relatives, which comprise respectively: “How the Body and Diabetes Work” (Map 1), “Healthy Eating and Physical Activity” (Map 2), "Medication Treatment and Blood Glucose Monitoring" (Map 3), and “Reaching the Goals with Insulin” (Map 4). In the intervention group (IG), a family caregiver who was considered a source of SS for the patient was included in the Education

Program in Diabetes Mellitus. The interventions for family caregivers were implemented through telephone contact and included the topics covered in the patient's educational program following an established protocol in accordance with Motivational Interviewing, SCT and communication theory principles. The main objectives of the family interventions were to stimulate dialogue between the patients and families, increase caregivers' interest in DM-related matters and DM care, and facilitate caregivers' assistance in achieving the goals established by the patients after each topic discussed in the educational groups. In the SCT, the probability of performing a particular behavior is a function of expectations and values, which are influenced by the combination of psychological (beliefs), environmental (physical resources) and social (relationships and social interaction) factors (Reaney et al., 2012). Life experiences result in new belief systems and consequently new behaviors. Thus, sharing the experiences with the peers and using the experiences of others as a model strengthens the belief of self-efficacy (SE), a key concept of the SCT, as the SE influences the performance of the behavior, and the observation of the behavior itself influences the sense of SE (Bandura, 1977). Social interaction contributes to psychosocial adjustment and improves SE perception (Strom & Egede, 2012). SS may promote self-care by SE, at first, and then behaviors influenced by SE can also be influenced by SS (Rosland et al., 2008). Thoits (1982, 1995) brings in the definition of SS that a person's basic social needs are met through interaction with significant people such as family, friends and colleagues. This same author refers to the multidimensionality of the SS concept, in which not only the amount of support is important, but also the types (emotional, instrumental and information) and sources of support (spouse, friends, relatives,

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

colleagues). In extending this concept, Martins (2005) highlights the functions performed by significant groups or individuals (family, friends or neighbors) for the benefit of the individual in the most varied circumstances of life, and also describes it as a complex and dynamic process involving Individuals and their social networks to meet their needs, provide and complement the resources they have and, in this way, face new situations. The family is, first of all, the individual's SS, is their primary resource for promoting health because it is a source of beliefs and behaviors related to it and for exercising a protective role in the face of everyday adversities (Ribeiro, 2007). In a previous study, developed in another sample of the same population of the present study, it was identified that the main sources of SS for people with T2DM were family and health professionals (Gomes-Villas Boas, Santos, Freitas, & Pace, 2009). In order to standardize telephone calls in a quantitative (connection time) and qualitative (content) mode, a protocol was developed for its implementation, which was submitted to the appreciation of professionals working in the area of attention in diabetes mellitus and in speech therapy. This protocol provides orientation from the initial greeting to the family to the closure of each call, and was directed to the topics addressed in the educational program with the participating patients. Except for the first map, which addresses the topic “How the Body and Diabetes Work,” the others explore contents related to self-care with DM and include the essential behavioral changes to disease control. Four telephone calls were made to the family member for each map, and in the calls that preceded each map, the patient's comments to the family-caregiver were rescued about their respective contents addressed at the educational meeting, as well as the possible doubts and the feelings of the person with T2DM about the subject, in order to encourage the dialogue and support of the family member to favor self-care (emotional social support). The two subsequent telephone calls were aimed at exploring and clarifying the doubts arising from the patient-family dialogue, reinforcing the concepts and behaviors intended and/or the guidelines provided in group educational sessions (social information support). On the fourth telephone call, the goals set by the patient were resumed, and then asked the family member about their compliance. When achieved, the family member was asked to congratulate the person on the success and, if not, the family member was motivated to persuade the person to fulfill them (emotional and instrumental social support). At the end of each call, were asked about the possible doubts of the subjects previously discussed, as well as informing the subject to be discussed with the patient, at the next meeting, and detailing the next telephone calls that would be made, related to the respective theme. Thus, patients and family members were similarly motivated to interact in matters of care with the disease. This aspect could be observed in the lines of both, when they said, for example, “I will pay attention because I will be questioned by my relative.” In total, 17 telephone calls were made to each family caregiver of the patients in the IG, each lasting an average of 8 min and occurring at a frequency of 10 to 30 days, based on the dates of the patient's return visits. To ensure that the guidance provided was documented, a digital phone recorder was used; at the end of each call, the families confirmed their interest in continuing the study. The researcher who made the telephone calls did not participate in the group educational interventions and was therefore unaware of which patients were receiving the study intervention. The telephone calls protocol was submitted to a Judges Committee for the evaluation of the appearance and the content, that is, the form, language, clarity and coherence of the information. At the same time, the researcher responsible for the links received training from the speech therapist to develop communicative skills on the telephone (attitude, quality of voice, ability to listen and interview strategies). To minimize possible biases arising from the interaction between participants in the group sessions, two telephone calls were made to

71

the families of participants in the control group (CG): one at the beginning and the other at the end of the study. The first telephone contact to both groups facilitated the blinding of participants at baseline, as this procedure helped ensure that the participants did not know to which group they had been assigned. 2.5. Study variables and data collection procedures Sociodemographic variables, time of diagnosis and lifestyle habits were collected using a structured instrument developed for this purpose. Blood pressure was measured with a digital apparatus with patients seated with their legs uncrossed and the cuff at heart level. Weight was measured using a Filizola digital scale after coats and shoes had been removed, and height was determined with a stadiometer in centimeters with patients barefoot and looking at the horizon. Body mass index (BMI) was calculated as weight/height2; waist circumference (WC) was measured on patients' bare abdomen at the line between the iliac crest and the costal margin. All measurements were taken twice, before and after the interview, and the mean of the two measurements was used. Laboratory data were extracted from the electronic records system of the local health unit based on a specific request for all study participants. Additionally, these data were collected from patient records, and information about changes in medications used for DM control during the study was collected. 2.6. Statistical evaluations Comparisons between groups and times were performed by orthogonal contrasts using a linear mixed effects model with repeated measures. This type of model is used to analyze data in which the responses of the same individual are grouped and the assumption of independence between observations within the same group cannot be assured (Schall, 1991). To use this model, it is necessary for the random effects to have a normal distribution with a zero mean and constant variance. The model was adjusted using the PROC MIXED procedure in SAS ® 9.2 software (SAS Institute Inc., SAS Campus Drive, Cary, North Carolina 27,513, USA). Laboratory data were analyzed using intentionto-treat analysis with the same model. Differences were considered significant when the significance level (p) was b 0.05. 2.7. Ethical considerations This study was part of a project entitled “Impact of a care program for people with Diabetes Mellitus focused on educational interventions and family and social support,” which was registered in Clinical Trials under the number NCT01387633; this project was approved by the Research Ethics Committee of the Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto (HCFMRP-USP), case No. 9510/2010, in accordance with resolution 466/2012 of the National Health Council. 3. Results At baseline, 222 participants were included in the sample; however, 58 (26.12%) discontinued participation (Fig. 2). According to the established criteria, study subjects who had ulcers, lower limb amputation, or cerebrovascular accidents and those who had started hemodialysis were excluded. 3.1. Sociodemographic characteristics, type of treatment, lifestyle habits and clinical data of the study participants The study sample consisted of adults and the elderly, with a mean age of 60.43 (SD = 8.38) years, and the participants had 4.86 (SD = 3.86) years of education on average. The following groups were more prevalent: women (56.7%), people living with a partner (71.3%),

72

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

Fig. 2. Composition of the study sample.

people who were retired or pensioners (52.5%), and people residing in a municipality close to where the study was implemented (59.1%). The average monthly family income was 1746.89 (SD = 1164.72) real. The Brazilian minimum wage in 2011 and 2012 was 545 reais (US $324.40) and 622 reais (US $318.97), respectively, with 1 real equaling $1.68 (2011) and $1.95 (2012). Currently, the dollar is equivalent to an average price of 3.39 reais, representing an average monthly family income of US $515.31. A combination of insulin and oral antidiabetic agents was the DM treatment modality most frequently used (68.3%); when considering non-medication treatment, 91 (55.5%) participants reported engaging in physical activity (regular or sporadic), and 118 (71.9%) reported following a diet. Smoking and alcohol consumption were reported by 11% and 13.4% of the participants, respectively. The two groups did not differ in sociodemographic data, type of treatment or lifestyle habits, showing similar 95% confidence intervals. The subjects had been diagnosed with DM for an average of 15.7 years (SD = 7.82), and both groups had average values above the level recommended by the American Diabetes Association (2016) for BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), WC, fasting plasma glucose (FPG) and HbA1c. Although the IG had slightly superior values than the CG for BMI, SBP, DBP, WC, FPG and HbA1c at T0, the groups did not significantly differ for these clinical variables based on the 95% confidence intervals (Table 1). 3.2. Characterization of family caregivers The family caregivers who participated in the educational intervention had an average age of 47.01 (SD = 13.52) years; 82.9% were female, 54.9% had a low educational level, 73.2% lived with a partner, and 57.3% Table 1 Descriptive analysis of the clinical and laboratory test variables before the intervention (T0). Ribeirão Preto, São Paulo, 2016. CG (n = 114)

IG (n = 108)

Variables

Average (SD)

Average (SD)

BMI SBP DBP WC FPG HbA1c Total cholesterol HDL cholesterol LDL cholesterol Triglycerides Urea Creatinine

31.78 (5.52) 143.22 (4.35) 75.541 (2.64) 106.731 (3.92) 157.43 (75.56) 9.40 (2.00) 176.37 (41.35) 35.81 (8.49) 103.07 (33.44) 200.45 (166.40) 47.23 (24.24) 1.10 (0.48)

32.4 (45.98) 144.18 (22.95) 77.02 (13.71) 109.7 (14.06) 161.00 (63.82) 9.47 (2.01) 173.20 (42.70) 35.46 (8.76) 202.95 (133.03) 174.20 (102.32) 50.68 (25.03) 1.31 (1.11)

were active in the labor market. The family caregivers were most commonly children (46.7%) of the study participants, followed by spouses (43.9%). Ten (12.2%) family caregivers reported having a diagnosis of DM, and 61 (74.4%) were overweight or obese. 3.3. Comparison of clinical variables between groups The clinical variables BMI, SBP, DBP and WC did not differ between groups one year after the intervention (T12) (Table 2). By comparing the results between time points within the study groups, SBP and DBP differed in relation to T0 in both the CG and the IG; however, the IG showed a greater reduction in these variables, indicating that the model of educational interventions developed in this study may have contributed to the improvement in these variables in the IG (Table 3). The clinical variables based on the results of the laboratory tests also showed no differences between groups at both T6 and T12 (Table 4). However, clinical improvement was evident in the comparison analysis between study times within groups; the IG showed a greater reduction in HbA1c at both T6 and T12, whereas the CG only showed a reduction at T12 that was smaller than that of the IG (Table 5). The results also showed that HDL at T12 increased in the CG and the IG but to a greater extent in the CG; triglycerides also increased in both groups, but to a greater extent in the IG. Urea did not change in the IG and increased in the CG, whereas creatinine did not change in the CG but increased in the IG (Table 5). The changes in drug treatment received during the educational interventions were analyzed in both groups, and the results suggested that these changes had no influence on the results of this study (data not shown). The results of the intention-to-treat analysis did not differ from those presented. 4. Discussion

Estimatea CI95%

−0.669 −0.983 −1.484 −2.608 −3.64 −0.07 2.66 0.04 4.89 1.93 −2.93 −0.15

LLCI

ULCI

−2.194 −7.127 −4.930 −6.216 −22.79 −0.58 −8.50 −2.51 −4.49 −36.85 −9.67 −0.38

0.856 5.161 1.963 1.000 15.51 0.44 13.82 2.59 14.27 40.71 3.81 0.08

CG: control group; IG: intervention group; SD: standard deviation; CI: confidence interval; LLCI: lower limit of the confidence interval; ULCI: upper limit of the confidence interval; SBP: systolic blood pressure; DBP: diastolic blood pressure; WC: waist circumference; FPG: fasting plasma glucose; HbA1c: glycated hemoglobin; HDL: high density lipoprotein; LDL: low density lipoprotein. a Difference between the averages of the study groups.

This sample consisted of adults and the elderly with an average age of 60 years and a higher proportion of participants who were female, had a low educational and socioeconomic level, had a long disease duration, were overweight or obese, and were receiving complex drug therapy. This profile is similar to that of a study conducted in the same location (Gomes-Villas Boas, Foss, Freitas, & Pace, 2012), as well as those of two national studies conducted in outpatient clinics (Stacciarini & Pace, 2014; Veras et al., 2014). Patients'sociodemographic characteristics are relevant to the process of teaching and learning, as they may influence the understanding of the disease and its treatment and thus the attainment of metabolic control. The literature suggests that a low education level is a barrier to self-care, as it can impede access to information and reduce learning opportunities (Grillo & Gorini, 2007; Pace, Ochoa-Vigo, Caliri, & Fernandes, 2006). Therefore, health education interventions for people with DM who have a low education level should consider this barrier and other individual characteristics, and the educational processes

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

73

Table 2 Comparison between groups for the clinical variables BMI, SBP, DBP and WC after the intervention (T12). Ribeirão Preto, São Paulo, 2016. Variables

Average (SD)

BMI SBP DBP WC

Estimated difference

CG (n = 82)

IG (n = 82)

31.32 (5,48) 136.16 (23,53) 71.65 (12,86) 106.02 (12,38)

33.05 (5,93) 137.55 (20,87) 71.06 (12,51) 110.22 (11,37)

implemented should be gradual, continuous and interactive (Cazarini, Zanetti, Ribeiro, Pace, & Foss, 2002). The family members who received the educational intervention by telephone also presented a profile similar to family caregivers of people with diseases/chronic health conditions in other studies (Gutierrez & Minayo, 2010). The intervention in this study used group education and was guided by the theoretical principles of the SCT, as these features were believed to enable a collective construction of knowledge about the disease and treatment based on the understanding and experiences of each person and professional; this approach was selected because adults and the elderly require self-directed learning, in which own life experiences are fundamental (Amendola, Oliveira, & Alvarenga, 2011). The literature shows that educational programs should be appropriate for the age, educational level and experience of the person with DM regarding his/her disease (diagnosis time), and group education may be the most advantageous and cost-effective treatment modality available (Sociedade Brasileira de Diabetes, 2016). Similarly, the interventions for family caregivers were developed using motivational interviewing techniques, which are based on cognitive principles and consist of skills related to asking, listening and informing (Rollnick, Miller, & Butler, 2009). In this approach, the health professional establishes an empathic and collaborative relationship with the listener (family member) (Melo, Oliveira, Araújo, & Pedroso, 2008) to allow his/her previous knowledge and experience to serve as a foundation for guiding the discussions. Significant differences were not observed between groups regarding the clinical variables. However, when compared over time, blood pressure was reduced in both groups, with a greater difference in DBP in the IG, although without statistical significance. Similar to this study, a randomized controlled trial conducted with 56 American adults with diabetes that compared face-to-face education (control group) with telemedicine education (intervention group) did not find significant differences between groups in BMI or blood pressure (Izquierdo et al., 2003). Another randomized controlled trial that compared individual education (control group) with group education (intervention group) in a sample of 104 Brazilian adults with T2DM also did not find differences in these variables according to educational mode (Torres, Franco, Stradioto, Hortale, & Schall, 2009). Two systematic reviews (Norris et al., 2001; Norris et al., 2002a) that aimed to evaluate the effectiveness of interventions for DM self-management showed inconclusive results regarding the effectiveness of teaching and collaborative interventions in improving the blood pressure of people with DM. In contrast, a recent systematic review found that culturally adapted interventions focusing on patients achieved a

Table 3 Estimates of the differences between T0 and T12 for each study group in BMI, SBP, DBP and WC. Ribeirao Preto, 2016. Variables

CG

IG

Difference

CI95%

Difference

CI95%

BMI SBP DBP WC

−0.232 6.002 3.677 −0.790

(−0.549–0.085) (1.498–10.505) (1.102–6.252) (−1.951–0.371)

−0.174 6.445 5.914 −0.964

(−0.495–0.147) (1.898–0.991) (3.314–8.514) (−2.140–0.211)

−0.611 −0.54 0.753 −2.782

P-value

CI95% LLCI

ULCI

−2.154 −7.383 −3.096 −6.482

0.932 6.303 4.603 0.918

0.435 0.876 0.70 0.14

significant reduction in BMI and WC, and a slight improvement in blood pressure and lipid profile was also observed (Bhurji et al., 2016). Therefore, the findings of this study can be considered positive, as hypertension and DM are often associated conditions. The United Kingdom Prospective Diabetes Study (UKPDS 38) was a randomized clinical study conducted in 20 hospital clinics in England, Scotland and Northern Ireland that followed 1148 people with hypertension and T2DM, 758 of whom were allocated to the IG (strict BP control) and 390 to the CG (less strict BP control), for a period of approximately nine years. That study showed that strict BP control led to a clinically important reduction in the DM mortality risk, DM complications, progression of diabetic retinopathy and deterioration in visual acuity (United Kingdom Prospective Diabetes Study Group, 1998). The current guidelines of the Sociedade Brasileira de Diabetes recommend that in people with diabetes who are b 60 years old, SBP should be reduced to b 140 mm Hg and diastolic pressure to values b80 mm Hg, with no advantages of lower targets, even in those with kidney disease (grade of recommendation: B); for those older than or equal to 60 years, blood pressure should be maintained at levels no N150/90 mm Hg (grade of recommendation: A) (Sociedade Brasileira de Diabetes, 2016). Statistically significant differences between groups were not observed in the laboratory data; however, when comparing the different study time points in each group, the IG presented better results than the CG for FPG in the first six months of follow-up (not statistically significant), HbA1c (after six and 12 months of follow-up), total cholesterol and triglycerides (both at six months of follow-up), and the reduction in the average values of these last three variables was significant. Considering the progressive nature of T2DM (Sociedade Brasileira de Diabetes, 2016), these findings are clinically relevant, especially for HbA1c, as even though the HbA1c values were above the control targets (N 7.0%) in T12, the average value found can be considered positive for delaying chronic complications. These findings also support the results of a previously cited systematic review (Norris et al., 2001), which showed that positive effects of educational interventions on glycemic control could be found in studies with follow-up times less than six months. For studies with a longer follow-up, positive effects on glycemic control were observed when the interventions incorporated regular reinforcements for participants. Interventions that used a collaborative approach (Norris et al., 2001) or that were culturally adapted (Bhurji et al., 2016) were more effective in improving glycemic control, body weight and lipid profile than educational interventions. Another systematic review (Chrvala et al., 2016) showed that the largest reduction in HbA1c occurred in studies with a follow-up time between six and 12 months and in those longer than 12 months that had a professional-patient contact time N10 h. A previously discussed randomized clinical trial (Torres et al., 2009) found similar results to this study regarding the reduction in HbA1c after six months of follow-up only, reinforcing the notion that clinical and metabolic improvement are post-intermediate outcomes of education for people with T2DM (Sociedade Brasileira de Diabetes, 2016). A meta-analysis of the effectiveness of DM education in attaining glycemic control among adults with T2DM showed an average reduction in HbA1c of 0.76% when measured immediately after follow-up (Norris et al., 2002b), similar to the findings of this research. Another metaanalysis found an average reduction of 0.36% in HbA1c (Minet et al., 2010).

74

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

Table 4 Comparison of laboratory data between the groups after six (T6) and twelve months (T12) of the educational intervention. Ribeirão Preto, São Paulo, 2016. Average (SD)

CI 95%

Variables

Time

CG

IG

Difference

LLCI

ULCI

FPG

T6 T12 T6 T12 T6 T12 T6 T12 T6 T12 T6 T12 T6 T12 T6 T12

159.56 (71.65) 161.32 (78.72) 9.34 (1.87) 8.94 (1.68) 167.36 (42.99) 165.4 (43.44) 34.9 (59.68) 36.28 (8.91) 96.58 (32.12) 94.27 (33.61) 200.45 (166.4) 182.78 (134.49) 44.58 (21.07) 46.99 (24.79) 1.1 (0.48) 1.16 (0.55)

154.71 (68.24) 161.53 (78.17) 9.09 (1.97) 8.73 (1.72) 164.72 (38.59) 166.71 (32.85) 34.59 (10.1) 34.85 (9.34) 94.66 (30.69) 95.83 (27.09) 174.2 (102.32) 210.94 (144.03) 48.47 (26.44) 55.23 (31.47) 1.28 (0.68) 1.49 (1.02)

4.15 −1.97 0.30 0.22 6.22 0.9 −0.01 0.52 4.52 3.76 25.95 −18.8 −4.16 −5.42 −0.23 −0.29

−16.3 −24.5 −0.23 −0.34 −5.26 −11.2 −2.66 −2.16 −5.46 −6.59 −14.1 −62.2 −11.1 −12.8 −0.46 −0.52

24.62 20.53 0.83 0.78 17.69 13.05 2.63 3.21 14.5 14.1 66.01 24.56 2.78 1.92 0.00 −0.05

HbA1c Total cholesterol HDL LDL Triglycerides Urea Creatinine

More recently, a third meta-analysis that analyzed the effectiveness of educational interventions in DM in achieving and maintaining healthy blood glucose levels found a greater average reduction in HbA1c in the intervention group (− 1.09%) than in the control group (−0.52%) (p b 0.01) (Klein et al., 2013). In the present study, although not statistically significant, the reduction in HbA1c was greater in the IG than in the CG (0.74% and 0.46%, respectively), suggesting that family SS significantly improved HbA1c; this improvement may have resulted from positive behavioral changes that were not influenced by the methods used in this study. A systematic review identified an average reduction in HbA1c of 0.74% for the IG and 0.17% for the CG, and the difference between the groups (0.57%) was considered clinically relevant (Chrvala et al., 2016). The reduction was significant for individuals in both groups who had poor glycemic control (N 9.0%) at baseline (Chrvala et al., 2016), and these findings were similar to those of the present study. A cross-sectional study conducted in Iran aimed to analyze the relationship between perceived family SS and glycemic control in 430 Iranian adults with DM (113 with T1DM and 317 with T2DM), using selfreport instruments to collect data. The mean perceived family SS score was higher for people with T1DM than with T2DM, but the difference was not significant. Perceived SS showed a significant correlation with HbA1c for people with T2DM only (p b 0.001) and was considered an Table 5 Estimates of the differences in laboratory test variables when comparing T0 with T6 and T12 for each study group. Ribeirão Preto, 2016. Variablesa

Time CG

IG

Estimate CI95%

Estimate CI95%

T6 T12 T6 T12 T6 T12 T6 T12 T6 T12

−3.03 −4.27 −0.02 0.41 6.02 6.17 0.53 −1.16 6.08 5.83

−18.38–12.33 −20.70–12.16 −0.29–0.26 0.11–0.71 −0.56–12.60 −0.85–13.19 −0.74–1.81 −2.47 – −0.15 −0.03–12.20 −0.42–12.08

4.77 −2.61 0.35 0.70 9.58 4.41 0.48 −0.68 5.72 4.70

Triglycerides

T6 T12

4.62 15.64

28.64 −5.10

Urea

T6 T12 T6 T12

2.20 −0.35 0.01 −0.07

−20.55–29.80 −11.78 – −43.06 −1.44–5.85 −4.28 – −3.58 −0.06–0.08 −0.15–0.00

−10.55–20.08 −19.54–14.32 0.08–0.63 0.39–1.01 3.02–16.14 −2.90–11.72 −0.79–1.76 −2.03 – −0.68 −0.30–11.73 −1.84 – −11.24 3.52–53.75 −33.35–23.15

0.97 −2.84 −0.07 −0.20

−2.55–4.50 −6.84–1.15 −0.14–0.00 −0.28 to −0.12

FPG HbA1c Total cholesterol HDL LDL

Creatinine a

Difference between the averages of study groups.

important psychosocial factor for glycemic control in T2DM (Tol et al., 2011). A controlled clinical trial that assessed the effectiveness of a familyfocused educational intervention on improving glycemic control has also been conducted among 243 Chilean adults with T2DM. That study used an adapted version of the Chronic Care Model as the theoretical framework, and the intervention involved family members of people with T2DM in educational activities through meetings with the healthcare team (both individual and group meetings) and home visits for 12 months. The CG maintained usual care. The IG showed a significant reduction of 0.9% in HbA1c. Compared with the CG, the reduction in HbA1c in the IG was significant only in the last six months of the study, when the activities with family members were more common (Garcia-Huidobro, Bittner, Brahm, & Puschel, 2011). Furthermore, an experimental study was conducted among American adults with DM to determine whether an educational intervention, based on the Theory of Interdependence and Family Systems Theory, resulted in better glycemic control when implemented with couples (people with diabetes and their spouses). Participants in that study were randomly assigned to three groups: couples intervention (“couples”: group of couples), individual intervention (group with only one member of each couple) and enhanced usual care, which consisted of two individual DM education sessions and a food plan review. Compared to the couples and enhanced usual care groups, the individual intervention group showed a greater reduction in HbA1c, total cholesterol and LDL cholesterol, as well as greater adherence to diet and monitoring of blood glucose (Trief et al., 2011). Another randomized controlled clinical trial conducted among 56 adults with T2DM (IG = 28 and CG = 28) at a community hospital in Taiwan compared the effects of an educational intervention with the patient's family with the effects of usual care. The intervention included individual and group educational sessions with “pairs” of patients and families, as well as a monthly telephone call to clarify any questions during the six-month study period. There were no statistically significant differences between groups after the intervention in terms of reduced HbA1c and lipid profile (Kang et al., 2010). These last two studies were similar to this study regarding the absence of statistically significant differences between groups. However, the reduction in HbA1c was clinically significant and shows the positive effect family SS can have on disease control. The absence of a significant difference in HbA1c between groups may have a similar explanation to the one provided for the meta-analysis cited above (Minet et al., 2010), whose objective was to evaluate the effectiveness of educational interventions in glycemic control. In that study, the authors found that the greatest reduction in HbA1c occurred in studies using educational interventions, compared to studies of

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

behavioral and psychosocial techniques, which may be more sensitive to intensity (duration) than didactic interventions (Minet et al., 2010). An additional explanation for that finding is that the results of an intervention may be underestimated when it is offered to the CG as “something beyond usual care,” which is consistent with the results reported in two other meta-analyses of DM education (Ellis et al., 2004; Klein et al., 2013). Although considered a limitation of this study, implementing group educational sessions in both groups provided the framework for the family caregiver interventions. The aforementioned meta-analysis (Ellis et al., 2004) and another systematic review (Warsi, Wang, Valley, Avorn, & Solomon, 2004) showed that the only predictor of reduced HbA1c was face-to-face education, which is in line with the findings of this study, as both groups showed a decrease in hemoglobin. On the other hand, the lack of differences between groups could be attributed to fluctuations in perceived SS, as this perception can vary over time, is influenced by personal characteristics, and is specific to a given condition (Finfgeld-Connett, 2005; Kadirvelu, Sadasivan, & Ng, 2012). Although most studies have focused on the positive aspects of SS, the negative aspects should also be considered (Croezen et al., 2012). When confronted with disease, family members are also emotionally, cognitively and behaviorally affected (Rintala et al., 2013). Accordingly, there may be family conflicts, particularly when there is an overload of familial tasks, when the work is not recognized by the person cared for and/or by other family members, or when the necessary resources for treatment are scarce or unavailable (Bicalho, Lacerda, & Catafesta, 2008; Diogo, Ceolim, & Cintra, 2005). Family disputes, in turn, can be a source of stress and present a barrier to self-care (Rintala et al., 2013). Another aspect to consider is the mode of intervention used with the family members because although the literature shows that phone contacts provide benefits for health outcomes, the studies using this strategy have directly contacted people affected by a disease or health condition. Therefore, its effectiveness when applied with family caregivers is uncertain. Moreover, the few studies involving family caregivers that used the same theoretical framework and that evaluated the same outcomes as this study limited further discussions of these results. 5. Conclusion From a clinical point of view, the results of this study suggest that the type of educational intervention received by the IG may have contributed to the reduced DBP and HbA1c because of the positive results, although not significant, regarding the greater reduction in these two variables in the IG compared to the CG. Another aspect worth noting is the possible positive effects that the professional SS provided during the educational interventions had on the studied parameters among people with T2DM; however, these effects could not be detected or compared between groups at study completion based on the parameters assessed. Furthermore, although the family was requested to motivate people with T2DM to engage in self-care, thus strengthening the guidelines received, this may not have occurred routinely throughout the study. Additionally, some people with T2DM may have perceived the “strengthening” of the family as a form of “control” or as a “burden”. The literature shows that it is not possible yet to infer a causal relationship between SS and DM related outcomes. Therefore, estimating the required “dose” of SS to improve clinical-metabolic control remains a challenge, since assessment of SS by self-report does not reveal the extent to which perceived support reflects actual behaviors of support. In addition, it is believed that psychological factors, such as health beliefs, may exert an interaction effect and influence both disease care and it control, as well as individual personality traits could influence the perception of SS. It is also relevant to investigate other structural or family dynamic factors that may be present and influence, directly or indirectly, disease management; these factors include family conflicts and

75

differences in beliefs regarding health/disease between family members. However, instruments that assess these latter variables and are validated for the Brazilian culture were not identified in the literature. Through the results of the present study, the importance of the insertion of the family member in the care of the person with DM is reiterated and, above all, include he/she in health care programs, especially those that may favor different modalities of social support, in order to strengthen the Bonding among family members, and to promote selfcare activities for the control of the disease and prevention/delay of its serious complications, due to bad control. Funding This work was supported by the São Paulo Research Foundation (FAPESP), process no. 2011/09037-6 and the National Council for Scientific and Technological Development (CNPq) 563598/2010-7, Public Notice MCT/CNPq/CT-SAÚDE/MS/SCTIE/DECIT no. 42/2010 – Diabetes. Declaration of interest none. References Amendola, F., Oliveira, M. A., & Alvarenga, M. R. (2011). Influence of social support on the quality of life of family caregivers while caring for people with dependence. Revista da Escola de Enfermagem da USP, 45, 884–889. American Diabetes Association (2016). Standards of medical care in diabetes-2016. Diabetes Care, 39, S1–S112. Bagnasco, A., Di Giacomo, P., Mora, D. R. D. R., Catania, G., Turci, C., Rocco, G., & Sasso, L. (2014). Factors influencing self-management in patients with type 2 diabetes: A quantitative systematic review protocol. Journal of Advanced Nursing, 70, 187–200. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Bhurji, N., Javer, J., Gasevic, D., & Khan, N. A. (2016). Improving management of type 2 diabetes in south Asian patients: A systematic review of intervention studies. BMJ Open, 6, 1–16. Bicalho, C. S., Lacerda, M. R., & Catafesta, F. (2008). Reflection on whom the family caregiver is. Cogitare Enfermagem, 13, 118–123. Cardoso, A. F., Queirós, P., & Ribeiro, C. F. (2015). Therapeutic self-care management interventions for individuals with diabetes mellitus: Systematic review. Revista Portuguesa de Saúde Pública, 33, 246–255. Cattani, R., & Girardon-Perlini, N. (2004). Taking care of the sick elder people at home in the speech of familial care taker. Revista Eletrônica de Enfermagem, 6, 254–271. Cazarini, R. P., Zanetti, M. L., Ribeiro, K. P., Pace, A. E., & Foss, M. C. (2002). Adherence of people with diabetes mellitus to an educational group: Percentage and causes. Medicina, 35, 142–150. Chen, M. -F., Wang, R. -H., Lin, K. -C., Hsu, H. -Y., & Chen, S. -W. (2015). Efficacy of an empowerment program for Taiwanese patients with type 2 diabetes: A randomized controlled trial. Applied Nursing Research, 28, 366–373. Chrvala, C. A., Sherr, D., & Lipman, R. D. (2016). Diabetes self-management education for adults with type 2 diabetes mellitus: A systematic review of the effect on glycemic control. Patient Education and Counseling, 99, 926–943. Cooper, H., Booth, K., Fear, S., & Gill, G. (2001). Chronic disease patient education: Lessons from meta-analyses. Patient Education and Counseling, 44, 107–117. Croezen, S., Picavet, H. S. J., Haveman-Nies, A., Verschuren, W. M. M., de Groot, L. C., & van't Veer, P. (2012). Do positive or negative experiences of social support relate to current and future health? Results from the Doetinchem cohort study. BMC Public Health, 12, 1–8. Diogo, M. J. D. E., Ceolim, M. F., & Cintra, F. A. (2005). Teaching program for elderly women who care for elderly relatives in their home: Report of experience. Revista da Escola de Enfermagem da USP, 39, 97–102. Eakin, E. G., Winkler, E. A., Dunstan, D. W., Healy, G. N., Owen, N., Marshall, A. M., ... Reeves, M. M. (2014). Living well with diabetes: 24-month outcomes from a randomized trial of telephone-delivered weight loss and physical activity intervention to improve glycemic control. Diabetes Care, 37, 2177–2185. Ellis, S. E., Speroff, T., Dittus, R. S., Brown, A., Pichert, J. W., & Elasy, T. A. (2004). Diabetes patient education: A meta-analysis and meta-regression. Patient Education and Counseling, 52, 97–105. Finfgeld-Connett, D. (2005). Clarification of social support. Journal of Nursing Scholarship, 37, 4–9. Francisco, P. M. S. B., Belon, A. P., Barros, M. B. D. A., Carandina, L., Alves, M. C. G. P., Goldbaum, M., & Cesar, C. L. G. (2010). Self-reported diabetes in the elderly: Prevalence, associated factors, and control practices. Cadernos de Saúde Pública, 26, 175–184. Gallant, M. P. (2003). The influence of social support on chronic illness self-management: A review and directions for research. Health Education & Behavior, 30, 170–195. Garcia-Huidobro, D., Bittner, M., Brahm, P., & Puschel, K. (2011). Family intervention to control type 2 diabetes: A controlled clinical trial. Family Practice, 28, 4–11.

76

L.C. Gomes et al. / Applied Nursing Research 36 (2017) 68–76

Gomes-Villas Boas, L. C., Foss, M. C., Freitas, M. C. F., & Pace, A. E. (2012). Relationship among social support, treatment adherence and metabolic control of diabetes mellitus patients. Revista Latino-Americana de Enfermagem, 20, 52–58. Gomes-Villas Boas, L. C., Santos, C. B., Freitas, M. C. F., & Pace, A. E. (2009). The relationship between social support and the social demographic characteristics of people with diabetes mellitus. Revista Gaúcha de Enfermagem, 30, 390–396. Grillo, M. D. F. F., & Gorini, M. I. P. C. (2007). Characterization of people with diabetes mellitus type 2. Revista Brasileira de Enfermagem, 60, 49–54. Gutierrez, D. M. D., & Minayo, M. C. D. S. (2010). Knowledge production on family health care. Ciência & Saúde Coletiva, 15, 1497–1508. Haas, L., Maryniuk, M., Beck, J., Cox, C. E., Duker, P., Edwards, L., ... Youssef, G. (2012). National standards for diabetes self-management education and support. Diabetes Care, 35, 2393–2401. Hoving, C., Visser, A., Mullen, P. D., & van den Borne, B. (2010). A history of patient education by health professionals in Europe and North America: From authority to shared decision making education. Patient Education and Counseling, 78, 275–281. Izquierdo, R. E., Knudson, P. E., Meyer, S., Kearns, J., Ploutz-Snyder, R., & Weinstock, R. S. (2003). A comparison of diabetes education administered through telemedicine versus in person. Diabetes Care, 26, 1002–1007. Kadirvelu, A., Sadasivan, S., & Ng, S. H. (2012). Social support in type II diabetes care: A case of too little, too late. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy, 5, 407–417. Kang, C. -M., Chang, S. -C., Chen, P. -L., Liu, P. -F., Liu, W. -C., Chang, C. -C., & Chang, W. -Y. (2010). Comparison of family partnership intervention care vs. conventional care in adult patients with poorly controlled type 2 diabetes in a community hospital: A randomized controlled trial. International Journal of Nursing Studies, 47, 1363–1373. Klein, H. A., Jackson, S. M., Street, K., Whitacre, J. C., & Klein, G. (2013). Diabetes self-management education: Miles to go. Nursing Research and Practice, 2013, 1–15. Lessa, I. (2004). Chronic non-communicable diseases in Brazil: A challenge for the complex task of surveillance. Ciência & Saúde Coletiva, 9, 931–943. Lynch, E. B., Liebman, R., Ventrelle, J., Avery, E. F., & Richardson, D. (2014). A self-management intervention for African Americans with comorbid diabetes and hypertension: A pilot randomized controlled trial. Preventing Chronic Disease, 11, E90. Martins, R. M. (2005). A relevância do apoio social na velhice. Millenium - Journal of Education, Technologies, and Health, 31, 128–134. Melo, W. V., Oliveira, M. S., Araújo, R. B., & Pedroso, R. S. (2008). Motivational interview in smokers: A theoretical review. Revista de Psiquiatria do Rio Grande do Sul, 30, 1–12. Minet, L., Møller, S., Vach, W., Wagner, L., & Henriksen, J. E. (2010). Mediating the effect of self-care management intervention in type 2 diabetes: A meta-analysis of 47 randomised controlled trials. Patient Education and Counseling, 80, 29–41. Norris, S. L., Engelgau, M. M., & Narayan, V. K. M. (2001). Effectiveness of self-management training in type 2 diabetes: A systematic review of randomized controlled trials. Diabetes Care, 24, 561–587. Norris, S. L., Lau, J., Smith, S. J., Schmid, C. H., & Engelgau, M. M. (2002a). Self-management education for adults with type 2 diabetes: A meta-analysis of the effect on glycemic control. Diabetes Care, 25, 1159–1171. Norris, S. L., Nichols, P., Caspersen, C., Glasgow, R., Engelgau, M., Jackjr, L., ... Garfield, S. (2002b). Increasing diabetes self-management education in community settings. American Journal of Preventive Medicine, 22, 39–66. Pace, A. E., Ochoa-Vigo, K., Caliri, M. H. L., & Fernandes, A. P. M. (2006). Knowledge on diabetes mellitus in the self care process. Revista Latino-Americana de Enfermagem, 14, 728–734. Pedro, I. C. D. S., Rocha, S. M. M., & Nascimento, L. C. (2008). Social support and social network in family nursing: Reviewing concepts. Revista Latino-Americana de Enfermagem, 16, 324–327.

Pocock, S. J. (1983). Clinical trials: A practical approach. Chichester: John Wiley & Sons. Reaney, M., Eichorst, B., & Gorman, P. (2012). From acorns to oak trees: The development and theoretical underpinnings of diabetes conversation map education tools. Diabetes Spectrum: A Publication of the American Diabetes Association, 25, 111–116. Ribeiro, C. (2007). Família, saúde e doença: o que diz a investigação. Revista Portuguesa de Clínica Geral, 23, 299–306. Rintala, T. M., Jaatinen, P., Paavilainen, E., & Astedt-Kurki, P. (2013). Interrelation between adult persons with diabetes and their family: A systematic review of the literature. Journal of Family Nursing, 19, 3–28. Rollnick, S., Miller, W. R., & Butler, C. C. (2009). Entrevista motivacional no cuidado da saúde: Ajudando pacientes a mudar o comportamento. Porto Alegre: Artmed Editora. Rosland, A., Kieffer, E., Israel, B., Cofield, M., Palmisano, G., Sinco, B., ... Heisler, M. (2008). When is social support important? The association of family support and professional support with specific diabetes self-management behaviors. Journal of General Internal Medicine, 23, 1992–1999. Schall, R. (1991). Estimation in generalized linear models with random effects. Biometrika, 78, 719–727. Sociedade Brasileira de Diabetes (2016). Diretrizes da Sociedade Brasileira de Diabetes 2015–2016. São Paulo: AC Farmacêutica. Stacciarini, T. S. G., & Pace, A. E. (2014). Translation, adaptation and validation of a selfcare scale for type 2 diabetes patients using insulin. Acta Paulista de Enfermagem, 27, 221–229. Strom, J. L., & Egede, L. E. (2012). The impact of social support on outcomes in adult patients with type 2 diabetes: A systematic review. Current Diabetes Reports, 12, 769–781. Thoits, P. A. (1982). Conceptual, methodological, and theoretical problems in studying social support as a buffer against life stress. Journal of Health and Social Behaviour, 23, 145–159. Thoits, P. A. (1995). Stress, coping, and social support processes: Where are we? What next? Journal of Health and Social Behaviour, 53–79 (Extra issue). Tol, A., Baghbanian, A., Rahimi, A., Shojaeizadeh, D., Mohebbi, B., & Majlessi, F. (2011). The relationship between perceived social support from family and diabetes control among patients with diabetes type 1 and type 2. Journal of Diabetes and Metabolic Disorders, 10, 21. Torres, H. C., Franco, L. J., Stradioto, M. A., Hortale, V. A., & Schall, V. T. (2009). Evaluation of a diabetes education program. Revista de Saúde Pública, 43, 291–298. Trief, P., Sandberg, J. G., Ploutz-Snyder, R., Brittain, R., Cibula, D., Scales, K., & Weinstock, R. S. (2011). Promoting couples collaboration in type 2 diabetes: The diabetes support project pilot data. Families, Systems & Health, 29, 253–261. United Kingdom Prospective Diabetes Study Group (1998). Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. British Medical Journal, 317, 703–713. Veras, V. S., Teixeira, C. R. S., Santos, M. A., Torquato, M. T. C. G., Rodrigues, F. F. L., & Zanetti, M. L. (2014). Glycemic profile of persons with diabetes mellitus in a home blood glucose self-monitoring program. Texto & Contexto Enfermagem, 23, 609–616. Warsi, A., Wang, P. S., Valley, L. M. P., Avorn, J., & Solomon, D. H. (2004). Self-management education programs in chronic disease: A systematic review and methodological critique of the literature. Archives of Internal Medicine, 164, 1641–1649. Wild, S., Roglic, G., Green, A., Sicree, R., & King, H. (2004). Global prevalence of diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care, 27, 1047–1053. Windrum, P., García-Goñi, M., & Coad, H. (2016). The impact of patient-centered versus didactic education programs in chronic patients by severity: The case of type 2 diabetes mellitus. Value in Health, 19, 353–362.