Synopsis/Appraisal Final Paper Overview and Instructions
davidstone742
Relationshipamongsocialsupporttreatmentadherenceandmetaboliccontrolofdiabetesmellituspatients.pdf
Rev. Latino-Am. Enfermagem Original Article 2012 Jan.-Feb.;20(l):52-8 www.eerp.usp.br/rlae
Relationship among social support, treatment adherence and metabolic
control of diabetes mellitus patients^
Lilian Cristiane Gomes-Villas Boaŝ
Milton Cesar Fosŝ
Maria Cristina Foss de Freitas''
Ana Emilia Pace'
This cross-sectional and quantitative study ainned to analyze the relationship among social
support, adherence to non-pharmacological (diet and physical exercise) and pharmacological
treatments (insulin and/or oral anti-diabetic medication) and clinical and metabolic control of
162 type 2 diabetes mellitus patients. Data were collected through instruments validated for
Brazil. Social support was directly correlated with treatment adherence. Adherence to non-
pharmacological treatment was inversely correlated with body mass index, and medication
adherence was inversely correlated with diastolic blood pressure. There were no associations
between social support and clinical and metabolic control variables. Findings indicate that social
support can be useful to achieve treatment adherence. Studies with other designs should be
developed to broaden the analysis of relations between social support and other variables.
Descriptors: Social Support; Patient Connpliance; Diabetes Mellitus; Nursing Care.
' Paper extracted from Master's Dissertation "Apoio social, adesao ao tratamento e controle metabólico de pessoas com diabetes mellitus"
presented to Escola de Enfermagem de Ribeirao Preto, Universidade de Säo Paulo, WHO Collaborating Centre for Nursing Research
Development, SP, Brazil.
^ RN, M.Sc. in Nursing, Adjunct Professor, Centro Universitario, Fundaçâo Educacional Guaxupé, MG, Brazil. E-mail:
^ Physician, Ph.D. in Medicine, Full Professor, Faculdade de Medicina de Ribeirao Preto, Universidade de Sâo Paulo, SP, Brazil. E-mail:
" Physician, Ph.D. in Medicine, Professor, Faculdade de Medicina de Ribeiräo Preto, Universidade de Säo Paulo, SP, Brazil. E-mail:
= RN, Ph.D. in Nursing, Associate Professor, Escola de Enfermagem de Ribeiräo Preto, Universidade de Säo Paulo, WHO Collaborating Centre for
Nursing Research Development, SP, Brazil. E-mail: [email protected].
Corresponding Author:
Ana Emilia Pace Universidade de Säo Paulo. Escola de Enfermagem de Ribeirao Preto Oepartamento de Enfermagem Geral e Especializada Av. dos Bandeirantes, 3900 Bairro: Monte Alegre CEP: 14040-902, Ribeirao Preto, SP, Brasil E-mail: [email protected]
53
Relaçâo entre apoio social, adesâo aos tratamentos e controle metabóllco de pessoas com diabetes mellitus
O presente estudo objetivou analisar a relaçâo entre apoio social, adesâo aos
tratamentos nâo medicamentoso (dieta e exercício físico) e medicamentoso (insulina
e/ou antidiabéticos oráis) e controle clínico-metabólico de 162 pessoas com diabetes
mellitus tipo 2. Constituiu-se em um estudo seccional, de abordagem quantitativa.
Os dados foram coletados por meio de instrumentos validados. O apoio social teve
correlaçao direta com a adesâo aos tratamentos. Observou-se correlaçio inversa entre
adesâo ao tratamento nâo medicamentoso e índice de massa corporal, bem como entre
adesâo medicamentosa e pressao arterial diastólica. Nao houve associaçôes entre apoio
social e variáveis de controle clínico-metabólico. Conclui-se que o apoio social poderá
ser útil para se obter a adesâo aos tratamentos. Estudos com outros delineamentos
devem ser desenvolvidos, a fim de se ampliar a análise das relaçôes entre apoio social
e outras variáveis.
Descritores: Apoio Social; Cooperaçâo do Paciente; Diabetes Mellitus; Cuidados de
Enfermagem.
Relación entre apoyo social, adhesión al tratamiento y control metabólico de personas con diabetes mellitus
El presente estudio objetivó analizar la relación entre apoyo social, adhesión a los
tratamientos no medicamentoso (dieta y ejercicio físico) y medicamentoso(insulina y/o
antidiabéticos orales) y control clínico-metabólico de 162 personas con diabetes mellitus
tipo 2. Se trata de un estudio seccional, de abordaje cuantitativo. Los datos fueron
recolectados por medio de instrumentos validados. El apoyo social tuvo correlación
directa con la adhesión al tratamiento. Se observó correlación inversa entre adhesión
al tratamiento no medicamentoso y índice de masa corporal, así como entre adhesión
medicamentosa y presión arterial diastólica. No hubo asociaciones entre apoyo social y
variables de control clínico-metabólico. Se concluye que el apoyo social podrá ser útil
para obtener la adhesión a los tratamientos. Estudios con otros delineamientos deben
ser desarrollados, a fin de ampliar el análisis de las relaciones entre apoyo social y otras
variables.
Descriptores: Apoyo Social; Cooperación del Paciente; Diabetes Mellitus; Atención de
Enfermería.
Introduction
Diabetes Mellitus (DM) stands out among non-
transmissible chronic diseases because it is considered a
public health problem due to its epidemic proportions. It
constitutes a challenge to the health system and society
because of the high financial and social costs to control
and treat its complications'''.
DM treatment aims to maintain metabolic control
and basically comprises non-medication and medication
therapy, the former of which is related to behavioral
changes associated with a healthy diet and physical
www.eerp.usp.br/rlae
exercise'^'. Clinical-metabolic control includes glucose
control, through glycated hemoglobin and fasting
plasma glucose measures, as well as blood pressure and
plasma lipid (triglycéride, total cholesterol and fractions)
control, as the latter two conditions generally coexist in
DM patients, constituting risk factors for cardiovascular
disease'^'.
In this context, medication and non-medication
treatment adherence represents a fundamental concept
in care delivery to DM patients, and the understanding
54 Rev. Latino-Am. Enfermagem 2012 Jan.-Feb.;20(l):52-8.
of barriers and facilitators for behavioral changes can
support nursing interventions with a view to furthering
or strengthening conditions that promote adherence.
This study highlights Social Support (SS) among the
factors that might influence treatment adherence. SS is
considered a complex and dynamic process that involves
individuals and their social networks, working to satisfy
their needs, provide and complement the resources they
have and, thus, cope with new situations'"'. The main
sources can be family members and health professionals.
Studies suggest that SS is associated with
adherence to medication and diet treatment. Others
studies involving adults with DM revealed that people
with a low SS perception presented significantly poorer
glycémie control when exposed to highly stressful
Although treatment adherence and SS have been
largely studied, little attention has been paid to the
relation between treatment adherence and perceived
social support, or to the need for this support among
people with DM<". Assessing SS is important to help
nurses to plan appropriate interventions that can
enhance people's adaptation to their disease* '̂ and,
consequently, improve treatment adherence.
This study was developed in view of the need to
know the perception of DM patients concerning SS and
its relation to metabolic control and medication and non-
medication treatment adherence.
Aim
To analyze the relation between SS, non-medication
treatment (diet and physical exercise) adherence,
medication treatment adherence and clinical-metabolic
control of type 2 Diabetes Mellitus patients under
outpatient follow-up.
Methods
This sectional and quantitative study was carried
out at a tertiary outpatient clinic in Ribeirâo Preto, SP,
Brazil between May and November 2008. The study
population was selected through a weekly search of the
medical profiles of people scheduled for appointments
at the unit and who met the following inclusion criteria:
minimum age of 40 years, medication treatment
including insulin, oral anti-diabetic medication and/or
associated medicines, absence of chronic complications
in advanced stages, and ability to dialog. People older
than 40 years were chosen because DM2 is most
frequently diagnosed after this age.
Three instruments were used for data collection:
Inventario da Rede de Suporte Sociai (IRSS), the
translated version of The Social Support Network
Inventory"', adapted and validated (a=0.95) for
the Brazilian culture'^', which serves to assess social
network variables (source and type of contact) and
perceived social support; Cuestionario das Atividades
de Autocuidado com a Diabetes (QAAD), the
translated version of the Diabetes Self-care Activities
Questionnaire'-^^'', adapted and validated (a=0.75)
for the Brazilian culture*'"', which assesses adherence
to diet and physical exercise recommendations; and
the Medida de Adesäo aos Tratamentos (MAT), the
translated version of the Morisky Test'"', adapted to
and validated (a=0.74) for the Portuguese language''^'
and readapted to Brazilian Portuguese by Faria'"', which
serves to assess medication treatment adherence.
The use of the instruments was previously
authorized by the authors of the original versions as well
as by the authors of the translated and adapted versions.
Socio-demographic and clinical data, and information
related to treatment and metabolic control (laboratory
data) and life style were collected through a structured
instrument that was tested in previous studies"^'.
The instruments were all read aloud, allowing the
necessary time for each participant to fill them out. Data
collection was carried out after the nature and objectives
of the study had been clarified and participants had
signed free and informed consent forms.
The collected data were stored in an Excel
database, with double entry and validation. Afterwards,
the database was exported to the Statistical Package
for Social Science® (SPSS), version 11.5 for exploratory
univariate and bivariate analyses with central trend
measures (average and median) and variability
(standard error) measures.
Numerical variables were submitted to the
Kolmogorov-Smirnov test for distribution analysis
and to the Levene test to verify the homogeneity of
variances. The relationship between these variables was
investigated through linear regression. The strength
of the relationship was measured through Person's
correlation coefficients and the degree of correlations
was verified following the classification proposed by
Zou, Tuncali and Silverman''^': weak (r < 0.3), moderate
(0.3 < r < 0.6), and strong (r > 0.6).
When categorical variables were evaluated in
relation to a numerical variable, they were submitted
to the previously mentioned tests of distribution and
homogeneity. For the analysis between two independent
www.eerp.usp.br/rlae
Gomes-Villas Boas LC, Foss MC, Freitas MCF, Pace AE. 55
samples. Student's t-test was applied; for more than
two independent samples, data were submitted to
variance analysis (ANOVA) followed by Tukey's test and
the Kruskal-Wallis test, followed by Dunn's test when
necessary, for parametric and non-parametric samples,
respectively. The level of significance adopted in this
study was 0.05.
Approval for this study was obtained from the
Institutional Review Board at the University of Säo Paulo
at Ribeirao Preto Medical School Hospital das Clínicas
on April 28* 2008, under protocol No 2049/2008, and
participants received and signed the Informed Consent
Term.
Results
During the study period, 1,004 people were
attended at the outpatient unit and, after analysis of
their medical records, 309 (30.8%) met the inclusion
criteria. Only 206 individuals could be contacted though.
Of these, nine were excluded due to physical/cognitive
limitations; 22 refused due to the following: concern
with the medical appointment or transportation, did
not see any benefit in participating in the study; 13 did
not attend the meeting on the scheduied day and time.
Therefore, the study sample comprised 162 people,
which is equivalent to 1 6 . 1 % of this study's baseline
population.
The sample's socio-demographic profile is as
follows: 94 (58%) participants were women and 68
(42%) were men. The average age was 59.4 (SE=8)
years, 114 (70.4%) were married, 67 (41.4%) were
pensioners, 129 (79.6%) were from Ribeirao Preto, SP,
Brazil and/or region, average education corresponded
to 5.36 (SE=3.90) years and the monthly average
family income was R$ 1,325.65 (SE=1,122.72), which
is equivalent to approximately 3.2 times the Brazilian
monthly minimum wage at the time of study"^'.
The diagnosis time was, on average, 14.8
(SE=7.72) years; 88 (54.3%) were obese and 58
(35.8%) were overweight; 55.5% practiced regular
or sporadic physical activity; 6.2% and 19.8% of the
participants reported smoking and alcohol consumption,
respectively. The most frequent complications/
comorbidities were dyslipidemia and hypertension;
104 (64.2%) participants were under mixed insulin
therapy and oral anti-diabetic medication; 143 (88.3%)
presented glycated hemoglobin higher than or equal
to 7% (average=9.1%; SE=1.82); average systolic
blood pressure was 140mmHg (SE=22.4) and diastolic
78mmHg (SE=11.84); average waist circumference
for men was 106.4cm (SE=12.87) and for women
106cm (SE=13.87); total cholesterol was 182 mg/dl
(SE=47.7); HDL for men was 38 mg/dl (SE=7.85) and
for women, 45 mg/dl (SE=9.16); triglycérides 215 mg/
dl (SE=237.3).
To check the reliability of the instruments used, in
this study, Cronbach's alpha coefficient was calculated.
The coefficients obtained for the IRSS, QAAD and
MAT were, respectively, 0.94, 0.68 and 0.66. A
psychometric property analysis review of instruments
to assess subjective phenomena found that Cronbach's
alpha coefficients higher than 0.50 are considered
reasonable"^'.
High levels of perceived SS were observed,
with relatives as the main source, foilowed by health
professionals. No statistically significant differences
were found in mean SS with regard to gender, marital
status and occupation. Weak but statistically significant
correlations were observed though, between SS and age
(r=0.20; p=0.01), as well as between SS and education,
the iatter of which was an inverse correlation ( r = - 0 . 2 3 ;
p=0.03). These data suggest that, the higher the age,
the higher the perceived SS. On the opposite, the higher
the education level, the lower the perceived SS''''.
Non-medication treatment (diet and physical
exercise) adherence was low in 6 9 . 1 % of participants
and an inverse and statistically significant correlation
was found with education ( r = - 0 . 1 8 ; p=0.02), suggesting
that, the higher this variable, the lower adherence. No
statistically significant differences were observed in self-
care adherence measures regarding gender, marital
status and occupation.
Medication treatment adherence was high for 95.7%
of participants. No statistically significant correlations
were found, though, between medication adherence and
socio-demographic variables.
To study the relations among the main study
variables, first, the relation between SS, medication and
non-medication (diet and physical exercise) treatment
adherence was analyzed. Direct but weak correlations
were observed between SS and non-medication
treatment adherence ( r = 0 . 2 1 ; p=0.01), as well as
between SS and medication treatment adherence
(r=0.18; p=0.02), suggesting that, the higher the
perceived SS, the greater adherence to diet and physical
exercise recommendations as well as to medication
therapy.
The analysis of the relation among the (medication
and non-medication) treatment adherence variables.
www.eerp.usp.br/rlae
56 Rev Latino-Am. Enfermagem 2012 Jan.-Feb.;20(l):52-8.
with significance set at p<0.05, revealed a direct and
statistically significant but weak correlation among the
scores (r=0.22; p=0.00).
Finally, the relation among the SS variables, non-
medication treatment adherence, medication treatment
adherence and clinical-metabolic control is shown in
Table 1.
Table 1- Relation between social support, non-medication (diet and physical exercise) treatment adherence and
medication adherence with clinical-metabolic control in the study sample. Ribeirâo Preto, SP, Brazil, 2008
Social support Adherence to non-medication
treatment Medication adherence
vanaDie scores
Systolic arterial pressure (mmHg)
Diastolic arterial pressure (mmHg)
Body mass index (kg/m^)
Waist circumference (cm)
Fasting plasma glucose (mg/dl)
Glycated hemoglobin (%)
Total cholesterol (mg/dl)
HDL cholesterol (mg/dl)
LDL cholesterol (mg/dl)
Triglycérides (mg/dl)
Statistical coefficient*
0.14
0.02
-0.15
-0.15
0.02
0.01
0.14
0.01
0.14
0.04
p-value
0.07
0.83
0.06
0.06
0.82
0.85
0.07
0.93
0.07
0.60
Statistical coefficient*
0.13
0.05
-0.23
-0.15
-0.01
-0.02
-0.08
0.00
-0.07
-0.06
p-value
0.09
0.51
0.00'
0.05
0.93
0.78
0.33
1.00
0.41
0.44
Statistical coefficient*
-0.06
-0.15
-0.04
-0.06
0.03
-0.03
-0.01
-0.04
-0.07
-0.03
p-value
0.40
0.04'
0.61
0.46
0.66
0.65
0.91
0.59
0.36
0.73
'The statistical coefficient corresponds to Pearson's or Spearman's correlation coefficient, according to the classification of the study sample. ' Statistical significance (p<0.05)
No Statistically significant associations are observed
between clinical-metabolic control and SS. With p set
at <0.05, an inverse and statistically significant but
weak correlation was observed between non-medication
treatment adherence and BMI, as well as between
medication adherence and diastolic blood pressure
(Table 1).
Discussion
The relation between SS and socio-demographic
characteristics in the study sample was explored in an
earlier study"". In that study, a direct correlation was
highlighted between SS and age, as well as an inverse
correlation with education.
SS was directly correlated with medication and non-
medication treatment adherence, in line with literature
findings''''''^"'. The influence of family members and
significant others may reinforce the health orientations
DM patients receive, which could lead to higher adherence
to diet and physical exercise recommendations as well
as to medication treatment. On the other hand, this
infiuence might conflict with health recommendations
and hinder adherence'"'.
In the study group, a direct correlation was
observed between the two adherence types under
analysis. As opposed to literature, this finding suggests
a close relation among different adherence aspects'"'.
This might be relevant for interventions to improve
adherence, that is, if -one behavior tends to predict
another, the same intervention might be an efficient
means to increase adherence in more than one aspect
of the treatment.
The correlation between SS and clinical-metabolic
control was not statistically significant. Similar results
were observed in two other studies. The first aimed
to analyze the relation between SS, medication and
non-medication treatment adherence and metabolic
control in North American adult DM patients'"', while
the second focused on the effects of SS on the health,
well-being and metabolic control of adult African DM
patients'"'. The later study, however, showed that SS
is an important determinant of DM patients' health
and well-beings. It also benefits at least one aspect of
disease management, which is blood pressure control.
This study's findings reveal a weak inverse
correlation between adherence to non-medication
treatment (diet and physical exercise) and the Body
Mass Index (BMI). This data reinforces the importance
of diet and physical exercise recommendations in BMI
control. A study that investigated the relation between
medication and non-medication treatment adherence
and DM control in Jamaican adults found similar
results'"'.
Regarding medication adherence, a weak inverse
correlation was observed between this variable and
www.eerp.usp.br/rlae
Gomes-Villas Boas LC, Foss MC, Freitas MCF, Pace AE. 57
mean diastolic blood pressure levels. In a study aimed
at determining medication adherence among North
American adult DM patients and its relation with the
number of drugs prescribed and metabolic control, lower
diastolic blood pressure levels were also identified among
participants with higher medication adherence, although
another instrument was used to assess adherence'"'.
For glucose control, assessed through glycated
hemoglobin levels, no statistically significant correlations
with SS were observed, similar to another study cited'"'.
No correlations were observed either between glycémie
control and adherence variables, as opposed to authors
who studied the relation between adherence and
metabolic control among Finish adult DM patients'^'. It is
highlighted that the latter study used another instrument
for SS analysis.
In summary, the analysis of the relations between
SS, medication and non-medication treatment adherence
and clinical/metabolic control revealed that this study's
findings are similar to those of studies carried out with
DM patients, focusing on the relation between SS and
adherence variables''''^^^"', but diverge from the results
of these same studies in terms of the relationship
between SS and glycémie control, as well as between
adherence variables and glycémie control.
On the other hand, the lack of association between
adherence variables and glycémie control was also
observed in another study'"', without ignoring its clinical
importance for care delivery to DM patients though.
Although a high glyeated hemoglobin level
suggests "something is wrong", this result does not
speeifieally indieate what is wrong with the medieal
reeommendations, or with the patient's disease
management, whieh suggests that other factors, like
the soeioeeonomie, eultural and occupation, exert
infiuenee'"'.
Metabolic control is a complex set of interactions, in
whieh adherenee is only one ofthe many related factors,
and that is why its use as an adherenee measure is of
limited value'22'.
Conclusion
Even though no statistically significant correlation
was observed between SS and elinieal/metabolie
control, one ean infer on the clinical importance of SS
when its effect on behavioral ehanges for self-care is
observed. This effect has been shown in the studied
literature, whieh emphasizes the influenee of SS on DM
management and the maintenance of preventive or
health proteetion behaviors.
The literature has presented eontroversial results
eoneerning the relation between adherence and
metabolic control, sometimes in line with, sometimes
differing from the present study. It is believed that
many other variables can influence metabolie eontrol.
Psyehologieal factors like health beliefs and soeioeultural
inequities might exert an interaetion effeet and, direetly
or indireetly, influence DM eontrol.
It should also be taken into aeeount that adherenee
to the therapeutie regimen, as opposed to metabolic
eontrol, is seen as a human behavior which, as such, is
subjeet to direet influenees from the soeial environment
individuals live in.
In conclusion, SS can be a valuable resouree,
capable of helping people to adapt better to the demands
the disease imposes, and is an important tool for health
professionals, espeeially nurses. Knowing the proeess
and strueture of SS permits a better understanding of
the way people's soeial relationships occur, and how
these relationships ean faeilitate or enhanee behaviors
that promote and/or protect health.
It is also highlighted that most studies aim to'
analyze the relations between psyehosoeial variables
and adherenee, or between psyehosoeial variables
and metabolie eontrol. Few studies, however, have
simultaneously investigated the relations between these
three sets of variables, whieh justifies further studies of
this kind.
Some limitations need to be taken into aeeount.
The first refers to the type of study. As it assesses SS
and treatment adherenee in a eertain time period only,
this does not permit eapturing possible infiuenees of vital
events on pereeived support and treatment adherence
behaviors. Knowledge on these infiuenees eould provide
important information for nursing intervention planning,
as it permits understanding people in their daily life
eontext.
Literature suggests that the evaluation of SS
through indirect methods (self-reports) does not reveal
the extent to whieh pereeived support refleets aetual
support behaviors, as individual personality traits ean
also infiuenee pereeptions. The same applies to treatment
adherenee, sinee evaluating these variables through
these same methods depends on the respondents'
memory eapaeities and on their personal willingness to
actually reveal their behaviors concerning therapeutie
reeommendations. In addition, researehers need to be
skilled in eolleeting information in a neither inductive nor
eoereive manner.
Therefore, further researeh in the field should be
eneouraged, with different designs and larger population
www.eerp.usp.br/rlae
58 Rev. Latino-Am. Enfermagem 2012 Jan.-Feb.;20(l):52-8.
samples, in order to broaden the analyses of the relations
between SS and other variables, and also to study the
infiuences of nursing interventions in the recognition,
change or reinforcement of social support to improve
patients' health-disease conditions.
References
1. American Diabetes Association. Economic Costs
of Diabetes in the U.S. in 2007. Diabetes Care.
2008;31(3):596-615.
2. Lerman I. Adherence to treatment: a key for advoiding
long-term complications of diabetes. Arch Med Res.
2005 May-Jun; 36(3):300-6.
3. American Diabetes Association. Standards of Medical
Care in Diabetes. Diabetes Care. 2011 Jan; 34:S11-S61.
Suppl. 1.
4. Martins RML. A relevancia do apoio social na velhice.
Millenium - Rev ISPV. 2005;(31): 128-34.
5. López DMM, López VMG, Ruiz MEG, López JHP, Escobar
AN. Disfunción familiar y control del paciente diabético
tipo 2. Rev Med (Mex). 2004;42(4):281-4.
6. Miller CK, Davis MS. The influential role of social
support in diabetes management. Topic Ciin Nutrit.
2005;20(2): 157-65.
7. Toljamo M, Hentinen M. Adherence to self-care and
social support. J Clin Nurs. 2001;10:618-27.
8. Lima EDRP, Norman EM, Lima AP. Translation and
adaptation of the Social Support Network Inventory in
Brazil. J Nurs Scholarsh. 2005;37(3):258-60.
9. Flaherty JA, Gaviria FM, Pathak DS. The measurement
of social support: The Social Support Network Inventory.
Compr Psychiatry. 1983;24(6):521-9.
10. Torres HC. Avaliaçao de um programa educativo para
controle de diabetes mellitus tipo 2 em Belo Horizonte,
MG [Tese de Doutorado]. Rio de Janeiro: Escola Nacional
de Saude Pública Sergio Arouca, Fundaçâo Oswaido
Cruz; 2004. 126 p.
11. Toobert DJ, Glasgow RE. Assessing diabetes self-
management: the summary of Diabetes Self-care
Activities Questionnaire. In: Bradley C, editor Handbook
of Psychology and Diabetes. Amsterdam: Harwood
Academic Publishers; 1994. p. 351-75.
12. Delgado AB, Lima ML. Contributo para a validaçâo
concorrente de uma medida de adesâo aos tratamentos.
Psicol Saude & Doenças. 2001;2(2):81-100.
13. Morisky DE, Green LW, Levine DM. Concurrent
and predictive validity of a self-reported measure of
medication adherence. Med Care. 1986;24(l):67-74.
14. Faria HTG. Fatores associados à adesâo do paciente
diabético à terapéutica medicamentosa [Dissertaçao de
Mestrado]. Ribeirao Preto: Escola de Enfermagem de
Ribeirâo Preto da Universidade de Sao Paulo; 2008. 146 p.
15. Pace AE, Ochoa-Vigo K, Caliri MHL, Fernandes
APM. O conhecimento sobre diabetes mellitus no
processo de autocuidado. Rev. Latino-Am. Enfermagem.
2006;14(5):728-34.
16. Zou KH, Tuncali K, Silverman SG. Correlation and
Simple Linear Regression. Radiology. 2003;227(3):617-
22.
17. Gomes-Villas Boas LC, Santos CB, Foss-Freitas MC,
Pace AE. A relaçâo entre o apoio social e as características
sociodemográficas das pessoas com diabetes mellitus.
RevGaúch Enferm. 2009;30(3):390-6.
18. Mota DDCF, Pimenta CAM. Avaliaçao e mensuraçâo
de variáveis psicossociais: desafio para pesquisa e clínica
de enfermagem. Rev Gaúch Enferm. 2007;28(3):309-
14.
19. Garay-Sevilla ME, Nava LE, Malacara JM, Huerta R,
Léon JD, Mena A, et al. Adherence to treatment and social
support in patients with non-insulin dependent diabetes
mellitus. J Diabetes Complications. 1995;9(2):81-6.
20. Wang CY, Fenske MM. Self-care of adults with non-
insulin-dependent-diabetes mellitus: infiuence of family
and friends. Diabetes Educ. 1996;22(5):465-70.
21. Davis MS. Variations in patients' compliance with
doctors' advice: an empirical analysis of patterns of
communication. Am J Public Health. 1968;58(2):274-88.
22. Johnson SB. Methodological issues in diabetes
research. Diabetes Care. 1992; 15(ll):1658-67.
23. Wilson W, Ary DV, Biglan A, Glasgow RE, Toobert
DJ, Campbell DR. Psychosocial predictors of self-care
behaviors (compliance) and glycémie control in non-
insulin-dependent diabetes mellitus. Diabetes Care.
1986;9(6):614-22.
24. Westaway MS, Seager J, Rheeder P, Van Zyl D.
The effects of social support on health, well-being and
management of diabetes mellitus: a black south African
perspective. Ethn Health. 2005;10(l):73-89.
25. Duff EM, O'Connor A, McFarlane-Anderson N, Wint
YB, Bailey EY, Wright-Pascoe RA. Self-care, compliance
and glycaemic control in Jamaican adults with diabetes
mellitus. West Indian Med J. 2006;55(4):232-6.
26. Grant RW, Devita NG, Singer DE, Meigs JB.
Polypharmacy and medication adherence in patients with
type 2 diabetes. Diabetes Care. 2003;26(5): 1408-12.
Received: Dez. IS'*" 2010
Accepted: June 27"' 2011
www.eerp.usp.br/rlae
Copyright of Revista Latino-Americana de Enfermagem (RLAE) is the property of Escola de Enfermagem de
Ribeirao Preto, Universidade de Sao Paulo and its content may not be copied or emailed to multiple sites or
posted to a listserv without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.
