Nephrol Dial Transplant (2003) 18: 1869–1873
DOI: 10.1093/ndt/gfg307
Original Article
Dietary self-efficacy: determinant of compliance behaviours and
biochemical outcomes in haemodialysis patients
Miklos Zrinyi1, Maria Juhasz2, Jozsef Balla2, Eva Katona2, Thomas Ben2, Gyorgy Kakuk1 and
Denes Pall2
World Health Organization, Geneva, Switzerland and 2Division of Nephrology, 1st Department of Medicine,
Medical and Health Center, University of Debrecen, Debrecen, Hungary
Abstract
Background. Despite the diversity of proposed theories, researchers are still unable to fully explain dietary
compliance behaviours of dialysis patients. Dietary
self-efficacy, a concept less studied in dialysis, has been
linked to positive compliance outcomes in the chronic
illness literature. Therefore, the aim of the present
research was to determine how dietary self-efficacy is
related to selected biochemical markers and selfreported behavioural outcomes in haemodialysis
patients.
Methods. 107 subjects participated in a cross-sectional
study. Four questionnaires assessed dietary selfefficacy, compliance attitudes and behaviours, and
staff–patient relationships. Laboratory outcomes were
retrospectively obtained from patients’ medical records
and averaged for the previous 6 months.
Results. Of the behavioural measures, only dietary
self-efficacy was associated with laboratory outcomes.
Dietary self-efficacy was also positively related to staff–
patient relationships and to patients’ self-reported
assessment of compliance behaviours. Women had
greater dietary self-efficacy than men. The number
of family members living with the respondent was
inversely related to dietary self-efficacy.
Conclusions. Results indicated that dietary self-efficacy
determined both behaviours and laboratory outcomes.
Patients with greater dietary self-efficacy had lower
serum potassium and weight gain, showed favourable
compliance attitudes and behaviours toward prescribed
regimens and fostered better relationships with staff.
Based on these findings we recommend an experimental
approach to clarify whether maximizing dietary selfefficacy efforts is without psychological burden to
patients and whether the positive effect of increased
Correspondence and offprint request to: Dr Miklos Zrinyi, World
Health Organization, 34 Avenue Blanc, 1202 Geneva, Switzerland.
Email: zrinyi_m@freemail.hu
dietary self-efficacy is maintained in long-term dialysis
patients.
Keywords: compliance behaviours; dietary selfefficacy; haemodialysis; health outcomes; patient–
staff relationships
Introduction
Patients’ compliance behaviours concerning therapy
and prescriptions have been extensively studied.
However, Kaveh and Kimmel [1] stated that despite
past efforts, no universal theory has been developed
that satisfactorily describes dialysis compliance behaviours. The need for such a theory is, however, pressing
so that we can develop more successful interventions
to improve therapeutic compliance in this population
[2–5].
Among the several cognitive–behavioural variables
identified, few proved to successfully explain compliance behaviours in dialysis patients [1,3,5,6]. Social
learning theory (self-efficacy), an emerging concept
proposed to explain behaviours, has been shown to
positively affect health outcomes and to improve
compliance in different chronic patient groups including renal patients [6–11].
According to Bandura [7], self-efficacy is mediated by
a person’s beliefs or expectations about his/her capacity
to accomplish certain tasks successfully or demonstrate
certain behaviours. Bandura postulates that these
expectations determine whether or not a certain behaviour or performance will be attempted, the amount of
effort the individual will contribute to the behaviour,
and how long the behaviour will be sustained when
obstacles are encountered.
Research to test whether manipulating self-efficacy
beliefs and expectations modifies dietary outcomes
has shown promise [12–16]. However, unlike fluid
ß 2003 European Renal Association–European Dialysis and Transplant Association
Downloaded from https://academic.oup.com/ndt/article/18/9/1869/1842053 by guest on 20 August 2022
1
1870
management self-efficacy [8,10,17], dietary selfefficacy has not been thoroughly investigated in haemodialysis patients. Therefore, our aim in the present
research was to determine how dietary self-efficacy is
related to selected care characteristics measured as
biochemical indices as well as to self-reported dietary
behaviours in haemodialysis patients.
Subjects and methods
(RAAQ) and the Renal Adherence Behaviour Questionnaire
(RABQ) to report patient behaviours and attitudes [19]. The
RAAQ is a 26-item scale measuring general attitudes toward
compliance. The scale is composed of Likert-type statements,
which measure a patient’s attitudes toward social restrictions,
well being, self-care/support and acceptance. The RABQ
consisted of a 25-item scale measuring self-reported dietary
(diet and fluid) compliance. Specific dimensions on the scale
were: compliance to fluid restrictions; compliance regarding
potassium and phosphate restrictions, compliance regarding
self-care; compliance regarding sodium intake; and compliance in times of particular difficulty.
Higher scores indicated better outcomes on each instrument. All instruments had pre-established validity and have
shown sufficient reliability in earlier use.
For ease of administration, we modified scoring on the SDS
by transforming responses to a 0–10 scale. Due to culturally
unsuitable statements, the number of items was also reduced
to 15 on the SDS. Validity and reliability (expressed as
Cronbach’s alpha) for the SDS was still preserved (0.92). For
the remaining instruments, reliability ranged between 0.84
and 0.92.
Descriptive statistics were used to describe sample characteristics. To obtain measures of association, we calculated
Pearson correlation coefficients and partial correlation coefficients. Independent sample t-tests and analysis of variance
were done to detect group differences. A hierarchical regression approach was employed to assess individual contributions of factors to patient reported compliance behaviours.
Statistical significance was set at 5%. We analysed data with
SPSS Windows version 8.0.
Results
Sample characteristics are presented in Table 1. The
average age of the sample was 57.6 (SD 14.03) years.
Patients have been diagnosed with kidney disease for
10.4 (SD 9.15) years on average and have been on
haemodialysis treatment for an average of 50.4 (SD
25.66) months. Our sample was gender and residence
balanced. The majority of the respondents lived with
a spouse, were unemployed and completed less than
high school education. Hypertension was dominant
and ischaemic heart disease was prevalent in our
sample.
Of the self-reported patient measures, only dietary
self-efficacy was associated with two of the biochemical
indices. These relationships remained unaffected even
when social, behavioural and treatment variables had
been controlled for. The greater dietary self-efficacy
was, the lower serum potassium (r ¼ 0.22; P < 0.001)
and weight gain (r ¼ 0.35; P < 0.001) had been. No
other behavioural measures correlated with any further
biochemical outcomes.
Dietary self-efficacy was also positively related to
staff-patient relationships (r ¼ 0.34; P < 0.05) and to
patients’ self-reported compliance attitudes (r ¼ 0.21;
P < 0.05) and behaviours (r ¼ 0.24; P < 0.05). Better
staff–patient relationships indicated greater dietary
self-efficacy. Greater dietary self-efficacy also resulted
in improved compliance attitudes and behaviours.
Downloaded from https://academic.oup.com/ndt/article/18/9/1869/1842053 by guest on 20 August 2022
The study was implemented with a cross-sectional design.
Subjects were selected from a pool of approximately 1700
patients in 20 dialysis centres. The eligible number of subjects,
applying the inclusion criteria explained below, was 1228.
For ease of administration (the current research considers a
pilot study of a larger follow-up), we aimed to sample 10%
of the eligible population. Therefore, a convenience quota
sampling of six subjects per each centre was employed.
Thirteen patients declined; our final sample consisted of 107
participants. Selection criteria included the following: (i) the
patient attends chronic haemodialysis treatment (haemodialysis and haemodiafiltration) for a minimum of 3 months
prior to the study; (ii) the patient is prescribed a minimum of
3 h treatment modality three times a week; (iii) treatment is
provided by polysulfone dialysers with a maintained blood
flow of at least 250 ml/min; the patient is at least 18 years of
age; the patient suffers no major mental or psychological
disorders and communicates in Hungarian. The institutional
ethical board approved the research. Signed individual
consents from all patients were obtained who participated
in the study.
Main clinical outcomes, such as serum potassium, serum
phosphorus, serum albumin and interdialytic weight gain
(SeK, SePO4, SeAlb and IWG) as well as treatment conditions and medical history were retrospectively assessed from
patients’ medical records. These indicators were selected
because they have been reported as proxy clinical measures of
patients’ dietary behaviours [1].
Self-reported assessments required patients to reflect on
their dietary efforts of the past month. Our goal was to match
these self-assessments with laboratory data obtained from
the previous month. However, the number of laboratory
measurement points of a single month did not provide
enough variance for meaningful analysis of behaviours. To
increase variance, we decided to average the above four
clinical outcomes for the past 6 months preceding the study.
A clinically trained psychologist assessed individual
behaviours and attitudes through patient interviews.
Subjects were asked to respond to four questionnaires and
to questions on a demographic sheet. Completing all
instruments took 45–50 min on average per each patient.
The Situational Dieting Self-Efficacy Scale (SDS) assessed
dietary self-efficacy [16]. This scale comprised 25 statements
about various eating situations combined with personal
moods. Respondents are asked to indicate on a scale of
0–100% their ability to resist a particular eating challenge.
Patient Reactions Assessment (PRA) measured staff–patient
relationships [18]. The instrument included 15 Likert-type
statements (1 ¼ strongly disagree, 5 ¼ strongly agree) on the
following three dimensions of staff–patient interactions:
provision of information, caring and personal communication. We used the Renal Adherence Attitudes Questionnaire
M. Zrinyi et al.
Dietary self-efficacy in haemodialysis patients
1871
Table 1. Sample characteristics
Mean
57.6
%
14.03
50.5/49.5
89.7/10.3
2.6
1.14
71/29
6.5/93.5
52.3/47.7
65.9
2.89
10.4
603.9
50.4
10.25
38.2
5.33
1.86
15.66
1.18
9.15
760.65
25.66
1.62
3.57
0.6
0.4
13.1/86.9
15/85
93.5/6.5
50.4/49.6
29/71
110
78.5
52.2
73.3
31.4
7.53
8.63
6.04
SD, standard deviation; SDS, Situational Dieting Self-Efficacy Scale; PRA, Patient Reactions Assessment; RAAQ, Renal Adherence
Attitudes Questionnaire; RABQ, Renal Adherence Behaviours Questionnaire.
Dietary self-efficacy was positively linked to the age
of the respondent (r ¼ 0.22; P ¼ 0.25); older people
exhibited more efficacy. The relationship between
dietary self-efficacy and duration of care (in months),
however, was not significant. Nor was dietary selfefficacy related to the number of years since the
diagnosis of the kidney disease.
We found no difference in dietary self-efficacy with
regards to education or employment of subjects
(F ¼ 0.96; P ¼ 0.44 and t ¼ 1.69; P ¼ 0.09). However,
significant gender differences were revealed (t ¼ 2.82;
P ¼ 0.006); women had greater dietary self-efficacy
than men.
Residual urinary output, despite expectations, did
not affect dietary self-efficacy. Dietary self-efficacy was
also unaffected by the severity of the patients’ physical
condition.
Of the social characteristics, living with or without a
spouse did not influence dietary self-efficacy (t ¼ 1.28;
P ¼ 0.22). However, the number of family members
living with the respondent was inversely related to
dietary self-efficacy (r ¼ 0.11; P ¼ 0.049); suggesting
that the increase in the number of family members
resulted in lower dietary efficacy beliefs and expectations for patients.
To assess the individual contribution of selected
variables predicting patients’ compliance behaviours, a
hierarchical regression analysis was run. We entered
demographic (age, gender), disease specific (time since
diagnosis of disease and since first treatment, hours
of treatment/week), therapeutic (IWG, SeK, SePO4,
SeAlb, residual urinary output) and behavioural
(dietary self-efficacy, patient–staff interactions and
attitudes toward therapy) factors in consecutive steps.
Results of the last hierarchical step are presented in
Table 2.
The final model accounted for 25% of the variance in
patient reported compliance behaviours (R2 ¼ 0.25;
F ¼ 4.925, P ¼ 0.035). The most significant contribution
to the model was attributed to residual urinary output
(increased explained variance by 12.7%), followed by
dietary self-efficacy, patient attitudes toward therapy and serum albumin. Demographic or any other
therapeutic factors reached no further significance in
the model.
Examining significant b weights informs about the
unit improvement in patient reported behaviours by
each factor. For example, an increase of 100 ml in
residual urinary output decreases the self-reported
compliance behavioural score with 12.2 points. That is,
the more urinary output a patient has, the less
compliant with therapy prescriptions the individual
will be. Similarly, a 10-point improvement in dietary
self-efficacy increased self-reported compliance behaviour scores with 20.25 points. In other words, increased
beliefs and expectations to be able to resist tempting
eating situations result in better patient compliance
with care.
Discussion
The aim of the current research was to explore how
dietary self-efficacy influenced biochemical treatment
Downloaded from https://academic.oup.com/ndt/article/18/9/1869/1842053 by guest on 20 August 2022
Age
Gender (female/male)
Spouse (yes/no)
Family members living together
Education (>high school/<high school)
Employment (yes/no)
Residence (urban/rural)
Dry weight (kg)
Interdialytic weight gain (kg/48 h)
Diagnosis of kidney diseases (years)
Residual urinary volume (ml)
Duration of dialysis (months)
Length of treatment (h/week)
Serum albumin (g/l)
Serum potassium (mmol/l)
Serum phosphorus (mmol/l)
AMI (myocardial infarct) (yes/no)
Stroke (yes/no)
Hypertension (yes/no)
Ischaemic heart disease (yes/no)
Diabetes (yes/no)
SDS (scores)
PRA (scores)
RAAQ (scores)
RABQ (scores)
SD
1872
M. Zrinyi et al.
a
Table 2. Hierarchical regression model, step 4
a
Beta
37.73
0.056
1.268
0.034
0.012
0.433
0.122
0.009
0.046
0.632
0.285
2.025
0.099
1.27
0.137
0.106
0.047
0.107
0.123
0.363
0.197
0.004
0.102
0.019
0.157
0.125
0.108
P
0.012
0.259
0.319
0.681
0.404
0.273
<0.001
0.166
0.974
0.003
0.867
<0.001
0.396
0.02
Note: dependent variable ¼ patient self-reported compliance behaviours (RABQ).
markers and compliance behaviours of haemodialysis patients. Results confirmed several relationships
between dietary self-efficacy and care and patient
characteristics. Patients with increased dietary selfefficacy had lower serum potassium and weight gain,
showed more favourable compliance attitudes and
behaviours toward prescribed regimens and fostered
better relationships with staff.
To our present knowledge, the current study is
among the few that have established a direct relationship between a specific dietary self-efficacy measure
and biochemical outcomes of dialysis care. Whether the
relationships between dietary efficacy and serum
potassium and weight gain may be translated into
effective interventions designed to enhance patient
compliance will be determined by future research.
Dietary self-efficacy capabilities were more linked to
the age of the respondent than to time since diagnosis of
kidney disease or duration of therapy. However,
expectation was that the longer the patient had been
affected by the dietary adjustment, the greater his/her
dietary self-efficacy should have been. Especially,
duration of therapy should have been more associated
with dietary self-efficacy. According to Bandura [7],
longer exposure time to the eating regulation problem
allows patients better master dietary efficacy experiences. Why this was not true in our study requires
further investigation.
As for social variables, the only significant difference
in dietary self-efficacy was found for women. Still, this
difference did not translate into better or improved
health outcomes for women in this research. How
gender differences in dietary self-efficacy may benefit
patients concerning health outcomes remains a question for longitudinal research.
Our results also confirmed the indirect effects of staff
interactions with patients on compliance attitudes and
behaviours as well as on laboratory outcomes. Positive
patient–staff relationships increased self-reported compliance efforts and behaviours of patients and enhanced
dietary self-efficacy capacities. Therefore, it appears
that through better relationships with their clients staff
can influence potassium intake and weight gain in
patients.
Similar results emerged for patients’ employment
status. Employment had no direct effect on dietary selfefficacy; however, those with a job had better relations
with staff, which increased dietary self-efficacy and
lowered serum potassium and extra weight.
The effect of social pressure on patients’ eating habits
was also present in our sample. As the number of family
members living with the patient increased so did the
patient’s expectation of his/her ability to resist tempting eating situations decrease. This relationship was
reflected in therapeutic outcomes as well; patients with
more family members had increased serum phosphorus
levels and more weight gain.
Results from the regression analysis confirmed the
positive effect of dietary self-efficacy on patient
reported compliance behaviours. Given that a patient
scored zero on the self-reported patient compliance
measure (no compliance behaviours at all), we
estimated that an approximately 60-point increase in
dietary self-efficacy could result in full patient
compliance behaviours. [Self-reported patient compliance was measured on a 25-item scale with a rating
between 1 and 5 for each item. The maximum possible
score was 125 (full compliance). As an increase of 1point in dietary self-efficacy results in a 2.025-point
improvement in the final patient compliance behaviour
score, increasing dietary self-efficacy with 62 points
yields a final patient compliance score of 125.55.] These
results support that designing clinical interventions
to modify dietary efficacy beliefs and expectations
can yield significant and prolonged improvements in
patients’ compliance behaviours. Whether such interventions are feasible and meaningful for the patients
should be determined by future research.
Evidence to promote the positive influence of
increased serum albumin on patient survival exists [1].
We also noted a positive relationship between serum
albumin and self-reported patient compliance behaviours in our regression analysis. We suspect that
increased serum albumin may have a beneficial effect
Downloaded from https://academic.oup.com/ndt/article/18/9/1869/1842053 by guest on 20 August 2022
Constant
Age
Gender
Time since diagnosis
Time since first treatment
Hours per week on treatment
Residual urinary output
Interdialytic weight gain
Serum potassium
Serum albumin
Serum phosphorus
Dietary self-efficacy (SDS)
Patient-provider relationship (PRA)
Patient attitudes toward therapy (RAAQ)
b weights
Dietary self-efficacy in haemodialysis patients
Acknowledgements. The authors would like to acknowledge the
support of the medical and nursing staff as well as the participation
of patients in this study.
Conflict of interest statement. None declared.
References
1. Kaveh K, Kimmel PL. Compliance in hemodialysis patients:
multidimensional measures in search of a gold standard. Am J
Kidney Dis 2001; 37: 244–266
2. Golper T. Patient education: can it maximize the success of
therapy? Nephrol Dialysis Transplant 2001; 16 [Suppl 7]: 20–24
3. Kutner G. Improving compliance in dialysis patients: does
anything work? Semin Dial 2001; 14: 324–327
4. Morgan L. A decade review: methods to improve adherence to
the treatment regimen among hemodialysis patients. Nephrol
Nurs J 2000; 27: 299–304
5. Shaw-Stuart NJ, Stuart A. The effect of an educational patient
compliance program on serum phosphate levels in patients
receiving hemodialysis. J Renal Nutr 2000; 10: 80–84
6. Thomas LK, Sargent RG, Michels PC, Richter DL, Valois RF,
Moore CG. Identification of factors associated with compliance
to therapeutic diets in older adults with end stage renal disease.
J Renal Nutr 2001; 11: 80–89
7. Bandura A. Self-Efficacy. The Exercise of Control. Freeman,
New York, NY: 1997: 259–313
8. Brady BA, Tucker CM, Alfino PA, Tarrant DG, Finlayson
GC. An investigation of factors associated with fluid adherence
among hemodialysis patients: a self-efficacy theory based
approach. Ann Behav Med 1998; 19: 339–343
9. Oka M, Chaboyer W. Influence of self-efficacy and other
factors on dietary behaviours in Japanese hemodialysis
patients. Int J Nurs Pract 2001; 7: 431–439
10. Rosenbaum M, Ben-Ari Smira K. Cognitive and personality
factors in the delay of gratification of hemodialysis patients.
J Personality Social Psychol 1986; 51: 357–364
11. Schneider MS, Friend R, Whitaker P, Wadhwa NK. Fluid
noncompliance and symptomatology in end-stage renal disease:
cognitive and emotional variables. Health Psychol 1991; 10:
209–215
12. AbuSabha R, Achterberg C. Review of self-efficacy and locus
of control for nutrition- and health-related behavior. J Am
Dietitian Assoc 1997; 97: 1122–1132
13. Schlundt DG, Pichert JW, Rea MR, Puryear W, Penha ML,
Kline SS. Situational obstacles to adherence for adolescents
with diabetes. Diabetes Education 1994; 20: 207–211
14. Schlundt DG, Rea MR, Hodge M et al. Assessing and
overcoming situational obstacles to dietary adherence in
adolescents with IDDM. J Adolescent Health 1996; 19: 282–289
15. Schlundt DG, Rea MR, Kline SS, Pichert JW. Situational
obstacles to dietary adherence for adults with diabetes. J Am
Dietitian Assoc 1994; 8: 874–879
16. Stotland S, Zuroff DC, Roy M. Situational dieting selfefficacy and short-term regulation of eating. Appetite 1991; 17:
81–90
17. Tanner JL, Craig CB, Bartolucci AA et al. The effect of a selfmonitoring tool on self-efficacy, health beliefs, and adherence
in patients receiving hemodialysis. J Renal Nutr 1998; 8:
203–211
18. Galassi J, Schanberg R, Ware WB. The patient reactions
assessment: a brief measure of the quality of the patientprovider medical relationship. Psycholgical Assess 1992; 4:
346–351
19. Rushe H, McGee HM. Assessing adherence to dietary
recommendations for hemodialysis patients: the Renal
Adherence Attitudes Questionnaire (RAAQ) and the Renal
Adherence Behaviour Questionnaire (RABQ). J Psychosom Res
1998; 45: 149–157
Received for publication: 12.6.02
Accepted in revised form: 18.4.03
Downloaded from https://academic.oup.com/ndt/article/18/9/1869/1842053 by guest on 20 August 2022
on compliance behaviours, which translates into better
survival of patients. However, from the available data
we were not able to support such a direct link.
Based on findings from the current research we
recommend an experimental approach to identify
methods of interventions that increase dietary selfefficacy in haemodialysis patients. We also recommend
further study of the relationships between dietary selfefficacy and patient compliance behaviours and
survival on haemodialysis care.
The authors acknowledge that results of this study
may be a function of the sample selection. Therefore,
generalizability is cautioned and should be limited to
centres utilizing identical care protocols.
This study employed only a selected set of cognitivebehavioural measures and accounted for some of
the technical aspects of care. Introducing additional
measures may have produced alternative outcomes.
Statistical power, due to a smaller number of subjects,
may have reduced the potential to identify significant
relationships in some analyses.
1873