Factors Influencing Behavior of Taking Medicine in Elderly Patients Undergoing Hemodialysis

doi:10.4236/ojneph.2012.23007

Paper Menu >>

Journal Menu >>

Open Journal of Nephrology, 2012, 2, 38-43

http://dx.doi.org/10.4236/ojneph.2012.23007 Published Online September 2012 (http://www.SciRP.org/journal/ojneph)

Factors Influencing Behavior of Taking Medicine in

Elderly Patients Undergoing Hemodialysis

Kumiko Ozawa1, Setsuko Niki1, Seiko Yamanouchi1, Hideaki Yamabe2, Toshiko Tomisawa2,

Maiko Kitajima2, Mayumi Urushizaka2, Kumiko Kawasaki2, Chieko Itaki2, Yuka Noto2, Morito Endo3

1School of Nursing, Hachinohe Junior College, Hachinohe, Japan

2Hirosaki University Graduate School of Health Sciences, Hirosaki, Japan

3Faculty of Human Health Science, Hachinohe University, Hachinohe, Japan

Email: [email protected]

Received July 19, 2012; revised August 27, 2012; accepted September 10, 2012

ABSTRACT

Purpose: The elderly are often faced with multiple diseases, in particular hemodialysis (HD) which requires many

kinds of medication. This study examined the factors influencing the behavior of taking medicine in elderly patients

undergoing HD. Subjects and Methods: The subjects comprised 70 outpatients >65 years undergoing HD (48 male

and 22 fe- male). The mean age of the patients was 72.5 ± 4.4 years. The mean duration of dialysis history was 6.1 ± 5.2

years. We performed a questionnaire survey using Medication Assessment Tool, The Kidney Disease Quality of Life

(QOL) —Short Form and Acceptance for Dialysis Instrument. Result: Two factors related to poor behavior of taking

medicine: dialysis history <2 years and poor acceptance of dialysis. The multiple regression analysis showed patient

satisfaction (



= −0.329, p < 0.01) and symptoms/problems (



= −0.273, p < 0.05) as significant independent variables

relating to behavior of taking medicine. Conclusion: Poor behavior of taking medicine is related to a short dialysis pe-

riod, poor acceptance and poor satisfaction of dialysis therapy in elderly patients.

Keywords: Behavior of Taking Medicine; Hemodialysis; Adherence; Elderly Patient

1. Introduction

The number of patients undergoing hemodialysis (HD)

for chronic kidney disease is increasing every year. The

mortality rate of HD patients is also increasing. In Japan,

the total number of dialysis patients was 297,126 and the

number of patients newly inducted to HD was 37,532 in

2010. For the causative diseases, the first position of the

new dialysis patient was diabetic nephropathy. In par-

ticularly, elderly patients newly inducted to HD are in-

creasing (mean age: 67.8 ± 13.3); the ratio of the patients

older than 65 years was 63.5% and older than 75 years

was 34.9% [1].

The elderly are often faced with multiple diseases,

such as diabetes mellitus, hypertension, heart disease, and

cerebrovascular disease. The processes of excretion of

waste material, adjustment of water and electrolyte bal-

ance, hormonal resolution and excretion and activation of

vitamin D work poorly in HD patients, requiring elderly

HD patients to take many kinds of medicine. Graveley et

al. have reported that the incidences of side effects and

interactions of medicines occur 2 - 3 times more often in

elderly people than in young people [2].

Other studies [3-6] have reported that adherence to

medicine is influenced by several factors: lifestyle, effect

of treatment, side effects, anxiety about the disorder,

confidence in treatment, simple explanation of the med-

icine, presence of social or family support, and relation-

ship between nurses and patients. Moreover, poor adher-

ence to medicine is frequent in elderly patients [7].

Nurses also have a role in ensuring patients carry out

self-care adequately. Therefore, we thought that the be-

havior of taking medicine should be properly evaluated.

There are few studies on the behavior of taking medi-

cine in elderly HD patients. The purpose of this study

was to elucidate the factors influencing the behavior of

taking medicine in elderly patients undergoing HD.

His template, created in MS Word 2003 and saved as

“Word 97-2003 & 6.0/95-RTF” for the PC, provides

authors with most of the formatting specifications needed

for preparing electronic versions of their papers. All

standard paper components have been specified for three

reasons: 1) ease of use when formatting individual pa-

pers; 2) automatic compliance to electronic requirements

that facilitate the concurrent or later production of elec-

tronic products; and 3) conformity of style throughout a

journal paper. Margins, column widths, line spacing, and

type styles are builtin; examples of the type styles are

provided throughout this document and are identified in

K. OZAWA ET AL. 39

italic type, within parentheses, following the example.

Some components, such as multi-leveled equations, grap-

hics, and tables are not prescribed, although the various

table text styles are provided. The formatter will need to

create these components, incorporating the applicable

criteria that follow.

2. Subjects and Methods

2.1. Subjects

Subjects were 70 patients >65 years undergoing HD.

Inclusion criteria were more than three months since the

induction of HD, no major change in treatment content

within the past three months and diagnosis of psychosis

or dementia.

We collected individual clinical records and carried

out a questionnaire survey. The questionnaire survey was

carried out either during dialysis or after dialysis. When

it was difficult for the patient to complete the survey

in-house, we had them complete it at home, and collected

it by collection box within a week. When it was difficult

for the patients to write by themselves, the survey was

taken verbally. The clinical record includes causative

disease, dialysis history, weight gain rate between dialyses,

clinical data, presence of diabetes, treatment content and

type of oral medication. Clinical data was the most re-

cently obtained data before the dialysis.

2.2. Behaviors of Taking Medicine

Medication Assessment Tool (MAT) [8] was used to

study the behavior of taking medicine by the patients.

MAT was developed to evaluate whether an outpatient

understands how to take oral medicine and it consists of

11 items (Table 1). As for the evaluation, each item was

evaluated on a 4-point scale from 1 to 4 with a total score

between 11 and 44. A high score indicates problems in

acceptance of taking the medicine.

2.3. Measuring QOL

The Kidney Disease Quality of Life-Short Form instru-

ment (KDQOL-SFTM; version 1.3) was used for this study.

Table 1. The items of the MAT.

1 Is it easy to understand how to take medicine?

2 Are you worried about side effect of the medicine taking now?

3 Do you understand why medicine is necessary by oneself?

4 Is the explanation of the medicine helpful to understand medicine?

5 Do you feel an effect of the medicine?

6 Is the medicine helpful to prevent sick deterioration?

7 Do you think that it is not good to depend on medicine?

8 Is the expense of the medicine a burden?

9 Do you think that there may be less medicine than now?

10 Is the work to take the medicine troublesome?

11 Does the taking medicine go well generally?

One item marks from 1 to 4 points.

It was developed to evaluate the quality of life of patients

with chronic kidney disease; its reliability and validity

have been confirmed [9]. It consists of the kidney disease-

specific instrument of 43 items and the Short-Form

Health Survey of 36 items (SF-36), which is a generic

instrument. It takes around 15 - 20 min to answer, which

is a burden for elderly people. Therefore, we changed the

SF-36 to the Short-Form Health Survey of 8 items

(SF-8TM), which can measure the same health concept

with a shortened questionnaire, reducing the burden on

elderly patients [10]. The score of all variables is high

when QOL is good. The content of KDQOL-SFTM and

SF-8TM is shown in Table 2.

Twenty eight items of KDQOL-SFTM consist of four

variables: symptoms/problems (KD1), effect of kidney

disease (KD2), burden of kidney disease (KD3) and sleep

(KD6); and six items consist of two variables: cognitive

function (KD4) and quality of social interaction (KD5).

There are also two items of social support (KD7) in

KDQOL-SFTM that evaluate the relationship between

patients and their family or friends, and three items,

which consist of two variables, dialysis staff encoura-

gement (KD8) and patient satisfaction (KD9), to evaluate

the relationship between the patient with medical staff.

Six items of SF-8TM consist of general health percep-

tions (SF8GH), physical functioning (SF8PF), role func-

tioning physical (SF8RP), bodily pain (SF8BP), vitality

(SF8VT), and social functioning (SF8SF) to evaluate the

physical state of the patient. The other two items of

SF-8TM consist of mental health (SF8MH) and role func-

tional emotional (SF8RE) to evaluate the psychological

state of the patient.

Table 2 .Variables in SF-8TM and KDQOL-SFTM.

Variable

SF-8TM

General Health Perceptions (SF8GH)

Physical Functioning (SF8PF)

Role Functioning Physical (SF8RP)

Bodily Pain (SF8BP)

Vitality (SF8VT)

Social Functioning (SF8SF)

Mental Health (SF8MH)

Role Functional Emotional (SF8RE)

Physical Component Summary (PCS-8)

Mental Component Summary (MCS-8)

KDQOL-SFTM

Symptoms/Problems (KD1)

Effect of kidney disease (KD2)

Burden of kidney disease (KD3)

Cognitive function (KD4)

Quality of social interaction (KD5)

Sleep (KD6)

Social support (KD7)

Dialysis staff encouragement (KD8)

Patient satisfaction (KD9)

K. OZAWA ET AL.

2.4. Evaluation of Acceptance of Dialysis

Acceptance of the dialysis instrument [11] consists of ten

items and was developed to evaluate the psychological

adjustment level of patients with chronic kidney disease,

from dialysis induction to the maintenance period (Table

3). In the evaluation, each item was evaluated on a

4-point scale 1 to 4, with a total score between 10 and 40.

A high score indicates that the receptive level is low.

2.5. Statistical Analysis

All data were expressed as means ± SD and the com-

parison between the mean of each item was carried out

using t-test, analysis of variance, U-test of Mann-Whitney.

The data were analyzed using SPSS Statistics 18.0 soft-

ware (SPSS, Chicago, IL, USA). We considered p < 0.05

as statistical significant.

2.6. Ethical Consideration

This study was approved by the Committee of Medical

Ethics of Hirosaki University Graduate School of Medicine,

Hirosaki, Japan. The subjects gave us informed consent.

3. Results

3.1. Relationship between Patients’ Characteristics

and Total MAT Score

The α trust coefficient of Cronbach’s of the total MAT

score was 0.777.

We examined the relationship between patients’ chara-

cteristics and MAT. The MAT score of the patients with

a dialysis history of less than 2 years was significantly

higher than those with more 2 years of history (p < 0.05).

The MAT score showed no relation with other character-

istics (Table 4).

3.2. Relationship between Patients’

Characteristics and Each Item on MAT

We compared the mean value of each item on MAT with

the patients’ characteristics. For the question, “Is it easy

Table 3. The items of the acceptance for dialysis instrument.

1Are you afraid that you think that you must undergo dialysis all

the time in future

2Are you anxious whether you can live by dialysis more how many

years?

3Are you anxious whether a physical complication does not occur?

4Are you anxious in future whether you can continue wor

(housework)?

5Are you anxious in future whether you can do life (economic

aspects)?

6Are you angry when you think why oneself must undergo dialysis?

7Do you realize the situation to have to undergo dialysis?

8Do you think that you had better die if you undergo dialysis, o

you do not want to undergo dialysis?

9How long do you receive explanation about the dialysis by the

chief physician?

10How long do you understand dialysis?

One item marks from 1 to 4 points.

Table 4. The comparison between mean value of the total MAT score and the patients’ characteristics.

Characteristic Group N (%) Mean ± SD p Value

Sex Male 48 (68.6) 21.7 ± 4.4 ns

Female 22 (31.4) 21.9 ± 5.3

Age (years) 65 ~ 74 46 (65.7) 21.8 ± 4.2 ns

≧75 24 (34.3) 21.5 ± 5.6

Housemate Lodger 61 (87.1) 21.6 ± 4.9 ns

No 9 (12.9) 22.7 ± 3.0

Education level Elementary school or Junior high school

graduation 37 (52.9) 22.0 ± 4.9 ns

University or High school graduation 33 (47.1) 21.5 ± 4.5

Causative disease Glomerulonephritis chronic 12 (17.1) 22.2 ± 2.0 ns

Diabetic nephropathy 37 (52.9) 21.7 ± 5.2

Nephrosclerosis 4 (5.7) 23.0 ± 5.7

A multiple cystic kidney 3 (4.3) 18.3 ± 7.0

Others 14 (20.0) 21.7 ± 4.4

<2 17 (24.3) 24.0 ± 3.9 *

Dialysis history (years) ≧2 53 (75.7) 21.1 ± 4.8

A water well-controlled group 55 (78.6) 21.4 ± 4.9 ns

Weight gain rate between the dialysis A water poor-controlled group 15 (21.4) 23.0 ± 3.5

Presence 37 (52.9) 21.7 ± 5.2 ns

Presence or absence of diabetes Absence 33 (47.1) 21.7 ± 4.1

An insulin injection 23 (62.2) 22.3 ± 5.4 ns

Diabetic treatment contents An oral antidiabetic drug 4 (10.8) 17.2 ± 7.0

Only as for the diet 10 (27.0) 22.4 ± 3.4

K. OZAWA ET AL. 41

Continued

HbA1c (%) <6 15(40.5) 20.8 ± 4.5 ns

6.0 - 6.9 13(35.1) 22.4 ± 6.6

≧7 9(24.3) 22.4 ± 4.5

<5 22(31.4) 20.6 ± 5.2 ns

The kind of the oral medicine(kinds) ≧6 48(68.6) 22.3 ± 4.5

Ht (%) <30 15(21.4) 22.3 ± 5.2 ns

30 - 32 22(31.4) 20.7 ± 5.8

≧33 33(47.1) 22.2 ± 3.6

＊p < 0.05, ns：not-significant. Data analysis：An U-test of Mann-Whitney (Age), A t-test(Sex, Housemate, Dialysis history, Weight gain rate between the dial-

ysis, Diabetic presence, The kind of the oral medicine), An analysis of variance (Education level, Causative disease, Diabetic treatment contents, HbA1c, Ht).

to understand how to take the medicine?” the water

well-controlled group showed a significantly lower score

than the water poor-controlled group (p < 0.01). For “Are

you worried about side effects of the medicine your tak-

ing now?” the group that had graduated university or

high school (continued education) showed a significantly

lower score than the group that had only graduated ele-

mentary school or junior high school (limited education)

(p < 0.05). The group with a dialysis history of ≥2 years

showed a significantly lower score than the <2 year

group (p < 0.05). For “Do you understand why you must

take medicine by yourself?” the continued education

group showed a significantly lower score than the limited

education group (p < 0.001). Moreover, for “Is the ex-

planation of the medicine helpful to understand the

medicine?” the continued education group showed a sig-

nificantly lower score than the limited education group (p

< 0.05). For “Is the medicine helpful to prevent the

worsening of your illness?” the group with a housemate

showed a significantly lower score than the group with-

out a housemate (p < 0.01). For “Is the expense of the

medicine a burden?” women showed a significantly

lower score than men (p < 0.05). For “Does taking the

medicine generally go well?” the group of the dialysis

history of ≥2 years showed a significantly lower score

than the <2 year group (p < 0.05). There was no signifi-

cant difference between the scores regarding age, causa-

tive disease and the number of types of oral medication.

3.3. Relationship between QOL and MAT Score

Relationship between QOL and MAT score was exam-

ined. The mean value of the total MAT score showed

significant negative correlations with SF8RE (r = −0.24,

p < 0.05), MCS-8 (r = −0.29, p < 0.05) in the SF-8TM,

KD1 (r = −0.38, p < 0.01), KD2 (r = −0.30, p < 0.01),

KD4 (r = −0.32, p < 0.01), KD5 (r = −0.30, p < 0.05),

KD6 (r = −0.31, p < 0.01), KD7 (r = −0.35, p < 0.01),

KD8 (r = −0.33, p < 0.01), and KD9 (r = −0.38, p < 0.01)

on the KDQOL-SFTM (Table 5). The multiple regression

analysis was used for the variables with a coefficient of

correlation of more than 0.3 to determine variables sig-

nificantly associated with MAT. Factors that influence

MAT based on the multiple regression analysis are

shown in Table 6. The mean value of the total MAT

score showed significant negative correlations with KD9

(



= −0.329, p < 0.01) and KD1 (



= −0.273, p < 0.05)

(Figure 1).

3.4. Relationship between MAT Score and

Acceptance of Dialysis Instrument

The mean value of the total MAT score showed a sig-

nificant positive correlation with the acceptance of the

dialysis instrument (r = 0.24, p < 0.05) (Figure 2). This

result means poor behavior of taking medicine relates to

the poor acceptance of the dialysis therapy.

Table 5. Factors of QOL influencing the MAT by the corre-

lation analysis.

Variable (item) r p

Role Functional emotional (SF8RE)a−0.24 *

Mental component summary (MCS-8)−0.29 *

Symptoms/problems (KD1)−0.38 **

Effect of kidney disease (KD2)−0.30 **

Cognitive function (KD4)−0.32 **

Quality of social interaction (KD5)a−0.30 *

Sleep (KD6)−0.31 **

Social support (KD7)−0.35 **

Dialysis staff encouragement (KD8)a−0.33 **

Patient satisfaction (KD9)a−0.38 **

*p < 0.05, **p < 0.01. a：Pearson’s product moment correlation coefficient,

Others：Spearman’s rank correlation coefficient.

Table 6. Factors of QOL influencing the MAT by the mult-

iple regression analysis.

Variable Standard β p value

Patient satisfaction (KD9) −0.329 **

Symptoms/problems (KD1) −0.273 *

Cognitive function (KD4) −0.189 ns

Sleep (KD6) 0.014 ns

Social support (KD7) −0.139 ns

Dialysis staff encouragement (KD8) −0.222 ns

Adjusted R2 = 0.219, *p < 0.05，**p < 0.01, ns: not-significant.

K. OZAWA ET AL.

Figure 1. The mean value of the MAT score showed in part-

icularly strong relation with Patient satisfaction (KD9) by

multiple regression analysis (β = −0.329, p < 0.01). This result

means poor behavior of taking medicine relates to the poor

satisfaction of dialysis therapy. The mean value of the MAT

score showed in particularly strong relation with Symptom/

Problem (KD1) by multiple regression analysis (β = −0.273,

p < 0.05). This result means poor behavior of taking medici-

ne relates to the poor healthy satisfaction of the symptom.

Figure 2. The mean value of the total MAT score showed a

significant positive correlation with acceptance of the dial-

ysis instrument (r = 0.24, p < 0.05). This result means poor

behavior of taking medicine relates to the poor acceptance

of dialysis therapy.

4. Discussion

The mean age of the patients in this study was 72.5 ± 4.4

years. The number of types of oral medication was 7.4 ±

3.1. Uejima et al. reported that patients older than 70

years take about five types of medication, which is twice

that of patients younger than 40 years in Japan [12]. Eld-

erly patients usually have multiple diseases and tend to

take many medications; these rates increase for elderly

HD patients. Moreover, elderly patients are more likely

to forget to take their medicine, with noon being the most

frequent time to forget [12].

We compared the mean value of the total MAT score

relative to patients’ characteristics. Patients with a dialy-

sis history of more than 2 years, showed significantly

better adherence to taking their medicine than those with

less history. Siegal et al. have reported that an elapse of

memory is more frequent in short-term HD patients than

long-term HD patients [13]. Cukor et al. also noted

memory lapse as a factor of worsened adherence of taking

medicine in HD patients [14], as supported by our results.

We compared the score of each item in MAT relative

to patients’ characteristics. In regards to understanding

how to take the medicine, water well-controlled patients

had significantly better understanding than water poor-

controlled patients. Lindberg et al. also reported that ad-

herence of taking medicine is influenced by the avoid-

ance of unnecessary fluid intake [15]. Water restrictions

may therefore be a factor of poorer adherence to taking

medicine.

Patients with less than 2 years of dialysis history also

felt more apprehension of side effects of the medicine

compared with those with more than 2 years of history,

which was consistent with the total MAT score results.

Patients with less than 2 years of dialysis history are

thought to be unstable physically and psychologically,

have poor disease acceptance, and apprehension for side

effects of the medicine, which may worsen adherence. It

is well known that side effects of medicine are associated

with poor adherence [7,15-17].

Patients with a high school or university education had

a better understanding of how to take their medicine than

those with less education. This result suggests that pa-

tients are more likely to take their medicine after having

understood the need of the medicine. We found no sig-

nificant difference for the causative disease and MAT

score, which is consistent with the results of Yuzawa [8].

The total MAT score showed significant negative cor-

relations with SF8RE, MCS-8, KD1, KD2, KD4, KD5,

KD6, KD7, KD8, and KD9. These results mean that if

daily activity of HD patients for cognitive function and

quality of social interaction are good, and the patients

were satisfied with social support and the dialysis care,

adherence is good in HD patients. Therefore, it is thought

that the behavior of taking medicine is strongly associ-

ated with the QOL of HD patients.

In addition, the MAT score showed a particularly

strong relationship with patient satisfaction and symp-

toms/problems by multiple regression analysis. Ferrans et

al. have also reported that patient satisfaction of dialysis

K. OZAWA ET AL.

care is significantly correlated with the mental state,

QOL, renal function and symptoms of the patient [18].

Therefore, it is important to improve patients’ satisfac-

tion, psychological adaptation and QOL.

As we showed a strong connection between behavior

of taking medicine and patients’ satisfaction for the di-

alysis care provided by medical staff, more attention

should be placed on the consultation situation of medical

staff when patients have doubts in taking medicine and

questions about their health, for further understanding of

the patient’s relationship with medical staff.

Moreover, the total MAT score showed a significant

positive correlation with acceptance of the dialysis in-

strument. Thus, more acceptable dialysis treatment may

lead to improved adherence. Much effort is still needed

to help patients accept dialysis therapy.

5. Conclusion

Poor behavior of taking medicine is related to a short

dialysis period and the degree of patient satisfaction for

dialysis therapy. It is important to improve patient satis-

faction for dialysis care, increase healthy satisfaction

regarding symptoms, and allow patients to better accept

their disease.

REFERENCES

[1] Japanese Society for Dialysis Therapy, “An Overview of

Regular Dialysis Treatment in Japan as of December 31,

2010,” Tokyo, 2011.

[2] E. A. Graveley and C. S. Oseasohn, “Multiple Drug

Regimens Medication Compliance among Veterans 65

Years and Older,” Resarch in Nursing & Health, Vol. 14,

No. 1, 1991, pp. 51-58. doi:10.1002/nur.4770140108

[3] J. M. Cargill, “Medication Compliance in Elderly People,

Influencing Variables and Interventions,” Journal of Ad-

vanced Nursing, Vol. 17, No. 4, 1992, pp. 422-426.

doi:10.1111/j.1365-2648.1992.tb01925.x

[4] A. W. Deborah, “Evaluation of Medication Interventions

for the Elderly,” Home Healthcare Nurse, Vol. 16, No. 9,

1998, pp. 612-617.

doi:10.1097/00004045-199809000-00008

[5] T. T. Fulmer, P. H. Feldman, T. S. Kim, B. Carty, M.

Beers, M. Molina and M. Putnam, “An Intervention

Study to Enhance Medication Compliance in Commu-

nity-Dwelling Elderly Individuals,” Journal of Gerontol-

ogy Nursing, Vol. 25, No. 8, 1999, pp. 6-13.

[6] A. A. Ryan, “Medication Compliance and Older People:

A Review of the Literature,” International Journal of

Nursing Study, Vol. 36, No. 2, 1999, pp. 153-162.

doi:10.1016/S0020-7489(99)00003-6

[7] J. Petermans, A. S. Suarez and T. V. Hees, “Therapeutic

Adherence in Elderly,” Revue Medical of Liege, Vol. 65,

No. 5-6, 2010, pp. 261-266.

[8] Y. Yuzawa, “Making and the Usefulness of the Medica-

tion Assessment Tool of the Outpatient,” Ochanomizu

Medical Journal, Vol. 50, No. 3, 2002, pp. 133-143.

[9] Y. Miura, J. Green and S. Fukuhara, “Manual of the

KDQOL Japanese Version 1.3,” Institute for Health Out-

comes & Process Evaluation Research, Kyoto, 2004.

[10] S. Fukuhara and Y. Suzukamo, “Manual of the SF-8

Japanese Version,” Institute for Health Outcomes &

Process Evaluation Research, Kyoto, 2004.

[11] I. Fukunishi, “Development of the Measure of the Psy-

chological Acceptance Level of the Dialysis Treatment,”

OFF TIME Chugai Pharmaceutical, Tokyo, 2002, p. 67.

[12] E. Uegima, S. Mikami and H. Morimoto, “Drug Compli-

ance in the Elderly,” Nippon Ronen Igakkai Zasshi, Vol.

29, No. 1, 1992, pp. 855-863.

doi:10.3143/geriatrics.29.855

[13] B. R. Siegal, R. J. Calsyn and R. M. Cuddihee, “The Re-

lationship of Social Support to Psychological Adjustment

in End-Stage Renal Disease Patient,” Journal of Chronic

Disease, Vol. 40, No. 4, 1987, pp. 337-344.

[14] D. Cukor, D. S. Rosenthal, R. M. Jindal, C. D. Brown and

P. L. Kimmel, “Depression Is an Important Contributor to

Low Medication Adherence in Hemodialyzed Patients

and Transplant Recipients,” Kidney International, Vol. 75,

No. 11, 2009, pp. 1223-1229. doi:10.1038/ki.2009.51

[15] M. Lindberg and P. Lindberg, “Overcoming Obstacles for

Adherence to Phosphate Binding Medication in Dialysis

Patients: A Qualitative Study,” Pharmacy World & Sci-

ence, Vol. 30, No. 5, 2008, pp. 571-576.

doi:10.1007/s11096-008-9212-9

[16] C. M. Hughes, “Medication Non-Adherence in the Eld-

erly,” Drugs & Aging, Vol. 21, No. 12, 2004, pp. 793-

811.

[17] R. Horne and J. Weinman, “Patients’ Beliefs about Pre-

scribed Medicines and Their Role in Adherence to

Treatment in Chronic Physical Illness,” Journal of Psy-

chosomatic Research, Vol. 47, No. 6, 1999, pp. 555-567.

[18] C. E. Ferranns, M. J. Powers and C. R. Kasch, “Satisfac-

tion with Health Care of Hemodialysis Patients” Research

of Nursing & Health, Vol. 10, No. 6, 1987, pp. 367-374.