Hemodialysis and Peritoneal Dialysis: Patients’ Assessment

of Their Satisfaction with Therapy and the Impact of the

Therapy on Their Lives
Erika Juergensen, Diane Wuerth, Susan H. Finkelstein, Peter H. Juergensen, Ambek Bekui,
and Fredric O. Finkelstein
Hospital of St. Raphael, Yale University and the Renal Research Institute, New Haven, Connecticut

This study was undertaken to examine patient satisfaction with peritoneal dialysis (PD) and hemodialysis (HD) therapies,
focusing attention on the positive and negative impact of the therapies on patients’ lives. Patients were recruited from a
free-standing PD unit and two free-standing HD units. A total of 94% (n ! 62) of eligible PD and 84% (n ! 84) of eligible HD
patients participated. HD patients were significantly older and had higher Charlson Comorbidity Index scores than the PD
patients, but there were no differences in duration of dialysis treatment, prevalence of diabetes, educational backgrounds, or
home situations. Patients were asked to rate their overall satisfaction with and the overall impact of their dialysis therapy on
their lives, using a 1 to 10 Likert scale. In addition, patients were asked to rate the impact of their therapy on 15 domains that
had been cited previously as being important for patients’ quality of life. The mean satisfaction score for PD patients (8.02 "
1.41) was higher than for HD patients (7.4 " 1.4; P ! 0.15). PD patients indicated that there was less impact of the dialysis
treatment on their lives globally (7.25 " 2.12 versus 6.19 " 2.83; P ! 0.019). In addition, PD patients noted less impact of the
therapy in 14 of the 15 domains examined. With the use of a proportional odds model analysis, the only significant predictor
of overall satisfaction and impact of therapy was dialysis modality (P ! 0.037 and P ! 0.021, respectively). Patients also were
asked to comment freely on the positive and negative effects of the dialysis treatments on their lives, and a taxonomy of
patient perceptions and concerns was developed. This study suggests that PD patients in general are more satisfied with their
overall care and believe that their treatment has less impact on their lives than HD patients.
Clin J Am Soc Nephrol 1: 1191–1196, 2006. doi: 10.2215/CJN.01220406

C
omparisons between hemodialysis (HD) and perito-
neal dialysis (PD) therapy for patients with ESRD
generally have focused on differences in morbidity
and mortality between these treatment modalities (1–3). Results
of these studies have been inconsistent, with different studies
indicating varying results. Problems that often have been cited
in interpreting these results relate to issues with patient selec-
tion for the different modalities, the differences in comorbidi-
ties in patients who select HD and PD, and problems with
relative utilization of PD and HD in defined geographic areas
(1– 4). Providing guidelines and advice for patients in terms of
modality selection must rely, to some extent, on factors other
than specific details of relative mortality rates of the different
therapies.
Recently, it has been recognized that patients’ quality of life
and patient satisfaction with care are important domains that
need to be understood better and addressed more fully (5–11).
This in part is because these domains have been associated with
a variety of defined medical outcomes, such as compliance with
Received April 13, 2006. Accepted August 4, 2006.
Published online ahead of print. Publication date available at www.cjasn.org.
Address correspondence to: Dr. Fredric Finkelstein, 136 Sherman Avenue,
New Haven, CT 06511. Phone: 203-787-0117; Fax: 203-777-3559; E-mail:
fof@comcast.net
care, morbidity, and mortality (7–9). Furthermore, satisfaction
with care and quality of life are being recognized increasingly
as valid end points to assess therapy (10,11).
Few studies have examined patient satisfaction with dialysis
therapy (6). Recently, increasing attention has been focused on
this area, as investigators have identified patients’ perception of
care as an important domain to examine (9 –11,12,13). Further-
more, it has been emphasized recently that dialysis patients
have strong preferences concerning their therapy (14). This
study was undertaken to examine patient satisfaction with PD
and HD, focusing attention on the impact of the therapies on
patients’ lives.
Materials and Methods
All patients who were receiving dialysis in three of the four dialysis
units that are affiliated with the Hospital of St. Raphael were consid-
ered to be candidates for participation in the study. To be eligible,
patients had to be 18 yr of age or older, maintained on the same dialysis
modality for a minimum of 6 mo, fluent in English, and medically
stable without acute medical problems for a minimum of 2 mo before
the study. In addition, patients who were unable to understand and
answer the questionnaires coherently were excluded. The three dialysis
units that were involved in the study were (1) New Haven CAPD, a
large, free-standing PD unit in New Haven that cares for an average of
100 PD patients and does the PD training and follow-up for all of our
groups’ ESRD patients; (2) the Milford Dialysis Unit, a 14-station,
free-standing HD unit that cares for 88 HD patients; and (3) Branford

Copyright © 2006 by the American Society of Nephrology ISSN: 1555-9041/106-1191

1192 Clinical Journal of the American Society of Nephrology
Dialysis Unit, an eight-station, free-standing unit that cares for an
average of 36 patients.
A total of 90% of the dialysis patients were considered to be eligible
to participate in the study. A total of 94% of the potentially eligible PD
patients and 84% of the potentially eligible HD patients agreed to
participate. All questionnaires were distributed by one investigator
(E.J.), who was not affiliated with any of the dialysis units. Question-
naires were given to PD patients at the time of the patient’s monthly
appointment and were completed during the patient’s visit. Question-
naires for the HD patients were completed during the dialysis session.
When patients had any questions about the meaning of any part of the
survey, one investigator (E.J.) was available to provide an explanation
of the question to the patient. In addition, when the patient could not
read, the survey was read to him or her by E.J.
Basic demographic data were collected on all participating patients.
Charlson Comorbidity Index (CCI) scores, using standard methods,
were calculated by a detailed review of the medical records and dis-
cussion with the primary nephrologist (15,16).
Questionnaire Development
The questionnaire (Appendix) was developed by a committee that
included two nephrologists, two social workers, one nephrology phy-
sician assistant, two dialysis nurses (all of whom had had extensive
experience working with HD and PD patients), and one physician
assistant student (E.J.) who was not affiliated with the dialysis centers.
In developing the questionnaire, committee members used data from
previous studies by our group concerning patients’ perceptions of the
impact of dialysis therapy on their lives (13,15,16). The questionnaire
was designed to focus attention on the patients’ perception of their care
and the impact of the dialysis therapy on their overall life as well as
specific domains of their life, cited in our previous work, as perceived
by patients to be important for their quality of life (13,17,18).
The questionnaire had four major sections. The first section asked
patients about basic demographic information. The second section
asked about patients’ overall satisfaction with their dialysis therapy
and the overall impact of the dialysis treatment on their life. The
questions were graded on a scale from 1 to 10, with 10 being the most
satisfaction with or least impact of the therapy and 1 being the least
satisfaction with and greatest impact of the therapy on their lives. The
third section inquired about the impact that dialysis therapy had on
specific aspects of the patients’ life, focusing attention on 15 specific
domains: Overall health, stress level, family life, social life, indepen-
dence, finances, mood, religion/spirituality, sex life, energy level, rec-
reation/hobbies, exercise ability, living arrangements, appetite, and
body image. The same rating scale was used. The domains used were
based on previous interviews that were conducted by two of the
investigators (D.W. and S.F.) in which patients had identified the
specific domains of their life that were important for them (13,17,18).
The fourth section was a free-text section that asked patients to list the
three most important positive and three most negative aspects of their
dialysis therapy. The questionnaires generally took between 10 and 15
min to complete.
Statistical Analyses
Surveys were collected and data were reviewed by three of the
investigators (E.J., D.W., and S.F.). Statistical analysis was performed
using an SPSS database (SPSS, Chicago, IL). Means were compared
between PD and HD using two-sided t test. For investigation of
whether CCI scores had an impact on satisfaction scores, a Pearson
correlation coefficient was calculated to compare the CCI scores with all
domains.
Clin J Am Soc Nephrol 1: 1191–1196, 2006
For the free-text answers, all PD and HD patients’ answers were
divided into positive and negative responses for each dialysis modality.
Responses were grouped into categories on the basis of the judgment of
the three reviewers. The primary investigator (E.J.) created the catego-
ries after a careful review of individual patient responses; these cate-
gories and the individual patient responses were reviewed by two
additional investigators (D.W. and S.F.). Consensus was reached as to
the appropriateness of each category and the assignment of individual
answers into its specified category.
Proportional odds models (Stata, version 8.2; StataCorp, College
Station, TX) were estimated using ordered logistic regression with
overall satisfaction and overall impact of therapy as outcome variables,
adjusting for age, gender, race, comorbidity (CCI), duration of dialysis
(in months), presence or absence of diabetes, marital status, education,
employment status, living arrangements, and mode of dialysis (PD or
HD).
Results
A total of 146 patients participated in this study; 84 patients
were on HD, and 62 were on PD. Patient demographics, shown
in Table 1, indicated that HD participants were older than PD
participants. A total of 47.6% of the HD and 51.6% of PD
patients had diabetes. The duration of dialysis was not different
between the PD and the HD patients. There were no significant
differences between the basic demographic data of the HD and
PD patients in terms of educational backgrounds or home
situations.
HD patients had a mean CCI score of 7.90 ! 1.87, which was
significantly higher than the CCI score of the PD patients
(5.80 ! 2.68; P " 0.0001). No significant correlation was noted
between the CCI score and any of the patient satisfaction or
impact scores.
The mean patient satisfaction score for PD patients (8.02 !
1.41) was higher than that for the HD patients (7.4 ! 1.41),
but these differences were not statistically significant on
univariate analysis (P # 0.15). The mean score of the ques-
tion concerning the overall impact of the dialysis therapy on
the patients’ lives was significantly higher for the PD pa-
tients (7.25 ! 2.12) than for the HD patients (6.19 ! 2.83; P #
0.019). The scores for the PD patients also were significantly
higher than the scores for the HD patients in the following
five domains: Family life (6.83 ! 0.71 versus 5.91 ! 2.83; P #
0.032), independence (6.18 ! 0.71 versus 5.14 ! 2.835; P #
0.016), religion/spirituality (7.02 ! 0.71 versus 5.97 ! 1.41;
P # 0.006), energy level (5.15 ! 2.12 versus 4.30 ! 4.95; P #
0.035), and living situation (6.70 ! 0.01 versus 5.40 ! 2.12;
P # 0.001). The PD scores were higher but not significantly
so in terms of impact on stress level, health, social life,
activity, energy level, recreation, appetite, and body image.
The only domain in which HD patients had higher scores
than the PD patients was in terms of sex life, but this differ-
ence was minimal (4.15 ! 2.12 versus 4.0 ! 2.83; Table 2).
With the proportional odds model analysis correlating
overall satisfaction as outcome variable, adjusting for age,
gender, race, comorbidity, duration in months of dialysis,
presence or absence of diabetes, marital status, education,
employment status, living arrangements, and mode of dial-
ysis, the only variable that was significantly associated with
Clin J Am Soc Nephrol 1: 1191–1196, 2006 Satisfaction with and Impact of PD and HD 1193

Table 1. Demographic dataa

Characteristic HD (n # 84) PD (n # 52) P

Mean age (yr) 69.6 ! 13.3 55 ! 14 "0.0001

Patients with diabetes (%) 47.6 51.6 NS
Duration of dialysis (mo) 32 ! 20 38 ! 18 NS
CCI score 7.9 ! 1.9 5.8 ! 2.7 "0.0001
Ethnicity (%) NS
white 80 77
black 20 17
Education (%) NS
"high school diploma 21 18
high school diploma 32 32
$high school diploma 48 50
Married (%) 60 57 NS
Living arrangements (%) NS
with family 78 70
ECF 10 8
a
CCI, Charlson Comorbidity Index; ECF, extended care facility; HD, hemodialysis; PD, peritoneal dialysis.

Table 2. Differences between PD and HD patients: Satisfaction with therapy and impact of therapy on their lives
Domain PD HD P

Global satisfaction 8.02 ! 1.41 7.4 ! 1.41 0.15

Global impact of therapy 7.25 ! 2.12 6.19 ! 2.83 0.02
Family life 6.83 ! 0.71 5.91 ! 2.83 0.03
Independence 6.18 ! 0.71 5.14 ! 2.83 0.02
Religion/spirituality 7.02 ! 0.71 5.97 ! 1.4 0.006
Energy level 5.15 ! 2.12 4.30 ! 4.95 0.04
Living situation 6.70 ! 0.01 5.40 ! 2.12 0.001
Stress level 5.91 ! 0.71 5.78 ! 0.711 NS
Overall health 6.57 ! 0.71 6.12 ! 0.71 NS
Social life 5.76 ! 1.41 5.00 ! 0.71 NS
Mood 5.88 ! 1.41 5.33 ! 3.54 NS
Exercise ability 4.85 ! 2.83 4.14 ! 2.83 NS
Recreation 5.12 ! 1.41 4.95 ! 2.12 NS
Appetite 5.98 ! 2.83 5.66 ! 2.10 NS
Body image 5.12 ! 1.43 5.02 ! 1.41 NS
Finances 4.96 ! 2.83 5.51 ! 2.12 NS
Sex life 4.01 ! 2.83 4.15 ! 2.12 NS

overall satisfaction was mode of dialysis (coefficient 1.0997, Satisfaction Taxonomy

P # 0.037). The odds ratio for improvement in overall satis-
faction score by one ordinal level was 3.003 (95% confidence
interval 1.0686 to 8.4412) for PD compared with HD as mode
of dialysis. A similar analysis using overall impact of therapy
on patients’ lives as the outcome variable demonstrated that
the only significant predictor was mode of dialysis (coeffi-
cient 1.2684, P # 0.021), with odds ratio for improvement in
overall impact score by one ordinal level of 3.5553 (95%
confidence interval 1.2077 to 10.4667).
Patient responses to the free-text portion of the questionnaire
concerning the dialysis therapy were codified into positive and
negative categories of HD and PD treatment. The most frequently
cited categories are summarized in Table 3. The most frequently
cited positive categories for HD (in order of highest frequency)
were staff interactions, being alive and well, frequency of medical
care, social interaction with other dialysis patients, and feeling
happier/healthier. The most frequently cited positive categories
for PD were being alive and well, feeling happier/healthier, abil-
1194 Clinical Journal of the American Society of Nephrology
Table 3. Most frequently cited impacts of dialysis therapy
HD
Positive Responses Negative Responses
Staff interaction Length of treatment
(n # 34) (n # 25)
Being alive and well Needle sticks/access
(n # 30) (n # 20)
Frequent medical care Fatigue/weakness
(n # 17) (n # 18)
Social interaction Cramping/sick after treatment
(n # 15) (n # 19)
Improved strength/energy Frequency of treatment
(n # 10) (n # 12)
Transportation to unit
(n # 12)
ity to do treatment at home, ability to do treatment while sleeping,
and increased independence. The most frequent negative HD
categories were length of treatment, needles/access, fatigue/
weakness, cramping, and frequency of treatment, whereas the
most frequently reported negative PD categories were problems
with supplies, bloating/pain, frequency of treatment, change in
daily routine, and interference with sleep.
Discussion
The purpose of our study was to compare patient satisfaction
with HD and PD and the impact of these therapies on patients’
lives. In addition, this study sought to describe the positive and
negative aspects that patients experience with each dialysis
modality and develop further a taxonomy of patient concerns
with their therapy.
Previous studies have indicated that PD patients in general
report a higher satisfaction with their therapy than HD pa-
tients. For example, the Choices for Healthy Outcomes in
Caring for ESRD (CHOICE) study showed that patients who
received PD were 1.5 times as likely to rate their dialysis care
as excellent compared with patients who received HD (6).
This difference was the largest when patients were asked
about the information that they were given about each mo-
dality and the amount of dialysis information given by the
staff (6). In addition, patients who are well informed about
each dialysis modality and choose PD are more likely to rate
their care higher and report greater satisfaction with their
care than are patients who choose HD (19). Furthermore, PD
patients are less likely to want to switch modalities than HD
patients (19). These studies focus attention on patients’ per-
ception of their care and not on the impact of the therapy on
their lives. Our study confirms the greater satisfaction of PD
patients with their therapy and also addresses the issue of
the impact of the therapy by asking patients to rate this
impact globally as well as on 15 different domains that were
cited previously by patients as being important for their
quality of life. It indeed is striking that PD patients indicate
that there is significantly less impact of the therapy globally
as well as in 14 of the 15 domains. Perhaps this is not so
Clin J Am Soc Nephrol 1: 1191–1196, 2006
PD
Positive Responses Negative Responses
Improved strength/energy Problems with supplies
(n # 23) (n # 14)
Being alive and well Frequency/length of treatment
(n # 16) (n # 14)
Do therapy at home Bloating/pain
(n # 14) (n # 12)
Do therapy while sleeping Interference with sleep
(n # 12) (n # 9)
Increased independence Change in daily routine
(n # 9) (n # 6)
surprising given that the treatment is done at home and not
in-center. However, PD is done nightly, whereas HD is done
three times per week.
The PD patients had significantly fewer comorbidities than
the HD patients and were significantly younger, confirming
findings from other studies (6). The CCI, which was used in our
study and has been used extensively in studies of dialysis
patients, includes 19 categories of comorbid illness and in-
cludes age in the index (15,16,20). It is interesting that the CCI
did not correlate with any of the patient assessments. Similarly,
in the CHOICE study, comorbidity scoring did not have an
impact on patient satisfaction ratings (6).
Our study has some shortcomings. All patients included in
the study were cared for by the same physicians, who are part
of one group. PD use in this group is high, with $30% of
prevalent patients with ESRD being maintained on PD. No
patients who were on home HD were included. Because only
prevalent patients were studied, it is possible that the results
reflect the effect of selection bias; that is, patients who remained
on PD were those with better outlooks. Some of the differences
between the PD and HD patients could be attributed to modal-
ity selection bias; that is, patients with a more positive outlook
were more likely to have chosen PD. The questionnaire that
was used was developed by the authors and has not been used
elsewhere. However, 90% of the patients who were cared for in
the three facilities were considered eligible to participate, and
between 84 and 94% of eligible patients participated The pa-
tient group seems to be a typical ESRD cohort. The patient
satisfaction data are similar to that reported in the CHOICE
study, lending support to the validity of the measurements.
Our study suggests that PD patients in general are more
satisfied with their overall care and believe that their treatment
has less impact on their lives than do HD patients. In addition,
a taxonomy of patient perceptions of their care, both positive
and negative, has been developed. We think that it is important
to extend these observations to a larger cohort of patients and
eventually to incorporate this information into education pro-
grams for patients with chronic kidney disease to assist with
modality selection.
Clin J Am Soc Nephrol 1: 1191–1196, 2006 Satisfaction with and Impact of PD and HD 1195

Appendix: Dialysis Survey

Date !!!!!!!!!!!!!!!!
In this survey, we are asking you about your satisfaction with your dialysis therapy and the impact that the dialysis treatment has had on
your life.

1. Please rate your overall satisfaction with your dialysis therapy using the 1 to 10 scale provided, %1" being completely dissatisfied and
"10" being completely satisfied.

1 2 3 4 5 6 7 8 9 10

2. Please rate the overall impact that dialysis has had on your life, %1" being very negative impact, "5" being NO impact, and "10" being
very positive impact.

1 2 3 4 5 6 7 8 9 10

In the questions below, we would like to ask you about the impact that dialysis therapy has had on several aspects of your life. Please
rate these questions on the 1 to 10 scale provided, %1" being very negative impact, "5" being NO impact, and "10" being very positive
impact.

3. Stress level? 1 2 3 4 5 6 7 8 9 10

4. Overall health? 1 2 3 4 5 6 7 8 9 10

5. Family life? 1 2 3 4 5 6 7 8 9 10

6. Social life? 1 2 3 4 5 6 7 8 9 10

7. Independence? 1 2 3 4 5 6 7 8 9 10

8. Finances? 1 2 3 4 5 6 7 8 9 10

9. Mood? 1 2 3 4 5 6 7 8 9 10

10. Religion/spirituality? 1 2 3 4 5 6 7 8 9 10

11. Sex life? 1 2 3 4 5 6 7 8 9 10

12. Energy level? 1 2 3 4 5 6 7 8 9 10

13. Recreation/hobbies? 1 2 3 4 5 6 7 8 9 10

14. Exercise ability? 1 2 3 4 5 6 7 8 9 10

15. Living arrangements? 1 2 3 4 5 6 7 8 9 10

16. Appetite? 1 2 3 4 5 6 7 8 9 10

17. Body image? 1 2 3 4 5 6 7 8 9 10

Any kind of dialysis treatment can have complications, disadvantages, and negative side effects. Please choose the top three items that
bother you most about your dialysis therapy and describe them in the spaces provided.
1.

2.

3.

Dialysis treatment also has positive aspects. Please choose the top three items that you like the most about your dialysis therapy and
describe them in the spaces provided.
1.

2.

3.

Comments?

Thank you for completing this survey. It will help us better understand your experience with your dialysis treatment and how
we can improve our program.
1196 Clinical Journal of the American Society of Nephrology
References
1. Jaar BG, Coresh J, Plantinga LC, Fink NF, Klag MJ, Levey
AS, Levin NW, Sadler JH, Kliger A, Power NR: Comparing
the risk for death with peritoneal dialysis and hemodialy-
sis in a national cohort of patients with chronic kidney
disease. Ann Intern Med 143: 174 –183, 2005
2. Vonesh EF, Snyder JJ, Foley RN, Collins AJ: The differen-
tial impact of risk factors on mortality in hemodialysis and
peritoneal dialysis. Kidney Int 66: 2389 –2401, 2004
3. Schaubel DE, Morrison HI, Fenton SS: Comparing mortal-
ity rates on CAPD/CCPD and hemodialysis. The Canadian
experience: Fact or fiction? Perit Dial Int 18: 478 – 484, 1998
4. Schulman G: Mortality and treatment modality of end-
stage renal disease. Ann Intern Med 143: 227–228, 2005
5. Polascheck N: Living on dialysis: Concerns of clients in a
renal setting. J Adv Nurs 41: 44 –52, 2003
6. Rubin HR, Fink NE, Plantinga LC, Sadler JH, Powe NR:
Patient ratings of dialysis care with peritoneal dialysis vs
hemodialysis. JAMA 291: 697–704, 2004
7. Paniagua R, Amato D, Vonesh E, Guo A, Mujais S: Health-
related quality of life predicts outcomes but is not affected
by peritoneal clearance: The ADEMEX trial. Kidney Int 67:
1093–1104, 2005
8. Mapes D, Lopes AA, Satayathum S, McCullough KP,
Goodkin DA, Locatelli F, Fukuhara S, Young EW, Kuro-
kawa K, Saito A, Bommer J, Wolfe FA, Held PJ, Port FK:
Health-related quality of life as a predictor of mortality and
hospitalization: The Dialysis Outcomes and Practice Pat-
terns Study (DOPPS). Kidney Int 64: 339 –349, 2003
9. Leggat JE: Adherence with dialysis: A focus on mortality
risk. Semin Dial 18: 137–141, 2005
10. Kimmel PL, Peterson RA: Depression in end-stage renal
disease patients treated with hemodialysis: Tools, corre-
lates, outcomes, and needs. Semin Dial 18: 91–97, 2005
Clin J Am Soc Nephrol 1: 1191–1196, 2006
11. Unruh ML, Weisbord SD, Kimmel PL: Health-related qual-
ity of life in nephrology research and clinical practice.
Semin Dial 18: 82–90, 2005
12. Concato J, Feinstein AR: Asking patients what they like:
Overlooked attributes of patient satisfaction with primary
care. Am J Med 102: 399 – 406, 1997
13. Wuerth D, Finkelstein SH, Finkelstein FO: Patient assess-
ment of quality of care in a peritoneal dialysis facility. Am J
Kidney Dis 35: 638 – 643, 2000
14. Bass EB, Wills S, Fink NE, Jenckes MW, Sadler JH, Levey
AS, Meyer K, Powe NR: How strong are patients’ prefer-
ences in choices between dialysis modalities and doses?
Am J Kidney Dis 44: 695–705, 2004
15. Charlson ME, Pompei P, Ales KL, McKenzie CR: A new
method of classifying prognostic comorbidity in longitudi-
nal studies: Development and validation. J Chronic Dis 40:
373–383, 1987
16. Bikbov MN, Tomilina FT: Some notes about the usage of
the Charlson co-morbidity index. Nephrol Dial Transplant
19: 2926 –2927, 2004
17. Wuerth DB, Finkelstein SH, Schwetz O, Carey H, Kliger
AS, Finkelstein FO: Patients’ descriptions of specific factors
leading to modality selection of chronic peritoneal dialysis
or hemodialysis. Perit Dial Int 22: 184 –190, 2002
18. Kliger AS, Finkelstein FO: Peritoneal dialysis: Improving
patients’ lives and learning what they want. Semin Dial 12:
363–364, 1999
19. Ahlmen J, Carlsson L, Schonborg C: Well-informed pa-
tients with end-stage renal disease prefer peritoneal dial-
ysis to hemodialysis. Perit Dial Int 13: 196 –198, 1993
20. Fried L, Bernadini J, Piraino B: Comparison of the Charlson
Comorbidity Index and the Davies score as a predictor of
outcomes in PD patients. Perit Dial Int 23: 68 –573, 2003