Open Journal of Nephrology, 2013, 3, 139-147
http://dx.doi.org/10.4236/ojneph.2013.33026 Published Online September 2013 (http://www.scirp.org/journal/ojneph)
Effectiveness of an Secondary Prevention Program in
Chronic Kidney Disease
Carlos Enrique Yepes Delgado*, Yanett Marcela Montoya Jaramillo, Beatriz Elena Orrego Orozco,
Paulina Bernal Ramírez, Luz Denise González, José Miguel Abad Echeverri,
María Patricia Arbeláez Montoya
University of Antioquia, Hospital Pablo Tobón Uribe, EPS SURA, Medellín, Colombia
Email: *caenyede@gmail.com
Received July 13, 2013; revised August 5, 2013; accepted August 14, 2013
Copyright © 2013 Carlos Enrique Yepes Delgado et al. This is an open access article distributed under the Creative Commons Attri-
bution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
ABSTRACT
Background: There are many programs which focus on late-stage chronic kidney disease (CKD), and it is considered
that further evidence needs to be generated regarding the effectiveness of the programs used before renal replacement
therapy. Study Design: A cohort study. Settings & Participants: Patients over 15 years of age who had been diagnosed
with CKD according to the KDOQI (Kidney Disease Outcomes Quality Initiative) guidelines and who had undergone
conventional treatment (CT) or a renal protection program (RPP). These were patients of two Colombian health insur-
ance companies. Predictors: Age, sex, marital status, comorbidities, CKD stage, and clinical indicators. Outcomes: First
CKD progression, and need for renal replacement therapy (RRT). Measures: Clinical marker. Results: The RPP is
structurally and functionally different from the CT. It offers the interdisciplinary management of patients, a greater
number of medical appointments, and patients start to receive treatment at younger ages and at earlier stages of their
condition. The clinical markers of the patients following the RPP are within adequate ranges, and their renal function is
less impaired, despite the differences in basal conditions. Upon finishing the study, we found that patients who received
CT had a higher risk of receiving nephrotoxic drugs and not receiving nephroprotective drugs. The explanatory vari-
ables for the first progression were age, stage, history of dyslipidemia, and hemoglobin, potassium, and albumin levels.
These variables, together with glycemia levels were also valid for RRT, except for history of dyslipidemia, as it was not
significant. Upon adjusting for the explanatory variables, it was found that belonging to the RPP and attending more
appointments had a protective effect in the process of controlling renal damage. Limitations: A possible selection bias.
Conclusions: Belonging to a structured renal protection program is an effective way to keeping the clinical markers as-
sociated with renal impairment within normal ranges.
Keywords: Chronic Kidney Disease; Renal Protection Program; Effectiveness; Clinical Markers; Progression of Renal
Damage; Renal Replacement Program
1. Introduction
Due to the rapid increase in its prevalence, its close
relationship with an increased risk of cardiovascular
disease, and the high costs of treatment, chronic kidney
disease (CKD) is now recognized as a public health
problem that affects people of all ages and alters their
quality of life [1,2]. Because of these characteristics, this
disease is becoming a priority for healthcare systems.
This has created an urgent need to develop effective
ways of measuring interventions to prevent its progress
[3,4].
According to the calculations of the National Health
and Nutrition Examination Survey (NHANES) III, it is
estimated that twenty million adults have CKD in the
United States. Currently, approximately 360,000 patients
have undergone renal replacement therapy (RRT) in the
European Union. In developed countries, it is estimated
that the number of people with end-stage CKD will
continue to increase at an annual rate of around 5% to
8%. This growth is driven by population aging and the
increased incidence of diseases related to renal failure
[5].
In Colombia, there are over 20,000 individuals with
stage 5 CKD, with an age-adjusted prevalence of 454
*Corresponding author.
C
opyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL.
140
patients per million people. This figure is lower than the
prevalence observed in the United States and higher than
the average prevalence in Latin American countries in
2005. Nevertheless, the prevalence of early-stage CKD is
only 0.87%. This figure is lower than the one reported in
other studies conducted on population samples and is
thought to be the result of underreporting [6].
In order to delay CKD progression in Colombia, the
Ministry of Social Protection implemented a model for
its prevention and treatment in 2006. The model aimed to
address the growing demand for services related to this
condition [7,8]. Furthermore, Act 1122 of 2007 created
the Cuenta de Alto Costo agency (CAC), a non-
governmental technical agency forming part of the
Colombian General System for Health-Related Social
Security. This is a strategy involving both public and
private institutions and seeking to address the high costs
generated by the condition itself, an issue that has a
significant impact on Colombia [9]. However, before the
appearance of such provisions, a Colombian health
insurance company (EPS) designed and implemented a
renal protection program (RPP) that was structurally and
functionally capable of providing its users with a parti-
cular type of medical care that was very different from
the conventional treatment (CT) that other institutions
continued to offer.
Although a number of national health institutions from
across the world claim that it is important to reduce
morbidity and mortality rates in the proactive manage-
ment of CKD by making early referrals of patients to a
multidisciplinary team, there are scant data within the
current literature to support such a recommendation [10].
The authors of a meta-analysis published in April 2010
concerning evidence regarding the clinical and cost
effectiveness of early referral strategies for managing
people with markers of renal damage reported that, in the
last two decades, only two studies in the world had
focused on assessing the effectiveness of interventions
carried out at early stages of CKD [11].
Not enough reports were found in the scientific litera-
ture that assessed RPPs which continuously monitored
patients for at least four years and observed the relation-
ship between the delay in kidney function impairment,
and the behavior of variables such as clinical parameters
(blood pressure, glycemia, lipid profile, etc.) and nutri-
tional treatment [12-16].
Since the mechanisms for identifying CKD patients in
the stages prior to RRT are not very effective, we are
missing out many opportunities to implement strategies
to reduce the progression speed of renal failure and thus
reduce the incidence of terminal renal failure [6]. That is
why this study aims to assess the effectiveness of the
renal protection program in monitoring the clinical
markers and their relationship with the progression of
renal damage.
2. Methods
2.1. Type of Study
An analytic cohort study with ambispective follow-up
given to two dynamic cohorts of patients over 15 years
old and diagnosed with CKD. Individuals were classified
by stage (according to the KDOQI guidelines) [17] in
order to compare the behavior of the clinical and renal
impairment indicators of the patients exposed to the RPP
and those of the patients undergoing CT.
2.2. Population
The study was conducted with patients from two health
insurance companies belonging to the same type of social
security regime and operating in Antioquia, Colombia.
Data was collected retrospectively from electronic me-
dical records from April 1, 2004 (the starting date of the
RPP) to the first half of 2007. After that date, informa-
tion was collected prospectively until April 30, 2008.
The total treatment time was 49 months. Figure 1 de-
scribes the participant selection process.
2.3. Treatment Comparison
The RPP is an interdisciplinary healthcare program based
on a protocol involving periodic appointments and
moni-toring of clinical and laboratory tests accompanied
by continuous education. The RPP is aimed at CKD pa-
tients from the initial stages of their condition and it is
divided into several levels. The first level is aimed at pa-
tients in stages 1 and 2. It offers medical appointments
Anal yze dMedical Records
9887
RPP:5589CT:4298
Pati entswhocompliedwiththecriteria
5663
RPP:4202CT:1461
(74.2%)(25.8%)
Pa tientesexcludedbecausethey
didn’tmeetthecriteria
4224
RPP:1387CT:2837
Lossofpa tientsfordifferent
reasons
1010(17.8%)
RPP:767CT:243
(18.3%)(16.6%)
Treatmentaba ndonment
509(9,0%)
RPP:391CT:118
(9.3%)(8.1%)
Deaths
312(5.5%)
RPP:224CT:88
(5.3%)(6.0%)
RPP:renalprotectionprogram.CT:conventionaltrea tme nt
Figure 1. Flowchart depicting the CKD patient selection
process in two Colombian private health insurance com-
panies. Medellin 2004-2008.
Copyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL. 141
with interists and nutritionists on an annual and biannual
basis, respectively. The second level is for patients in
stages 3 and 4, and it offers medical appointments with
internists, nephrologists, and nutritionists on a three-per-
year and bimonthly basis, respectively. Laboratory tests
are per-ormed one month prior to the medical appoint-
ment with a specialist.
When patients request a medical appointment because
they feel the need to do it, conventional treatment (CT)
either provides primary level attention or refers patients to
a specialist, depending on the general practitioner’s cri-
teion.
No significant differences were found between the RPP
and CT groups in terms of treatment abandonment, loss of
patients for different reasons, and death (p = 0.157; p =
0.163; p = 0.318 respectively).
2.4. Variables
The variables considered as criteria for CKD diagnosis
were: creatinine clearance of less than 60 ml/min, al-
rations observed in renal ultrasound tests, proteinuria
levels higher than 150 mg/day, alterations in urinary
deposits, and -for diabetic patients- micro-albuminuria
levels higher than 30 mg.
The variables included in the study were: age, sex,
comorbidities, marital status, CKD stage according to
the KDOQI guidelines, and clinical and laboratory
variables as measured upon diagnosis and whenever
stage progression took place. Some quantitative vari-
ables were categorized based on the clinical definitions
for cutoff in accordance with the values reported in the
literature
2.5. Analysis Technique
A comparison was carried out between the guidelines
for clinical monitoring and care of patients with renal
alterations used by the institution offering the RPP, and
the treatment process used by the institution offering
CT. The objective was to identify differences in the
healthcare process, tests carried out, frequency of tests
and appointments, types of professionals involved in
the process, etc.
The population was characterized when CKD was
diagnosed through frequency distributions and propor-
tions for the qualitative variables. Quantitative varia-
bles were characterized via measures of dispersion and
descriptive statistics. The Kolmogorov-Smirnov or
Shapiro Wilk tests were also used. The qualitative
variables of the two groups under study were then
compared using the Pearson Chi2 test or the Chi2 test
for trends, depending on the case. If the quantitative
variables had a normal distribution, they were com-
pared using the t-Student test. If the distribution was
not normal, then the U Mann-Whitney test was used.
The odds ratios (OR) and their corresponding confi-
dence intervals were calculated to be 95%.
A propensity score (PS) was used to assess ef-
fectiveness in maintaining the values for clinical mar-
kers and exposure to nephrotoxic and nephroprotective
drugs within normal ranges. This statistical technique
calculates the conditional probability of having been
exposed to the variable of interest based on the sub-
ject’s baseline features [18]. The PS is composed of the
following variables: age (>65 years/<=65 years), sex
(female/male), marital status (no partner/ with perma-
nent partner), arterial hypertension (yes/no), diabetes
mellitus (yes/no) and dyslipidemia (yes/no). The PS
created with these variables shows how they explain
the probability of joining a RPP or undergoing CT by
means of a logistic regression model that aims to
predict exposure, and manages to group the various
confounding variables into a single one [19]. These
models also included the initial values of each variable.
Finally, two logistic regression models were created.
Their dependent variables were the first progression
into a later stage and the need for RRT. The variables
included in the models were selected based on the
Hosmer Lemeshow test. The quantity of data available
for each variable was determined using the stepwise
technique to obtain the final model. These models
excluded patients who joined the study at stage 5 of
their condition.
2.6. Avoidance of Bias
A potential bias could have arisen when we were forming
the cohort and encountered participants with different
features which increased the probability of having any of
the outcomes. We therefore conducted multivariate ana-
lyses that adjusted for the features and basal conditions
of the subjects. To avoid selection bias, we ensured that
patients complied with the inclusion criteria. Possible
migration bias could have occurred when the conditions
of those who abandoned treatment were different from
the conditions of those who stayed in it, and when such
conditions were associated with the outcome itself. To
prevent this, we inquired about censoring, either by
reviewing the records of the health insurance companies
or by asking via telephone. We thus located 80% of the
patients and found no differences between the groups.
A possible reporting bias might have occurred when it
was more likely to find more cases in the cohort being
monitored for a longer time period (the RPP). Another
bias resulted from the fact that the RPP actively searches
for patients, thus its clinical records had more data than
those of the patients following CT (underreporting). This
bias was avoided by triangulating information and
crossing the records for the missing data when possible.
Copyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL.
142
3. Results
When comparing the number of appointments attended
by patients from both programs, it was observed that
patients belonging to the RPP had more appointments
with internists, as well as educational and nutritional ap-
pointments. The highest percentage of appointments for
patients following CT was with general practitioners and
ophthalmologists, as shown in Table 1.
There was a proportional difference in CKD stage at
the moment of diagnosis between the patients from the
two health insurance companies. More than 30% of the
patients who joined the RPP were in the first two stages
of the condition, compared to only 16.7% of patients who
underwent CT. In other words, CT had a higher number
of patients in the final stages of CKD. Patients undergo-
ing CT are more likely to be diagnosed with CKD when
the condition is already in the final stages of its devel-
opment, as shown by the OR represented in Table 2.
According to this, a patient following CT is approxima-
tely 50% more likely to be diagnosed with late-stage
CKD (stages 4 and 5) than a patient from the RPP.
Additionally, CKD patients following CT were signi-
ficantly older (Me:70 years old) than RPP patients (Me:
66 years old). This was observed in both males and fe-
males.
Upon comparing the morbidities of the patients fol-
lowing CT and those of the RPP patients, it was observed
that prevalence values were similar for both groups
(Figure 2), except for arterial hypertension (p < 0.001),
dyslipidemia (P = 0.023), diabetes mellitus (p = 0.014),
cerebrovascular disease (0.002) and obesity (p = 0.017).
After analyzing the values observed for the various
clinical markers at the moment of diagnosis, we found
significant differences between the CT and RPP patients
regarding serum creatinine, creatinine clearance, triglyc-
erides, and potassium. These values were less affected or
even within normal ranges for the RPP patients. The
values for the high- and low-density lipoproteins were
more favorable for the patients undergoing CT. No sig-
nificant differences were found between the values for
systolic arterial pressure, glycated hemoglobin, 24-hour
proteinuria, total cholesterol, hemoglobin, phosphorus,
and parathyroid hormone.
Likewise, upon analyzing the values at the moment of
diagnosis, we found that the risk of receiving nephrotoxic
drugs such as non-steroidal anti-inflammatory drugs or
aminoglycoside antibiotics is 1.7 times greater for pa-
tients following CT (CI95%: 1.4 - 2.1) than for patients
in the RPP. We also found that the risk of not receiving
nephroprotective drugs is 2.1 times higher (CI95%: 1.8 -
2.4) for patients with CT than for those following a RPP.
The risk of having out-of-range values for the clinical
markers and for the variable “final exposure to nephro-
toxic and nephroprotective drugs” at the end of the
treatment period was analyzed (Table 3). It was then
found that CT patients had a higher risk of having clini-
cal markers which were out of range for SAP, fasting
blood glucose, glycated hemoglobin, and potassium. The
risk of having out-of-range clinical markers for the cho-
lesterol variable was higher in the RPP patients. No dif-
ferences were found in albumin, calcium, and hemoglo-
bin. Similarly, the risk of receiving nephrotoxic drugs for
patients undergoing CT was 0.63 times greater than that
of the patients in the RPP. The risk of not receiving
nephroprotective drugs was 0.35 times higher for CT
patients.
The models described below show the effect of the
“first progression” and “need for RRT” variables on re-
nal function impairment. The variables are adjusted for
patient features (Table 4).
Regarding the variable representing the first progress-
sion into the next stage of CKD, we found that the pa-
tients in the RPP and those who attended more than eight
medical appointments had a lower risk of CKD progress-
sion. Their protective fraction was 51.7% and 71.3%
respectively. Likewise, a history of dyslipidemia, to-
gether with out-of-range levels of hemoglobin, potassium,
albumin, and glycemia, behaves as a risk factor for CKD
progression. On the other hand, out-of-range hemoglobin
and albumin levels represent the highest risk (2.91 and
1.53 respectively). The association was not significant
for glycemia. In terms of age, we found that the older the
patient, the higher the risk of CKD progression.
Regarding the risk of CKD progression based on the
condition’s stage at the moment of diagnosis (and taking
stage 4 patients as a reference point) we found that the
risk of having the first progression increases as the stage
at the moment of diagnosis decreases. Hence, the risk of
progression for patients in stage 1 is 38.75 times higher
than the risk for those in stage 4.
The model used to analyze the need for RRT also
showed that patients belonging to the RPP have 44.9%
less probability of requiring renal replacement therapy.
Similarly, attending more than 8 medical appointments
means 66.2% less probability of requiring RRT.
Likewise, it was found that patients with out-of-range
levels of glycemia, hemoglobin, potassium, and albumin
had a higher risk of needing RRT than those with levels
within the acceptable range. Out-of-range hemoglobin
posed the highest risk for needing RRT (3.64), followed
by albumin (2.06). Similarly, patients with a history of
dyslipidemia had a higher risk of needing RRT, but the
association was not significant. Additionally, it was
found that being younger is a protective factor against the
risk of requiring RRT.
Regarding the CKD stage at the moment of diagnosis,
it was observed that patients in stage 4, had a 9.35 times
greater risk of requiring RRT than patients in stages 1, 2, and 3.
Copyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL.
Copyright © 2013 SciRes. OJNeph
143
Table 1. Number of appointments for CKD patients in the RPP. Appointments are arranged by appointment type and treat-
ing professional. Medellín 2004-2008.
RPP CT
Clinical Indicators N (%) Md (Q1 - Q3) N (%) Md (Q1 - Q3)
Educational appointments 450 (10.7) 4 (2 - 8) 55 (3.8) 1 (1 - 1)
Nutritional appointments 1630 (38.8) 2 (1 - 3) 320 (21.9) 1 (1 - 2)
Ophthalmological appointments 226 (5.4) 1 (1 - 2) 184 (12.6) 1 (1 - 1)
Psychological appointments 10 (0.2) 1 (1 - 6) 22 (1.5) 1 (1 - 3)
General practitioner 2401 (57.1) 3 (1 - 5) 1022 (70.0) 3 (1 - 5)
Internist 3909 (93.0) 4 (2 - 6) 773 (52.9) 3 (1 - 5)
Nephrologist 1925 (45.8) 3 (1 - 5) 4 (0.3) 3.5 (1.3 - 6.5)
Total appointment s 4186 (99.6) 8 (4 - 15) 1316 (90.1) 8 (4 - 13)
Table 2. Percentage distribution and probability of belonging to each CKD stage among patients benefiting from two types of
intervention. Medellín, 2004-2008.
RPP CT
Stage of the condition
upon diagnosis n (%) n (%) OR CI 95%
1 403 (9.6) 148 (10.1) 1.00
2 885 (21.1) 96 (6.6) 0.30 (0.22 - 0.4)
3 2354 (56.0) 911 (62.4) 1.05 (0.86 - 1.3)
4 493 (11.7) 267 (18.3) 1.47 (1.17 - 1.91)
5 67 (1.6) 39 (2.7) 1.59 (1 - 2.51)
P value of the Chi-square test for tendency <0.001
Figure 2. Comorbidities of the groups of CKD patients following a RPP and undergoing conventional treatment TC. Medellín,
2004-2008. HTN: arterial hypertension, DM: diabetes mellitus, AMI: acute myocardial infarction, C. Disease: Coronary dis-
ease, CVD: cerebrovascular disease, UTI: urinary tract infection, ARF: acute renal failure.
C. E. YEPES D. ET AL.
144
Table 3. Clinical conditions and exposure to nephrotoxic and nephroprotective drugs at the end of the treatment for the CKD
patients in the renal protection program, RPP, and for those undergoing conventional treatment, CT. These variables are
adjusted for the basal conditions (Propensity Score). Medellín, 2004-2008.
CI 95.0%
Clinical Markers B S.E. WaldglP value OR IL SL
SAP
Intervention through CT 0.180 0.0726 10.013 1.197 1.0391.379
Initial systolic arterial pressure outside of normal range 1.013 0.063259 10.000 2.754 2.4343.116
PS 0.745 0.2857 10.009 0.475 0.2720.830
Constant 0.713 0.080 78 10.000 0.490
Fasting blood glucose
Intervention through CT 0.264 0.1304 10.042 1.302 1.0101.680
Initial glycemia outside of normal range 2.424 0.131342 10.000 11.291 8.73414.595
PS 0.965 0.5603 10.085 0.381 0.1271.143
Constant 1.742 0.151134 10.000 0.175
Hba1c
Intervention through CT 1.322 0.24429 10.000 3.751 2.3276.046
Initial Hba1c levels outside of normal range 2.911 0.202209 10.000 18.377 12.38027.276
PS 1.650 0.9653 10.087 0.192 0.0291.274
Constant 1.995 0.282 50 10.000 0.136
Total cholesterol
Intervention through CT 0.252 0.1165 10.030 0.778 0.6200.975
Initial total cholesterol level outside of normal range 2.064 0.097452 10.000 7.879 6.5149.530
PS 0.962 0.4375 10.028 0.382 0.1620.901
Constant 1.473 0.131126 10.000 0.229
Hemoglobin
Intervention through CT 0.228 0.1204 10.056 0.796 0.6301.006
Initial hemoglobin level outside of normal range 2.376 0.151248 10.000 10.763 8.00514.470
PS 1.450 0.46610 10.002 4.262 1.71010.619
Constant 1.391 0.127120 10.000 0.249
Albumin
Intervention through CT 0.694 0.4602 10.131 2.001 0.8134.930
Initial albumin level outside of normal range 2.925 0.220176 10.000 18.639 12.10128.708
PS 0.299 1.0940 10.785 0.742 0.0876.333
Constant 2.960 0.293102 10.000 0.052
Calcium
Intervention through CT 0.111 0.2480 10.656 0.895 0.5511.455
Initial calcium level outside of normal range 1.827 0.177107 10.000 6.218 4.3998.788
PS 1.258 0.7153 10.078 3.519 0.86714.285
Constant 1.985 0.203 95 10.000 0.137
Potassium
Intervention through CT 0.391 0.1616 10.015 1.479 1.0782.029
Initial potassium level outside of normal range 1.523 0.16981 10.000 4.586 3.2956.384
PS 0.268 0.5480 10.625 1.307 0.4473.825
Constant 1.642 0.147124 10.000 0.194
Nephrotoxic drugs
Intervention through CT 0.491 0.12316 10.000 1.633 1.2832.080
Initial nephrotoxic drug consumption 2.811 0.118564 10.000 16.624 13.18220.966
PS 0.590 0.4851 10.224 1.803 0.6974.668
Constant 3.426 0.144567 10.000 0.033
Copyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL.
Copyright © 2013 SciRes. OJNeph
145
Continued
Nephroprotective drugs
Intervention through CT 0.300 0.099 9 1 0.002 1.350 1.1121.638
No initial nephroprotective drug consumption 2.642 0.102 674 1 0.000 14.045 11.50517.147
PS 3.365 0.359 88 1 0.000 28.935 14.32558.442
Constant -4.263 0.131 1.0591 0.000 0.014
Table 4. Logistic regression models for the variables “first progression” and “need for renal replacement therapy” amongst
the CKD patients in the RPP and those following conventional treatment (CT), 2004-2008.
First progression1 Need for RRT2
CI 95% CI 95%
Variables in the model P value OR IL SL
P value OR IL SL
Treatment with RPP* 0.000 0.483 0.336 0.696 0.010 0.551 0.345 0.882
Attending more than 8 appointments 0.000 0.287 0.219 0.376 0.000 0.338 0.208 0.550
Age 0.000 1.032 1.023 1.041 0.020 0.984 0.972 0.997
History of dyslipidemia 0.010 1.354 1.078 1.701 0.250 1.243 0.857 1.801
Stage of the condition upon diagnosis
Stage 4 0.000 1 (Reference) 0.000 10.354 7.02 15.273
Stage 3 0.000 2.018 1.468 2.775
Stage 2 0.000 10.5006.857 16.090 1 (Reference)**
Stage 1 0.000 39.75022.996 68.707
Clinical Indicators
Glycemia outside of normal range 0.870 1.025 0.757 1.389 0.010 1.899 1.212 2.977
Hemoglobin level outside of normal range 0.000 2.917 2.307 3.689 0.000 3.647 2.531 5.256
Potassium level outside of normal range 0.010 1.458 1.116 1.903 0.020 1.616 1.098 2.380
Albumin level outside of normal range 0.010 1.535 1.096 2.148 0.000 2.060 1.314 3.231
Constant 0.000 0.026 0.000 0.088
1. First progression: 2Nagelkerke R = 27.55%; 2. Need for RRT: 2 Nagelkerke R = 32.8%; *Reference intervention; **Stages 1, 2, and 3 were grouped as a ref-
erence in order to compare them with stage 4 in model 2.
4. Discussion
The Colombian healthcare system now considers CKD as
a catastrophic and very costly pathology. Thanks to this,
patients now have more and better access to the ser-
vices designed to address it. This has also helped health
insurance companies and other institutions reduce the
impact of the costs associated with this condition. Conse-
quently, improvements in users’ health are expected, as
well as a better use of the health system’s resources [8].
However, a great deal of studies have focused on the
final stages of CKD, both in the world and in Colombia,
and it is considered that more evidence should be gener-
ated regarding the effectiveness of the programs used
before RRT. In this sense, this study demonstrated the
effect of a predialysis intervention program complying
with strategies for generating scientific evidence [8,20],
such as early patient uptake, comprehensive care, and
treatment of comorbidities.
The fact that patients are started on the renal protection
program when they are at the initial states of their condi-
tion (which is demonstrated by the fact that 30% of the
patients join the program at stages 1 and 2 in comparison
with CT patients) demonstrates the significant effect of
this program’s active search for patients, something
which should be included in all secondary prevention
strategies.
The results show that the RPP patients receive more
interdisciplinary care. This is evidenced by the propor-
tion of patients seen by doctors and by the higher number
of medical appointments involving different health pro-
fessionals. This is compliant with the interdisciplinarity
proposed by a number of studies, which show differences
in CKD outcomes when comorbidities are treated by
multiple disciplines [21-23].
The higher percentage of educational appointments
and the comprehensive care provided by the RPP show
how it can be used for prevention. One of its positive
C. E. YEPES D. ET AL.
146
effects could be a lower exposure to nephrotoxic drugs
and higher rates of nephroprotective drugs usage.
The high number of general medical appointments for
patients undergoing CT may be due to the particular
characteristics of the on-demand type of health care pro-
vided by this kind of intervention. Additionally, the
greater quantity of ophthalmology appointments ob-
served in patients following CT may signify that this
target organ is more affected in their case. This could be
linked to the fact that their arterial hypertension, glyce-
mia, or glycated hemoglobin levels were not within nor-
mal ranges at the end of their treatment.
The literature recommends that patients in stage 3 and
onwards should be managed by nephrology professionals
[24,25]. Despite this, only 4 individuals following CT
had this kind of treatment, making it impossible to com-
pare the groups studied.
In terms of clinical indicators, it’s worth noting that,
although the logistic regression model showed that hav-
ing out-of-range hemoglobin and albumin levels repre-
sented a high risk of having the first CKD progression
and needing RRT, no significant differences were found
between the RPP and CT patients regarding the man-
agement of these markers upon adjusting for the basal
levels and PS. This shows how necessary it is to imple-
ment or improve the actions aiming at keeping the levels
of these indicators within normal ranges. This would
greatly contribute to reducing the risk of renal function
impairment [26].
Likewise, it is also worth mentioning that the RPP
shows shortcomings regarding the management of total
cholesterol when compared to CT—it is therefore neces-
sary to identify the cause of such shortcomings, since
inappropriate management of this indicator (i.e. high
cholesterol values) is a risk factor for renal function im-
pairment. Moreover, high cholesterol is also a risk factor
for cardiovascular disease, which in turn is the main
cause of morbidity and mortality among CKD patients
[27].
Patients diagnosed with stage 4 CKD had a high risk
of needing RRT compared with those diagnosed at earlier
stages of their condition. This demonstrates the impor-
tance of early detection of CKD in order to prevent or
slow down the deterioration of renal function and the
resulting deterioration of patients’ quality of life, 1-2 as
well as the economic consequences for the various ele-
ments of the General System for Health-Related Social
Security [28,29].
Although the results obtained in this study in most
cases showed differences favoring the RPP, it can also be
observed that the proportion of patients attending ap-
pointments offered by the program and having clinical
markers within the normal ranges could be improved.
However, the findings regarding the protection levels
provided by the RPP against the studied outcomes ex-
ceed 50% and explain more than 25% of its variability,
thus suggesting that it is possible to have more effec-
tiveness and a higher effect on the delay of CKD pro-
gression.
Implementing strategies for early intervention for
CKD patients is an effective contribution to help keep the
clinical markers associated with renal damage within
normal ranges. It also has a positive effect on the treat-
ment of the condition. Therefore, encouraging the im-
plementation of such strategies optimizes the use of re-
sources and improves patient prognosis.
5. Acknowledgements
The authors would like to thank Colciencias, the Univer-
sity of Antioquia, the EPS Sura health insurance com-
pany, and the insurance company offering CT. Study
sponsor not had any role in study design; collection,
analysis, and interpretation of data; writing the report;
and the decision to submit the report for publication. We,
the authors, declare that there were no conflicts of inter-
est during this study.
REFERENCES
[1] R. C. Atkins, “The Epidemiology of Chronic Kidney Dis-
ease,” Kidney International Supplement, Vol. 94, 2005,
pp. S14-18. doi:10.1111/j.1523-1755.2005.09403.x
[2] R. C. Atkins, “The Changing Patterns of Chronic Kidney
Disease: The Need to Develop Strategies for Prevention
Relevant to Different Regions and Countries,” Kidney In-
ternational Supplement, Vol. 98, 2005, pp. S83-85.
doi:10.1111/j.1523-1755.2005.09815.x
[3] R. S. Barsoum, “Chronic Kidney Disease in the Develop-
ing World,” The New England Journal of Medicine, Vol.
354, No. 10, 2006, pp. 997-999.
doi:10.1056/NEJMp058318
[4] S. L. White, A. Cass, R. C. Atkins and S. J. Chadban, “Chro-
nic Kidney Disease in the General Population,” Advances
in Chronic Kidney Disease, Vol. 12, No. 1, 2005, pp. 5-
13. doi:10.1053/j.ackd.2004.10.009
[5] N. Lameire, K. Jager, W. Van Biesen, D. de Bacquer and
R. Vanholder, “Chronic Kidney Disease: A European
Perspective,” Kidney International Supplement, Vol. 99,
2005, pp. S30-38. doi:10.1111/j.1523-1755.2005.09907.x
[6] Cuenta de Alto Costo, “Contenidos Mínimos Indispen-
sables Para la Gestión Del Riesgo Renal en un Programa
de Atención a Pacientes Adultos con Enfermedad Renal
Crónica, Sin Terapia de Reemplazo Renal en Colombia,
Definiciones Técnicas Basadas en Evidencias,” 2011.
http://www.cuentadealtocosto.org/patologias/ERC/docs/c
onsenso/programas_ERC.pdf
[7] Ministerio de la Protección Social, República de Colom-
bia, “Guía Para el Manejo de la Enfermedad Renal Cró-
nica y Modelo de Prevención y Control de la Enfermedad
Renal Crónica. Componente de un Modelo de Salud Re-
Copyright © 2013 SciRes. OJNeph
C. E. YEPES D. ET AL.
Copyright © 2013 SciRes. OJNeph
147
nal,” 2011.
http://pos.gov.co/Documents/Archivos/ERC.pdf
[8] F. Martínez and M. Valencia “Modelo en Prevención y
Control de la Enfermedad Renal Crónica. Componente de
un Modelo de Salud Renal,” Bogotá, 2005.
[9] Ministerio de Protección Social, República de Colombia,
“Cuenta de Alto Costo,” 2011.
http://www.cuentadealtocosto.org/CAC.html
[10] B. M. Curtis, P. Ravani, F. Malberti, F. Kennett, P. A.
Taylor, O. Djurdjev and A. Levin, “The Short- and Long-
Term Impact of Multi-Disciplinary Clinics in Addition to
Standard Nephrology Care on Patient Outcomes,” Ne-
phrology Dialysis Transplantation, Vol. 20, No. 1, 2005,
pp. 147-154. doi:10.1093/ndt/gfh585
[11] C. Black, P. Sharma, G. Scotland, K. McCullough, D.
McGurn, L. Robertson, N. Fluck, A. MacLeod, P. Mc-
Namee, G. Prescott and C. Smith, “Early Referral Strate-
gies for Management of People with Markers of Renal
Disease: A Systematic Review of the Evidence of Clinical
Effectiveness, Cost-Effectiveness and Economic Analy-
sis,” Health Technology Assessment, Vol. 14, No. 21,
2010, pp. 1-184.
[12] R. Pellicano, P. G. Kerr and R. C. Atkins, “What Have
We Learned from Clinical Trials on Prevention?” Kidney
International Supplement, Vol. 94, 2005, pp. 101-106.
doi:10.1111/j.1523-1755.2005.09424.x
[13] L. D. B. Gray, “Outcomes Research in Nutrition and
Chronic Kidney Disease: Perspectives, Issues in Practice,
and Processes for Improvement,” Advances in Chronic
Kidney Disease, Vol. 12, No. 1, 2005, pp. 96-106.
doi:10.1053/j.ackd.2004.10.012
[14] M. M. Slowik, “Early Education of Patients with Chronic
Renal Insufficiency: The Healthy Start program. Case
Study of the Anemic Patient,” Nephrology Nursing Jour-
nal, Vol. 28, No. 6, 2001, pp. 643-646.
[15] K. Tungsanga, C. Ratanakul, W. Pooltavee, N. Maha-
tanan, A. I. Na Ayuthaya and S. Rodpai, “Experience
with Prevention Programs in Thailand,” Kidney Interna-
tional Supplement, Vol. 94, 2005, pp. 68-69.
doi:10.1111/j.1523-1755.2005.09417.x
[16] R. T. Gansevoort, Van der Heij, C. A. Stegeman, F. T. De
Charro, M. G. Nieuwenhuizen and D. De Zeeuw, “Trends
in the Incidence of Treated End-Stage Renal Failure in
the Netherlands: Hope for the Future?” Kidney Interna-
tional Supplement, Vol. 92, 2004, pp. 7-10.
doi:10.1111/j.1523-1755.2004.09202.x
[17] National Kidney Foundation, “K/DOQI Clinical Practice
Guidelines for Chronic Kidney Disease: Evaluation,
Classification, and Stratification,” Recommendations for
clinical Performance Measures, 2007.
http://www.kidney.org/professionals/kdoqi/guidelines_ck
d/p8_cpm.htm
[18] M. Katz, “Multivariable Analysis. A Practical Guide for
Clinicians,” 2nd Edition, Cambridge, 2006.
[19] A. Ruiz and L. Morillo, “Epidemiología Clínica. Investi-
gación Clínica Aplicada,” Editorial Panamericana, Bo-
gotá, 2004.
[20] H. Tazeen, “The Growing Burden of Chronic Kidney Dis-
ease in Pakistan,” The New England Journal of Medicine,
Vol. 9, No. 10, 2006, pp. 995-997.
[21] B. J. Barrett, “Applying Multiple Interventions in Chronic
Kidney Disease,” Seminars in Dialysis, Vol. 16, No. 2,
2003, pp. 157-164.
doi:10.1046/j.1525-139X.2003.16032.x
[22] D. C. Mendelssohn, “Coping with the CKD Epidemic:
The Promise of Multidisciplinary Team-Based Care,”
Nephrology Dialysis Transplantation, Vol. 20, No. 1,
2005, pp. 10-12. doi:10.1093/ndt/gfh636
[23] R. Pellicano, R. C. Atkins and P. G. Kerr, “What Have
We Learned from Clinical Trials on Prevention?” Kidney
International Supplement, Vol. 94, 2005, pp. 101-106.
doi:10.1111/j.1523-1755.2005.09424.x
[24] B. J. Pereira, “Overcoming Barriers to the Early Detec-
tion and Treatment of Chronic Kidney Disease and Im-
proving Outcomes for End-Stage Renal Disease,” Ameri-
can Journal of Managed Care, Vol. 8, No. 4, 2002, pp.
S122-S135.
[25] “Chronic Kidney Disease: Stating the Managed Care
Case for Early Treatment Discussion and Consensus of
Presentations of Economic Analyses, Managed Care Or-
ganization Case Studies, and Opportunities for Interven-
tion in a Managed Care Setting,” American Journal of
Managed Care, Vol. 8, No. 4, 2002, pp. S114-S120.
[26] A. Levin, “The Need for Optimal and Coordinated Man-
agement of CKD,” Kidney International, Vol. 68, No. 99,
2005, pp. S7-S10.
[27] W. F. Owen, “Patterns of Care for Patients with Chronic
Kidney Disease in the United States: Dying for Improve-
ment,” Journal of the American Society of Nephrology,
Vol. 14, No. 2, 2003, pp. 76-80.
doi:10.1097/01.ASN.0000070145.00225.EC
[28] Grupo de Economía de la Salud, “Evaluación Económica
en Salud: Tópicos Teóricos y Aplicaciones en Colombia,”
Observatorio de la Seguridad Social, 2006.
[29] B. Manns, B. Hemmelgan, M. Tonelli, Au Flora, T. C.
Chiasson, J. Dong and S. Klarenbach, “Population Based
Screening for Chronic Kidney Disease: Cost Effective-
ness Study,” BMJ, Vol. 341, 2010, p. 5869.