The authors declare no conflict of interest
Cardiovascular disease is the leading cause of death in the world and approximately half of these are due to coronary heart disease. The most affected countries are those of medium and low income, as Colombia. There are interventions such as cardiac rehabilitation for these patients, however 7.5%-29.0% of coronary patients are referred and of these 40%-55% manage to finish and adhere to the treatment.
To determine the factors associated with non-adherence of patients with coronary heart disease to CRP (Cardiac rehabilitation program) in a Cali clinic between 2017 and 2019.
Retrospective study of cases and controls where a bivariate and multivariate analysis of 99 patients was performed. 46 cases were not adherent to the PRC and 56 controls adherent to it.
21% of the patients correspond to the male gender with an average age of 63.2 ±10.25 years; an association was evidenced between the anxiety variable evaluated with the HADS scale and non-adherence to the CRP with an OR of 0.26 (CI 95%: 0.07-0.84; p: 0.0114).
The presence of anxiety is an associated factor in patients with non-adherence to CRP. It is considered highly relevant to have an interdisciplinary team in the CRP. The correct application of questionnaires such as the HADS allow identifying these types of situations at a personal level helping to improve adherence to the CRP and thus achieve the multiple benefits evidenced in the programs.
Las enfermedades cardiovasculares son la principal causa de muerte en el mundo y aproximadamente la mitad de ellas se deben a enfermedad coronaria. Los países más afectados son los de ingresos medios y bajos como el nuestro. Existen intervenciones como la rehabilitación cardíaca para estos pacientes, sin embargo 7.5%-29.0% de los pacientes coronarios son referidos y de estos el 40%-55% logran finalizar y adherirse al tratamiento.
Determinar los factores asociados a la no adherencia de pacientes con enfermedad coronaria al PRC (Programa de rehabilitación cardiaca) en una clínica de Cali entre 2017 y 2019.
Estudio retrospectivo de casos y controles donde se realizó un análisis bivariado y multivariado de 99 pacientes, 46 casos no se adhirieron al PRC y 56 si lo hicieron.
el 21% de los pacientes correspondieron al sexo masculino con una edad promedio de 63.2 ±10.25 años; se evidenció asociación entre la variable ansiedad evaluada con la escala HADS y la no adherencia a la PCR con una OR de 0.26 (IC 95%: 0.07-0.84; p: 0.0114).
La presencia de ansiedad es un factor asociado en pacientes con no adherencia al PRC y por eso se considera de gran relevancia que el programa cuente con un equipo interdisciplinario. La correcta aplicación de cuestionarios como el HADS permiten identificar este tipo de situaciones a nivel personal ayudando a mejorar esta adherencia y así lograr los beneficios propuestos en el programa.
Objective:
To determine the factors associated with non-adherence of patients with coronary heart disease to the CRP (Cardiac Rehabilitation Program) in a clinic in Cali between 2017 and 2019.
Study design:
Retrospective case-control study
Source of information:
Records of the cardiac rehabilitation program of a private clinic in the city of Cali, Colombia, in the period between 2017-2019.
Records of 99 patients
Principle findings
There was evidence of an association between the anxiety variable assessed with the HADS scale and non-adherence to CRP with an OR of 0.26 (95% CI: 0.07-0.84; p: 0.0114).
Cardiovascular diseases have been, and are expected to continue to be, the leading cause of death in the world. Each year, they cause 17.7 million deaths worldwide and are projected to increase, reaching 75% by the year 2030. Thus, about 23.6 million people will die from this cause
At the national level, deaths due to cardiovascular diseases are considered a public health problem. They correspond to 28.7% of total registered deaths, of which 74% were attributed to coronary heart disease and cerebrovascular disease. Likewise, between 2010 and 2016, there was an increase in deaths due to cardiovascular causes (7,038 deaths); in addition, during this period, an annual economic expenditure between 4,277 and 4,846 dollars per patient over 20 years of age with coronary heart disease was generated. This is a very high expense compared to the 602 annual healthcare dollars generated by a patient over 20 years old without cardiovascular disease. The increase in life expectancy is represented by a change in the population pyramid, leading secondarily to an increase in mortality from these diseases
Colombia is facing a population change that impacts its demographic and epidemiological transition. In 2005, 63% of the population was between 15 and 64 years old, while in 2018, this population reached 68.2%, and the group aged ≥65 years went from 6.3% to 9.1% of the total population, which increased the prevalence of chronic noncommunicable diseases
In this sense, cardiac rehabilitation programs are part of the care pathway for patients with coronary heart disease from the moment the individual presents the clinical event. Thus, cardiac rehabilitation programs improve patients' quality of life and functional status and reduce long-term mortality by 20-30%
About the above, adherence to treatment is defined as "the capacity and/or attitude that each person has to modify lifestyles focused on good health, which is related to the indications given by the therapist and/or physician in charge of the treatment, in order to generate a desired preventive or therapeutic result"
Socioeconomic factors, including gender, age, socioeconomic stratum, unemployment, culture, distance from the treatment centre, and educational level. Factors related to the health care system or equipment that refers to the health care system. Disease-related factors such as the severity of symptoms, degree of disability, the severity of disease, and comorbidities. Treatment-related factors such as duration of treatment and previous treatment failures. Patient-related factors such as psychosocial stress, forgetfulness, anguish, low motivation, lack of knowledge of the treatment as well as the patient's perception of its effect
These factors associated with adherence to the cardiac rehabilitation program have also been grouped into intrapersonal, clinical, interpersonal, logistical, cardiac rehabilitation program, and health system factors
This research aimed to determine the factors associated with nonadherence to cardiac rehabilitation in patients with coronary artery disease in a clinic in Cali, Colombia, between 2017 and 2019.
A retrospective case-control study was conducted using the records of patients with coronary heart disease who were part of the cardiac rehabilitation program of a private clinic in the city of Cali, Colombia, between 2017 and 2019. The variables taken from these records were adherence, gender, age over 60 years, origin, schooling, employment status, and marital status. Socioeconomic stratum, admission diagnosis, type of acute myocardial infarction (AMI) and/or angina, number of coronary vessels involved, type of coronary intervention and number of vessels intervened, myocardial ejection fraction after the procedure, body mass index at the beginning of the program, abdominal perimeter, qualitative result HADS scale, qualitative result HRQOL questionnaire, qualitative result Patient Health Questionnaire (PHQ-9) and smoking.
The case was defined as a patient with coronary artery disease linked to the cardiac rehabilitation program of the Cali clinic classified as nonadherent (not completing 80% or more of the 36 cardiac rehabilitation sessions). The control was the registry of patients with coronary artery disease who are linked to this program and classified as adherent (completing 80% or more of the 36 cardiac rehabilitation sessions). Inclusion criteria considered records of patients with disease linked for the first time in the PRC of the clinic in Cali.
Meanwhile, records of patients with coronary artery disease older than 79 years, diagnosed with mental illness, and under 18 years of age were excluded. The total population that met the selection criteria was included.
Anxiety and depression were assessed with the Hospital Anxiety and Depression Scale (HADS) questionnaire, applied in a targeted manner on admission to the cardiac rehabilitation program
In turn, the Patient Health Questionnaire (PHQ-9) is a diagnostic tool for mental health disorders used by health professionals and is self-administered by the patient. It consists of 10 questions, each with 4 response options, inquiring about the person's motivation to do some things in life in the two weeks prior to the application of the survey. Each question has a severity index corresponding to: 0= "never", 1= "some days", 2= "more than half of the days" and 3= "almost every day". At the end, there is a question about the difficulty these nine items generated in daily activities (in case you checked yes to any of them).
The original MacNewQLMI questionnaire is a specific questionnaire to assess the quality of life of patients who have had a myocardial infarction. It was initially developed at MacMaster University in Ontario (Canada) to identify the main problems that appear after infarction. It has three dimensions: emotional, physical, and social
The data belonged to a secondary source, taken from a database (DB) of patients with coronary artery disease attending a cardiac rehabilitation program in Cali. They were elaborated on in the Excel program. The total patient data included was 99 (46 cases and 53 controls). It was decided to include all the data and not to perform a sample calculation with the database in order not to reduce the number of these further; however, it should be noted that these 99 patients’ data do correspond to a sample over time because only those who were active in the program between 2017 and 2019 were included. The identity of the patients is coded for confidentiality assurance reasons.
In addition, to ensure the quality of the information, a random sampling of 10% of the data and verification of duplicate data was performed. In this way, the variables included in the statistical analysis were selected, and the quality of their coding was verified. The data were exported to the Stata 14.0® program (Stata Corp, 2014, College Station, TX, USA). Likewise, an exploratory data analysis was performed to evaluate the distribution of the variables and univariate analysis. Qualitative variables were presented with frequencies and percentages; quantitative variables, according to the result of the Shapiro Wilk and Kolmogorov Smirnov tests, and their normality behaviour were represented with means and standard deviation or median and their interquartile range.
The evaluation of the association between each of the exposure variables and the outcome variable was performed based on a bivariate analysis using the Chi test2 or Fisher's exact test for values in the boxes below 5. The OR and its 95% confidence intervals were used to measure association. The independent qualitative variables were contrasted with the dependent variable using contingency tables. In this sense, the means of the quantitative variables were compared using Student's t-tests or the Mann-Whitney U test. With this, a multiple analysis was performed, and a saturated model was constructed where the independent variables that presented in their bivariate analysis a coefficient of p <0.25 and the clinically relevant ones were included in the model, considering the control of confounding and the identification of effect modification. Subsequently, the exclusion of each of them was evaluated based on their significance and the model's plausibility using the Backward
According to Resolution 8430 of 1993 of the Ministry of Health of Colombia in its article 11, this research is classified at a no-risk level. It was approved by the ethics committee of the Universidad Libre sectional Cali in approval minutes of April 16, 2020.
Ninety-nine records that met the inclusion and exclusion criteria were analysed; 46 (47%) corresponded to cases and 53 (53%) to controls. Furthermore, for this sample of 99 patients with 47% adherence, the power of this study is 37.6%.
The mean age was 63.2 ±10.25 years, and 54% presented an age over 60 years. Seventy-two percent of the patients were male, which prevailed in both cases and controls. Eighty-seven percent of the patients were from Cali. Likewise, 77% had incomplete secondary schooling up to postgraduate level, and 90% were not working then. Regarding socioeconomic strata, the highest percentage belonged to the low stratum, and the lowest percentage to the medium/high stratum (
Sociodemographic Factors
Frequency
Percentage
Genre
Female
28
28
Male
71
72
Age Years
Years completed
63.2
±10.25
Age >60
No
46
46
Yes
53
54
Source
Cali
83
84
Outside Cali
12
12
No Data
4
4
Schooling
Up to elementary school
22
22
From incomplete high school to postgraduate studies
75
76
No Data
2
2
Laboral
Does not work
10
10
Labora
89
90
Marital Status
No Stable Union
75
76
Stable Union
22
22
No Data
2
2
Socioeconomic stratum
Medium/high
31
31
Low
61
62
No Data
7
7
Of the total number of patients, 61 had 1-2 vessel involvement; 95 of them had coronary lesions, 29 underwent total lesion intervention, 46 underwent percutaneous transluminal coronary angioplasty (PTCA) with medicated STENT implantation, followed by 21 who underwent surgical myocardial revascularization (SMR). Eighty-three percent of the population had left ventricular ejection fraction (LVEF) greater than or equal to 40%. According to body mass index (BMI), 63% of the population is obese, with a median of 26 kg/m2 RI (24-28), with a mean abdominal circumference of 95 ±8.66. Only 4% of the patients presented active smoking at the time of being in the CRP (
Clinical Factors
Description
Frequency
Percentage
Non-adherence
No
53
53
Yes
46
47
Admission diagnosis
IAM Con St
35
35
Stable Angina
7
7
Unstable Angina
1
1
Sudden Death
0
0
IAM Sin St
55
56
Not available
1
1
Coronary vessel involvement
0
4
4
1-2
61
62
≥3
30
30
Not available
4
4
Intervention of the totality of the involved vessels
No
66
67
Yes
33
33
Medical surgical procedure
Catheterization
7
7
PTCA + Conventional stent
7
7
PTCA + Medicated stent
46
46
Myocardial revascularization
21
21
Medical management
10
10
PTCA+Medicated stent+Rvm
3
3
Not available
5
5
Catheterization
No
92
93
Yes
7
7
PTCA Conventional stent
No
92
93
Yes
7
7
PTCA Medicated stent
No
53
54
Yes
46
46
RVM
No
78
79
Yes
21
21
Medical management
No
89
90
Yes
10
10
PTCA Medicated stent RVM
No
96
97
Yes
3
3
PTCA medicated stent and RVM
No
98
99
Yes
1
1
LVEF ≤40
No
64
65
Yes
13
13
Not available
22
22
Smoking
No
93
94
Yes
4
4
No data
2
2
Quantitative LVEF
%
53
RI (25-58)
Quantitative BMI
K/cm2
26
RI (24-28)
Quantitative abdominal circumference
cm
95.4
SD ±8.66
Questionnaires
Qualitative BMI
Normal
36
36
Obese
62
63
No Data
1
1
HADS Depression
No
90
91
Yes
9
9
HADS Anxiety
No
77
78
Yes
22
22
CVRS
Good
74
75
Mala
25
25
PHQ Interpretation
Minimum
62
63
Slight
22
22
Moderate
11
11
Moderately Severe
1
1
Serious
3
3
PHQ Grouped
No intervention for depression
96
97
Intervention is required for depression
3
3
Of the total number of participants, nine patients were at risk of depression, and 22 were at risk of Anxiety, according to the Hospital Anxiety and Depression Scale (HADS). According to the Patient Health Questionnaire (PHQ-9), 96 patients did not require intervention for depression, and the three that did require it were found to be of severe severity. Health-related quality of life (HRQoL), according to the Myocardial Infarction Questionnaire (QLMI), was good for 75% of the population (
The bivariate logistic analysis showed an association between Anxiety assessed with HADS where the OR was 0.26 (95% CI: 0.086-0.77), and this association was statistically significant (p= 0.0114). In the case of the treatment-related variables, none showed an association with nonadherence in this analysis (
Feature
Description
N
Case
Control
OR
IC
p-value
Genre
Male
71
33
38
1.00
0.38-2.66
0.996
Female
28
13
15
Age >60
Yes
53
21
32
0.55
0.25-1.23
0.144
No
46
25
21
Source
Cali
83
38
45
1.66
0.41-7.16
0.4157
Outside Cali
12
7
5
Laboral
Does not work
10
3*
7
2.18
0.46-3.80
0.2709
Labora
89
43
46
Schooling
Up to elementary school
22
10
12
1.05
0.36-3.08
0.9201
From incomplete high school to postgraduate studies
75
35
40
Marital status
Stable Union
75
35
40
0.95
0.32-2.74
0.9201
No Stable Union
22
10
12
Socioeconomic stratum
Medium/high
31
15
16
0.85
0.36-2.02
0.710
Low
61
27
34
IAM
With st elevation
35
16
19
0.94
0.40-2.20
0.885
No st elevation
55
30
30
Non-ST-elevation AMI
Yes
55
26
29
1.08
0.48-2.38
0.857
No
44
20
24
Stable Angina
Yes
7
3*
4*
0.85
0.18-4.03
0.843
No
92
43
49
Coronary vessel involvement
0 vessels
4
1*
3*
1
1 or 2 vessels
61
25
36
2.08
0.20-21.2
0.535
≥3 vessels
30
17
13
3.92
0.36-42.2
0.259
Intervention of the totality of the involved vessels
Yes
33
18
15
1.63
0.70-3.78
0.256
No
66
28
38
Cardiac catheterization
Yes
7
3*
4*
0.85
0.18-4.03
0.843
No
92
43
49
PTCA Conventional stent
Yes
7
2*
5
0.44
0.08-2.36
0.336
No
92
44
48
PTCA Medicated stent
Yes
46
20
26
0.80
0.36-1.77
0.579
No
53
26
27
RVM
Yes
21
10
11
1.06
0.36-3.11
0.905
No
78
36
42
Medical Management
Yes
10
6
4*
1.84
0.40-9.43
0.365
No
89
40
49
PTCA medicated stent and RVM
Yes
3
2*
1*
2.36
0.21-26.9
0.488
No
96
44
52
LVEF ≤40
Yes
13
8
5
2.34
0.69-7.94
0.174
No
64
26
38
Qualitative BMI
Yes
62
30
32
1.17
0.51-2.67
0.706
No
36
16
20
HADS Depression
Yes
9
4*
5
0.91
0.23-3.63
0.899
No
90
42
48
HADS Anxiety
Yes
22
5
17
0.26
0.08-0.77
0.0114
No
77
41
36
CVRS
Yes
25
8
17
0.45
0.17-1.16
0.0980
No
74
38
36
PHQ Grouped
Yes
3
1*
2*
0.57
0.05-6.46
0.6470
No
96
45
51
Smoking
Yes
4
2*
2*
1.16
0.16-8.6
0.8830
No
93
43
50
Feature
Description
Central Trend
Dispersion
OR
IC
p-value
Age years
Years completed
63.2
SD ±10.25
0.98
0.94-1.02
0.337
Quantitative BMI
K/cm2
26
RI (24-28)
1.10
0.99-1.23
0.086
Quantitative abdominal circumference
Cm
95.4
SD ±8.66
1.05
1.00-1.10
0.063
Quantitative LVEF
Percentage
53
RI (25-58)
1.00
0.98-1.02
0.716
Multiple logistic regression was performed where the variables with a significance <0.25 in the bivariate analysis and with clinical relevance were included in the model. Additionally, the anxiety variable was obtained as the only variable associated with nonadherence to the program, using the Backward strategy based on the individual significance of each variable and evaluating the likelihood of the model, which was evaluated with the HADS questionnaire and with an adjusted OR of 0.26 (95% CI: 0.086-0.77) and had an initial model likelihood of 44.5 and a final model likelihood of 65.00. The rest of the variables entered the multivariate are considered confounders, making the association found in the bivariate analysis no longer significant. Possibly explained by the small sample size (99 patients) (
n
Cases
Controls
OR adjusted
Adjusted OR CI
n= (%)
n= (%)
Age >60
No
46
25
21
0.55
0.23-1.32
0.1429
0.51
0.18-1.42
YES
53
21
32
Involvement of 1-2 coronary vessels
No
38
21
17
0.56
0.23-1.38
0.1659
0.67
0.04-10.28
YES
61
25
36
Involvement ≥3 coronary vessels
No
69
29
40
1.80
0.070-4.71
0.1796
2.56
0.87-7.52
YES
30
17
13
LVEF ≥40%.
No
64
26
38
2.34
0.59-10.05
0.1662
2.61
0.68-10.04
YES
13
8
5
HADS Anxiety
No
77
41
36
0.26
0.07-0.84
0.0114
0.30
0.08-1.13
YES
22
5
17
Quantitative BMI
Kg/cm2
-
-
-
1.1
0.99-1.23
0.086
1.04
0.85-1.27
Quantitative Abdominal Perimeter
cm
-
-
-
1.05
1.00-1.10
0.063
1.05
0.96-1.15
n
Cases
Controls
n=(%)
n=(%)
HADS
No
77
41
36
0.0114
0.26
0.07-0.83
Anxiety
Yes
22
5
17
This research evaluated the factors associated with nonadherence to CRP in patients with coronary artery disease in a clinic in Cali between 2017 and 2019. The analysis of the 99 records, 46 cases, and 53 controls show an association in the bivariate analysis with demographic and economic variables such as age >60 years, with mean age 63.2 ±10.25 years and crude OR 0.55. In a systematic review of cohorts that included 43 studies in 10 countries and 63,425 patients, authors reviewed prospective cohorts, and the factors associated with non-participation and/or abandonment of cardiac rehabilitation programs. In eight of these studies, advanced age was found to be a factor associated with nonadherence to the programs (OR range: 1.01-4.76); however, in another four studies, young age was reported as a factor associated with nonadherence (OR range: 1.04- 1.72)
In a cohort study in 38 facilities with 4,413 patients with a mean age of 65 ±12 years, the association between organizational and patient factors and adherence to the cardiac rehabilitation program was demonstrated. Likewise, patients aged ≥65 years have the highest probability of adherence (OR: 1.56, 95% CI: 1.24- 1.97)
In the case of the present study, the multivariate analysis did not reveal an association between age and nonadherence, which could be influenced by the small sample used for the analysis. In the case of those >6 years of age, some are employed or are pensioners, others have time available to adhere to the program but are economically dependent on their family and/or caregiver. Those of younger age who are in their productive years also present a significant nonadherence.
On the other hand, in the case of disease-related variables initially, the association was found between nonadherence and involvement of 1-2 vessels (OR: 0.56), involvement of 3 or more vessels (OR: 1.80) and LVEF ≤40% (OR: 2.34), an association that disappears in the multivariate. There is a relationship between variables related to the disease, treatment, and nonadherence, finding higher attrition in patients with a history of cardiovascular disease and greater participation of patients with surgical myocardial revascularization (MCR) with an OR range of 0.02-0.49
The differs from the prospective study, which evaluated program dropout in patients with heart disease, in which no statistically significant difference was found in the group that performed more than 23 sessions and the group that performed ≤23 sessions (mean: 59 ±16 and 58 ±16, respectively)
Regarding the patient-related variables, the abdominal perimeter and body mass index (BMI) measurement correlated with nonadherence (OR 1.05 and 1.10, respectively). Using BMI, a significant difference is evident in the mean for the group that performed more than 23 sessions of 27±4 and for the group that performed ≤23 sessions of 27±5 (p >0.05)
The increased symptom associated with initiating cardiac rehabilitation increases the risk of dropout
Thus, after multivariate analysis, the anxiety variable evaluated with the HADS questionnaire showed a significant association with nonadherence to the program (adjusted OR: 0.26). Several authors report similar results in their systematic review, showing that only one study has an association between Anxiety and participation in the program
Thus, it is shown that the result of the HADS questionnaire positive for Anxiety has a more significant association with nonadherence to the program. Cardiac rehabilitation programs in which quality of life and emotional state questionnaires are applied help to diagnose emotional states that require specialized intervention, such as psychology and/or psychiatry, which makes more comprehensive care possible. In recent years, little research has been done on the role of Anxiety in patients in these programs. However, it is known that the prevalence can be as high as 50% in patients suffering from cardiovascular complications
The present study showed that Anxiety was a protective factor for nonadherence to the program. Thus, patients identified as anxious tend to drop out of the program less, so there are two possible explanations for this; the first is that the anxious patient receives treatment for their state of Anxiety, helping them to control and overcome it, for this must be taken into account that the program has an interdisciplinary team that includes a psychologist located in the same IPS where the cardiac rehabilitation therapy is performed, the comprehensive management of the patient makes the patient have clinical results motivating them to complete the treatment. The second possibility is that patients with Anxiety caused by their state of health are more adherent to the program since their Anxiety and concern drive them to seek, accept, and comply with the indications provided by the health personnel. Therefore, it should be emphasized that an acute coronary event is a situation that can be very significant in a person's state of mind.
The results found in this study show that patients who presented Anxiety have a higher risk of nonadherence to CRP. The support of an interdisciplinary team in developing cardiac rehabilitation programs and the knowledge and application of questionnaires such as the HADS to personally identify this type of situation is considered of great relevance. All this allows us to have elements to improve adherence to the program and provide the benefits of the program.
It is recommended that future research be carried out to include data from a more significant number of patients, where different institutions and patients from the subsidized and contributory health regimes can be included. In addition, evaluating a more significant number of characteristics related to the treatment in such a way that the program's quality can be improved, focusing on those variables that show statistically significant associations, is necessary. It is suggested that the statistical analysis be repeated with at least 152 data and 152 controls for a total of at least 304 patient data, and, in this way, the power of the study should be increased to 80% with a nonadherence rate of 40%.
Since this is a secondary source of information, errors could be found in the tabulation of the data, for which data quality control and coding were carried out, as well as verification of duplicate data. Another information bias can occur when creating dichotomous variables to make the calculations of the bivariate and multivariate analysis; this was controlled because the calculations were repeated 3 times by different operators. Confounding bias was controlled using logistic regression to define the variables associated with nonadherence. Selection bias was controlled by including both cases and controlling only the records of patients entering the program for the first time.
Considering that the study was only carried out in a clinic that serves the contributory system population, it is impossible to extrapolate the data to the general population. The amount of patient data meeting the inclusion criteria is small despite collecting data over two years. Therefore, a power of 37.6% was obtained for this study. In addition, the lack of information meant that it was impossible to include variables referring to the pathological history of the patients.
The source of information generated so that the clinic can make decisions regarding strategies to improve the adherence of patients who present factors associated with nonadherence to the program can be the basis for generating new studies to evaluate strategies to improve adherence. The data quality is excellent, and it is evident that no typing errors were found during quality control.
To the clinic for providing us with the data of the patients of their PRC, to the patients who contributed their information for the analysis of this research.
Muñoz EBE, López LMF, Betancourt PJ, Avila VJC, Alzate SRA. Factors associated with non-adherence to cardiac rehabilitation in patients with coronary disease iJEPH. 2021; 4(2): e-7071. Doi: 10.18041/2665-427X/ijeph.2.7071.