Introduction
Heart disease is a significant reason for disability and a leading cause for death in the United States by the statistics of the Centers for Disease Control and Prevention. Coronary heart disease is the commonest of the heart diseases (CHD) and often presents as a heart attack. About 785,000American were predicted to have a new coronary attack in 2009. About every 25 seconds, an American will have a coronary event and about every minute, an American will die from it. Many Americans suffer from a heart attack and they need to recover from it. Therefore, the purpose of this literature review is to identify the benefits of cardiac rehabilitation programs for CHD patients in terms of their psychosocial and physiological recovery.
An advanced nurse practitioner (ANP) could recommend Cardiac Rehabilitation (CR) to CHD patients who go through a cardiac event in order to support their recuperation from the illness while providing the evidence why the CR is beneficial. Evidence-based nursing, based on the current best evidence available from researches, is the best quality of nursing care for the cardiac patient (Polit &Beck, Nursing Research, 2009). Healthcare professionals including nurses should acknowledge evidence of Cardiac Rehabilitation benefits and provide their patients with high quality of care. The American Heart Association (AHA) states that the core components of Cardiac Rehabilitation include baseline patient assessment, nutritional counseling, risk factor management (lipids, blood pressure, weight, diabetes mellitus, and smoking), psychosocial interventions, and physical activity counseling and exercise training.
Review of the Literature
Literature published between 2003 and 2009 was searched through multiple electronic databases; PubMed, CINAHL and Ovid. The search was conducted using keywords such as benefits of cardiac rehabilitation, Coronary Heart Disease, quality of life, risk factors modification and depression. The searches revealed some interesting facts.
Kardis et al (2005) conducted a study to compare quality of life assessments after a 3-month Phase II cardiac rehabilitation program following a cardiac intervention. Fast et al (2009) studied the effects of attending Phase II cardiac rehabilitation on patient versus spouse (proxy) quality-of-life perceptions. Both of them investigated the benefits of cardiac rehabilitation on quality of life, but Fast et al (2009) compared the perception of quality of life between patients and their spouse in addition. While Snow et al (2005) conducted the research of effects of cardiac rehabilitation on lipids in Coronary Artery Disease (CAD), Adams et al. (2007) focused on a comparison study of cardiovascular risk factor outcome including lipid levels between exercise-based cardiac rehabilitation, traditional care, and an educational workshop among CAD patients. Milani et al (2007) highlighted depression and its associated mortality among coronary artery patients and used cardiac rehabilitation programs as an intervention. In all five research articles, the authors described their CR and the programs have the same core components of CR since the programs follow the AHA recommendations.
Research Design
Kardis et al (2005) used one group pretest-posttest design which is a weaker quasi-experimental or pre-experimental design because of an absence of a control group. Fast et al (2009) used descriptive, prospective pretest-post test design for their study. Since descriptive, prospective pretest-post test design and one group pretest-posttest design are not experimental, their study findings were not conclusive. Therefore, the cause- and -effect inferences were less convincing (Polit & Beck, Nursing Research, 2009).
A retrospective cohort comparison design and a non-equivalent, three-group, and observational design were utilized for the study of Snow et al (2005) and the study of Adams et al (2007) respectively. In a retrospective design, a researcher tries to link a phenomena existing in present to phenomena that occurred in the past before a study was initiated (Polit &Beck, Nursing Research, 2009). The observational design and the retrospective cohort comparison were not experimental studies; the research design was aimed at inferring the causality only (Polit & Beck, Nursing Research, 2009).
Milani et al (2007) utilized a relatively strong research design compared with other studies described above. They used a controlled trial to explore the effect of cardiac rehabilitation on a psychiatric illness of depression along with the related mortality. Although a true experimental design, which involves a manipulation (intervention), control and randomization, yields the highest-quality evidence regarding intervention effects, this design is unsuitable for some studies (Polit &Beck, Nursing Research, 2009). Milani et al (2007) mentioned that they did not conduct a true experimental study because it was unethical to assign a randomized control group who were not receiving cardiac rehabilitation when this therapy is a clinically proven method for reducing morbidity and mortality in CAD patients.
Sampling and Setting
Sampling is the process of selecting a part of the population to represent the whole population so that inferences about the population can be made; a sample is a subset of population component (Polit & Beck, Nursing Research, 2009). A critical consideration in assessing a sample on a quantitative study is its representation of similar characteristics of the population (Polit & Beck, Nursing Research, 2009). In order to eliminate a sampling bias which is the systematic over-representation or under-representation of some portion of the population, a research needs to use random sampling which is probability sampling (Polit & Beck, Nursing Research, 2009). Inclusion and exclusion criteria are used to further define study population in order to enhance study findings, but those criteria limit external validity of the findings. External validity concerns the generalizability of causal inferences to other populations that have not been studied (Polit & Beck, Nursing Research, 2009).
Kardis et al (2005) chose a non probability and convenience sampling of 302 patients who completed cardiac rehabilitation program in the Inova Heart Institute in Northern Virginia in their study. Fast et al (2009) recruited 42 pairs of non probability samples of patients and their spouses who were referred to Midwestern phase II cardiac rehabilitation. In the research of Fast et al (2009), inclusion criteria were patients older than 21 years, who had been hospitalized with a cardiac event and received a physician referral to phase II CR. In addition, they had an identified spouse older than 21 years, and the ability to read English. Both patient and spouse had to be willing to participate in the study. Both had a possible sampling bias because non probability samples were used. Furthermore, “the convenience sampling is the weakest form of sampling” (Polit & Beck, Nursing Research, 2009).
Snow et al (2005) selected non probability sample of 766 patients in a large urban comprehensive Cardiac Rehabilitation (CR) program from January 2000 to August 2003 and the sample was divided into 2 groups. Patients who had no changes in the type, number, or dosage of lipid medication at enrollment or during CR were included in Group1 (613) and patients who did have changes in the lipid medication regimen were included in Group2 (153). Adams et al (2007) also chose non probability sample of 217 patients from Baylor University Medical Center. The patients were split into a traditional care group (114), CR group (78) and Leap for Life group (25) according to financial and geographic constraints, physician preferences, and participant choices; thus it is not a random assignment. The researchers described the inclusion criteria as patients who were able to read and understand English and functioning independently. The exclusion criteria were that patients who had undergone organ transplant surgery recently or were on a transplant waiting list or were taking any immunosuppressive regimen.
Milani et al (2007) conducted an experimental study and their study had a control group of 179 patients who did not complete CR. They entered CR but dropped out within 2 weeks of entry, as a result all control group patients received less than 5 sessions of rehabilitation. The experimental group had 522 consecutive patients who were referred to, attended and completed phase II CR. They were all enrolled in the CR program for 2-6 weeks following a coronary event including acute MI, coronary bypass surgery and percutaneous coronary intervention. Although their study lacks random assignment of their sample into the experimental and control group, it has the strongest design to infer the causality of the study.
Data Collection
Both Kardis et al (2005) and Fast et al (2009) investigated benefits of cardiac rehabilitation on quality of life, but they used a different instrument to measure their study participants’ quality of life. Dartmouth COOP Charts and SF-36 were utilized for the research of Kardis et al (2005) and Fast et al (2009) respectively. Both instruments claim to measure patients’ Quality of Life (QoL). Physical, Emotional, Daily Activities, Social Activities, Pain and Overall Health were measured by the Dartmouth COOP Charts to evaluate the participants’ health status. In this instrument, lower scores indicate increased QoL, thus the scores and QoL are negatively corelated. The participants were assessed at their entrance to the cardiac rehabilitation (baseline) and again at 3 months via the Dartmouth COOP Charts in the study of Kardis et al (2005).
The SF-36 is a generic instrument that measures Health-Related Quality of Life (HRQL) by addressing 8 subscales, which are frequently used in health survey and most influenced by disease and treatment. The 8 subscales evaluate patient ability to engage in robust activities, perform activities of daily living, and involve in family, social, and occupational role. These 8 subscales are summated into 2 categories: component summaries with physical and mental components. The scores of SF-36 are standardized on a 1 to 100 scale and higher scores on the 8 dimensions; 2 summary measures mean better physical and mental functioning. Reliability coefficients of 8 subcategories have exceeded 0.80 and 0.90 for the physical and mental summary measures.
In the study of Fast et al (2009), the purpose of their study was to evaluate the level of agreement-between-patient-and-spouse score on the Short Form-36 version 2(SF-36) measured before and after patient participation in phase II cardiac rehabilitation (CR), thus the patient rated his or her own Health Related Quality of Life (HRQL) on the SF-36 and the spouse rated the patient’s HRQL on the SF-36 which often stated as a “proxy” rating. Patients and their spouses completed the first SF-36 at the orientation visit (T1). The second survey (T2) was administered by trained CR staff and it was completed by both the patient and spouse after 6 weeks of phase II CR.
Snow et al (2005) and Adams et al (2007) studied outcome of cardiovascular risk factors focusing on lipid levels following cardiac rehabilitation programs. The Healthy Promoting Lifestyle Profile II (HPLP II) and the Katz index were used in the study Snow et al (2005) and Adams et al (2007) respectively as their study instrument. Additionally, the 21-item self-reported Beck Depression Inventory and the 53-item Brief Symptom Inventory were utilized to evaluate symptom of depression in the study of Adams et al (2007). The HPLP II is a 52-items scale consisting of 4-point responses which evaluate health activity in the following 6 dimensions such as nutrition, exercise, health responsibility, interpersonal support, spiritual growth, and stress management. The Katz index was used to evaluate functional ability in bathing, continence, dressing, feeding, toileting and transfers. Each question can score from 0 to 3 and the highest possible score indicating independence is 18.
In the research of Snow et al (2005), patients were regarded to be on a lipid-altering agent (LAA) if they were on any of the following categories of medication at enrollment: statins, fibrates, niacin, bile acid sequestrants, or hormone replacement therapy. The lipid treatment goal was set according to the National Cholesterol Education Adult Treatment Panel (NCEP ATP) III guidelines which are Low Density Lipoprotein Cholesterol (LDL-C) <100 mg/dL, High Density Lipoprotein Cholesterol (HDL-C) >40mg/dL for males and > 50mg/dL for females, Triglyceride (TG) <150mg/dL, and non-HDL-C <130mg/dL. All information that used statistical analyses was captured in an electrical medical record and it has been evaluated for reliability of demographic, co-morbid, and laboratory data. Snow et al (2005) cross-validated all medication and dosage for their study. On the registration to the large urban comprehensive CR program from 2000 to 2003, all study participants were being treated with LAA and pre-registration and postregistration lipids were obtained. Demographic, HPLP II, co-morbidity, and laboratory findings were routinely gathered at patient enrollment and at completing of CR. In the study of Adams et al (2007), a non-equivalent, three-group designs was chosen to evaluate the differences in physiological and psychological variables following cardiac events (coronary artery bypass graft surgery, MI, or transcatheter procedures) by applying three types of rehabilitation interventions.
Baseline data were collected 6 to 8 weeks after the cardiac event and follow-up data were obtained 3, 6, 9, and 12 months after the initial evaluation. During each of five sessions, the participants’ weight, height, body mass index, blood pressure, heart rate, and plasma lipid levels such as fasting total cholesterol, triglyceride, high-density lipoprotein and low-density lipoprotein were gathered. At each examination, the participants’ scores on the Beck Depression Inventory, the Brief Symptom Inventory, and the Katz index were collected and their medication use was also assessed. Milani et al (2007) evaluated 522 patients in the treatment group who completed phase II CR and 179 patients in the control group who enrolled in the CR but dropped out within 2 weeks of the enrollment between January 2000 and July 2005. All participants completed the Keller Symptom Questionnaire and the medical Outcomes Short Form 36 (SF-36) survey before the CR and the treatment group completed those questionnaires after the CR. In order to evaluate behavioral characteristics, including symptoms of depression, anxiety, somatization, and hostility, the Keller Symptom Questionnaire was used. When the depression score exceeded 6(>7for hostility and anxiety, >8 for somatization), depression symptoms were regarded to be significant. Even though depression symptoms were recognized in 139 patients on enrollment into the CR (26% controls and 17 % treatment; p=0.005), all of the study participants were not taking antidepressives. The Medical Outcomes SF-36 survey was utilized to evaluate QoL, with a high score demonstrating a more favorable QoL.
Furthermore, the participants’ survival status was recorded on January 1st , 2006 after a mean follow-up of 1296±551 days from the National Death Index. The researches assessed the participants’ height, weight, body mass index, age, sex, fasting blood lipids, high sensitivity C-reactive protein(hs-CRP), and peak oxygen consumption(peak VO2) at baseline and again 1 week after completing the CR.
Milani et al (2007) stated their study protocol which is related to treatment fidelity (Polit & Beck, Nursing Research, 2009).The treatment fidelity indicates the extent to which the administration of an intervention (treatment) is faithful to its plan; thus study participants’ treatment adherence can be enhanced and contamination of treatments can be avoided by using a study protocol (Polit & Beck, Nursing Research, 2009). Milani et al (2007) make an effort to strengthen their study findings by utilizing their study protocol.
Snow et al (2005), Adams et al (2007) and Milani et al performed the group comparison in their study. Snow et al (2005) compare their two cohorts based on participants’ demographic, co-morbidity (diabetes and body mass index) and CR indications; HPLP II scores were compared between two groups. Adams et al (2007) compared participants’ demographic variables of income, age, physical independence, exercise habits and social isolation scores. Katz index scores were analyzed in order to compare the participants’ physical independence capability. Milani et al (2007) described the baseline characteristic differences of treatment and control groups. The scores of Keller Symptom Questionnaire of both groups were compared in their study.
Key Findings
In the study of Kardis et al (2005), the authors tested categorical variables by utilizing the Cochran Mantel Haenszel Score statistic in order to test the hypothesis that severity scores at baseline and 3 months were the same and continuous variables were tested by the paired t test. Continuous data was showed as mean ±SD and categorical data was displayed as frequency (n) and percentage. Differences were regarded statistically meaningful for p<0.05.
Their study findings were as followed. The largest decreases from baseline to 3 months were seen in the measurement of Fitness (-22.5%, p<0.0001) and Daily Activities (-22.4%, p<0.0001). No statistically significant changes from baseline to month 3 were measured in the categories of Changes in Health (-6.8%, p<0.082) and Feeling (-5%, p<0.119). The measurement of social support was worse at 3 months than at baseline (+9.7%, p<0.046). The subscale of Quality of Life scored to 2.06 at baseline and 1.90 at month 3, thus the score decreased by -0.16(p<0.0005). Overall, 8 of 9 measured dimensions in Dartmouth COOP Charts improved.
Furthermore, Kardis et al (2005) stratified QoL scores on 3 baseline clinical univariate risk factors such as gender, smoking and sedentary lifestyle. The authors defined smoking and sedentary lifestyle in their study. The study subjects consisted of 81 % of males with average age of 61.9±10.8 years and 19% females with age of average 1 year younger than their male counterpart. The 53.4 % and 48.3% of subjects described themselves as sedentary and overweight respectively and 28.6% of subjects described themselves as tobacco users. Gender differences from baseline to 3 months were minimal over most QoL dimensions, but Changes in Health and Pain. Compared with male participants, the female made the largest gains in Changes in Health (-22.7%vs-2.4%) and compared with female participants, the male made large gains in Pain (-15.9%vs-5.6%). While male subjects had more improvement on physical dimensions, female subjects in this study had a tendency to have increased QoL scores on psychosocial dimensions. Smokers reported notably better QoL changes on the category of Changes in Health (-12.8%vs-1.5%), Feelings (-13.0%vs-1.1%), and Overall QoL (-26.2%va-12.7%) compared with non-tobacco users at 3 months. Despite stratification by those above risk factors, improvements in QoL remained intact.
Fast et al (2009) made group comparisons between subjects and within subjects in order to evaluate the extent of agreement between CR patients and spouse rating of HRQL. A paired-samples t test was utilized to assess difference between the means of the 8 subscales and 2 summary measures of SF-36(physical component and mental component summary) for patients and spouses at T1 and T2. In order to compare patients at T1 verses T2 and spouses at T1 verses T2, a paired t test was conducted and significance was set at p<0.05.
The study findings show that statistical significant improvement from T1 to T2 across 7 of the 8 health measures including physical functioning, role-physical, bodily pain, vitality, social functioning, role-emotional(all p<0.001), and mental health(p<0.01) among patients(n=42) in terms of means of patient scores on each subscales ,and 2 summary measures of the SF-36. Statistically important differences were identified in both physical and mental component summary measures (p<0.001) in patients’ scores. Spouse (n=42) perception of patient HRQL revealed statistically significant gains across 7 of the 8 subscales including physical functioning, role-physical, bodily pain, vitality, social functioning, mental health(all p<0.001), and role-emotional(p<0.01) in terms of means of spouse scores on each subscales ,and 2 summary measures of the SF-36. Statistically significant improvements were showed in the physical component summary (p<0.001) and in the mental component summary (p<0.01) in spouse’s proxy rating.
Furthermore, differences between patient scores in contrast with spouse scores on subscales and 2 summary measures of the SF-36 were examined at T1 and T2. Results show that statistically significant findings between patient and spouse scores at T1 for 2 subscales including vitality and mental health (both p<0.01). Statistically notable differences between patient and spouse scores were detected in 3 of the 8 subscales including vitality (p<0.001), physical functioning, and mental health (both p<0.05) at T2. Although differences between patient and spouse scores in role-physical, bodily pain, general health, social functioning, or role emotional were not statistically meaningful at T2, there was a statistically significant difference between patient and spouse scores in the mental component summary (p<0.05). No statistically significant difference was found between patient and spouse scores in physical component summary at T2. Overall, the study results of Fast et al (2009) suggest that HRQL of patients and spouses’ perception of patient HRQL have improved significantly after the 6 weeks of phase II CR. However, there are significant differences between patient scores versus spouse scores at T 1 for 2 subscales which are vitality and mental health. At T2, patient scores versus spouse scores show statistically significant differences on subscales of vitality, physical functioning, and mental health. The perception of HRQL differs between the patient and spouse slightly in their levels of agreement but the improvement of patient HRQL appears to be similarly perceived by the patient and spouse.
Kardis et al (2005) and Fast et al (2009) conducted the research of benefits of cardiac rehabilitation on quality of life and their study findings show similar results that the cardiac rehabilitation helps patients’ QoL after a cardiac event. In addition, Fast et al (2009) investigated levels of agreement between patients and their spouses in terms of the perception of HRQL and they found that the perceptual differences between the patients and their spouses persist as the patient moves through the recovery. Therefore, Fast et al (2009) suggested that cardiac rehabilitation professionals should consider the spouse contribution to patient recovery because spouse attitude influences rehabilitation effectiveness of patient. In the study of Snow et al (2005), statistical analyses were performed on two groups. 613 participants were enrolled on a lipid-altering agent (LAA) with no medication adjustments (Group1) and 153 participants were enrolled on an LAA with medication adjustments (Group2). The percentage of patients who were directed towards achieving the lipid goals constituted the χ2 for dichotomous variables. The continuous variables referred to the absolute changes in the lipid levels and were indicated using the t test. Their study results were as follows. Group1 achieved significant improvement in all lipid fractions after completion of the CR program.
The greatest improvement in lipid showed in HDL-C levels with an absolute increase of 11.2 % of patients achieving goal (p=0.0001). 251 out of 613 in group 1 met HDL-C goal (>40mg/dl) at enrollment to the CR and 320 out of 613 in group 1 met HDL-C goal (>40mg/dl) at the end of CR (p=<0.0001). 68.5 % of patients in group 1 met LDL-C goal (<100 mg/dl) at entry to the CR and 74.9 % of patients in group 1 met LDL-C goal (<100 mg/dl) at the end of CR (p=0.01). Group 2 demonstrated significant percentage increases in patients at goal for LDL-C and Non-HDL-C. In group 2, the patents who met the Non-HDL-C goal (<130mg/dl) was 41.2% at pre-CR program and the Non-HDL-C goal revealed absolute increase of 22.2 % at post-CR (p<0.001). 63 out of 153 patients in group 2 met the Non-HDL-C goal (<130mg/dl) at the beginning of the CR and 97 out of 153 patients in group 2 met the Non-HDL-C goal at the end of the CR (p <0.001).The significant improvement in lipid showed in LDL-C levels with an absolute increase of 20.3% of patients achieving goal (p <0.0001). The 43% of participants in group 2 met the LDL-C goal (<100mg/dl) at the beginning of the CR and the 63.4 % of participants in group 2 achieved the LDL-C goal at the completion of CR. In group 2, all differences in lipid levels at the pre-post CR were meaningful statistically with the exception of HDL-C and TG.
Snow et al (2005) compared the groups regarding demographic, co-morbid information, and indications for CR and provided their statistical findings between the two groups. The researchers stated that no significant group differences were found based on their group analysis. Snow et al (2005) concluded that although changes in type or dosage of LAA during CR program result in a large improvement in lipid levels, subjects in CR without LAA adjustment also revealed significant benefits which are independent to pharmacologic changes in terms of improvement of their lipid levels.
In the study of Adams et al (2007), the researchers performed a comparison among three different intervention groups (traditional care group, CR group and Leap for Life group). The researchers did not find significant differences among the three groups in terms of their demographic variables of income, age, physical independence, exercise habits, and social isolation scores. Social isolation and the physical independence capabilities of participants were compared between the three groups. The comparison did not reveal significant group differences in terms of these variables. The psychological and physiological results and the results of three groups’ characteristics were summarized. An analysis of covariance was performed in order to compare the three non equivalent groups. The Statistical Package for the Social Sciences (SPSS) was utilized for statistical analysis.
The psychological and physiological findings were as follows. Adams et al (2007) analyzed different groups on depression, anxiety, or the Global Severity Index (a summary of overall level of psychological distress) by using the Brief Symptom Inventory (BSI). The difference of score of BSI was not statistically meaningful among the three groups between initial and the final exanimation. The participants’ depression was evaluated by the Beck Depression Inventory (BDI) and three groups were compared regarding their BDI scores. The group scores did not show any differences. The physiological variables such as body weight, body mass index and blood pressure did not differ notably between the baseline and the final 1-year follow-up measurement among the groups. The researchers performed analysis of covariance in order to evaluate any differences in groups in physiological variables, but no significant differences were shown. The only important finding in the study of Adams et al. (2007) was High-Density lipoprotein (HDL) level between the cardiac rehabilitation and traditional care groups in the subset of participants with initial HDL levels (< 40 mg/dL) and with age as a covariance. HDL level increased from 30.17mg/dL to 33.67 mg/dL and from 30.54 mg/dL to 37.48 mg/dL in the traditional care group and in the cardiac rehabilitation group respectively (F=4.577, P=0.035).
Based on their research findings, Adams et al (2007) concluded that the significant improvement in HDL levels in the cardiac rehabilitation group was the result of exercise training. Although the researchers predicted that the CR participants who receive the most intensive intervention would have better overall outcomes, the differences were limited. The researchers stated that the baseline physiological values tended to be within normal limits because they were taken 6 to 8 weeks after discharge. Thus, the normal value of baseline data affected their study negatively. Despite their minimal research findings, Adams et al (2007) suggested that more individually customized intense CR programs focus on reducing cardiovascular risk factors for cardiac patients since their study goal was to determine the best intervention among their study subjects who attended exercise-based cardiac rehabilitation, traditional care, or Leap-For-Life workshops.
In the study of Milani et al (2007), the 91 out of 522 patients comprising the treatment group was depressed on entry into rehabilitation based on the Kellner Symptom Questionnaire. In the study of Milani et al (2007), statview software 5.0.1 was utilized for statistical analysis. Results were expressed as mean ±SD or frequencies explicated as percentages. The 91 out of 522 patients comprising the treatment group and the 48 out of 179 patients comprising the control group were depressed respectively on entry into rehabilitation based on the Kellner Symptom Questionnaire. 10.7 ± 3.4 was the depression score of 91 patients (17% of treatment group) and 11.3 ±3.6 was the depression score of 48 patients (26% of control group).In treatment group, the score of 10.7 ± 3.4 changed to 3.9±4.2 at the end of CR program, which meant that the prevalence of depressive symptoms dropped 63% from 17% on enrollment to 6% on upon completion of the CR program (p<0.0001). All-cause mortality was assessed in the treatment group according to the presence or absence of depression symptoms at the completion of rehabilitation. The mortality rate for depressed patients was 22 % and 5% for non depressed patients (p=0.0004). This finding suggests that depressed participants have more than 4 times of chance of likelihood of dying compared to non depressed patients (22% vs 5%, p=0.0004).
In the control group, all-causes of mortality were further evaluated based on whether the subjects had depression symptom or not. 30 % and 11 % were the mortality rates for depressed patients and non depressed patients respectively. The depressed participants had an approximately 3-fold higher mortality compared with the participants without depression (30% vs 11%, p=0.003). Furthermore, Milani et al (2007) evaluated mortality in treatment and control groups of depressed patients identified at entry into cardiac rehabilitation in order to analyze the potential effect of rehabilitation on mortality in depressed participants. The 91 subjects in treatment cohort have 8 % of the mortality rate and the 48 subjects in control cohort have 30 % of the mortality rate. Therefore, the depressed patients who dropped out of the CR program had a nearly 4-fold increase in mortality compared with depressed patients who completed the CR program (30% vs 8%, p=0.0005). Milani et al (2007) tried to isolate the impact of the exercise component of cardiac rehabilitation upon depression symptoms by classifying the participants into 3 groups according to changes in their peak VO2 over the course of the program and by comparing the three cohorts. The 3 groups were VO2 loss group (n=102) or those who have no gain or loss in peak VO2, mild gain VO2 group (n=135) or those who a mild improvement (≤ 10 %) in peak VO2 and high VO2 group (n=285) or those who increased peak VO2 by>10%. The depression rate was assessed in each of these groups at entry and completion of rehabilitation and the mortality was evaluated subsequently.
The results were as follows. The prevalence of depression symptoms changed from 18 % to 14% on the completion of the CR in the group 1 (p=NS), 17% to 5% in group 2 (p=0.004), and 18% to 5% in group 3 (p<0.0001). The mortality was 6% and 4% in group 2 and group 3 respectively. The 15% of relatively high mortality was shown in group1 compared with that of group 2 and group 3. Therefore, the mortality was low in the patients who improved their exercise capacity measured by their peak VO2. Milani et al (2007) concluded that survival mirrored a decrease in depression symptoms with the best outcomes taking place in cohorts who improved peak VO2.
The research findings of Milani et al (2007) suggest that the prevalence of depressive symptoms in the CHD patients after a major coronary event and patients who have depressive symptoms influence a noticeable increase in all-cause mortality risk over time. CR is associated with a significant reduction in the prevalence of depressive symptoms resulting in subsequent improvement in survival. Modest gain of exercise capacity that was achieved by CR affects a decrease in depressive symptoms corresponding to increase in survival.
Study Limitations
Kardis et al (2005), Fast et al (2009), Snow et al (2005), Adams et al (2007) and Milani et al (2007), stated a problem of generalization since their study include a non probability sample without randomization. Probability sampling (random sampling) includes the haphazard selection of elements from a population and it involves a selection process in which each element in the population has an equal and independent opportunity of being selected. Thus, more confidence can be placed in this sampling in terms of the representation of the population (Polit & Beck, Nursing Research, 2009).
Kardis et al (2005) conducted their study in one institution which is Inova Heart Institute. Thus their study findings only represented those who enrolled in Inova Heart Institute Cardiac Rehabilitation Program. Also they pointed out 40 % of attrition rate which is very high. Despite no absolute standard for acceptable attrition rate available, biases usually emerge as a concern if the attrition rate exceeds 20% (Polit &Beck, Nursing Research, 2009).
Milani et al (2007) pointed out selection bias and a problem with their instruments’ reliability as their study limitations. Selection bias is the pre-existing differences between control and experimental group and there is always a possibility that the groups are not equivalent when subjects are not assigned randomly into the two groups (Polit & Beck, Nursing Research, 2009).
Since the researches used a controlled trial without randomization as their study design, selection bias exists in their research. Milani et al (2007) stated that the Keller Symptom Questionnaire which is one of their study instruments was not well established to evaluate depression symptom. In fact, the researcher did not mention the instrument’s reliability such as internal consistency, test-retest reliability and inter-rater reliability in their study. Selection bias and instrument reliability influence internal validity of study that is the existence of true causal relationship between the independent variable and dependent variable (Polit & Beck, Nursing Research, 2009).
In summary, CR programs improve patients’ QoL based on the findings from Kardis et al (2005) and Fast et al (2009). The research outcomes of Snow et al (2005) and Adams et al (2007) reveal that CR program has a positive impact on cardiovascular risk factors focusing on lipid levels. The study results of Milani et al (2007) show that CR programs help to reduce the prevalence of depressive symptom and its associated mortality among CAD patients. However, this literature review fails to address other issues such as why some patients drop out of CR programs and how healthcare professionals can help these patients to adhere to CR programs that were clinically proven to be beneficial to CAD patients in terms of their recovery after a cardiac event. Another literature gap is a gender difference in terms of recovering from a cardiac event. According to the research findings from Kardis et al (2005), compared with male gender, the female gender made the largest gains in Changes in Health (-22.7%vs-2.4%) and compared with female gender, the male gender made larger gains in Pain (-15.9%vs-5.6%). Male subjects had more improvement on physical dimensions and female subjects had a tendency to have increased QoL scores on psychosocial dimensions in their study. Therefore, more detailed studies are necessary to focus on gender difference in how opposite sexes recover from cardiac events differently and gender-specific interventions including CR for CAD patients based on findings.
References
Adams, J. L., Nuss, T.,Banks, C., Hartman, J., Segrest, W.,Spears J.,Yout P., & Bryant L. (2007). Risk Factor Outcome Comparison between exercise-based Cardiac Rehabilitation, Traditional Care and an Educational Workshop. The Journal of Continuing Education in Nursing 38 (2):83-88.
Fast, Y. J., Steinke, E. E., & Wright, D. W. (2009). Effects of Attending Phase II Cardiac Rehabilitation on Patients Verses Spouse (Proxy) Quality –of –Life Perceptions. Journal of Cardiopulmonary Rehabilitation and Prevention 2009; 29: 115-120.
Kardis, P.,Bruce, A., Michaels, J., & Barnett, S.D.(2005).Quality-of- Life Changes Following the Completion of Phase II Cardiac Rehabilitation. Journal of Nursing Care Quality 20(2):161-166.
Milani, R. V.& Lavie, C. J.(2007). Impact of cardiac rehabilitation on depression and its associated mortality.The Amercan Journal of Medicine 120,799-806.
Snow, R., Lalonde, M., Hindman, L., Falko, J., and Caulin-Glaser, T. (2005). Independent Effect of Cardiac Rehabilitation on Lipids in Coronary Artery Disease. Journal of Cardiopulmonary Rehabilitation 25(5):257-261.