1
Running head: META-ANALYSIS
Cognitive Behavioural Interventions and the Immune System:
A Meta-analysis of Randomized Control Trials
Sinem Turan
student ID 10280219
Department of Psychology, University of Amsterdam,
Nieuwe Achtergracht 129-B,
1018 WT Amsterdam,
The Netherlands
Sinem_turan@hotmail.com
2
Table of Contents
Abstract 3
Cognitive Behavioural Interventions and the Immune System: 3 A Meta-analysis of Randomized Control Trials
Method 7
Results 9
Discussion 19
References 22
3 Cognitive Behavioural Interventions and the Immune System: A Meta-analysis of
Randomized Control Trials
Abstract
A prior meta-analysis has demonstrated the efficacy of five different psychological
interventions (i.e., stress–management training, relaxation training, disclosure, hypnosis, and conditioning), on the human immune system, leaving out the most common therapy method out: Cognitive Behavioural Therapy (CBT). The aim of this paper was therefore to provide a meta-analysis of studies that explored the efficacy of CBT interventions on the immune system. Based on our literature search, 954 studies were identified of which ultimately seven were selected. The selected studies comprised 490 participants from clinical samples, with either an HIV diagnosis, cancer diagnosis and/or an panic disorder. Meta-analytic results revealed no significant immunological benefits for participants who received CBT when compared to control groups. The methodological limitations and potential future research are discussed.
Keywords: Cognitive behavioural therapy, immune system, diagnosis, disorder.
Numerous studies have investigated the complex relationship between psychological stress and functioning of the human immune system, which is studied by the field of
psychoneuroimmunology. Segerström and Miller undertook a meta-analysis in 2004 of more than 300 papers published over de course of 30 years which describe a relationship between psychological stress and parameters of the human immune system. This meta-analysis showed that, as a general rule, short-term stress (lasting minutes) to be strengthening of the bodily resources such as Lymphocyte cells and Cytokine’s, and the long-term stress to lead to a suppression of the immune system response, such as a weaker response to vaccination and diminished DNA repair function (McEwen, 1998; Segerström & Miller, 2004; Cohen, Marshall Jr., Cheng, Agarwal & Wei, 2012). Examples of research findings in this area are:
4 • Family members caring for long term disease victims suffer chronic stress that
negatively effects their immune system (Kiecolt, Glaser Shuttelworth, Dyer, Ogrocki & Spiecher, 1987)
• Breast cancer patients suffer from significantly decreased immunity due to illness-related stress factors (Levy, Herberman, Lipman & d’Angelo, 1987)
• Students’ immune systems perform significantly worse during stressful exam periods (Cohen et al., 2000).
• HIV infected individuals with elevated levels of psychological distress experience relatively more immunosuppression and disease progression (Antoni, Pereira, Marion, Peake, Rose, Mc Calla, Somin, Fletcher, Lucci, Efantis-Potter, & O’Sullivan, 2008). Based on these and other examples, evidence seems to indicate that clinical as well as non-clinical samples show decreased levels of immune functioning when psychological distress levels are elevated.
Description of the Immune System
The immune system is made of a network of cells, proteins, tissues and organs, and serves as the body’s chief protection against harmful ‘foreign’ agents, such as toxins and
micro-organisms. When functioning properly, the immune system identifies and attacks these agents to minimize the chances of developing an illness or disease. As shown in Figure 1, the organs of the immune system are positioned throughout the body and are called Lymphoid organs as they are home to Lymphocytes. Lymphocytes are small white blood cells that are the key players in the immune system. They are spread out though the body and take different cellular forms. The main types of Lymphocytes are the CD4+ or helper T cells, which regulate
immune functions through the secretion of hormone-like substances (cytokines) and CD8+ (cytotoxic T cells) and CD56+ (natural killer cells), which both kill virus-infected and cancerous cells (Morgan, Dudley, Wunderlich, Hughes, Yang, Sherry, Royal, Topalian, Kammula, Restifo, Zheng, Nahvi, de Vries, Rogers-Freezer, Mavroukakis & Rosenberg, 2006). As immune cells travel between tissues through the bloodstream, the numbers and proportions of Lymphocytes in the circulation provide an important representation of the state of Lymphocytes distribution in the body. Stress-induced increases in plasma corticosterone are accompanied by a significant decrease in numbers of Lymphocytes. Furthermore, the stress-induced changes in the Lymphocytes numbers are rapidly reversed upon the cessation
5 of stress. Moreover, chronic stress is associated with the decrease of cell activities that affect processes such as immune surveillance of tumours and genomic stability (Reiche, Vargas-Nunes & Morimoto, 2004).
Figure 1
Source: National Cancer Institute, USA
https://www.aids.gov/hiv-aids-basics/just-diagnosed-with-hiv-aids/hiv-in-your-body/immune-system-101/
Cognitive - Behavioral Interventions and Immunity
As stress can negatively influence the immune system, methods directed to managing stress adequately, could possibly enhance immune functioning. A common approach to reducing stress-related symptoms is cognitive behavioural therapy (CBT). It can be described as an
6 umbrella term that encompasses various psychological interventions that emphasize the modification of dysfunctional thinking and behaviour patterns in a structured and time-limited manner to lower psychological distress (Morin, Vallières, Guay, Ivers, Savard, Mérette, Bastien & Baillargeon, 2009). The aim of modifying dysfunctional thoughts may be reached by uncovering underlying causes of worries and fears, and help develop more adequate coping and problem-solving skills. During treatment behavioural patterns are restructured by focusing on, for example, physical activation and exposure techniques. Table 1 provides an overview of commonly used CBT intervention types.
Tatrow and Montgomery conducted a meta-analysis in 2006 where they examined CBT interventions for distress and pain specifically in breast cancer patients. Their study included 20 studies that used CBT interventions with breast cancer patients and their findings revealed effect sizes of d = 0.31 for distress (p < 0.05) and .49 for pain (p < 0.05), indicating that 62 and 69% of breast cancer patients in the CBT intervention treatment groups had less distress and less pain (respectively) relative to the control groups. Moreover, CBT has proven to be at least as successful as antidepressants in the treatment of (moderate and severe) depression and it is associated with a better maintenance of therapeutic achievements over time, in comparison to short –term antidepressant treatment (Savard, Simard, Giguere, Ivers, Morin, Maunsell, Gagnon, Robert & Marceau, 2006).
The Meta-Analysis
A variety of studies have been conducted to explore the relationship between stress reduction techniques and immune functioning. However, to our knowledge, a limited number of meta-analyses have been performed aimed at drawing general conclusions regarding the presumed effectiveness of these techniques on immune functioning. Meta-analyses offer a
systematic and quantitative approach to synthesizing empirical evidence and which can be used to assess the clinical effectiveness of interventions by combining the findings from independent studies. A good meta-analysis aims for a complete coverage of all relevant studies to avoid biased conclusions and examines an overall effect size between the variables of interest (Higgins & Thompson, 2002). Meta-analyses therefore typically involve a
comprehensive search strategy which interrogates several electronic databases, e.g., PubMed. Miller and Cohen (2001) conducted a meta-analytic review in which they investigated the modification of immune system functioning in individuals by focusing on the effects of various psychological treatments: i.e., stress-managing interventions, relaxation interventions, disclosure interventions, hypnosis with immune suggestion interventions, and classical
7 conditioning interventions. The meta-analysis yielded no convincing evidence for
improvement of immune system functioning by the described treatments. It is notable that CBT was not included in the meta-analysis, although CBT has proven to be highly efficacious in stress reduction (Koszycki, Benger, Shlik & Bradwejn, 2007).
In light of the preceding introduction, the aim of the current paper is to investigate whether CBT - as a powerful stress reducing method - has a positive effect on immune functioning in individuals with psychological or physical illness. If so, CBT can be a helpful tool to enhance illness recovery in clinical samples.
Method
Identification of Studies
The literature search focuses on randomized control trials involving CBT interventions and the immune system. To identify studies of interest, a computerized literature search of the PubMed database was performed. The search was executed in March 2015, and the following combination of keywords was used: (psychotherapy OR conditioning OR disclosure OR hypnosis OR relaxation OR meditation OR biofeedback OR "stress management" OR stress-management OR mindfulness OR "cognitive behavioral therapy" OR "cognitive therapy") AND (immune OR immunology OR leukocytes OR lymphocyte OR lymphocytes OR “natural killer cell” OR “natural killer cells” OR psychoneuroimmunology OR “immune function” OR inflammation OR IL-6 OR CRP OR CD8 OR CD4 OR CD56 OR vaccine OR vaccination OR immunization OR allergy OR “wound healing” OR hypersensitivity OR herpes OR virus OR antibody) AND (“randomized controlled trial” OR “controlled trial” OR RCT OR “randomized controlled”). The search yielded a pool of 954 articles for potential inclusion. Articles chosen to be included in this meta-analysis had to meet the following criteria: the publication date of the articles of interest was between 2001 and 2015. This was done knowing that the last meta-analysis regarding the effects of psychological interventions on the immune system was conducted by Miller et al. in 2001. The second criterion was that all participants had a psychological or a physical diagnosis. This criterion was included to maximize possible effect sizes. The third criterion was that participants were randomly
assigned to either the CBT condition or a control condition. The fourth criterion was that each study had to provide before and after CBT treatment statistics. The last criterion was that immunological parameters were measured for all participants.
8
Meta-Analytic Procedure, Presentation of Statistical Analyses
Effect Size
An effect size is denoted by the symbol “d”, It is the mean difference between groups in standard score forms i.e. the ratio of the difference between the means to the standard deviation (SMD). Cohens rule for interpretation of the SMD statistics would mean that a value of 0.2 is indicating a small effect, a value of 0.5 indicating a medium effect and a value of 0.8 or larger indicating a larger effect. (Moher, Liberati, Tetzlaff & Altman 2009).
Correlations between the independent variable psychological outcomes and dependent
variable immunological outcomes were collected and used as effect size in this meta-analysis. Heterogeneity
Heterogeneity in meta-analysis refers to the variation in study outcomes between studies. A test which is commonly used is Cochrane’s Q, a statistics based on the chi-square test. The presence or absence of heterogeneity influences the subsequent model (fixed or random) of an analysis. (Heckman, Urzua & Vytlacil, 2006). The heterogeneity of selected studies was assessed to determine whether the between-studies variability was compatible with chance alone.
Forest Plot
A forest plot is a convenient way of presenting a number of means and their confidence intervals in a graphic manner. It is a usual way of displaying data from a meta-analysis (Neyeloff, Fuchs & Moreira, 2012). A forest plot was computed that incorporated the effects sizes of all seven studies as well as an averaged overall effect size.
Radial Plot
Radial plot, also known as Galbraith plot, is an alternative to the forest plot. It is a way of displaying several estimates of the same quantity that have different standard errors. (Harbord, Egger & Sterne, 2006). As lastly a radial plot was computed to display the heterogeneity results.
All statistical calculations were computed using MedCalc 11.3.1.0.
Meta-Analytic Procedures, Before and After CBT treatment phases
The meta-analysis of the treatment effect was analyzed in two phases, these phases are called “Before CBT treatment” and “After CBT treatment”. Regarding the first phase, to ensure that both treatment and control participant groups were not significantly different from one
another and thus comparable, effect sizes (SMD) were computed before treatment. In the before CBT treatment phase, the random effects model was used. Here we prefer to use
9 random effects model as the result of the test for heterogeneity were significant (see Table 6). A pictorial representation displaying data from the meta-analysis where done by presenting a forest plot as well as a radial plot.
In the After CBT treatment phase, effect sizes were computed to certify whether or not CBT had a more favorable effect on immunological measures compared to control groups. In this phase, the random effects model was once again used as test of heterogeneity was found to be significant (see Table 9). Lastly, once again, a forest plot and radial plot were computed to present a pictorial overview of the results.
Results
Literature search results
The literature searches of the PubMed database yielded 954 studies. As shown in Figure 2, after application of inclusion criteria, a pool of seven studies were left regarding CBT interventions and its influence on the immune system.
Figure 2 Flowchart of Study Selection 954 studies identified by PubMed
221 studies retrieved
Included in the meta-analysis: 7 studies including CBT and
immune system
733 studies excluded after reading title and abstracts
Unrelated topic Review papers No CBT intervention
No measures of the immune system
214 studies excluded after reading the paper
No control groups No random assignment
10 A close scrutiny of all seven studies showed that all CBT interventions improved mood in individuals participating in the studies in comparisons to control group participants. Now that this assumption has been confirmed, we continue exploring whether CBT has a positive effect on immune functioning in individuals with psychological or physical illness.
Table 1 provides an overview with definitions of different types of CBT interventions that were used in the selected studies. In addition, the studies in which these specific CBT interventions were used, along with the total number of participants, are showcased in this table. Table 2 offers a description of the different psychological measurements and
immunological variables which were investigated in the included studies. An overview of study characteristics, interventions and outcome measures of all seven studies is displayed in Table 3.
Table 1 Definition of CBT Interventions and Number of Participants in the Selected Studies
CBT Interventions Definition Nst Np
Cognitive Therapy (CT)
Cognitive therapy is based upon identifying and restructuring unhelpful or inadequate thinking.
1 21
Cognitive Behavioural Therapy
(CBT)
CBT is based upon identifying and restructuring unhelpful or inadequate thinking and associated behaviour patterns.
1 21
Cognitive Behaviour Stress Management
(CBSM)
CBSM is a short-term CBT approach that focuses on stress management techniques and relaxation skills.
3 135
Mindfulness-based Cognitive Behaviour
Therapy (MBCT)
MBCT uses traditional CBT methods and adds mindfulness and meditation in order to being aware of incoming thoughts and feelings without responding to them.
1 19 Cognitive - Behavioural Relaxation Training (RLXN)
RLXN combines traditional CBT methods, relaxation and coping strategies in order to modify stressor appraisals.
1 65
11
Table 2 Instruments and Outcomes in the Selected Studies
Psychological Measurement Instruments
Instrument Description
Beck Anxiety Inventory (BAI)
The BAI is a 21-item self-report inventory that is used for measuring the severity of anxiety in children and adults.
Beck Depression Inventory (BDI)
The BDI includes 21 items evaluating the severity of depressive symptoms.
Dealing With Illness Scale (DIS)
The DIS assesses stress levels and coping patterns. It is a 20-item questionnaire with Stress subscales comprised of Stress and Coping.
Hospital Anxiety and Depression Scale (HADS)
The HADS is a 14-item questionnaire divided into two sub-scales: depression and anxiety.
Health Behaviours Questionnaire (HBQ)
The HBQ is a 60-item self-report inventory that assesses recent health behaviours that may potentially confound the relationship between psychological aspects and immune function.
Hamilton Depression Rating Scale
(HDRS)
The HDRS is a clinical interview which contains 17 items assessing the severity of depression symptoms.
Hamilton Rating Scale for Anxiety (HRS-A)
The HRS–A is a questionnaire used to rate the severity of an individual’s anxiety through 14 items.
Impact of Event Scale (IES)
The IES is a self-report measure that assesses an individual’s distress caused by traumatic events. Respondents are asked to identify a specific stressful life event and then indicate how much they were distressed during the past seven days by a 15-item questionnaire.
12
Life Experiences Survey (LES–10)
The LES–10 is used to measure level off stress. It contains 10 items where individuals are asked to respond whether events had a positive or negative impact on them.
The Profile of Mood States (POMS)
The (POMS), a 65-item scale, assesses six different mood states during the past week using a variety of adjectives that
participants are asked to rate. Symptom Checklist-90
Revised (SCL–90–R)
The CLS-90-R is a self-report questionnaire with 90 items. It is designed to evaluate a broad range of symptoms of psychopathology as well as measuring the progress and outcome of psychiatric treatments.
Immunological Variables
Outcome Description
CD Cells The cluster of differentiation (CD) is a protocol used for the identification and investigation of cell surface molecules providing targets for
immunophenotyping of cells. The CD system is commonly used as cell markers in immunophenotyping, Combining markers such as CD4 and CD8 cells in blood is often used to monitor the progression of HIV infection.
Cell Cytotoxicity Demolition of a target cell by specific Lymphocytes, such as cytotoxic T Lymphocytes or NK cells.
Lymphocyte Proliferation Morphologic alteration of small B Lymphocytes or T Lymphocytes in culture into large blast-like cells able to synthesize DNA and RNA and divide mitotically. It is induced by interleukins, mitogens and by specific antigens.
13
Table 3 Study Characteristics, Interventions and Outcome Measures of Selected Studies
Study Sample characteristcs Interventions Psychological outcomes Immunological outcomes Duration intervention 1) McCain et al. (2008) Individuals with HIV infection N = 122 Female : 40 % Male : 60 % Mean age = 42,2 RLXN (N = 65) vs. CTRL (N = 57)
IES, DIS, Lymphocyte proliferation 10 2) Koh et al. (2004) Individuals with the DSM diagnosis Panic disorder N = 63 Female : 46 % Male : 54% Mean age = 35, 2 CBT+ BEN (N = 21) vs. BEN (N = 42) SCL-90-R, HRS-A Lymphocyte proliferation 6 3) McCain et al. (2003) Individuals with a HIV infection N = 89 Female : 20 % Male : 80 % Mean age = 39, 4 CBSM ( N = 53) vs. WLC (N = 36)
DIS, IES, Cell cytotoxicity 8
4) Gonzalez et al. (2014) Individuals with a HIV infection Total N = 39 Female : 0 % Male : 100 % Mean age = 49, 4 MBCT (N = 19) vs. CTRL (N = 20) BDI-II, BAI CD4+ cell count 8 5) Antoni et al. (2008) Cervical intraepithelial neoplasma and HIV infection A 10-week CBSM (N = 21) vs. a 1-day LES-10 CD4+ CD3+ cell count 10
14 N = 39 Female : 100 % Male : 0 % Mean age = 31, 2 CBSM (N = 18) 6) Savard et al. (2006) Breast cancer N = 37 Female : 100 % , Male : 0 % Mean age = 51,5 CT (N =21) vs. WLC (N = 16) HADS, BDI, HDRS, HBQ Lymphocyte proliferation 8 7) Antoni et al. (2006) HIV infection. Total N = 101 Female : 0 % Male 100 % Mean age = 41,6 CBSM + MAT (N = 61) vs. MAT (N = 40) POMS, BDI CD4+ CD8+ cell count 10
BAI= Beck Anxiety Inventory, BD-II = Beck Depression Inventory, BEN = Benzodiazepine, CBSM = Cognitive Behaviour Stress Management, CBT= Cognitive Behavioural Therapy, CTRL = Wait-listed Control Group, DIS= Dealing With Illness Scale, ELISA = standardized enzyme-linked immunosorbent, HADS = Hospital Anxiety and Depression Scale, HBQ = Health Behaviors Questionnaire, HDRS = Hamilton Depression Rating Scale, HIV = Human Immunodeficiency Virus, HRS-A = Hamilton Rating Scale for Anxiety, IES = Impact of Event Scale, LES – 10 = Life Experiences Survey, MAT = Medication Adherence Training, MBCT = Mindfulness-based Cognitive Behaviour Therapy, MEMS = medication event monitoring system, POMS = The Profile of Mood States, rHCMSS = revised HIV Center Medical Staging Scale, SCL-90-R = Symptom Checklist-90 Revised, WLC = Waiting List Control.
Interpreting the Meta-Analytic Findings
In the present meta-analysis, of the seven studies included for analysis, a total of 490
individuals participated. Of these participants 261 belonged to the CBT group, and 229 were assigned to the control group. Table 4 provides an oveview of meta-analysis statistics of control and CBT groups before treatment status, where the Immunological outcome
represents the dependent varible. Table 5 provides an overview of the effect sizes (SMD) of the CBT group before CBT treatment, where the Immunological outcome represents the dependent varible. The results show an effect size before treatment of -0.542 which confirms that the treatment groups at the beginning of the research were not significantly different from eachother. The statistical results from the effect size calculation shows only a slight preferene to the CBT treatment group. Table 6 presents the heterogeneity of the seven studies included in this meta-analysis, which shows heterogeneity to be significant Q=62,95 with p<0.0001. A random effects model was preferred due to significant heterogeneity scores. This effect size is displayed in the forest plot and radial plot of mean difference beneath Table 6.
15 changes in the CBT treatment group and the control groups, where the Immunological
outcome represents the dependent varible.
The random effect model was used to determine the effect size. These results show that CBT was not superior to a control group regarding it’s effects on the immune system.
Tabel 8 provides an overview of the effect sizes (SMD) of the CBT group after CBT treatment for the seven studies where the Immunological outcome represents the dependent varible. Table 9 presents the heterogeneity of the seven studies in cluded in this meta-analysis, which show heterogeneity to be significant Q = 177,42 with p<0.0001. A random effects model was prefferred due to significant heterogeneity scores. The total effect size using the random effects model was 0,105. As according to Cohen’s rule, a result below 0.2 shows that there is no evidence to support a positive effect of CBT on the immune measures included in the analyses when compared to the control group (Moher et. al 2009). This effect size is displayed in the forest plot and radial plot of mean difference beneath Table 8. There is a slight but an insignificant effect detected between the CBT and control group.
Before CBT Treatment Statistics
Table 4 Meta-Analysis Statistics of Control and CBT groups Before Treatment status where the Immunological Outcome represents the Dependent Varible.
Studies Total N M SD Total N M SD
CBT CBT CBT Control Control Control 2008, McCain, Gray, Elswick Jr. 65 6.79 1.25 57 6.55 1.28
2004, Koh, Lee 21 3.93 0.37 42 4.18 0.29
2003, McCain, Munjas, Munro 53 11.4 11.0 36 13.5 6.6
2003, Gonzalez-Garcia, Ferrer, Borras 19 5.55 1.98 20 4.93 2.23
2008, Antoini, Pereira, Marion 21 4.14 2.93 18 4.15 1.85
2006, Savard, Simard, Giguere 21 1.31 0.14 16 1.54 0.17
2006, Antoni, Carrico, Duran 61 13.23 3.09 40 22.98 7.34
Table 5 Meta-Analysis Statistics of Effect Sizes Before Treatment status where the Immunological Outcome represents the Dependent Varible.
16
Studies Total N Total N Total SMD 95% CI
CBT Control N 2008, McCain, Gray, Elswick Jr. 65 57 122 0,188 -0,172 to 0, 548
2004, Koh, Lee 21 42 63 -0,775 -1,328 to -0,222
2003, McCain, Munjas, Munro 53 36 89 -0,22 -0,650 to 0,211 2003, Gonzalez-Garcia, Ferrer, Borras 19 20 39 0,288 -0,365 to 0,940 2008, Antoini, Pereira, Marion 21 18 39 -0,00354 -0,654 to 0,647 2006, Savard, Simard, Giguere 21 16 37 -1,465 -2,232 to -0,699 2006, Antoni, Carrico, Duran 61 40 101 -1,864 -2,347 to -1,381 Total (fixed effects) 261 229 490 -0,446 -0,635 to -0,257 Total (random effects) 261 229 490 -0,542 -1,173 to 0,0884 Table 6 Test for Heterogeneity of the seven studies; Before Treatment
Q 62.9461
DF 6 Significance level P<0.0001
Forest Plot for Before Treatment status of the seven studies
17
After CBT Treatment Statistics
Table 7 Meta-Analysis Statistics of Control and CBT groups After Treatment status where the Immunological Outcome represents the Dependent Varible.
Studies Total N M SD Total N M SD
CBT CBT CBT Control Control Control 2008, McCain, Gray, Elswick Jr. 65 7.797 1.1.85 57 5.163 1.211
2004, Koh, Lee 21 4.05 0.3 21 4.01 0.36
2003, McCain, Munjas, Munro 21 4.2 7.8 19 1.4 3.6
2003, Gonzalez-Garcia, Ferrer, Borras 19 6.14 1.83 20 5.16 2.38
2008, Antoini, Pereira, Marion 21 3.845 2.972 18 4.768 2.327
2006, Savard, Simard, Giguere 16 1.46 0.15 16 1.4 0.17
2006, Antoni, Carrico, Duran 61 5.632 3.054 40 31.956 16.218
Table 8 Meta-Analysis Statistics of Effect Sizes After Treatment status where the Immunological Outcome represents the Dependent Varible.
-10.0 -7.0 -4.0 -1.0 2.0 0.0 5.0 10.0 15.0 20.0 1 2 3 4 5 6 7
Radial Plot of Mean Difference
1/(Standard Error) Z S ta ti s ti c
18
Studies Total N Total N Total SMD 95% CI
CBT Control N
2008, McCain, Gray, Elswick Jr. 65 57 122 2,186 1,730 to 2, 643
2004, Koh, Lee 21 21 42 0,118 -0,506 to 0,743
2003, McCain, Munjas, Munro 21 19 40 0,444 -0,205 to 1,094
2003, Gonzalez-Garcia, Ferrer, Borras 19 20 39 0,451 -0,208 to 1,109
2008, Antoini, Pereira, Marion 21 18 39 -0,336 -0,991 to 0,320
2006, Savard, Simard, Giguere 16 16 32 0,365 -0,364 to 1,094
2006, Antoni, Carrico, Duran 61 40 101 -2,499 -3,037 to -1,961
Total (fixed effects) 224 191 415 0,209 -0,0112 to 0,430
Total (random effects) 224 191 415 0,105 -1,104 to 1,315
Table 9 Test for Heterogeneity of the seven studies; After Treatment
Q 177,4203
DF 6
Significane level P<0,0001
19
Radial Plot for After Treatment status
Discussion
Based on my literature search, the present meta-analysis was the first to investigate whether CBT has a positive effect on immune functioning in individuals with psychological or physical illness. The present meta-analysis included seven studies, summarized in Table 3, in which CBT intervention effects were compared to control groups. The random effects analysis performed at after CBT treatment showed an effect size of 0,105 for the correlation between psychological outcomes and immunological outcomes. The results indicate that a CBT intervention did not have a superior effect, compared to control groups, regarding its efficacy on the immune system. Using statistical analyses, I could conclude that the
Lymphocyte levels did not get elevated by a CBT treatment as there were a before treatment analysis performed for both the CBT group and control group participants.
However, the results of this study need to be interpreted in light of certain limitations. Firstly, our meta-analysis included a relatively small sample size compromised of seven studies which limits our possibilities to draw general conclusions of our results - a low statistical power is assumed. In this case a statistical power analysis could have been of
-15.0 -7.5 0.0 7.5 15.0 0.0 5.0 10.0 15.0 20.0 1 2 3 4 5 6 7
Radial Plot of Mean Difference
1/(Standard Error) Z S ta ti s ti c
20 benefit to know more certain how many studies it would have been needed to not make a Type II error, thus concluding there is no effect when, in fact, there is one (Lieberman & Cunningham, 2009).
Secondly, the overall gender distribution was skewed, showing that more men than women participated in the total of seven studies. When analysing gender sample sizes in those studies, it caught my attention that significantly less women than men with a HIV diagnosis participated, while all the participants with a cancer diagnosis were female. This could imply that the statistical results of our meta-analysis are not generalizable for women and men with the same diagnosis. Furthermore is there a gender difference question rising regarding the reporting of psychological complains. Previous research have shown that there is a significant gender difference in the way women and men report ‘worry’, with women often reporting more worry than men. Women have also reported a more negative problem orientation and engaging in more thought suppression (Robichoud, Dugas & Conway, 2003). This might imply that positive CBT results are easier to achieve in male participants than female participants leaving us with a gender difference in reported therapeutic effects and results.
Moreover did the duration of a CBT treatment not exceed 10 weeks in the selected studies, which indicates that longer therapy periods could provide different results. Consequently are there questions about how many CBT session and treatment length there should be delivered to be able to achieve a clinical efficacy in an adult population. There seem to be no recent studies conducted regarding this question. Future studies may precede on the research conducted by Shapiro, Barkham, Rees, Hardy, Reynolds, Startup in 1994, where the
researches found patients diagnosed with a severe depression to benefit more from a 16 week CBT treatment in comparison to patients with a severe depression receiving 8 weeks of CBT treatment.
Fourthly, in the selected studies there are variations within CBT that differ in degree of emphasis on cognitions or behavioural components. This raises questions about their comparability and our outcomes. Additionally, the distinctive features between the CBT interventions were not clearly mentioned which may suggest that the major therapeutic effects of the interventions occur through common factors. This in turn limits our understanding of which intervention truly have had the best effect on reducing stress and as result, enhance the immune system.
21 Analysing the seven studies did however unveil 3 remarkable studies where CBT did display a superior effect on the immune system when compared to a control group. One of those studies was conducted by Gonzalez et. al (2014) where the researchers investigated the effectiveness of MBCT on patients aging with HIV infection. Another study presenting a CBT intervention to have a superior effect when compared to a control group was the study by McCain et. al. (2003) where the researchers investigated the effects of CBSM in patients with HIV infection. Noticeable factors with these two studies were both studies having significantly more male participants, Gonzalez et. al including 100% male participants and McCain et. al. including 80% male participants, and all participants in both studies being treated for HIV infection. Furthermore did both studies have positive scores at follow-up at 5 and 6 months.
5 years after McCain et. al conducted there CBSM study on HIV-infected men, Antoni et. al (2008) examined the effects of CBSM on HIV-infected women. Their participants were also diagnosed with Cervical intraepithelial neoplasia (CIN). Their results, just as Antoni et. al, were positive as CBSM showed to have a significant positive effect on the immune system when compared to a control group. This was once again seen at a follow-up at 9 months. These positive results may suggest men and women infected with HIV to react better to a CBT intervention when compared to participants with cancer or a panic disorder, and perhaps show CBSM to be one of the more effective CBT interventions for HIV infected participants. Future meta-analyses could benefit from studies with bigger sample sizes in which males and females are more equally distributed within the clinical group of interest. In addition, the duration of CBT treatment could be prolonged in future studies to investigate if immune system functioning improves with CBT treatment periods exceeding ten weeks. Moreover, future meta-analyses may in addition focus on separating the types of CBT interventions used when analysing their efficiency on the immune system. This could in turn provide more detailed insight into the significance of various CBT interventions on immune system functioning in individuals with physical or psychological illness.
If we in the future can draw conclusions about specific relationships between types of CBT interventions and its durations and immune functioning in specific clinical populations, then CBT could be of an alternative treatment method in enhancing our immune system when faced with a clinical diagnosis, particularly since as many as 60% of individuals living with HIV disease are seeking non-traditional therapies (McCain et.al 2008). CBT may thus be
22 considered as a helpful tool to expand life span, shorten recovery periods and enhance mood in a clinical samples.
References
Antoni, M., H., Carrico, A., W., Duran, R.,. E., Spitzer, S., Penedo, F., Ironson, G., Flethcer, M.,A., Kilmas, N. & Schneiderman, N. (2006) Randomized clinical trial of cognitive behavioural stress management on human immunodeficiency virus viral load in gay men treated with highly active antiretroviral therapy. Psychosomatic medicine 68,
143-151.
Antoni, M., H., Pereira, D., B., Marion, I., Ennis, N., Peake, M., Rose, R., Mc Calla, J., Somin., T., Fletcher, M., A., Lucci, J., Efantis-Potter, J. & O’Sullivan, M., J. (2008). Stress management effects on perceived stress on cervical intraepithelial neoplasia in low income HIV infected women. Psychosomatic medicine 65 (4), 289-401.
Cheng, Y., Agarwal, A. & Wei, Q. (2012). DNA repair capacity in healthy medical students during and after exam stress. Journal behavioural medicine 23 (6), 531-544.
23 Cohen, S., Marshall Jr., D., Cheng, L., Agarwal, S.,K. & Wei, Q. (2000). DNA Repair
Capacity in Healthy Medical Students During and After Exam. Stress Journal of
behavioural medicine 23, (6), 531-544.
Gonzalez-Garcia, M., Ferrer, M., J., Borras, X., Munoz-Moreno, J., A., Miranda, C., Puig, J., Perez-Alvarez, N., SOler, J., Feliu-Solar, A., Clotet, B & Fumaz, C., R..(2014). Effectiveness of mindfulness-based cognitive therapy on the quality of life, emotional status, and CD4 cell count of patients aging with HIV infection. AIDS behave 18, 676
-685.
Harbord, R., M., Egger, M., & Sterne, J., A., C. (2006). A modified test for small-study effects in meta-analyses of controlled trial with binary endpoints. Statistics in medicine 25
(20), 3443-3457.
Heckman, J., J., Urzua, S. & Vytlacil, E. (2006). Understanding Instrumental Variables in Models with Essential Heterogeneity. Economics and statistics 88 (3), 389-432). Higgins, J., P., T. & Thompson, S., G. (2002).Quantifying heterogeneity in a meta-analysis.
Statistics in medicine 21 (11), 1539–1558.
Kiecolt-Glaser, J.,R., Glasser, R., Shuttelworth, E.,C., Dyer, C.,S., Ogrocki, P & Speicher, C.,E. (1987). Chronic Stress and Immunity in Family Caregivers of Alzheimer’s Disease Victims. Psychosomatic medicine 49, 523-535.
Koh, K., B & Lee, Y (2004). Reduced anxiety level by therapeutic interventions and cell mediated immunity in panic disorder patients. Psychother psychosom73, 286-292. Koszycki, D., Benger, M., Shlik, J & Bradwejn, J, (2007). Randomized trial of meditation
based stress reduction program and cognitive behaviour therapy in generalized social anxiety disorder. Behaviour research and therapy 45 (10), 2518-2526.
Levy, S., Herberman, R., Lippman, M. & d'Angelo, T. (1987). Correlation of Stress Factors with Sustained Depression of Natural Killer Cell Activity and Predicted Prognosis in Patients with Breast Cancer. Journal of clinical oncology 5 (3), 348-353.
Lieberman, M., D. & Cunningham, W., A. (2009). Type I and Type II error concerns in fMRI research: re-balancing the scale. Social cognitive and affective neuroscience
24 McCain, N., J., Gray, D., P., Elswick Jr., R., K., Robins, J., W., Tuck., I., Walter, J.,M.,
Rausch, S.,M. & McKinney Ketchum, J. (2008). A randomized clinical trial of
alternative stress management interventions in persons with HIV infection. Journal of
consulting and clinical psychology 76 (3), 431-441.
McCain, N.,L., Munjas, B., A., Munro, C.,L., Elswick Jr., R.,K., Wheeler Robins, J.,L., Ferreira-Gonzalez, A., Baliko, B., Kaplowitz., L.,G., Fisher, E., J., Garrett, C., T., Brigle, K.,e., Kendall, L., C., Lucas., V & Cochran, K., L. (2003). Effects of stress management on PNI based outsomes in persons with HIV disease. Nurse health 26
(2), 102-117.
Mc.Ewen, B.,S. (1998) Stress, Adaptation, and Disease: Allostasis and Allostatic Load.
Annals of the New York academy of sciences, 840, 33–44.
Miller, G. and Cohen, S. (2001). Psychological interventions and the immune system: A meta analytic review and critique. Health psychology, 20 (1), 47-63.
Moher,D., Liberati, A., Tetzlaff, J. & Altman, D., G. (2009). Preferred reporting items for systematic reviews and meta – analyses: The PRISMA statement. PLoS Med 6 (6):
e1000097. Doi: 10.1371/journal.pmed1000097.
Morgan, R., A., Dudley, M., E., Wunderlich, J.,R., Hughes, M., S., Yang, J., C., Sherry, R., M., Royal, R., E., Topalian, S., L., Kammula, T.,U., S., Restifo, N.,P., Zheng, Z., Nahvi, A., de Vries, C., R., Rogers-Freezer, L., J., Mavroukakis, S.,A and Rosenberg, S., A. (2006). Cancer Regression in Patients After Transfer of Genetically Engineered Lymphocytes. Science immunology 316, 126-129.
Morin,C., M., Vallières, A., Guay, B., Ivers, H., Savard, J., Mérette, C., Bastien, C. & Lucie Baillargeon (2009). Cognitive Behavioral Therapy, Singly and Combined With Medication, for Persistent Insomnia A Randomized Controlled Trial. The journal of
the American medical association,301 (19), 2005-2015 .
Neyeloff,J., L., Fuchs, S., F. & Moreira, L., B. (2012). Meta-analyses and Forest plots using a
Microsoft excel spreadsheet: step-by-step guide focusing on descriptive data analysis. BMC research notes 5 (52), Doi: 10.1186/1756-0500-5-52.
25
Reiche, E., M., V., Vargas-Nunes, S., O. & Morimoto, H., K. (2004) Stress, depression, the
immune system, and cancer. The Lancet oncology 4, (10), 617-625.
Robichoud, M., Dugas, M. & Conway, M. (2003). Gender differences in worry and
associated cognitive-behavioral variables. Journal of anxiety disorders 17 (5) 501-516. Savard,.,J., Simard, S., Giguere, I., Ivers, H., Morin, C., H., Maunsell, E., Gagnon, P.,
Robert., J. & Marceau, D. (2006). Randomized clinical trial on cognitive therapy for depression in women with metastatic breast cancer. Psychological an immunological effects. Palliative and supportive care 4, 219-237.
Segerström, S., C. & Miller, G., E. (2004). Psychological Stress and the Human Immune System: A Meta-Analytic Study of 30 Years of Inquiry. Psychological bulletin, 130
(4), 601-630.
Shapiro,D., A., Barkham, M., Rees, A., Hardy, G., E., Reynolds, S. & Startup, M. (1994). Effects of treatment duration and severity of depression on the effectiveness of cognitive behavioural and pschyodynamic-interpersonal psychotherapy. Journal of
consulting and clinical psychology 62 (3), 522-534.
Tatrow, K. & Montgomery, G., H. (2006). Cognitive Behavioral Therapy Techniques for Distress and Pain in Breast Cancer Patients: A Meta-Analysis. Journal of behavioural
