The Effect of Daily Self-Measurement of Pressure Pain Sensitivity Followed by Acupressure on Depression and Quality of Life versus Treatment as Usual in Ischemic Heart Disease: A Randomized Clinical Trial

Background Depressive symptoms and reduced quality of life (QOL) are parts of the chronic stress syndrome and predictive of adverse outcome in patients with ischemic heart disease (IHD). Chronic stress is associated with increased sensitivity for pain, which can be measured by algometry as Pressure Pain Sensitivity (PPS) on the sternum. Aim To evaluate if stress focus by self-measurement of PPS, followed by stress reducing actions including acupressure, can decrease depressive symptoms and increase psychological well-being in people with stable IHD. Design Observer blinded randomized clinical trial over 3 months of either intervention or treatment as usual (TAU). Statistical analysis: Intention to treat. Methods Two hundred and thirteen participants with IHD were included: 106 to active treatment and 107 to TAU. Drop-out: 20 and 12, respectively. The active intervention included self-measurement of PPS twice daily followed by acupressure as mandatory action, aiming at a reduction in PPS. Primary endpoint: change in depressive symptoms as measured by Major depression inventory (MDI). Other endpoints: changes in PPS, Well-being (WHO-5) and mental and physical QOL (SF-36). Results At 3 months PPS decreased 28%, to 58, in active and 11%, to 72, in TAU, p<0.001. MDI decreased 22%, to 6.5, in active group vs. 12%, to 8.3 in TAU, p = 0.040. WHO-5 increased to 71.0 and 64.8, active group and TAU, p = 0.015. SF-36 mental score sum increased to 55.3 and 53.3, active and TAU, p = 0.08. Conclusions PPS measurements followed by acupressure reduce PPS, depressive symptoms and increase QOL in patients with stable IHD. Trial Registration ClinicalTrials.gov NCT01513824


Summary
Background: Stress and depression has been found to be mutually linked as well as to have an independent prognostic influence on the survival of heart patients. A new biological measure measuring the Pressure Pain Sensitivity (PPS) of the chest bone has been found to be a reliable marker for physiological stress and to be linked to depression and quality of life. An intervention which uses the PPS measure as a biological feedback marker for stress has been found to reduce the level of stress as well as the level of depression and increase life quality.

Objectives:
The primary objective of the study is to examine, if a simple individualized self-care based intervention, using the PPS as a biofeedback marker for stress, reduces the level of depression and stress and improves life quality in patients with ischemic heart disease.
Design: A prospective, randomized single-blinded trial in which the effect of self-care based intervention program in addition to that of best practice in cardiac rehabilitation (REHAB + ULL) is compared to a control group who only receives the best practice in cardiac rehabilitation (REHAB + CARE) over a 3 month period. The study will be open to the patient and the treating professionals but blinded towards the evaluating researchers.
Patients: 300 patients with ischemic heart disease, who have been through a cardiac rehabilitation program at Gentofte University Hospital and Herlev University Hospital respectively.

Effect variables:
The primary effect variable is the degree of depression. The secondary effect variables are the level of stress measured by PPS and quality of life.
Treatment: All patients have been through best practice Cardiac Rehabilitation. Half of them, allocated by randomization, receive the self-care based intervention program in addition, which focuses on daily PPS measurement for cognitive reflection and nerve stimulation for stress reduction.
Ethics: All patients continue with their standard cardiac rehabilitation programme as before. The self-care based intervention programme is documented as a risk-free intervention.
Project organization and flow: Jens Faber is the principal investigator. The medical evaluation of the patients will be conducted by an unbiased and blinded nurse in collaboration with a doctor. Total project period is one year.
Perspective: If the present study meets the success criteria, it is concluded that the

Stress, depression and ischemic heart disease
In recent years there has been an increased focus on the link between heart disease, depression and chronic stress (Grippo and Johnson, 2009). Evidence of a coincidence of significant metabolic changes for the following conditions has emerged: hypertension, heart failure, obesity, diabetes mellitus, metabolic syndrome, chronic stress, depression and other mental disorders (Koschke et al., 2009;Szczepanska-Sadowska et al., 2010).
The prevalence of depression, as measured as a MDI score ≥ 20, was found to be approximately 20 % in patients with ischemic heart disease (Olesen 2003;Bech 2001).

Stress measurement
So far there is no international consensus on biochemical diagnostic methods for measurement of transient and persistent stress (Ekman and Lindstedt, 2002;Noble, 2002). This is partly because it is difficult to identify a single physiological measure of stress as stress involves several factors, each of which that has been used as markers of stress in research. These are: 1) objective environmental stimuli, 2) the individual's perception of stimuli, 3) coping mechanism and 4) physiological reactions. These four factors are not necessarily present simultaneously, which has complicated the measurement of stress further (Holmes et al., 2006).

Biological warning systems and stress (the PPS measure)
Biological warning systems have always been essential for the survival of living organisms (Kumazawa, 1998a). The ability in animals to discover something in the environment that is a potential threat is developed to a nociceptive system, which involves a reflex making the animal pull back from potential threatening stimuli. The nociceptive system is based on a polymodal receptor, an undifferentiated nerve cell of the same kind as throughout the evolutionary chain from fish to higher vertebras and humans (Sneddon et al., 2003). The receptor is stimulated by mechanic pressure, temperature and acidity.
The sensitivity of the receptor is susceptible to various forms of modulation. One example is stress-induced analgesia (pain relief) where the pain sensation is suppressed) (Lewis et al 1980). This analgesia helps the wounded animal suppress general pain sensations, thereby letting the animal continue its fight or flight.
Another example is stress-induced hyperalgesia (pain hypersensitivity) where the pain sensation is increased. This is reflected in animal studies as an increased sensitivity in paw-pressure tests and tail flick test (Meagher et al., 2001;Pilcher and Browne, 1983).
Clinical observations of heart patients show that an increased tonus in the sympathetic nerve system causes an increased tenderness in certain areas on thorax. A pilot study of 250 healthy people showed a significant connection between the degree of tenderness in certain areas on thorax and the numbers of clinical stress signs. These observations have been confirmed later in a crosssection study .
Such observations have not previously been described as a part of the warning system in humans, but from an evolutionary biological perspective the survival potential improves by both aspects of stress-induced pain modulation -and through a mechanism that can be detected as far back in the evolution as the fish (Sneddon et al., 2003).

1.4.1.1
The sensor, the PPS measure and known markers of stress Polymodal sensory nerves are responsible for nociception and send their impulses through small myelinized A-delta fibers or unmyelinized C-fibres (Kumazawa, 1998b). They are sensitive to sympathetic input (Nilius et al., 2004). The polymodal sensory nerves in the chest bone have proved to increase their sensitivity during stress (Ballegaard et al., 2009). This can be measured by a new instrument that registers the sensitivity to pain caused by pressure (Pressure Pain Sensitivity, PPS).
The PPS measure is simple and has proved reproducible both when measured by the individual and the professional (Ballegaard et al., 2009), and repeated measures do not seem to cause systematic changes conditional to the habituation of the measurement technique . This means that the PPS measure can become a new tool for stress management: Daily measurements followed by reflection and action as when diabetes patients measure their blood sugar. An intervention program for personal PPS based biofeedback guided stress management has been developed (see points 1.5 and 9.2.2).
The new medical measure has been shown to be linked to known markers of transient stress such as pulse, blood pressure, salivary cortical and work of the heart measured as Pressure-rate-Product (PRP) (Ballegaard et al., 2009). PRP has been shown to be linked to stress in ischemic heart disease patients with ischemic heart disease (Jain et al., 2001).
In 292 office workers the PPS measure has proven to be linked to both the number of clinical stress signs and the prevalence of depression, as measured by the IC 10 Major depression Inventory (MDI) regarding markers of persistent stress . Depression has been shown to have a prognostic influence on ischemic heart disease patients (Surtees et al., 2008). PPS was also linked to general mental and physical health measured by the SF-36 questionnaire ) -a questionnaire that reflects the quality of life in patients with ischemic heart disease (Dempster and Donnelly, 2000;Oldridge et al., 2002).

Validations studies of the PPS measurement itself
In collaboration with researchers from The National Research Center for the Working Environment, Columbia University in New York and University of Copenhagen, a number of validation studies of the PPS measurement have been made in order to enlighten the following: 1. The reproducibility; the ability to reproduce the same measurement values in repeated measurements, was examined in two distinct situations with respectively 5 seconds or a full day between the two measurements.
2. Categorization ability and its reproducibility; the ability to identify groups of people with a higher health risk than another group and the ability to reproduce this categorization.
3. The PPS measure's specificity and sensitivity when other health markers are used as categorization variables.

4.
Risk assessment regarding to the systematic changes in the PPS measure caused by habituation to the measurement technique.
Ad 1: Reproducibility a) Studies with 5 seconds between the two PPS measurements a. In 181 patients in an outpatient clinic the measurements were conducted by a professional and showed high consistency between the first and the second measurement (correlation coefficient r = 0.97, p < 0.001) with a mean difference of 0.3 units (p > 0.5), and a standard deviation (SD) of 5.7 units.

b.
A high consistency between the first and second measurement was also found when ten different professionals measured 308 office workers (r = 0.90, p < 0.001; mean difference 1.32 units, SD = 7.40 units) .
c. In a study of 35 opera singers who measured themselves two times daily for a two week period, a total of 698 sets of PPS measurements (r = 0,95, p < 0.001; mean difference = -0,32 units, SD = 7,6 units).
b) A Bland-Altmann analysis on the studies mentioned above (a, b and c) shows an uniform measurement difference between the first and second measurement on the entire scale .
c) Studies with a full day between the two PPS measurements: a. There was a significant correlation between a total of 318 sets of daily measurements (morning and same-day evening) in the study with 35 opera singers who measured themselves twice a day for two weeks; correlation coefficient r= 0.91 (p < 0.0001); mean difference: 2.2 units (SD = 10,5 units)(p < 0.001).
b. In another study where 60 office workers measured themselves at home for three months, there was also a significant correlation between a total of 1854 sets of daily measurements: correlation coefficient r = 0.87 (P < 0.0001), mean difference + 2.2 units (SD = 8.9 units) (p < 0.001).

Ad 2: Categorization
a) The PPS measurement's ability to make a 2-parted classification of people in "persistent stressed" (PPS ≥ 60) and "non-persistent stressed" ( PPS < 60) was examined through a ROC analysis of the 292 office workers  compared to an analog 2-parted classification "having an elevated health risk marker" and "not having an elevated health risk marker" for the following effect variables and based on the following discrimination criteria: 1) the number of clinical stress signs (less than 10/ 10 or ii. The probability of an equivalent uniform classification among the 318 day measurements made by 35 opera singers during two weeks was 83 %.

Ad 3: The PPS measurement's specificity and sensitivity
When the discrimination criteria mentioned above are used for a two-parted categorization regarding grouping from a PPS measurement that is 60 units, the specificity was 0.79 and the sensitivity 0.32 was found in terms of the prevalence of clinical stress signs, mental and physical health measured by the SF-36 and the prevalence of depression measured by the MDI. The result is deemed acceptable since the high specificity ensures that a large percentage of the people not in need of treatment are not offered. The low sensitivity can cause some people to be offered treatment even though the need is not present, but since the treatment is harmless and without side-effects, it is deemed acceptable.

Ad 4: Risk assessment in regard to habituation bias of repeated PPS measurements
The studies hitherto implemented cannot definitely exclude that measure adaption affects the measurement result when PPS is measured daily over a period of time, but the following studies suggest that such influence does not have decisive significance in patients' everyday clinical practice.
a) A review of the scientific database PUBMED was unable to provide scientific evidence that the phenomenon exists, which in itself obviously does not mean that the phenomenon does not exist.
b) In two case-control database studies in patients with ischemic heart disease and stroke respectively and at 3 to 4 years of observation, the individual patient used the daily PPS measurement as a biological biofeedback marker of stress in regard to cognitive reflection on his/hers stress level. The results showed a significant decrease in the use of health services and an improved survival rate in both studies (Ballegaard et al., 2004;Magnusson et al., 2010).
c) In two randomized intervention studies with respectively 50 opera singers and 308 office workers and with respectively 6 and 3 months of observation, a decrease in PPS was linked to significant and clinical relevant improvements in known risk factors such as blood pressure, pulse rate, work of the heart and serum cholesterol .

Compliance of domestic PPS measuring
Compliance was 90 % in the study with 35 opera singers, who measured themselves twice a day for two weeks, whereas it was 20 % among the 60 office workers, who measured themselves daily for three months. Given the results achieved among the 60 stressed office workers, it does not seem a concern that their measure compliance was only 20 % measured over a three month period The reasons for the lower compliance can be many: 1) that not all measurements are entered, 2) that measurements are not entered when they have been low for a long time (e.g. < 40), 3) that in the long term it is not necessary to measure daily, but maybe just a few times per week or whenever the need for cognitive reflection is present.
SF-36 is a questionnaire that measures quality of life and has shown to be a valuable prognostic tool in patients with ischemic heart disease (Dempster and Donnelly, 2000;Nishiyama et al., 2005;Oldridge et al., 2002). In addition, the SF-36 has been used in large population studies to assess the long term health damaging effects of work related stress (Stansfeld et al., 1998). A connection between SF-36 and early retirement from the labor market has also been found (Harkonmaki et al., 2006).
WHO-5: Measures well-being. A study has shown that those heart patient who have a low score (< 50) on the WHO-5 questionnaire has a significant higher mortality within six year than patients with a high score (> 50) (Birket-Smith et al., 2009).

Metabolic variables and stress
Persistent stress has been associated to the development of Metabolic Syndrome R 2004). MES is a cluster of risk factors for the development of cardiovascular diseases including 1) reduced glucose tolerance/diabetes type II, 2) abdominal fat deposition, 3) dyslipidemia (disturbed lipid metabolism, and 4) hypertension (high blood pressure). These risk factors are considered a significant health problem in the western world and affect 40 million Americans (Farmer, 2004;Ford, 2005).
Insulin resistance is regarded as one of the key factors in the development of MES.
Moreover, a review by Sutherland et al. 2004 points that MES is associated or caused by chronic low-grade inflammation due to activation of the immune system.
Chronic low-grade inflammation is probably also a relevant factor in the development of atherosclerosis disease in the heart (Pradhan and Ridker, 2002).
Insulin resistance, MES and chronic low-grade inflammation are all associated with endothelial dysfunction, which is characterized by blood vessels becoming stiffer and changes in the blood vessels' structure. Seematter et al. (2004), finds that these changes are connected with the development of atherosclerosis.

Blood pressure as a marker of stress
Obesity, insulin resistance, hypertension and dyslipidemia (all seen in MES), are each a risk factor for cardiovascular disease and early death (Farmer, 2004).
The increase in blood pressure often seen with increasing age has shown to be related to persistent stress (Timio et al., 1997).

Heart-rate variability and stress
Decreased heart-rate variability (HRV) has proven to be associated with increased stress and bad prognosis for heart disease. In studies with a stress reducing intervention, HRV has improved (Del Pozo et al., 2004;Nolan et al., 2008;Routledge et al., 2010).

Pulse and Pressure-rate Product as stress markers
Resting pulse has proven to have a prognostic value in ischemic heart disease (Lanza et al., 2006), and the link has proven to be the degree of persistent stress (Rogowski et al., 2007). Pressure-rate Product (PRP) is the product of the systolic blood pressure and the pulse, which is used in cardiology as an index of heart's ability to obtain oxygen and therefore reflects the work of the heart. The work of the heart is mainly controlled by beta-adrenergic catecholamine receptors (Opie, 2001), which makes the PRP suitable as an indicator of the sympathetic tonus in the heart (Noble, 2002) and suitable as a prognostic marker of ischemic heart disease (Villella et al., 1999).
A previous study has shown a significant link between resting pulse, PRP and the PPS measure (Ballegaard et al., 2009), yet we have opt to use blood pressure, pulse, Pressure-rate product and serum cholesterol as primary and secondary variables in the Songheart study because the study's primary target group are receiving medication that affects these variables.

Control system (DNIC)
Changes in DNIC might in part explain the increased pain sensation among chronically stressed people (over reaction of the ascending stimulating nerve signal in combination with a reduced descending and inhibitory nerve signal (Arndt-Nielsen et al. 2010). This fits with a plausible hypothesis for the effect of daily nerve stimulation (acupressure), where the persons own stimulation followed by control for reducing tenderness seems to result in a restitution of the lost DNIC function, typically for the chronically stressed person.

The effect of the intervention that uses PPS as a biofeedback marker of stress
The use of PPS as a biofeedback marker in combination with a multifaceted stress handling program has been validated in two bloc-randomized controlled experiments; one smaller with 50 opera singing students  and a larger with 292 office workers .
In the first study 50 opera singing students from two different schools were randomized either to an active treatment group or a control group. The active treatment is comparable to the treatment that is used in this study, while the control group had a two hour lecture on general stress handling. Compared to the control group the active group obtained a decrease in pulse (14 %), a decrease in work of the heart measured as PRP (14 %), a decrease in the number of clinical stress signs (67 %), in PPS (35 %) and in the total number of elevated health risk markers (60 %) (all p < 0.05) .
In the subsequent larger study on 292 office workers, the participants were divided into two groups based on a PPS measurement; very stressed (PPS ≥ 60) and little to moderate stressed (PPS <60). The very stressed were bloc-randomized and placed in either an active treatment group where they receive a PPS measuring instrument plus a personal guide to the programme in a three month period, or a control group. The results showed that when the active group was compared to the control group, the intervention had had a clinically useful and scientific significant effect in blood pressure (10 %), work of the heart (10 %), serum cholesterol (13 %) and PPS (42 %) (all p < 0,001) and with an 80 % response rate .
In a prospective uncontrolled study on a random selected group of women with breast cancer, the effect of the intervention gave the following results on PPS (Axelsson et al., 2011). The average change was from 78 to 56 in PPS measure, compared to an average change from 74 to 49 among the 60 office workers in the active group of the previously discussed study . These women also experienced a significant improvement in level of anxiety and depression, and the improvement in PPS was linked to the improvement in level of depression. Furthermore the feasibility was high, more than 90 % of the women were still using the program six months after start (Axelsson et al., 2011).
A prospective, uncontrolled, clinical case-control database study of 10 years with 160 patients with ischemic heart disease (all were candidates for invasive treatment), who used finger palpated tenderness on the sternum as a biofeedback measure of PPS in combination with a comprehensive self-care based stress management program, showed that the intervention had a long lasting beneficial effect on mortality and healthcare expenses (Ballegaard et al., 2004). The three year risk of dying was significant lower than the general Danish population's and there was a considerable decrease in use of medication, number of hospitalization days, and a decreasing demand of invasive cardiologic treatment, which overall led to an approximately 70.000 DKK reduction of the yearly health expenses per patient (based on the costs ten years ago). These patients started with three weeks of acupuncture treatment while the self-care part was the most important contribution for the rest of the observation period (Ballegaard et al., 2004;1996).    It is expected that about 450 will accept. These are convened to an examination at Herlev Hospital Medical Department O.
All patients sign the consent form and go through the physiological preliminary examination.
Patients with PPS < 60 are ended immediately, and a subset of the first 50 patients with PPS < 45 are invited to a measuring of their body composition.
Since it is expected that 2/3 of patients have an elevated PPS measure (≥ 60), 300 patients with PPS ≥ 60 are randomized to either an active or control group with 150 in each: Control group: Continue their own efforts in extension of the already given cardiac rehabilitation and receive a textbook on general stress management (REHAB + CARE).
Active group: the same plus the Ull treatment program (REHAB + ULL).
All included patients will at baseline be offered an additional measurement of body composition.
Both the control and active group continue with the given medical treatment. If there is a change in heart condition including possible side effects of medication, the medication is adjusted according to the procedures in the rehabilitation programs at the hospital the patient is assigned to (Gentofte or Herlev Hospital). If possible, the medication should remain stable during the 3 months of the observation. The patient and the patient's own doctor are informed that the aim of the study is that any adjustment of cardiovascular medicine happen after contact with the project nurse (perhaps through a telephone hot-line).

Control/Active group:
1. Control group: receive a book about general stress management 2. Active group: receive in addition to the above mentioned: Instruction in self-measurement of PPS by a PPS instructor at home. The patient is instructed in PPS measurement, nerve stimulation and the use of the web platform in order to be able to enter their own measurements. They are not instructed in stress management. There is a 45 minute home visit in the beginning plus two visits after one and two months respectively for additional technical instruction if the PPS measure is ≥ 60. In addition there are 5 telephone dialogues á 15 minutes solely for technical instruction after 1, 3, 5, 6 and 10 weeks respectively. The patients receive an Ull meter, a written user's manual and an instructional DVD.
The patients record their PPS measurements either on the web platform (www.songheart.org) or on paper at least once a week If their PPS records show a PPS ≥ 60 for 7 consecutive morning measurements, the patient is contacted regarding the need for additional technical instruction.
After three months of participation in the project all patients (both groups) are called to a repetition of the baseline visit at Herlev Hospital. Those who at baseline opted to have made body composition will have the same offer again if, at the physiological examination, they have lost 3 kg or more.
The two examinations at Herlev are conducted by the same blinded nurse at baseline and the 3 month follow-up respectively.
If the success criteria for the study are met, the subsequent analysis are carried out after the 12 month observation: Use of health services over the last 12 months before the start of the study compared to the 12 months after commencement of participation.
The prevalence of clinical events. Here a Kaplan-Meier plot analysis will be used.
The Minimal Important Difference (MID) between the active and control group is set at a 10 % improvement in the active group versus the control group.

Randomization procedure
The treatment is assigned randomized and with an equal number of patients in the control group and for active intervention. An independent research company electronically conducts the practical randomization (www.larix.dk).
The treatment is allocated by a minimization procedure with built-in randomness (Pocock and Simon, 1975). The minimizing will ensure an approximate balance in the number of patients in both treatment groups with respectively low and high values of the baseline parameters taken into account in the procedure. Pocock and Simon's minimization method (Pocock and Simon, 1975) will be used with a randomness parameter of 0.8. The following baseline parameters will be included: Age (

Th
The ba 1.

2.
3. self-care in the form of PPS measuring, cognitive reflection and nerve stimulation, which is linked to professional technical help depending on need.
The elements of the self-care program An individual start-up of a 45-minutes face-to-face consultation at home with an Ull Care instructor who explains the use of the instrument, the interpretation of the PPS measurement, nerve stimulation and the web platform.
A personal Ull Meter instrument.
A personal web journal for track recording of the participant's PPS measure's development (patients without internet access will receive printed registration sheets).
A booklet that describes the self-care program in details.
Two additional face-to-face meetings with additional technical instruction should the PPS measure remain ≥ 60 after 1 and 2 months respectively. 5 x 15 minutes of telephone consultations after 1, 3, 5, 6 and 10 weeks respectively.
Only the following four subjects may be discussed: the PPS measure, interpretation of the measurements, nerve stimulation and the use of the web journal. This is to ensure a correct use of the Ull Meter instrument and a correct management of the nerve stimulation technique.
Online assessment of the participant's PPS measurements. This assessment is made automatically on the website. If the morning PPS measure has been ≥ 60 for more than seven days in a row, the participant is offered additional technical support.
The daily steps of the self-care program Four elements constitute the daily self-care system: measuring, cognitive reflection, action and control.
Measuring: PPS is measured twice a day as a fully private measure and is entered in a personal web journal, so that the person can track his/hers own development, while at the same time there is a regular extern professional control.  -Moberg, 1997a). If possible, the spouse or cohabitant will be instructed to perform nerve stimulation on the patient's back, which may independently contribute to stress-relieving effects through a separate oxytocin release caused by the human care (Carter et al., 2007;Grewen et al., 2005;Light et al., 2005;Uvnãs-Moberg, 1997b). If a spouse or cohabitant is not available, the patient can perform the treatment himself with a ball (the person will receive instructions in the technique).

Joint intervention for both the active and control group:
All participating patients receive a book on general stress management ("Stress, Laegens Bord" Netterstrøm, Bo, DR, 2002).

Blood sampling and analysis:
Blood samples are taken locally at Herlev Hospital where they are stored at -80 degrees until analysis.
After analysis, and by June 1st 2013, the debris is destroyed.
All blood tests will be performed at Medical Department O, Endocrinology Unit, Research Lab 54O4, Herlev Hospital, either as a routine or as a research analysis through commercial analysis-kits. All samples are analyzed in the same assay to avoid inter-assay variation, i.e. the analyses are first performed after completion of the study.
The following is analyzed (baseline study and intervention study

Body composition:
The body composition is done using whole-body DEXA scan (apparatus: Hologic) for the determination of body composition in fat, muscle and bone mass.
Furthermore, the visceral fat is calculated using new software to distinguish between android and gynoid fat. A well-calibrated x-ray is used for the scanning of the patient.

HRV post test:
The subjects get their short-term HRV (5 minutes) measured by an EKG (heart electro diagram) in supine rest. The measurement is done with the participant lying on a folding bed. After the rest, the participant is tilted without using his or her own muscle power to a 70 degrees (upright) position. EKG is registered in this position.
The difference between resting and standing represents the reactivity of the autonomic nervous system. Lower reactivity in the autonomic nervous system is associated with persistent stress. The measurement of HRV is non-invasive and is not associated with discomfort or any risk for the participants.

Algometry on the Tibia bone with an established Algometer
Pain threshold will be measured at the Tibia bone with both the PPS measure device used in this study as well as with a known and established algometer (Arndt-Nielsen et al. 2010). Fifty randomly selected patients will be studied during their follow-up visit after 3 months.

Time frame for the study
• The first patient is expected to begin 3 months after the financial resources has been provided, and after approval by the Scientific Ethics Committee and the Danish Data Protection Agency.  the active and control group respectively will be compared to analyze a possible treatment effect. All secondary and tertiary effect variables will be analyzed after the intervention through variance and regression analysis. If the premises for the proposed statistical analysis cannot be met, non-parametric analysis is used.

P
Quality-of-life data will be analyzed to meet the directions of each of the questionnaires. All effect data will be presented a unifying table and primary, secondary and tertiary effect variables will be listed.

Cost-Utility analysis (CUA)
A cost-utility analysis will be carried through after 12 months of intervention for comparison of active versus control treatment. Relevant data, meaning consumption of healthcare services such as hospital visits, ambulatory visits, use of medication, consumption of invasive procedures etc. will be gathered during the study. Data from SF-36 questionnaires will be used to generate OALY's based on the fact that

Additional studies
It is the intention to repeat the Songheart study with an identical design in Israel with the purpose of subsequently being able to add data from the two studies together and thus increase the overall statistical power.

Analysis of substudies:
Separate analyses will be performed with regard to: The effect of intervention in the following subgroups: a. Age (<65, ≥65) b. Gender c. Heart failure (yes,no) d. Diabetes (yes,no) e. Time since last MI or invasive procedure: <2 yr, ≥2 yr.

Success criteria
The result of the study will be concluded as positive, if the following results are reached: Compared to the control group, the active group obtains a significant improvement in MDI depression score and with a therapeutic effect size of at least 0,4 when the effect in the active group is compared to the control group (see point 11.1 for exact calculation of the effect size).
When compared to the control group, the active group obtains a significant improvement in mean PPS measurement measured by a professional instructor and with a MID of 10 units on the PPS scale.
The patient does not experience any increased risk in connection with the intervention measured as no increase in blood pressure or serum cholesterol after treatment, Both parts are measured as mean values for the two variables in the active group when the baseline values are compared at the end of the treatment period.

Usefulness of the study and treatment in a broader perspective
By participating in the study, each patient can get a general profit from the opportunity of an additional comprehensive health check. If the findings of the study are positive, it means that the patients in the active group have an extra benefit in terms of reduced stress, improved mood and some concrete tools for better stress management. If renewed fund applications allow it, the control group will then be offered the active treatment. The study gives the surrounding society the possibility of testing the value of a new and economically inexpensive treatment in heart patients: A treatment that is expected to have a measurable impact on the patients' future health, use of health care services and patient survival. Other scientific studies have shown that if stress and depression is reduced in a heart patient, their survival is improved. As the intervention represents a paradigm shift in the treatment of chronic illness with an emphasis on a far greater self-insight and self-effort, a positive outcome will expectedly lead to a test of the intervention in people with other chronic illnesses. The study is also designed to uncover possible links between persistent stress and heart disease and which mechanisms that is responsible for the effect that a positive outcome will show.

Ethics
The study depends on an approval from the Scientific Ethics Committee. After approval, the study will be reported to www.clinicaltrials.gov. The study is also reported to the Danish Data Protection Agency. The study complies with the Helsinki Declaration.
The study will meet Danish medical guidelines of good clinical practice.
All subjects are informed orally and in writing about the study. The investigator will obtain informed written consent. All subjects receive the booklet "Your rights as a test subject in a biomedical research project", written "Participant information" and an offer of a personal oral information session in the initial letter with an invitation for participation.
There will be emphasis on the fact that the information session is about a query of participation, and that the subject has the right to bring an accompanying person.
The subject's right not to receive information on his/hers health condition and disposition to disease will be reviewed and respected.
If during the conduct of the trial new information on effect, risks and side effects appear, the subjects will be informed and a new informed consent may be obtained.
If the subject gives written permission, a spare blood sample glass (5 ml) will be consented and what is referred to in this protocol will follow the usual guidelines, including a renewed reporting to the ethics committee.
Approximately 40 ml of blood per time (two times total) is taken. This includes the spare blood sample glass.
There will be no financial compensation to subjects.

Risks and rights of the participants
The patients are not deprived of other known effective treatment that are offered to similar patients at hospitals in Copenhagen, and participation in the study does not interfere with the patient's current treatment.
Blood samples are done through vein puncture and will be performed with standard equipment, which leads to a known and limited discomfort.
The DEXA scan uses low x-rays. The radiation dose is approximately 0,1 mSv per examination, which is the same as approximately 3 % of the natural background radiation every year.
By measuring the pressure threshold, some soreness or bruises may occur if the participant does not follow the instructions and says stop or stops in time.
The HRV position testing is non-invasive and without discomfort or risk to the person. The given active intervention is risk-free. Both control and active group receive a book on general stress management.
The subjects are informed that they have the opportunity to complain to the Patient Complain Board and seek compensation through the Patient Insurance Association.

Conflicts of interest
Søren Ballegaard is a shareholder in Ull Care A/S and is the inventor of the Ull Meter.
To minimize possible bias, he shall not participate in 1) patient selection, 2) patient instruction and contact (he well only supervise the treatment performed by others),

3) evaluation of the treatment
The responsible researcher is Professor Jens Faber, Medical Department O, Herlev Hospital, and he is associated with Ull Care A/S as an independent, unpaid and external scientific advisor. Jens Faber does not own shares or other financial interests in Ull Care A/S and has never received funding/fees from the company.
No other participants are associated with Ull Care A/S and nobody has any economic benefit form the study.

Publications
Both positive and negative results will be compiled in one or more scientific articles and sent to international peer-reviewed publications regarding publication. The steering group will be co-authors of all the scientific articles. Affiliated researchers will be co-authors on the articles where they have contributed.

Use of data and notification of the Danish Data Protection Agency
All data rights regarding scientific evaluation and publication are owned by the project responsible researcher, and the project is reported to the Danish Data Protection Agency after the approved guidelines of Herlev Hospital.
Raw data will in anonymized form and after appointment with the principal investigator be accessible for all the affiliated institutions for use in research projects. All publications must be confirmed by the primarily responsible researcher Jens Faber.

Economy
The study seeks funding by both private and public funds. The total budget is 3.5 million DKK.

Organization and responsibility
Jens Faber is the principal investigator.
The steering group consists of Per Bech, Åke Hjalmarson and Søren Ballegaard.
Søren Ballegaard is included in the steering committee because of his special competence in relation to the active intervention included in the study. Jens Faber is an adjunct member of the steering group and therefore has no voting rights.