| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the School of Psychology, University of St. Andrews, St. Andrews, Fife, Scotland.
Address reprint requests to: Marie Johnston, PhD, School of Psychology, University of St. Andrews, St. Andrews, Fife, KY16 9JU, Scotland.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: A randomized controlled trial with follow-up to 1 year was conducted with 100 patients recruited within 72 hours of a first MI and their partners: a Control group received normal care; an Inpatient group received cardiac rehabilitation from a nurse counselor while in hospital; and an Extended group received the same cardiac rehabilitation as the Inpatient group, but with additional sessions continuing up to 6 weeks after discharge from hospital. The scales for main outcome measures were 1) knowledge of heart disease and treatment (correct, misconceptions, and uncertainty); 2) mood (Hospital Anxiety and Depression Scale); 3) satisfaction; 4) disability (Functional Limitations Profile).
RESULTS: Inpatient cardiac counseling and rehabilitation resulted in more knowledge, less anxiety, less depression, and greater satisfaction with care in both patients and partners and in less disability in patients, with effects enduring to 1 year. There was some evidence of additional benefit from the Extended program. Both nurse counselors achieved benefits on all outcome variables.
CONCLUSIONS: This Inpatient cardiac counseling and rehabilitation program resulted in significant and enduring benefits of clinical value. It is likely that it would be acceptable to most post-MI patients, many of whom are not offered or are unable to
Key Words: cardiac rehabilitation, • cardiac counseling, • myocardial infarction, • anxiety, • depression, • partners, • disability.
Abbreviations: MI = myocardial infarction;; HADS = Hospital Anxiety and Depression Scale;; CCU = coronary care unit;; FLP = Functional Limitations Profile;; WHO = World Health Organization;; ANOVA = analysis of variance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Arguments for including the patients’ partners are both practical and therapeutic. Partners can be incorporated in the patients’ program with little additional cost. It is likely that the partner’s attitudes to the patient’s MI can affect recovery through being over concerned and protective (8), and, particularly for male patients, the spouse’s views on diet may well affect the reduction of cardiovascular risk factors. Perhaps as important as the possible gains for the patient are the benefits for the spouse. As one would expect, they are distressed after the partner’s MI (2) and might well benefit from the support, information, and enhanced feelings of control that they might experience by being included in the rehabilitation.
The objectives of the current research were to evaluate the effectiveness of a cardiac counseling and rehabilitation program run by a nurse for patients and their partners on a comprehensive range of psychological and functional outcomes. An additional aim was to examine whether an inpatient-only program was as effective as an extended program involving both inpatient and outpatient contact. Outcomes were examined in both patients and partners. The main outcome measures were knowledge of matters relating to heart disease, satisfaction with care, anxiety, and depression. Limitations in activities were also examined in the patients.
|
Research Questions |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
[Benefit being defined as increased knowledge, improved mood, greater satisfaction with care, and less disability/limitation in patients’ activities over the year after the MI.]
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Randomization
Simple randomization to cardiac rehabilitation and normal care groups was not possible because of the possibility of contamination effects and perceived "deprivation" of control patients mixing with cardiac rehabilitation patients on the same ward. On leaving the Coronary Care Unit, patients, therefore, were allocated to wards in accordance with current practice (ie, by gender and by day of the week) and the six wards were randomly allocated to have one of the two cardiac rehabilitation programs or to a Control group. To minimize the confounding of groups with wards and to ensure that interviewers were kept blind to the patient’s allocation, the ward allocation was changed at variable intervals of approximately 2 months. Each randomization involved a clearance period during which time the patients on the earlier regimen were discharged. Randomization was directed by M.J. who was not involved in patients’ care or outcome assessment.
The Cardiac Counseling and Rehabilitation Programs
There was no formal program available for the normal care Control group. The patients in the cardiac rehabilitation group and their partners were offered a program of education and counseling that began within 3 days of admission and was planned to involve up to five sessions for the Inpatient group with an additional possible eight sessions over the 6 weeks after discharge for Extended group patients. Data on actual sessions indicated that the Inpatient group received an average of 5.55 sessions (SD = 0.82) with an average total of 3.69 hours (SD = 0.19) whereas the Extended group received an average of 9.55 sessions (SD = 0.57) averaging a total of 8.43 hours (SD = 0.48).
A Nurse Counselor introduced the program after baseline assessment by the research assistant who was blind to group allocation. Two nurse counselors were involved: one for the first 11 months of the project and the second for the remaining 7 months, so that the feasibility of introducing the methods to the second counselor using the counselor’s manual could be assessed and any counselor differences in effectiveness be evaluated. The manual (written by J.F.) outlined the methods and content areas [explanation of a heart attack, risk factors and their modification, emotional effects, recovery period, investigations, and treatment] and procedures designed to enhance patient control.
The program was introduced as nonjudgemental counseling, with a focus on helping the patient to achieve their objectives. The counselor arranged appointments to include the patients’ partner when possible; this was usually achieved after the first meeting between the patients and the counselor. Although the possibility of sessions being shared between patients had been built into the program, this was rarely possible in practice. The program was designed to address issues as and when they became relevant for the patient. The counselor provided information, guided action plans, gave advice, and provided leaflets and videos that were already available, ie, they were not produced for this program. The patients were encouraged to take control by contributing to their own assessment and plans. For example, the choice of topic was guided by the patient’s choice from the following written menu:
Menu
Information from the second nurse’s notes for 21 patients indicate that the explanation of MI and discussion of resumption of activities were selected by all patients except one who did not choose to have the explanation, perhaps because he had already received adequate information. Nearly all patients chose emotional effects and risk factors. The first topic chosen was most commonly the explanation of a heart attack (for 14 patients), but for a substantial number (6 patients) it was discussion of emotional effects. Risk factors discussed and the number of patients are: stress, 16; smoking, 13; cholesterol, 9; inactivity/exercise/sedentary lifestyle, 7; weight/obesity, 6; family history, 6; blood pressure, 2; diabetes, 1; and alcohol, 1. In the Extended program, topics discussed in hospital were similar to those discussed in the Inpatient group, but after discharge from hospital, explanation of MI was rarely selected and the focus was on resumption of normal activities and on risk factors.
The notes also indicate that an important part of the counseling was listening to current concerns of the patients, allowing them to express emotions, and then offering them emotional support. This was frequently part of the counselor’s initial discussion with partners.
Subjects and Setting
The subjects were consecutive patients admitted to the CCU at Ninewells Hospital Dundee, between January 1992 and February 1993. They were transferred from CCU to one of six medical admission wards serving the geographical region. The wards were identical in design and management and provided the standard nursing care that was common to all groups. The study was approved by Tayside Health Board Medical Ethics Committee and written informed consent was obtained. Inclusion criteria were: presence of a first MI (WHO criteria), aged 
70 years, fluent in English, able to participate and giving informed consent to inclusion in a randomized trial within 72 hours of admission. Partners (defined as the person the patient identified as being most involved in their recovery) were also invited to participate. Thirteen patients refused to participate and five who were initially recruited were subsequently excluded when the diagnosis changed (MI not confirmed by cardiac enzymes and/or EKG). Of 117 patients included and randomly allocated, 10 withdrew (6 Control, 3 Inpatient, 1 Extended); of these, 5 withdrew before 2 months assessment, 3 before 6 months, and 2 before 1 year. Seven died (2 Control, 5 Inpatient), leaving 100 patients who constitute the study sample. Those who withdrew or died did not differ from the remainder on age, gender, Norris Index (9), days in hospital, anxiety, or depression, although there was tendency for those who died to be older and have higher Norris scores. The reasons given for withdrawing included (one patient for each reason): difficulties at home, family illness, patient on renal dialysis, upset by questionnaires, moved away, unavailable, confusion over time of interview, did not wish to continue. Two patients withdrew with no reason given.
Group allocation and descriptive clinical and baseline data on the three groups are presented in Table 1. The severity of the participant’s myocardial infarction was assessed by the Norris Coronary Prognostic Index and the level of their general risk of CHD by the Dundee Coronary Risk-Disk. The Norris Index (9) is an empirically derived index that assesses the likelihood of short-term and 3-year survival after a MI. Short-term risk is determined from age, size and site of infarction, heart size, systolic blood pressure on admission, degree of congestion of the lungs, and a history of ischemia. Blood pressure and size and site of the infarction are excluded from the 3-year index. The Dundee Coronary Risk-Disk (10) is a standardized measure of modifiable risk factors. It is based on smoking, blood pressure, and serum cholesterol and provides age and gender standardized scores running from 1 = very high risk to 100 = low risk. Social class was categorized, using the Registrar General’s Classification (11).
|
Mood: Hospital Anxiety and Depression Scale (at Baseline, Discharge, 2, 6, and 12 Months).
The HADS (12) is a brief measure of mood designed to avoid confounding of mood with somatic symptoms. Anxiety and depression are each measured on seven 4-point scales giving scores ranging from 0 to 21. The scales are widely used in studies of patients with physical complaints and have some evidence of validity.
Satisfaction With Care (at Discharge, 2 Months).
Participants rated "How satisfied do you feel generally about the advice that you received after your/your partner’s heart attack?" from 1 to 10, 10 indicating "extremely satisfied" and 1 "not at all satisfied."
Disability and Resumption of Normal Activities: Functional Limitations Profile (at Discharge, 2, 6, and 12 Months).
The FLP (13) is the British version of the widely used Sickness Impact Profile questionnaire and assesses limitations or disablement due to the respondent’s health. It provides a total score as well as two subscale scores labeled the physical dimension and the psychosocial dimension.
Statistical Methods
Equivalence of groups on baseline measures was analyzed using one-way ANOVA. To investigate differences between groups over the study period, repeated-measures ANOVA was performed. If there was an interaction effect, this was the primary focus and, using simple effects and simple comparison analyses, the locus of this interaction was explored. Simple effects analysis was made at each time point with the overall means being studied while holding time constant. If this was significant, simple comparisons were performed to pinpoint where the differences in the means lay. Only if there was no significant interaction (or if it was comparatively weak) was the focus on the main effects and main comparisons, due to the possible confounding effect if there was an interaction effect. Nonnormal variables were transformed where appropriate. The Greenhouse-Geisser correction was calculated where the assumption of sphericity was violated (indicated "g").
The hypotheses led to two planned contrasts:
On knowledge and mood, where prerandomization baseline data were available, a group by time interaction was hypothesized, whereas on disability, where the first assessment occurred after discharge, a main effect of group was hypothesized.
On examination of the data, some post hoc tests of the Control compared with the Extended group were conducted where the omnibus test was significant. To account for the planned and unplanned contrasts and multiple comparisons, a Bonferroni adjustment was made to the critical value (14).
Where significant effects of cardiac rehabilitation were found, additional analyses were undertaken to assess whether the effects were obtained by each nurse counselor. For each nurse, planned contrasts were used to compare the combined cardiac rehabilitation groups (Inpatient and Extended) with the Control group (2-tailed).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Patient Outcomes
Knowledge.
There were significant group by time interactions for all three measures [correct, F(4,166) = 7.94, p < .0005; misconceptions, F(4,166) = 4.73, p = .001; uncertainty, F(4,166) = 4.04, p = .004]. For items correct (see Figure 1, A), both Inpatient and Extended groups had higher scores than the Control group at discharge (p = .004 and .002, respectively) and at 2 months (p = .003 and <.0005, respectively). For misconceptions, the Inpatient group had significantly higher scores than the Extended group at recruitment and so initial levels were used as covariate; results showed more misconceptions in the Control group than in the other two groups (p = .001 for Inpatient and p = .004 for Extended). Both cardiac rehabilitation groups had less uncertainty than the Control group at 2 months (p = .013 for Inpatient and p = .004 for Extended comparison). The Extended group had higher correct scores than the Inpatient group (p = .016) at 2 months.
|
|
|
|
|
Mood.
There were significant group by time interactions for both depression [F(5.7,134.3) = 3.08 g, p = .008] (Figure 2, B) and anxiety [F(5.7,133.1) = 2.28, p = .043] (Figure 3, B). On depression, the Inpatient group had lower levels than the Control group at 6 months (p = .02), whereas the Extended group had lower levels than the Control group at 2 months (p = .002) and 6 months (p = .0008). On anxiety, the Inpatient group had lower scores than the Control group at discharge (p = .016), and at 2 months (p = .02) and the Extended group had lower levels than the Control group at 2 months (p < .0005), 6 months (p = .0005), and 12 months (p = .002). The Extended group had lower anxiety than the Inpatient group at 2 months (p = .02) and 6 months (p = .02).
Satisfaction With Care.
There was a significant main effect of group [F(2,55) = 32.12, p < .0005] (Figure 4, B). Cardiac rehabilitation groups had higher levels of satisfaction than the Control group (p < .0005). The interaction was not significant.
Nurse Counselor effects
Cardiac rehabilitation groups had significantly better outcomes than the Control group for each of the two nurse counselors for all outcome variables for patients (correct, misconceptions, uncertainty, depression, anxiety, FLP total, FLP psychosocial, FLP physical) and partners (correct, misconceptions, uncertainty, depression, anxiety).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
The results of this study are similar in some respects to those reported by Thompson and Meddis (15, 17) who used an inpatient intervention with male MI patients and their wives. Treatment lowered anxiety in both patient and wife. It is possible to compare the studies directly, because Thompson and Meddis also measured mood with the HADS. The interventions had beneficial effects on anxiety and depression in the patients in both studies. However, the pattern of results was different. The main benefit seen in the current study was in preventing the increase in anxiety and depression seen in control patients between discharge from hospital and 2 months after their MI. Anxiety and depression were not high in patients when in hospital. Patients in Thompson and Meddis had higher anxiety immediately after the MI. The main effects of the intervention was an accelerated reduction in distress (most of which occurred before discharge). It is not clear why distress was less immediately after the MI in this study. Although it is tempting to believe that this relates to good hospital practice, the ratings of satisfaction with care were not high in control subjects at discharge, when distress was low. Nor is it clear why anxiety increased in the control patients after discharge. We do not know when this first occurred; if it was at or near our measurement point of 2 months post-MI, it may relate to the patients’ return to work or the need to undertake a widening range of activities.
It is clear that the anxiety and depression at baseline were broadly comparable in the partners in the two studies and that the inpatient interventions had similar beneficial effects on anxiety (17). However, the extended intervention seems to have a greater effect on partners’ anxiety and depression, indeed depression in partners was not affected by the inpatient intervention in the Thompson and Meddis study.
The positive findings from this study contrast with the results of three recent studies, all of which report no benefit from psychological interventions after a MI (18–20). However, the interventions were quite different in these studies from what we have reported [or indeed normal cardiac rehabilitation (21, 22)]. Two of the studies (19, 20) used an intervention directed at detecting significant distress by telephone interview and then offering either advice or referral to other agencies, whereas the other (18) relied on an educational approach offered in a group format. Our approach was one-to-one, conducted by a trained nurse working from a treatment manual, initiated in hospital, and targeted at the main concerns of post-MI patients. We also included a partner in the intervention. This may increase the effectiveness of the intervention for the patient as well as the partner, eg, by increasing knowledge, communication and partner expectations.
Although other studies provide data on mood, less frequently reported are measures of knowledge and disability. Knowledge may have implications for risk reduction and measures of disability are critical in assessing the return to a full active life, especially where simplistic indices such as return to work are inappropriate. Patients in both rehabilitation conditions had greater knowledge than controls and were markedly less disabled, both psychosocially and physically. Self-ratings by the patient were confirmed by the partners’ ratings.
This study not only involved partners in the rehabilitation program, it also evaluated the effects on the partners. Partners had high levels of anxiety and depression compared with norms for the HADS (on which the cutoff for caseness is 7/8 in both cases, (12)). The intervention lowered anxiety and depression, whereas the control group continued to have high levels, even 12 months after the patient’s infarction.
Like the patients, partners were also more knowledgeable after the interventions. Because the majority of partners were women who probably controlled the patients’ diets, their increased knowledge of heart disease and coronary disease prevention could have an impact on the health of the patients (and perhaps other family members). Partners in the control group gave very low ratings of satisfaction with care and those in the intervention groups were markedly more satisfied. Thus, the partner of the MI patient receiving normal care is in a very unenviable position—anxious, depressed, ill-informed, and very dissatisfied with the care that is being given. The rehabilitation program offered in this study markedly improved their situation.
There were few instances in which the extended rehabilitation group had reliably better outcomes than those receiving only the inpatient program. The inpatient program had a substantial effect on mood, knowledge, and behavior of the patients and there was little, if any, direct gain for the patients by participation in the extended program. However, for the partners, the extended program led to reliably lower levels of anxiety, with a similar pattern on depression and satisfaction. The extended program, much of which occurred in the family home, may have been more effective with the partners, either because it was longer and they needed more input than the patient, or, more probably, because the brief program focused more on the patient’s concerns (perhaps because of lack of time, the recency of the MI, and the effects of the hospital environment).
The design of this study was somewhat unusual. The study involved an inpatient intervention conducted on the ward, hence individual patients could not be randomly allocated to treatment, because of the likelihood of contamination between treatments. To avoid this, wards were allocated at random to treatment, so that all study patients on the same ward at the same time received the same intervention. Wards were randomly reallocated to treatment several times during the study to avoid confounding of ward and intervention. Allocation of patients to wards was determined primarily by the day the patient was admitted to hospital and is essentially random. This method of randomization seemed to work well; between-treatment contamination was avoided and the three conditions were well matched on virtually all important measures at baseline.
A recent survey (23) has shown that cardiac rehabilitation services in England and Wales are very incomplete and only offered to highly selected patients. We have shown that a brief intervention given before discharge from hospital, and hence likely to be acceptable to many patients currently untreated, markedly reduces the likelihood of the patient showing undue distress in the subsequent year and leads to a rapid and enduring reduction in the high levels of distress experienced by partners. Extending the program after discharge enhances these valuable gains, especially for partners.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Joan Foulkes, who initiated this project and was principal grant holder, died during the project. Received for publication September 29, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONResearch QuestionsMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. Auer, J. Gaume, N. Rodondi, J. Cornuz, and W. A. Ghali Efficacy of In-Hospital Multidimensional Interventions of Secondary Prevention After Acute Coronary Syndrome: A Systematic Review and Meta-Analysis Circulation, June 17, 2008; 117(24): 3109 - 3117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Clark, L. Hartling, B. Vandermeer, and F. A. McAlister Meta-Analysis: Secondary Prevention Programs for Patients with Coronary Artery Disease Ann Intern Med, November 1, 2005; 143(9): 659 - 672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Grace, S. E. Abbey, R. Pinto, Z. M. Shnek, J. Irvine, and D. E. Stewart Longitudinal Course of Depressive Symptomatology After a Cardiac Event: Effects of Gender and Cardiac Rehabilitation Psychosom Med, January 1, 2005; 67(1): 52 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R Thompson and R. J P Lewin CORONARY DISEASE: Management of the post-myocardial infarction patient: rehabilitation and cardiac neurosis Heart, July 1, 2000; 84(1): 101 - 105. [Full Text] [PDF]
|
|
|
|
|
|
R. A. Mayou, D. Gill, D. R. Thompson, A. Day, N. Hicks, J. Volmink, and A. Neil Depression and Anxiety As Predictors of Outcome After Myocardial Infarction Psychosom Med, March 1, 2000; 62(2): 212 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Sherrard Counselling after a myocardial infarction improved mood for patients and their partners and decreased patient functional limitations Evid. Based Nurs., January 1, 2000; 3(1): 21 - 21. [Full Text]
|
|
|
|
 |
|
 Other articles noted Evid. Based Ment. Health, November 1, 1999; 2(4): 104 - 104. [Full Text]
|
|
|
|
|
|
D W JOHNSTON Lifestyle changes after a myocardial infarction Heart, November 1, 1999; 82(5): 543 - 544. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
