Abstract
Background
Public access automatic external defibrillators (AEDs) can save lives, but most deaths from out-of-hospital sudden cardiac arrest occur at home. The Home Automatic External Defibrillator Trial (HAT) found no survival advantage for adding a home AED to cardiopulmonary resuscitation (CPR) training for 7001 patients with a prior anterior wall myocardial infarction. Quality of life (QOL) outcomes for both the patient and spouse/companion were secondary endpoints.
Methods
A subset of 1007 study patients and their spouse/companions was randomly selected for ascertainment of QOL by structured interview at baseline and 12 and 24 months following enrollment. The primary QOL measures were the Medical Outcomes Study 36-Item Short-Form (SF-36) psychological well-being (reflecting anxiety and depression) and vitality (reflecting energy and fatigue) subscales.
Results
For patients and spouse/companions, the psychological well-being and vitality scales did not differ significantly between those randomly assigned an AED plus CPR training and controls who received CPR training only. None of the other QOL measures collected showed a clinically and statistically significant difference between treatment groups. Patients in the AED group were more likely to report being extremely or quite a bit reassured by their treatment assignment. Spouse/companions in the AED group reported being less often nervous about the possibility of using AED/CPR treatment than those in the CPR group.
Conclusions
Adding access to a home AED to CPR training did not affect quality of life either for patients with a prior anterior myocardial infarction or their spouse/companion but did provide more reassurance to the patients without increasing anxiety for spouse/companions.
Keywords: Cardiopulmonary resuscitation, sudden cardiac arrest, automated external defibrillator, quality of life, emergency medical services
INTRODUCTION
Automated external defibrillators (AEDs) can be used to identify and promptly terminate ventricular fibrillation and ventricular tachycardia in patients with sudden cardiac arrest (SCA). They may thereby allow bystanders to administer potentially life-saving therapy critical minutes earlier than can be accomplished by emergency medical personnel. As a consequence, AEDs are increasingly being deployed in public places, such as airports and shopping malls, and available evidence suggests that they do increase survival to hospital discharge following a SCA.1–4 However, the impact of public access AEDs on overall morbidity and mortality following SCA is limited, as approximately seventy-five percent of out-of-hospital SCAs occur in the home.1,5
We conducted the Home Automatic External Defibrillator Trial (HAT), an international, multicenter, randomized clinical trial, to compare the effect of placing an AED in the home versus no home AED in 7001 stable patients with a prior anterior myocardial infarction. Spouse/companions in both treatment groups were taught cardiopulmonary resuscitation (CPR) and instructed to call emergency medical services (EMS) promptly. As previously reported, home access to an AED had no significant effect on all-cause mortality.5
Previous studies on the psychosocial and other quality of life effects of home AEDs for patients and their spouse/companions have been small and inconclusive. In a randomized trial of 158 patients with a recent acute ischemic event, home AED access was associated with trends toward worse quality of life on most subscales of the Short Form-36, particularly for social functioning.6 Two smaller interview studies with patients and their spouse/companions suggested that access to a home AED improved feelings of security and control.7,8 We examined the quality of life effects of home access to AEDs in a randomly selected subset of patients and their spouse/companions enrolled in HAT.
METHODS
Patient Population and Study Overview
Between January 23, 2003 and October 20, 2005, 7001 clinically stable patients with a history of anterior wall Q-wave or non-Q-wave myocardial infarction and who had a spouse or companion who was willing and able to call EMS, perform CPR, and use an AED were enrolled.5,9 Patients were recruited from sites in the United States, Canada, United Kingdom, Australia, New Zealand, Ireland, Germany and the Netherlands and were randomly assigned to receive an AED for home use or no AED. All spouses/companions of patients in the trial were trained to promptly call EMS and perform CPR, and those randomized to the home AED were also trained to use the device and instructed to do so prior to calling emergency services and starting resuscitation if SCA occurred. Details of the methods used to train the spouse/companion and complete inclusion and exclusion criteria have been previously published.5,9 The primary endpoint of the study was all-cause mortality. No overall survival advantage was observed for the home AED arm over a mean follow-up of 37.3 months.5
Quality of life was a prespecified secondary endpoint of the HAT trial.
Funding of Quality of Life Study and Human Subjects Protection
The HAT trial was funded by grant U01 HL67972 from the National Heart, Lung, and Blood Institute. Additional funding came from Philips Medical Systems and Laerdal Medical as a subsidiary distributor. The corporate sponsors were not involved in trial design, data collection or analysis, or manuscript preparation. All patients and spouse/companions provided informed consent and the study protocol was approved by each site’s institutional review board or ethics committee. The authors had complete independence in the conduct of the analyses, interpretation of the data, and preparation of this paper and are solely responsible for its content.
Quality of Life Data Collection Methods
Of the 7001 patients and their spouse/companions, 1007 (14%) were selected for the QOL assessment using permuted block randomization designed to balance the number of patients selected from each of the HAT treatment arms within each clinical site. Germany (14 patients enrolled in HAT) and the Netherlands (2 patients enrolled) did not participate in the QOL study. A random sample was used for the QOL portion of HAT for budgetary reasons and because it was projected that there would be >80% power to detect differences of one quarter of a standard deviation (SD) or greater in the SF-36 scales with >700 patients. Selected patients and their spouse/companions participated in a structured QOL interview at baseline and then at 12 and 24 months post-randomization. Patients and spouses/companions were interviewed separately by each site’s study coordinators. Overall collection of QOL data was coordinated by the Seattle Institute for Cardiac Research, with support during follow-up from the Duke Clinical Research Institute (DCRI). Data were transferred to the DCRI for analyses.
Quality of Life Measures for Patients
Only limited data were available regarding the QOL effects of home AED therapy to guide the selection of a primary QOL measure when HAT was designed. However, related work on the effects of CPR training suggested that psychological distress/well-being was the QOL domain most likely to demonstrate an effect, should one be present. The QOL battery used for patients in HAT consisted of 4 instruments:
The Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36). We used the SF-36 Mental Health Inventory-5 (MHI-5) as measurement of anxiety/depression symptoms and the 4-item vitality subscale from the SF-36 for energy/fatigue. Both sub-scales were scored from 0–100, with higher scores representing better psychological well-being and more energy. A change of 5 points or greater is considered clinically significant. The MHI-5 has shown excellent validity against more comprehensive measures of anxiety and depression.10 We used the remaining six subscales from the SF-36 to assess general health, physical function, role functioning-physical, role functioning-emotional, social functioning, and bodily pain. In addition, we calculated the two summary scores for the SF-36 covering physical and mental health.11 The Physical Health summary scale is a composite of the SF-36 physical functioning, role-physical, bodily pain, and general health subscales. The SF-36 Mental Health summary scale is a composite of the SF-36 vitality, social functioning, role-emotional, and mental health subscales. For the two summary scales, we report results using the more recently devised scoring method in which the standard 0–100 scores are transformed to a norm-based score where the reference score has a mean of 50 and a standard deviation of 10. There are no official standards for what constitutes a clinically significant difference in the SF-36 scores. To facilitate interpretation, we have used a one-quarter standard deviation difference between groups as the benchmark for the minimum clinically significant difference (corresponding to approximately 2.5 points for the summary scales and 5 points for the others).12,13
The Influential Relationships Questionnaire (IRQ): This is a 37-item measure that assesses perceptions of three aspects of interpersonal relationships: care, overprotection, and criticism.14 Higher scores indicate less caring, more overprotection, and more criticism. Defining a clinically significant difference as one quarter of a SD translates to 1.5 points on the IRQ scales.
We used a modification of a questionnaire developed by Dracup and colleagues to examine how reassured patients felt knowing that CPR or an AED would be used if needed.15
The EuroQoL-5D (EQ-5D) consists of two parts: a) a visual analog scale self-rated measure of generic health status where 0 represents the worst imaginable and 100 the best imaginable health state and b) the EQ-5D index, which summarizes and expresses the assessment of health status in five dimensions.16,17 The EQ-5D was collected on all 7001 HAT patients.
Quality of Life Measures for Spouses/Companions
Three sets of scales were collected from spouses/companions:
The SF-36 MHI-5 and the vitality subscales were used to assess psychological distress/well-being and energy/fatigue, respectively.
The Influential Relationships Questionnaire, described above.
Questions developed by Dracup and colleagues were used to examine attitudes of the spouse/companion toward the AED and CPR.15 These questions elicited overall feelings about the assigned intervention (CPR training/AED access), sense of responsibility attributed to the treatment assignment, anxiety about the prospect of using the treatment, and reassurance derived from the assigned treatment.
Statistical Analyses
As with the primary analyses of HAT, comparisons of QOL were undertaken on an intention-to-treat basis, so that the initial random treatment assignment was used to define the patient’s treatment group regardless of subsequent management. Continuous or ordinal variable comparisons between groups used the Wilcoxon rank-sum test, and group comparisons of categorical variables used a chi-square test. Summary statistics for continuous variables are reported as medians with interquartile ranges or means with standard deviations. Discrete variables are summarized using frequencies and percentages. Treatment group comparisons with p<0.05 were considered statistically significant.
RESULTS
Patient Population and Baseline Characteristics
Of the 1007 patients randomly selected for participation in the QOL substudy of HAT, 948 (94%) provided baseline QOL data and formed the study population (Figure 1). Baseline characteristics for the patients participating in the QOL assessment and their spouse/companions were quite similar to those of the overall HAT study cohort, indicating that the random selection process had accomplished its objective (Appendix Table A1). In addition, baseline characteristics of patients in the two treatment arms were well balanced (Table 1). The median age of the QOL cohort was 62 years, 17% were female and 12% represented racial minorities. Median time from the qualifying anterior MI to enrollment was 2 years. Seventy-five percent had a history of previous coronary revascularization. Median ejection fraction among the 80% of patients in whom it was known was 45%. Sixty-five percent of patients were in New York Heart Association Class I and 30% were in Class II. Among spouse/companions, 89% were spouses, 4% were companions, and 7% were other (e.g., sibling, child), and median spouse/companion age was 58 years.
Figure 1.
Data collection rates in the HAT QOL substudy.
Appendix Table A1.
Baseline Characteristics of Overall HAT Patient Population versus Quality of Life Subset
| Overall HAT (n=7001) | HAT QOL Substudy* (n=948) | |
|---|---|---|
| Patients | ||
| Age, years median (interquartile range) | 62 (54, 70) | 62 (54, 70) |
| Female | 17% (1220) | 17% (162) |
| Racial or ethnic minority | 13% (906) | 12% (114) |
| Time since most recent anterior MI | ||
| ≤1 Month | 10% (566) | 11% (87) |
| >1 Month to 3 months | 10% (590) | 9% (72) |
| >3 Months to 6 months | 9% (554) | 9% (78) |
| >6 Months to 1 year | 11% (670) | 10% (80) |
| >1 Year | 60% (3578) | 62% (509) |
| Employment Status | ||
| Full time | 33% (2284) | 32% (302) |
| Part time | 10% (697) | 11% (104) |
| Not employed | 57% (4020) | 57% (542) |
| Estimated daily time alone at home – hr median (interquartile range) | 1.5 (0.5, 4.0) | 1.5 (0.5, 4.0) |
| Estimated daily time away from home – hr median (interquartile range) | 4.0 (2.0, 8.0) | 3.5 (2.0, 7.5) |
| Previous procedures - no. (%) | ||
| Percutaneous coronary revascularization | 53% (3742) | 55% (526) |
| Coronary artery bypass grafting | 27% (1867) | 27% (257) |
| Coexisting conditions - no. (%) | ||
| Hypertension | 54% (3769) | 52% (496) |
| Diabetes | 21% (1504) | 22% (206) |
| Hypercholesterolemia | 79% (5557) | 80% (762) |
| Stroke | 6% (437) | 7% (66) |
| Ejection fraction measured –% (no.) | 80% (5624) | 80% (763) |
| Left ventricular ejection fraction - % median (interquartile range) | 45 (35, 55) | 45 (37, 55) |
| Body mass index – kg.m−2 median (interquartile range) | 27.7 (25.0, 30.9) | 27.7 (25.3, 30.7) |
| NYHA Class – no. (%) | ||
| I | 65% (4570) | 65% (614) |
| II | 29% (2053) | 30% (282) |
| III | 5% (367) | 5% (52) |
| IV | <1% (11) | 0% (0) |
| Type of myocardial infarction – no (%) | ||
| Anterior Q-wave | 64% (4509) | 66% (625) |
| Anterior non-Q-wave | 36% (2491) | 34% (323) |
| Use of medication – no. (%) | ||
| Beta-blocker (other than sotalol or amiodarone) | 79% (5531) | 80% (759) |
| ACE inhibitor or angiotensin-receptor blocker | 82% (5719) | 81% (765) |
| Statin | 89% (6241) | 89% (844) |
| Daily use of aspirin | 87% (6063) | 87% (823) |
| Digoxin | 7% (489) | 6% (60) |
| Warfarin | 18% (1249) | 18% (169) |
| Any antiarrhythmic drug | 4% (298) | 5% (47) |
| Spouse or companion | ||
| Relationship to patient – no. (%) | ||
| Spouse | 88% (6150) | 89% (844) |
| Companion | 4% (303) | 4% (40) |
| Other | 8% (548) | 7% (64) |
| Age – yr median (interquartile range) | 58 (50, 66) | 58 (50, 66) |
| Employment Status – no. (%) | ||
| Full time | 32% (2223) | 31% (295) |
| Part time | 17% (1196) | 16% (149) |
| Not employed | 51% (3582) | 53% (504) |
| Estimated daily time away from home – hr median (interquartile range) | 4.0 (2.0, 7.5) | 3.0 (2.0, 7.0) |
| Completed secondary school – no. (%) | 81% (5682) | 82% (779) |
Defined as any QOL substudy-randomized patient that did not refuse to complete at least one of the two patient-oriented questionnaires at baseline
Table 1.
Baseline Characteristics by Intention to Treat
| EMS Call/CPR (n=473) | AED + EMS Call/CPR (n=475) | |
|---|---|---|
| Patients | ||
| Age, years median (interquartile range) | 62 (54, 70) | 62 (55, 69) |
| Female | 18% (86) | 16% (76) |
| Racial or ethnic minority | 15% (70) | 9% (44) |
| Time since most recent anterior MI | ||
| ≤1 Month | 10% (39) | 12% (48) |
| >1 Month to 3 months | 10% (42) | 7% (30) |
| >3 Months to 6 months | 9% (38) | 10% (42) |
| >6 Months to 1 year | 9% (38) | 10% (42) |
| >1 Year | 62% (252) | 61% (257) |
| Employment Status | ||
| Full time | 33% (157) | 31% (145) |
| Part time | 8% (38) | 14% (66) |
| Not employed | 59% (252) | 56% (264) |
| Estimated daily time alone at home – hr median (interquartile range) | 1.1 (0.3, 4.0) | 2.0 (0.5, 4.0) |
| Estimated daily time away from home – hr median (interquartile range) | 3.5 (2.0, 8.0) | 3.5 (2.0, 7.0) |
| Previous procedures - % (no.) | ||
| Percutaneous coronary revascularization | 57% (269) | 54% (257) |
| Coronary artery bypass grafting | 26% (125) | 28% (132) |
| Coexisting conditions - % (no.) | ||
| Hypertension | 54% (255) | 51% (241) |
| Diabetes | 22% (105) | 21% (101) |
| Hypercholesterolemia | 72% (387) | 79% (375) |
| Stroke | 6% (30) | 8% (36) |
| Ejection fraction measured –% (no.) | 82% (386) | 79% (377) |
| Left ventricular ejection fraction - % median (interquartile range) | 48 (38, 59) | 45 (37, 53) |
| Body mass index – kg.m−2 median (interquartile range) | 27.5 (25.2, 30.9) | 27.7 (25.3, 30.7) |
| NYHA Class – % (no.) | ||
| I | 67% (318) | 62% (296) |
| II | 28% (132) | 32% (150) |
| III | 5% (23) | 6% (29) |
| Type of myocardial infarction – % (no.) | ||
| Anterior Q-wave | 63% (299) | 69% (326) |
| Anterior non-Q-wave | 37% (174) | 31% (149) |
| Use of medication – % (no.) | ||
| Beta-blocker (other than sotalol or amiodarone) | 80% (381) | 80% (378) |
| ACE inhibitor or angiotensin-receptor blocker | 81% (384) | 80% (381) |
| Statin | 89% (422) | 89% (422) |
| Daily use of aspirin | 87% (412) | 86% (411) |
| Digoxin | 7% (31) | 6% (29) |
| Warfarin | 15% (72) | 20% (97) |
| Any antiarrhythmic drug | 4% (17) | 6% (30) |
| Spouse or companion | ||
| Relationship to patient – % (no.) | ||
| Spouse | 88% (416) | 90% (428) |
| Companion | 5% (26) | 3% (14) |
| Other | 7% (31) | 7% (33) |
| Age – yr median (interquartile range) | 58 (49, 66) | 59 (50, 66) |
| Employment Status – % (no.) | ||
| Full time | 32% (153) | 30% (142) |
| Part time | 15% (70) | 17% (79) |
| Not employed | 53% (250) | 54% (254) |
| Estimated daily time away from home – hr median (interquartile range) | 3.0 (2.0, 7.0) | 3.0 (2.0, 7.0) |
| Completed secondary school – % (no.) | 82% (389) | 82% (390) |
Quality of Life Outcomes for Patients
From a total of 2846 expected patient contacts, 2473 QOL questionnaires were collected, representing 94% of patients eligible for this assessment at baseline and 83% of those eligible at 1 and 2 years of follow-up (Figure 1). Follow-up for QOL was more complete in the AED arm at 1 year (87% of eligible provided data versus 79% of the CPR only arm) and 2 years (86% versus 81%) (Table 2). These differences in response rates were part of a pattern seen in the clinical follow-up of the entire trial.
Table 2.
Collection of QOL Data by Treatment Group Over the 2-Year Follow-up
| All | EMS Call/CPR | AED + EMS Call/CPR | |
|---|---|---|---|
| Patients enrolled in HAT | 7001 | 3506 | 3495 |
| Randomized to QOL | 1007 (14%) | 507 (14%) | 500 (14%) |
| Provided Baseline responses | 948 (94%) | 473 (93%) | 475 (95%) |
| Excluded between Baseline and Year 1 | 20 (2%) | 14 (3%) | 6 (1%) |
| died | 15 | 9 | 6 |
| withdrew consent | 5 | 5 | 0 |
| Eligible for Year 1 | 928 (98%) | 459 (97%) | 469 (99%) |
| Provided Year 1 responses | 768 (83%) | 362 (79%) | 406 (87%) |
| Excluded between Year 1 and Year 2 | 17 (2%) | 7 (2%) | 10 (2%) |
| died | 10 | 3 | 7 |
| withdrew consent | 7 | 4 | 3 |
| Eligible for Year 2 | 911 (98%) | 452 (98%) | 459 (98%) |
| Provided Year 2 responses | 757 (83%) | 364 (81%) | 393 (86%) |
The median MHI-5 score was 80 in both treatment groups at baseline, did not change during follow-up and did not differ in follow-up between treatment groups (all p values>0.10) (Table 3). At baseline, median vitality scores were 60 in the AED plus CPR arm and 65 in the CPR-only arm (p=0.21). At 1 and 2 years follow-up, median vitality scores were 60 in each treatment group and these did not differ by treatment (each p>0.10).
Table 3.
Primary HAT Quality of Life Measures – Patients
| EMS Call/CPR | AED + EMS Call/CPR | P-value | |
|---|---|---|---|
| MOS SF 36 Mental Health Inventory 5 | |||
| Baseline | 0.38 | ||
| N | 452 | 459 | |
| Median (25th, 75th) | 80.0(65.0, 88.0) | 80.0(68.0, 88.0) | |
| Mean ± SD | 75.7 ± 16.9 | 76.9 ± 15.7 | |
| Year 1 | 0.55 | ||
| N | 355 | 393 | |
| Median (25th, 75th) | 80.0(68.0, 92.0) | 80.0(68.0, 88.0) | |
| Mean ± SD | 76.7 ± 17.1 | 76.7 ± 15.5 | |
| Year 2 | 0.96 | ||
| N | 347 | 383 | |
| Median (25th, 75th) | 80.0(64.0, 88.0) | 80.0(64.0, 88.0) | |
| Mean ± SD | 75.4 ± 17.7 | 75.7 ± 16.6 | |
| MOS SF 36 Vitality Scale | |||
| Baseline | 0.21 | ||
| N | 457 | 463 | |
| Median (25th, 75th) | 65.0(50.0, 75.0) | 60.0(45.0, 75.0) | |
| Mean ± SD | 60.4 ± 20.5 | 58.9 ± 20.5 | |
| Year 1 | 0.19 | ||
| N | 357 | 396 | |
| Median (25th, 75th) | 60.0(50.0, 75.0) | 60.0(45.0, 75.0) | |
| Mean ± SD | 59.7 ± 21.3 | 58.5 ± 19.9 | |
| Year 2 | 0.98 | ||
| N | 353 | 385 | |
| Median (25th, 75th) | 60.0(50.0, 75.0) | 60.0(50.0, 75.0) | |
| Mean ± SD | 58.9 ± 21.0 | 59.0 ± 20.1 | |
| EQ-5D Visual Analog Scale | |||
| 0–100 scale; 0=worst imaginable health, 100=perfect health | |||
| Baseline | |||
| No. of patients | 504 | 499 | 0.13 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 85.0) | |
| Mean ± standard deviation | 76.2 ± 16.2 | 74.9 ± 16.4 | |
| Year 1 | |||
| No. of patients | 451 | 463 | 0.25 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 78.2 ± 17.0 | 77.5 ± 16.3 | |
| Year 2 | |||
| No. of patients | 415 | 426 | 0.37 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 77.9 ± 16.3 | 77.1 ± 15.9 | |
| EQ-5D Single Summary Index (Health State) | |||
| Baseline | |||
| No. of patients | 504 | 499 | 0.20 |
| Median (interquartile range) | 0.87 (0.73, 1.00) | 1.00 (0.78, 1.00) | |
| Mean ± standard deviation | 0.83 ± 0.22 | 0.86 ± 0.19 | |
| Year 1 | |||
| No. of patients | 450 | 466 | 0.50 |
| Median (interquartile range) | 1.00 (0.73, 1.00) | 0.85 (0.73, 1.00) | |
| Mean ± standard deviation | 0.84 ± 0.23 | 0.84 ± 0.21 | |
| Year 2 | |||
| No. of patients | 422 | 434 | 0.36 |
| Median (interquartile range) | 0.85 (0.73, 1.00) | 1.00 (0.74, 1.00) | |
| Mean ± standard deviation | 0.82 ± 0.25 | 0.85 ± 0.20 | |
Secondary Quality of Life Comparisons for Patients
No clinically and statistically significant differences were found between groups for the SF-36 Physical Health and Mental Health summary scales or the SF-36 subscales for general health perceptions, physical functioning, role functioning-physical, role-functioning-emotional, and social functioning (Appendix Table A2). While two comparisons yielded p values <0.05 (SF-36 General Health Perceptions and SF-36 Physical Health Summary at 1 year), in neither case did the observed difference also meet criteria for a clinically significant difference (Appendix Table A2).
Appendix Table A2.
Secondary Quality of Life Measures by Intention to Treat - Patient
| EMS Call/CPR | AED + EMS Call/CPR | P-value* | |
|---|---|---|---|
| MOS Short Form-36 Subscales | |||
| General Health Perceptions | |||
| Baseline | 0.21 | ||
| N | 457 | 461 | |
| Median (25th, 75th) | 65.0(47.0, 80.0) | 62.0(45.0, 77.0) | |
| Mean ± SD | 62.5 ± 21.8 | 60.7 ± 21.9 | |
| Year 1 | 0.025 | ||
| N | 353 | 390 | |
| Median (25th, 75th) | 67.0(52.0, 80.0) | 64.0(45.0, 77.0) | |
| Mean ± SD | 64.3 ± 21.8 | 60.6 ± 22.6 | |
| Year 2 | 0.52 | ||
| N | 342 | 381 | |
| Median (25th, 75th) | 67.0(50.0, 77.0) | 62.0(47.0, 80.0) | |
| Mean ± SD | 62.6 ± 21.4 | 61.4 ± 22.5 | |
| Physical Functioning | |||
| Baseline | 0.19 | ||
| N | 464 | 469 | |
| Median (25th, 75th) | 80.0(55.0, 95.0) | 75.0(50.0, 90.0) | |
| Mean ± SD | 70.4 ± 26.5 | 68.1 ± 27.4 | |
| Year 1 | 0.06 | ||
| N | 356 | 393 | |
| Median (25th, 75th) | 80.0(58.0, 95.0) | 80.0(50.0, 90.0) | |
| Mean ± SD | 72.3 ± 27.4 | 69.8 ± 26.1 | |
| Year 2 | 0.13 | ||
| N | 354 | 388 | |
| Median (25th, 75th) | 80.0(50.0, 95.0) | 75.0(50.0, 90.0) | |
| Mean ± SD | 71.3 ± 26.8 | 68.8 ± 27.6 | |
| Role Functioning - Physical | |||
| Baseline | 0.90 | ||
| N | 424 | 437 | |
| Median (25th, 75th) | 75.0(0.0, 100.0) | 75.0(0.0, 100.0) | |
| Mean ± SD | 57.4 ± 41.6 | 57.3 ± 41.7 | |
| Year 1 | 0.48 | ||
| N | 343 | 384 | |
| Median (25th, 75th) | 75.0(25.0, 100.0) | 75.0(25.0, 100.0) | |
| Mean ± SD | 67.1 ± 39.2 | 64.8 ± 39.8 | |
| Year 2 | 0.93 | ||
| N | 335 | 378 | |
| Median (25th, 75th) | 75.0(25.0, 100.0) | 100.0(25.0, 100.0) | |
| Mean ± SD | 65.5 ± 39.8 | 65.5 ± 40.9 | |
| Role Functioning - Emotional | |||
| Baseline | 0.52 | ||
| N | 422 | 425 | |
| Median (25th, 75th) | 100.0(67.0, 100.0) | 100.0(67.0, 100.0) | |
| Mean ± SD | 79.0 ± 35.0 | 77.4 ± 36.1 | |
| Year 1 | 0.58 | ||
| N | 336 | 381 | |
| Median (25th, 75th) | 100.0(67.0, 100.0) | 100.0(67.0, 100.0) | |
| Mean ± SD | 82.2 ± 32.7 | 80.1 ± 34.7 | |
| Year 2 | 0.93 | ||
| N | 331 | 376 | |
| Median (25th, 75th) | 100.0(67.0, 100.0) | 100.0(67.0, 100.0) | |
| Mean ± SD | 80.7 ± 34.9 | 82.2 ± 31.9 | |
| Social Functioning | |||
| Baseline | 0.92 | ||
| N | 469 | 471 | |
| Median (25th, 75th) | 88.0(63.0, 100.0) | 88.0(63.0, 100.0) | |
| Mean ± SD | 77.7 ± 25.3 | 77.5 ± 25.9 | |
| Year 1 | 0.78 | ||
| N | 357 | 397 | |
| Median (25th, 75th) | 88.0(63.0, 100.0) | 88.0(63.0, 100.0) | |
| Mean ± SD | 79.9 ± 25.2 | 80.0 ± 23.5 | |
| Year 2 | 0.64 | ||
| N | 355 | 392 | |
| Median (25th, 75th) | 88.0(63.0, 100.0) | 88.0(63.0, 100.0) | |
| Mean ± SD | 79.2 ± 24.9 | 78.8 ± 24.2 | |
| Bodily Pain | |||
| Baseline | 0.92 | ||
| N | 469 | 471 | |
| Median (25th, 75th) | 74.0(51.0, 100.0) | 72.0(51.0, 100.0) | |
| Mean ± SD | 68.9 ± 26.1 | 68.8 ± 25.9 | |
| Year 1 | 0.37 | ||
| N | 357 | 397 | |
| Median (25th, 75th) | 74.0(52.0, 100.0) | 74.0(51.0, 100.0) | |
| Mean ± SD | 71.8 ± 26.7 | 70.9 ± 24.6 | |
| Year 2 | 0.90 | ||
| N | 356 | 392 | |
| Median (25th, 75th) | 74.0(51.0, 100.0) | 74.0(51.0, 100.0) | |
| Mean ± SD | 70.7 ± 25.4 | 70.2 ± 26.1 | |
| SF-36 Summary Scales | |||
| Physical Health Summary Scale | |||
| Baseline | 0.60 | ||
| N | 393 | 407 | |
| Median (25th, 75th) | 46.7(36.9, 53.6) | 46.3(36.1, 53.5) | |
| Mean ± SD | 44.8 ± 10.8 | 44.4 ± 11.1 | |
| Year 1 | 0.037 | ||
| N | 325 | 364 | |
| Median (25th, 75th) | 49.6(40.2, 55.5) | 47.6(36.7, 54.1) | |
| Mean ± SD | 46.8 ± 10.9 | 45.3 ± 10.9 | |
| Year 2 | 0.36 | ||
| N | 305 | 356 | |
| Median (25th, 75th) | 48.6(38.9, 55.0) | 47.8(37.4, 54.2) | |
| Mean ± SD | 46.2 ± 10.4 | 45.2 ± 11.4 | |
| Mental Health Summary Scale | |||
| Baseline | 0.78 | ||
| N | 393 | 407 | |
| Median (25th, 75th) | 53.6(47.2, 57.7) | 53.7(46.7, 57.9) | |
| Mean ± SD | 51.3 ± 9.0 | 51.4 ± 9.0 | |
| Year 1 | 0.73 | ||
| N | 325 | 364 | |
| Median (25th, 75th) | 54.0(47.8, 58.1) | 53.8(47.1, 57.7) | |
| Mean ± SD | 51.8 ± 8.7 | 51.7 ± 8.3 | |
| Year 2 | 0.86 | ||
| N | 305 | 356 | |
| Median (25th, 75th) | 54.3(47.1, 58.0) | 53.6(47.3, 57.8) | |
| Mean ± SD | 51.2 ± 9.5 | 51.7 ± 8.6 | |
| Reported Health Transition | |||
| Baseline | 0.09 | ||
| N | 467 | 469 | |
| Much better than one year ago | 13% (60) | 10% (47) | |
| Somewhat better than one year ago | 16% (75) | 20% (95) | |
| About the same as than one year ago | 43% (203) | 46% (217) | |
| Somewhat worse than one year ago | 22% (103) | 17% (79) | |
| Much worse than one year ago | 6% (26) | 7% (31) | |
| Year 1 | 0.39 | ||
| N | 356 | 395 | |
| Much better than one year ago | 17% (60) | 15% (61) | |
| Somewhat better than one year ago | 22% (80) | 26% (104) | |
| About the same as than one year ago | 48% (172) | 49% (195) | |
| Somewhat worse than one year ago | 11% (40) | 8% (30) | |
| Much worse than one year ago | 1% (4) | 1% (5) | |
| Year 2 | 0.20 | ||
| N | 355 | 391 | |
| Much better than one year ago | 7% (25) | 10% (41) | |
| Somewhat better than one year ago | 19% (68) | 21% (83) | |
| About the same as than one year ago | 60% (213) | 58% (225) | |
| Somewhat worse than one year ago | 12% (43) | 8% (33) | |
| Much worse than one year ago | 2% (6) | 2% (9) | |
| Influential Relationships Questionnaire | |||
| Care | |||
| Baseline | 0.20 | ||
| N | 443 | 445 | |
| Median (25th, 75th) | 9.0(5.0, 12.0) | 9.0(4.0, 12.0) | |
| Mean ± SD | 9.0 ± 5.3 | 8.5 ± 5.3 | |
| Year 1 | 0.85 | ||
| N | 345 | 385 | |
| Median (25th, 75th) | 10.0(5.0, 12.0) | 10.0(4.0, 12.0) | |
| Mean ± SD | 9.0 ± 5.8 | 8.9 ± 5.4 | |
| Year 2 | 0.42 | ||
| N | 329 | 363 | |
| Median (25th, 75th) | 10.0(4.0, 12.0) | 9.0(4.0, 12.0) | |
| Mean ± SD | 9.1 ± 5.7 | 8.7 ± 5.5 | |
| Criticism | |||
| Baseline | 0.42 | ||
| N | 438 | 440 | |
| Median (25th, 75th) | 10.0(5.0, 13.0) | 9.5(4.0, 13.0) | |
| Mean ± SD | 9.4 ± 5.6 | 9.0 ± 5.5 | |
| Year 1 | 0.86 | ||
| N | 344 | 382 | |
| Median (25th, 75th) | 10.0(4.0, 13.0) | 10.0(4.0, 13.0) | |
| Mean ± SD | 9.0 ± 5.5 | 9.0 ± 5.5 | |
| Year 2 | 0.31 | ||
| N | 325 | 362 | |
| Median (25th, 75th) | 10.0(4.0, 13.0) | 9.0(4.0, 13.0) | |
| Mean ± SD | 9.1 ± 5.6 | 8.7 ± 5.8 | |
| Overprotection | |||
| Baseline | 0.62 | ||
| N | 438 | 442 | |
| Median (25th, 75th) | 12.0(7.0, 14.0) | 11.5(7.0, 15.0) | |
| Mean ± SD | 10.9 ± 5.3 | 10.7 ± 5.2 | |
| Year 1 | 0.56 | ||
| N | 342 | 383 | |
| Median (25th, 75th) | 12.0(6.0, 14.0) | 12.0(7.0, 15.0) | |
| Mean ± SD | 10.6 ± 5.4 | 10.8 ± 5.5 | |
| Year 2 | 0.74 | ||
| N | 328 | 362 | |
| Median (25th, 75th) | 12.0(6.0, 14.0) | 11.0(7.0, 14.0) | |
| Mean ± SD | 10.4 ± 5.5 | 10.4 ± 5.6 | |
P-value for continuous variables is based on Wilcoxon Rank-Sum test. P-value for categorical variables is based on Chi-square test.
Continuous variables shown as median (25th to 75th percentiles).
Categorical variables are shown as percentages.
Means are shown with standard deviations.
The Influential Relationships Questionnaire revealed no significant difference by treatment group in perceptions of the patients or their spouse/companions in the areas of caring, criticism, and overprotection. No differences by treatment group were found in the EQ-5D visual analog scale or single summary index.
At 12 months, 72% of the AED group versus 61% of the CPR-only group reported feeling “extremely” or “quite a bit” reassured knowing they would be treated with CPR/AED if needed (p=0.0014). At 24 months, these rates were 72% in the AED arm and 55% in the CPR-only group (p<0.0001) (Table 4).
Table 4.
Patient and Spouse/Companion Attitudes about CPR/AED
| EMS Call/CPR | AED + EMS Call/CPR | P-value | |
|---|---|---|---|
| Patient | |||
| CPR/AED Attitude: How reassured are you knowing that you will be treated with CPR/AED if needed? | |||
| Baseline | <0.0001 | ||
| N | 449 | 462 | |
| Not at all | 8% (37) | 6% (29) | |
| A little bit | 9% (42) | 7% (32) | |
| Moderately | 21% (94) | 11% (50) | |
| Quite a bit | 33% (149) | 38% (177) | |
| Extremely | 28% (127) | 38% (174) | |
| Year 1 | 0.0014 | ||
| N | 352 | 392 | |
| Not at all | 7% (24) | 7% (27) | |
| A little bit | 11% (37) | 6% (25) | |
| Moderately | 21% (75) | 15% (59) | |
| Quite a bit | 37% (130) | 35% (136) | |
| Extremely | 24% (86) | 37% (145) | |
| Year 2 | <0.0001 | ||
| N | 348 | 380 | |
| Not at all | 7% (26) | 8% (31) | |
| A little bit | 11% (40) | 6% (24) | |
| Moderately | 26% (89) | 13% (51) | |
| Quite a bit | 29% (101) | 35% (133) | |
| Extremely | 26% (92) | 37% (141) | |
| Spouse/Companion | |||
| CPR/AED: | |||
| What are your feelings about learning [CPR/AED]? | |||
| Baseline | |||
| N | 456 | 468 | 0.40* |
| Very positive | 91.2% (416) | 93.4% (437) | |
| Ambivalent | 8.6% (39) | 6.2% (29) | |
| Negative/Very negative (combined due to small cell sizes) | 0.2% (1) | 0.4% (2) | |
| Year 1 | |||
| N | 345 | 383 | 0.0015* |
| Very positive | 80.6% (278) | 88.3% (338) | |
| Ambivalent | 18.8% (65) | 10.2% (39) | |
| Negative/Very negative (combined due to small cell sizes) | 0.6% (2) | 1.6% (6) | |
| Year 2 | |||
| N | 328 | 363 | 0.0073* |
| Very positive | 80.5% (264) | 88.7% (322) | |
| Ambivalent | 18.0% (59) | 10.7% (39) | |
| Negative/Very negative (combined due to small cell sizes) | 1.5% (5) | 0.6% (2) | |
| CPR/AED: | |||
| What is your sense of the responsibility placed on you by learning [CPR/AED]? | |||
| Baseline | |||
| N | 455 | 468 | 0.16 |
| Too much | 3.3% (15) | 2.8% (13) | |
| Right amount | 86.8% (395) | 90.0% (421) | |
| Very little | 4.8% (22) | 4.9% (23) | |
| None | 5.1% (23) | 2.4% (11) | |
| Year 1 | |||
| N | 344 | 379 | 0.58 |
| Too much | 5.2% (18) | 3.4% (13) | |
| Right amount | 86.1% (296) | 87.1% (330) | |
| Very little | 5.8% (20) | 5.5% (22) | |
| None | 2.9% (10) | 4.0% (15) | |
| Year 2 | |||
| N | 332 | 365 | 0.57 |
| Too much | 4.5% (15) | 4.4% (16) | |
| Right amount | 84.0% (279) | 87.1% (318) | |
| Very little | 8.4% (28) | 5.8% (21) | |
| None | 3.0% (10) | 2.7% (10) | |
| How nervous are you about the possibility of using [CPR/AED] if needed? | |||
| Baseline | |||
| N | 458 | 471 | 0.32 |
| Not at all | 23.4% (107) | 23.8% (112) | |
| A little bit | 47.4% (217) | 52.4% (247) | |
| Moderately | 17.7% (81) | 15.5% (73) | |
| Quite a bit | 9.6% (44) | 7.2% (34) | |
| Extremely | 2.0% (9) | 1.1% (5) | |
| Year 1 | |||
| N | 346 | 383 | 0.0357 |
| Not at all | 18.5% (64) | 21.4% (82) | |
| A little bit | 45.7% (158) | 53.3% (204) | |
| Moderately | 21.1% (73) | 15.7% (60) | |
| Quite a bit | 12.1% (42) | 7.3% (28) | |
| Extremely | 2.6% (9) | 2.4% (9) | |
| Year 2 | |||
| N | 334 | 367 | 0.0258 |
| Not at all | 18.6% (62) | 20.2% (74) | |
| A little bit | 42.2% (141) | 51.8% (190) | |
| Moderately | 24.6% (82) | 16.4% (60) | |
| Quite a bit | 11.1% (37) | 9.5% (35) | |
| Extremely | 3.6% (12) | 2.2% (8) | |
Continuous variable p-values are based on Wilcoxon Signed-rank test. Categorical variable p-values are based on Likelihood Ratio Chi-Square, except for those indicated with (*) which used Fishers Exact test.
Quality of Life Comparisons for Spouses/Companions
Among spouses/companions, no significant differences were found in the MHI-5, SF-36 vitality scale, or IRQ scores between groups with up to 2 years of follow-up (Appendix Table A3).
Appendix Table A3.
Quality of Life Measures by Intention to Treat – Spouse/Companion Responses
| EMS Call/CPR | AED + EMS Call/CPR | P-value | |
|---|---|---|---|
| MOS Short Form-36 Scales | |||
| Mental Health Inventory 5 | |||
| Baseline | |||
| N | 453 | 464 | 0.84 |
| Median (25th, 75th) | 80.0 (68.0, 88.0) | 80.0 (68.0, 88.0) | |
| Mean ± SD | 76.6 ± 13.8 | 76.3 ± 14.0 | |
| Year 1 | |||
| N | 345 | 375 | 0.92 |
| Median (25th, 75th) | 80.0 (68.0, 88.0) | 80.0 (68.0, 88.0) | |
| Mean ± SD | 75.5 ± 14.8 | 75.2 ± 15.5 | |
| Year 2 | |||
| N | 327 | 360 | 0.24 |
| Median (25th, 75th) | 76.0 (68.0, 88.0) | 80.0 (68.0, 88.0) | |
| Mean ± SD | 74.8 ± 15.8 | 76.3 ± 14.9 | |
| Vitality | |||
| Baseline | |||
| N | 455 | 466 | 0.27 |
| Median (25th, 75th) | 70.0 (60.0, 80.0) | 70.0 (55.0, 88.0) | |
| Mean ± SD | 67.1 ± 16.6 | 65.7 ± 17.7 | |
| Year 1 | |||
| N | 346 | 376 | 0.46 |
| Median (25th, 75th) | 65.0 (55.0, 75.0) | 65.0 (50.0, 75.0) | |
| Mean ± SD | 63.5 ± 17.3 | 62.4 ± 18.0 | |
| Year 2 | |||
| N | 330 | 361 | 0.99 |
| Median (25th, 75th) | 65.0 (53.3, 75.0) | 65.0 (50.0, 75.0) | |
| Mean ± SD | 62.8 ± 17.6 | 62.7 ± 18.1 | |
| Influential Relationships Questionnaire – without imputation | |||
| Care | |||
| Baseline | |||
| N | 398 | 414 | 0.41 |
| Median (25th, 75th) | 10.0 (6.0, 13.0) | 10.0 (5.0, 13.0) | |
| Mean ± SD | 9.9 ± 5.7 | 9.4 ± 5.4 | |
| Year 1 | |||
| N | 317 | 342 | 0.27 |
| Median (25th, 75th) | 11.0 (6.0, 13.0) | 10.0 (6.0, 13.0) | |
| Mean ± SD | 10.5 ± 5.9 | 10.0 ± 5.5 | |
| Year 2 | |||
| N | 291 | 326 | 0.27 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 11.0 (6.0, 13.0) | |
| Mean ± SD | 10.6 ± 6.1 | 10.0 ± 5.8 | |
| Criticism | |||
| Baseline | |||
| N | 376 | 402 | 0.67 |
| Median (25th, 75th) | 10.0 (5.0, 13.0) | 10.0 (4.0, 13.0) | |
| Mean ± SD | 9.6 ± 5.5 | 9.3 ± 5.9 | |
| Year 1 | |||
| N | 309 | 336 | 0.15 |
| Median (25th, 75th) | 11.0 (5.0, 14.0) | 10.0 (4.0, 13.0) | |
| Mean ± SD | 10.1 ± 5.8 | 9.6 ± 6.0 | |
| Year 2 | |||
| N | 287 | 320 | 0.37 |
| Median (25th, 75th) | 11.0 (5.0, 14.0) | 10.0 (4.0, 14.0) | |
| Mean ± SD | 10.0 ± 6.0 | 9.6 ± 6.3 | |
| Overprotection | |||
| Baseline | |||
| N | 397 | 411 | 0.64 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 11.0 (6.0, 14.0) | |
| Mean ± SD | 10.2 ± 5.6 | 10.0 ± 5.3 | |
| Year 1 | |||
| N | 315 | 338 | 0.52 |
| Median (25th, 75th) | 12.0 (6.0, 14.0) | 12.0 (6.0, 14.0) | |
| Mean ± SD | 10.7 ± 5.6 | 10.4 ± 5.9 | |
| Year 2 | |||
| N | 294 | 327 | 0.91 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 12.0 (5.0, 14.0) | |
| Mean ± SD | 10.5 ± 6.0 | 10.3 ± 6.0 | |
| Influential Relationships Questionnaire – with imputation | |||
| Care | |||
| Baseline | |||
| N | 440 | 456 | 0.57 |
| Median (25th, 75th) | 11.0 (6.0, 13.0) | 10.0 (5.0, 13.0) | |
| Mean ± SD | 10.0 ± 5.6 | 9.6 ± 5.5 | |
| Year 1 | |||
| N | 335 | 372 | 0.73 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 11.0 (7.0, 13.0) | |
| Mean ± SD | 10.5 ± 5.9 | 10.3 ± 5.7 | |
| Year 2 | |||
| N | 310 | 347 | 0.17 |
| Median (25th, 75th) | 11.0 (7.0, 14.0) | 11.0 (6.0, 13.0) | |
| Mean ± SD | 10.8 ± 6.1 | 10.1 ± 5.9 | |
| Criticism | |||
| Baseline | |||
| N | 439 | 450 | 0.72 |
| Median (25th, 75th) | 11.0 (5.0, 13.0) | 11.0 (4.0, 14.0) | |
| Mean ± SD | 9.8 ± 5.6 | 9.6 ± 5.8 | |
| Year 1 | |||
| N | 337 | 372 | 0.46 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 11.0 (5.0, 14.0) | |
| Mean ± SD | 10.2 ± 5.7 | 10.1 ± 6.2 | |
| Year 2 | |||
| N | 310 | 343 | 0.38 |
| Median (25th, 75th) | 11.0 (5.0, 14.0) | 10.0 (4.0, 14.0) | |
| Mean ± SD | 10.2 ± 6.1 | 9.7 ± 6.3 | |
| Overprotection | |||
| Baseline | |||
| N | 436 | 453 | 0.99 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 11.0 (6.0, 14.0) | |
| Mean ± SD | 10.2 ± 5.6 | 10.2 ± 5.4 | |
| Year 1 | |||
| N | 335 | 368 | 0.57 |
| Median (25th, 75th) | 12.0 (6.0, 14.0) | 12.0 (6.0, 14.0) | |
| Mean ± SD | 10.8 ± 5.7 | 10.6 ± 6.1 | |
| Year 2 | |||
| N | 306 | 344 | 0.76 |
| Median (25th, 75th) | 11.0 (6.0, 14.0) | 12.0 (5.5, 14.0) | |
| Mean ± SD | 10.5 ± 5.9 | 10.3 ± 6.0 | |
Spouses/companions assigned to the home AED group responded more positively toward to their assigned treatment mode than did spouses/companions who were instructed in using only CPR, with 88–89% of the AED group feeling “very positive” about learning their intervention compared with 81% of the CPR-only group (p=0.0015 at 12 months, p=0.0073 at 24 months) (Table 3). No significant differences were found between the groups in their response to the sense of responsibility placed on the spouse/companion from learning CPR or to use an AED. Spouses/companions in the home AED group reported less apprehension about the possibility of using their assigned intervention than spouses/companions in the CPR-only group. At 1 year, 75% of the AED spouse/companions were “not at all” or only “a little bit” nervous about the possibility of using their assigned resuscitation mode if needed compared with 64% of the CPR-only group (p=0.036). Corresponding rates at 2 years were 72% and 61%, respectively (p=0.026).
DISCUSSION
Our study has 3 important findings. First, home placement of an AED for stable patients with a previous anterior MI did not impair psychological well-being or any other aspects of QOL in either the patients or their spouse/companions out to 2 years of follow-up. Secondly, home placement of an AED did not adversely affect the interpersonal relationship between patients and their spouse/companions compared with having CPR training only. Finally, home placement of an AED provided more reassurance to the patient than training the spouse/companion to perform CPR alone, and the additional responsibility did not increase spouse/companion anxiety.
The Home Automatic External Defibrillator Trial was the first large-scale randomized test of the hypothesis that providing home AED access to patients at increased risk for SCA could lower all-cause mortality. Previous studies have focused solely on cardiac arrest mortality. As reported previously, the trial did not demonstrate a survival benefit with home AED access, probably for several reasons.5 First, the mortality rate observed in HAT and the incidence of SCA were much lower than expected based on prior experience and the trial was thus underpowered. Second, although almost three-quarters of tachyarrhythmic SCAs occurred in the home, as expected, less than half were witnessed. Even in those witnessed in the AED arm, an AED was not always applied. While the home AED does save lives when used appropriately in a timely fashion and although there was no documented instance of an inappropriate shock, HAT shows that it is an inefficient means of addressing the important public health problem of SCA in the population and in the manner of implementation we selected.
Assessing the effects of a home AED on patient QOL differs from most clinical trial assessments of QOL, since placing an AED in the home of an at risk patient does not require anything specific to be done to the patient beyond some explanation and training of household members. An AED that is stored somewhere in the house and is rarely thought about would presumably have no more detectable effects on patient or spouse/companion QOL than a home fire extinguisher. A reasonable presumption, therefore, is that effects of the AED on QOL most likely would be mediated primarily through the explanations given by study personnel when the patients were randomized or the sense of dependency or responsibility inculcated in the participants. Prior to randomization, patients may not have been aware of their potential for SCA and finding out that their doctors felt that the risk was significant enough to consider participation in HAT could have induced a number of different psychological reactions. Similarly, the unblinded treatment assignment might have augmented or mitigated those reactions in different ways, depending on knowledge about cardiac arrests and their treatment, prior experience with CPR and AEDs, and individual personality factors. Since the effectiveness of home AED therapy is critically dependent on the spouse/companion, as well as patient factors, this therapy also has the potential to affect household interpersonal relationships more than most other therapies given to patients with coronary heart disease.
Prior to the HAT trial, there was very little empirical data relating to psychological or other QOL effects that home placement of an AED might generate. In a small study of post-myocardial infarction patients, interviews with patients and with spouse/companions suggested that access to an AED would enhance feelings of security and control and would neither increase patient stress or anxiety nor augment relationship stress.7 More recently, 158 patients and family members with a recent acute ischemic cardiac event were randomly assigned to received CPR training with or without home AED access.6 QOL was assessed with the SF-36 at 2 weeks, 3 months, and 9 months and scores were worse in the AED arm on most scales at all follow-up points. However, most of the differences were not large enough to be considered clinically significant, and significant imbalances in some baseline scores combined with a small sample size made the results of the study inconclusive. Earlier studies of the psychological effects of CPR training on the patient and spouse/companion also suggested the potential for adverse consequences.18 In this context, our much larger study provides important reassurance that home access to an AED as implemented in the HAT trial does not cause a general increase in anxiety or feelings of social isolation relative to CPR training alone. Although the presence of an AED and what it signifies might cause patients to become dependent on the security of home and the presence of the spouse/companion, we did not see any evidence that patients in the AED arm or their spouses spent less time out of the home than the CPR subjects (data not shown). Neither did we observe any increase in stress between patients and their spouse/companions related to the interdependence created by the presence of the AED in the home.
Several caveats should be kept in mind when evaluating our results. First, our follow-up response rate for patients in the QOL portion of HAT was lower than expected (83%) and was better in the AED arm at 1 and 2 years. This may have led to undetected biases, although it seems unlikely that a clinically important difference in QOL was obscured by this follow-up difference. Secondly, the assigned study intervention was not masked, and patient and spouse/companion biases about CPR and about AEDs may have influenced our findings. Third, we did not have sufficient numbers of patient/spouse pairs with a SCA event in the QOL subset to examine the effects of the intervention on spouse/companion post-event QOL and attitudes. Finally, two countries (Germany and the Netherlands) entered HAT late (16 patients total enrollment) and were not included in the QOL portion of the study, in part due to the added expense of obtaining valid translations of the study instruments for those countries.
In conclusion, placement of an AED in the home of patients at increased risk for SCA due to a prior anterior myocardial infarction did not affect quality of life either for them or their spouse/companion but did provide more reassurance to patients without increasing anxiety for spouse/companions compared with CPR training alone.
Appendix Table A4.
EuroQoL for Overall HAT Population by Intention to Treat
| EMS Call/CPR | AED + EMS Call/CPR | P-value | |
|---|---|---|---|
| EuroQoL Visual Analog Scale | |||
| 0–100 scale; 0=worst imaginable health, 100=perfect health | |||
| Baseline | |||
| No. of patients | 3496 | 3485 | 0.24 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 76.2 ± 15.9 | 75.6 ± 16.4 | |
| Year 1 | |||
| No. of patients | 3090 | 3229 | 0.39 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 78.1 ± 16.1 | 78.5 ± 15.7 | |
| Year 2 | |||
| No. of patients | 2840 | 3011 | 0.82 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 78.0 ± 16.1 | 78.0 ± 16.4 | |
| EuroQoL Single Summary Index (Health State) | |||
| Baseline | |||
| No. of patients | 3491 | 3478 | 0.82 |
| Median (interquartile range) | 0.88 (0.73, 1.00) | 0.88 (0.76, 1.00) | |
| Mean ± standard deviation | 0.85 ± 0.21 | 0.85 ± 0.20 | |
| Year 1 | |||
| No. of patients | 3103 | 3235 | 0.56 |
| Median (interquartile range) | 0.88 (0.73, 1.00) | 0.88 (0.76, 1.00) | |
| Mean ± standard deviation | 0.84 ± 0.22 | 0.85 ± 0.21 | |
| Year 2 | |||
| No. of patients | 2872 | 3044 | 0.88 |
| Median (interquartile range) | 1.00 (0.74, 1.00) | 1.00 (0.74, 1.00) | |
| Mean ± standard deviation | 0.85 ± 0.22 | 0.85 ± 0.21 | |
Appendix Table A5.
EuroQoL Scores of Overall HAT Patient Population versus Quality of Life Subset
| Overall HAT | Non-QoL HAT | QoL HAT | P-value comparing Non-QOL to QOL groups | |
|---|---|---|---|---|
| EuroQoL Visual Analog Scale | ||||
| 0–100 scale; 0=worst imaginable health, 100=perfect health | ||||
| Baseline | ||||
| No. of patients | 6981 | 5978 | 1003 | 0.61 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 75.9 ± 16.2 | 76.0 ± 16.2 | 75.5 ± 16.3 | |
| Year 1 | ||||
| No. of patients | 6319 | 5405 | 914 | 0.79 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 78.3 ± 15.9 | 78.4 ± 15.8 | 77.8 ± 16.6 | |
| Year 2 | ||||
| No. of patients | 5851 | 5010 | 841 | 0.27 |
| Median (interquartile range) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | 80.0 (70.0, 90.0) | |
| Mean ± standard deviation | 78.0 ± 16.2 | 78.1 ± 16.3 | 77.5 ± 16.1 | |
| EuroQoL Single Summary Index (Health State) | ||||
| Baseline | ||||
| No. of patients | 6969 | 5966 | 1003 | 0.65 |
| Median (interquartile range) | 0.88 (0.74, 1.00) | 0.88 (0.75, 1.00) | 0.88 (0.73, 1.00) | |
| Mean ± standard deviation | 0.85 ± 0.20 | 0.85 ± 0.20 | 0.85 ± 0.21 | |
| Year 1 | ||||
| No. of patients | 6338 | 5422 | 916 | 0.42 |
| Median (interquartile range) | 0.88 (0.74, 1.00) | 0.88 (0.74, 1.00) | 0.88 (0.73, 1.00) | |
| Mean ± standard deviation | 0.85 ± 0.21 | 0.85 ± 0.21 | 0.84 ± 0.22 | |
| Year 2 | ||||
| No. of patients | 5916 | 5060 | 856 | 0.23 |
| Median (interquartile range) | 1.00 (0.74, 1.00) | 1.00 (0.74, 1.00) | 0.88 (0.73, 1.00) | |
| Mean ± standard deviation | 0.85 ± 0.21 | 0.85 ± 0.21 | 0.84 ± 0.23 | |
Acknowledgments
We thank Diane Minshall-Liu for additional follow-up QOL coordination and site management and Melanie R. Daniels for editing this manuscript. We are also greatly indebted to the site coordinators at each of the participating HAT sites who enrolled and followed the study patients. Finally, we thank the participating patients and spouse/companions who agreed to be a part of this study.
FUNDING SOURCES
Supported by grant (UO1 HL67972) from the National Heart, Lung, and Blood Institute/National Institutes of Health, Bethesda, MD. Additional funding came from Philips Medical Systems and Laerdal Medical as a subsidiary distributor.
Supported by grant (UO1 HL67972) from the National Heart, Lung, and Blood Institute/National Institutes of Health, Bethesda, MD.
Footnotes
Clinical Trials Registration Number: ClinicalTrials.gov number, NCT00047411 http://clinicaltrials.gov/ct2/show/NCT00047411?term=HAT&rank=9
DISCLOSURES
Dr. Mark reports receiving research support from Medtronic; Dr. Smith, having stock options in Cameron Health; Dr. Poole, receiving research support from Biotronik, participating in the speaker’s bureaus for and receiving honoraria from Medtronic, St. Jude Medical, and Boston Scientific; Dr. Bardy, consulting for Philips Medical Systems.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Hallstrom AP, Ornato JP, Weisfeldt M, Travers A, Christenson J, McBurnie MA, Zalenski R, Becker LB, Schron EB, Proschan M. Public-access defibrillation and survival after out-of-hospital cardiac arrest. N Engl J Med. 2004;351:637–46. doi: 10.1056/NEJMoa040566. [DOI] [PubMed] [Google Scholar]
- 2.Page RL, Joglar JA, Kowal RC, Zagrodzky JD, Nelson LL, Ramaswamy K, Barbera SJ, Hamdan MH, McKenas DK. Use of automated external defibrillators by a U.S. airline. N Engl J Med. 2000;343:1210–6. doi: 10.1056/NEJM200010263431702. [DOI] [PubMed] [Google Scholar]
- 3.Valenzuela TD, Roe DJ, Nichol G, Clark LL, Spaite DW, Hardman RG. Outcomes of rapid defibrillation by security officers after cardiac arrest in casinos. N Engl J Med. 2000;343:1206–9. doi: 10.1056/NEJM200010263431701. [DOI] [PubMed] [Google Scholar]
- 4.Caffrey SL, Willoughby PJ, Pepe PE, Becker LB. Public use of automated external defibrillators. N Engl J Med. 2002;347:1242–7. doi: 10.1056/NEJMoa020932. [DOI] [PubMed] [Google Scholar]
- 5.Bardy GH, Lee KL, Mark DB, Poole JE, Toff WD, Tonkin AM, Smith W, Dorian P, Packer DL, White RD, Longstreth WT, Jr, Anderson J, Johnson G, Bischoff E, Yallop JJ, McNulty S, Ray LD, Clapp-Channing NE, Rosenberg Y, Schron EB. Home use of automated external defibrillators for sudden cardiac arrest. N Engl J Med. 2008;358:1793–804. doi: 10.1056/NEJMoa0801651. [DOI] [PubMed] [Google Scholar]
- 6.Cagle AJ, Diehr P, Meischke H, Rea T, Olsen J, Rodrigues D, Yakovlevitch M, Amidon T, Eisenberg M. Psychological and social impacts of automated external defibrillators (AEDs) in the home. Resuscitation. 2007;74:432–8. doi: 10.1016/j.resuscitation.2007.01.028. [DOI] [PubMed] [Google Scholar]
- 7.Chen MA, Eisenberg MS, Meischke H. Impact of in-home defibrillators on postmyocardial infarction patients and their significant others: an interview study. Heart Lung. 2002;31:173–85. doi: 10.1067/mhl.2002.124344. [DOI] [PubMed] [Google Scholar]
- 8.Haugk M, Robak O, Sterz F, Uray T, Kliegel A, Losert H, Holzer M, Herkner H, Laggner AN, Domanovits H. High acceptance of a home AED programme by survivors of sudden cardiac arrest and their families. Resuscitation. 2006;70:263–74. doi: 10.1016/j.resuscitation.2006.03.010. [DOI] [PubMed] [Google Scholar]
- 9.Bardy GH, Lee KL, Mark DB, Poole JE, Toff WD, Tonkin AM, Smith W, Dorian P, Yallop JJ, Packer DL, White RD, Longstreth W, Anderson J, Johnson G, Bischoff E, Munkers CD, Brown A, McNulty S, Ray LD, Clapp-Channing NE, Rosenberg Y, Salive M, Schron EB. Rationale and design of the Home Automatic External Defibrillator Trial (HAT) Am Heart J. 2008;155:445–54. doi: 10.1016/j.ahj.2007.12.008. [DOI] [PubMed] [Google Scholar]
- 10.Berwick DM, Murphy JM, Goldman PA, Ware JE, Jr, Barsky AJ, Weinstein MC. Performance of a five-item mental health screening test. Med Care. 1991;29:169–76. doi: 10.1097/00005650-199102000-00008. [DOI] [PubMed] [Google Scholar]
- 11.Ware JE, Kosinski M. Interpreting SF-36 summary health measures: a response. Qual Life Res. 2001;10:405–13. doi: 10.1023/a:1012588218728. [DOI] [PubMed] [Google Scholar]
- 12.Mark DB, Anstrom KJ, Sun JL, Clapp-Channing NE, Tsiatis AA, vidson-Ray L, Lee KL, Bardy GH. Quality of life with defibrillator therapy or amiodarone in heart failure. N Engl J Med. 2008;359:999–1008. doi: 10.1056/NEJMoa0706719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mark DB, Pan W, Clapp-Channing NE, Anstrom KJ, Ross JR, Fox RS, Devlin GP, Martin CE, Adlbrecht C, Cowper PA, Ray LD, Cohen EA, Lamas GA, Hochman JS. Quality of life after late invasive therapy for occluded arteries. N Engl J Med. 2009;360:774–83. doi: 10.1056/NEJMoa0805151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Baker B, Kazarian S, Marquez-Julio A. Perceived interpersonal attitudes and psychiatric complaints in patients with essential hypertension. J Clin Psychol. 1994;50:320–4. doi: 10.1002/1097-4679(199405)50:3<320::aid-jclp2270500303>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
- 15.Dracup K, Moser DK, Guzy PM, Taylor SE, Marsden C. Is cardiopulmonary resuscitation training deleterious for family members of cardiac patients? Am J Public Health. 1994;84:116–8. doi: 10.2105/ajph.84.1.116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.EuroQol--a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy. 1990;16:199–208. doi: 10.1016/0168-8510(90)90421-9. [DOI] [PubMed] [Google Scholar]
- 17.Rosser RM, Stintonen H. The EuroQoL quality of life project. In: Walker SR, Rosser RM, editors. Quality of Life Assessment: Key Issues in the 1990s. Dordrecht: Kluwer Academic Publishers; 1993. pp. 197–9. [Google Scholar]
- 18.Dracup K, Moser DK, Taylor SE, Guzy PM. The psychological consequences of cardiopulmonary resuscitation training for family members of patients at risk for sudden death. Am J Public Health. 1997;87:1434–9. doi: 10.2105/ajph.87.9.1434. [DOI] [PMC free article] [PubMed] [Google Scholar]