3.1 Characteristics of the group
T1 and T2 data were obtained from 222 MI survivors. Their sociodemographic and clinical characteristics are presented in Table 1. The men (n = 163) and women (n = 59) were similar with respect to most of the tested characteristics. At T1, men were more likely to be professionally active and to have been assigned to intensive cardiac training than women. At T3, the number of participants dropped to 140 participants (42 women and 98 men). Men and women did not differ in professional activity, as the number of men in work had fallen. The attrition analysis revealed that T1 and T3 samples were similar in respect of the majority of the sociomedical characteristics, except that the T3 participants had fewer comorbidities at baseline (χ²(2, n = 222) = 10.31; p = 0.006) and were more likely to be engaged in low-level cardiac training (χ²(2, n = 222) = 8.58; p = 0.02).
Table 1
Gender differences in personality traits and coping strategies
Scale
|
|
Time 1
(n = 222)
|
|
Time 2
(n = 222)
|
|
Time 3
(n = 140)
|
|
|
M (SD)
|
F (p)
|
M (SD)
|
F (p)
|
M (SD)
|
F (p)
|
Personality
|
|
|
|
|
|
|
|
Anxiety-trait
|
W
|
25.14 ± 4.90
|
18.78 (< .001)
|
—
|
—
|
—
|
—
|
|
M
|
22.02 ± 4.67
|
—
|
—
|
—
|
—
|
Anger-trait
|
W
|
25.07 ± 6.19
|
1.57 (.211)
|
—
|
—
|
—
|
—
|
|
M
|
23.88 ± 6.27
|
—
|
—
|
—
|
—
|
Curiosity-trait
|
W
|
29.9 ± 3.98
|
4.26 (.040)
|
—
|
—
|
—
|
—
|
|
M
|
31.14 ± 3.95
|
—
|
—
|
—
|
—
|
Optimism
|
W
|
14.71 ± 3.77
|
1.19 (.277)
|
—
|
—
|
—
|
—
|
|
M
|
15.36 ± 3.98
|
—
|
—
|
—
|
—
|
Coping strategy
|
|
|
|
|
|
|
|
Reinterpretation
|
W
|
35.44 ± 6.08
|
4.74
(.031)
|
35.20 ± 5.25
|
5.91
(.016)
|
35.62 ± 6.76
|
97.00
(.327)
|
|
M
|
33.18 ± 6.14
|
33.32 ± 5.84
|
34.47 ± 6.15
|
Religion
|
W
|
10.97 ± 3.3
|
30.13
(< .001)
|
10.86 ± 3.34
|
24.11
(< .001)
|
9.86 ± 4.09
|
2.38 (.125)
|
|
M
|
8.17 ± 3.38
|
8.31 ± 3.45
|
8.79 ± 3.62
|
Humor
|
W
|
10.28 ± 3.5
|
10.1 (.002)
|
11 ± 3.22
|
4.38 (.038)
|
11.00 ± 3.22
|
1.17 (.281)
|
|
M
|
12.04, ± 3.68
|
12.17 ± 3.84
|
12.17 (3.84)
|
Resignation
|
W
|
7.58 ± 2.17
|
2.12 (.147)
|
7.98 ± 2.24
|
4.47 (.036)
|
7.62 ± 2.4
|
3.76 (.054)
|
|
M
|
7.12 ± 2.04
|
7.23 ± 2.29
|
6.87 ± 1.96
|
Social support
|
W
|
18.97 ± 5.04
|
2.08 (.151)
|
19.88 ± 4.27
|
2.96 (.087)
|
18.45 ± 4.43
|
.04 (.842)
|
|
M
|
17.93 ± 4.64
|
18.6 6 ± 4.8
|
18.3 ± 4.16
|
Problem solving
|
W
|
23.69 ± 5.41
|
.30 (.584)
|
24.51 ± 4.5
|
1.05 (.307)
|
23.38 ± 5.42
|
.52 (.472)
|
|
M
|
23.22 ± 5.78
|
23.7 ± 5.43
|
24.06 ± 4.97
|
Substance use
|
W
|
4.88 ± 1.98
|
.96 (.330)
|
5.12 ± 2.25
|
1.31 (.255)
|
4.67 ± 1.9
|
1.70 (.195)
|
|
M
|
5.18 ± 2
|
5.53 ± 2.44
|
5.32 ± 2.98
|
M, mean; SD, standard deviation; W, women; M, men. |
3.2 Gender differences in personality, coping and HRQoL after MI
There were no gender differences in anger and optimism. Women reported higher trait anxiety and lower trait curiosity than men (Table 2).
Table 2
Gender and HRQoL at subsequent timepoints
HRQoL
dimension
|
Gender
|
Time 1
|
Time 2
|
Time 3
|
Mdn
|
U (p)
|
Mdn
|
U (p)
|
Mdn
|
U (p)
|
|
|
Physical
(symptoms)
|
W
|
3
|
6121.5 (< .001)
|
2
|
5777.5 (.015)
|
3
|
2284.5 (.288)
|
M
|
1
|
0
|
1
|
Mental
(discomfort)
|
W
|
5
|
6340.5 (< .001)
|
4
|
6039
(.003)
|
3
|
2378
(.142)
|
M
|
2
|
1
|
2
|
Mdn, median; U, the result of a Mann–Whitney U test; M, men; W, women |
At T1 and T2, women made more frequent use of reinterpretation and religion than men, whereas men made more use of humour than women. At T2, women also made more use of resignation. At T3, there were no gender differences in coping strategies (Table 2). At T1 and T2, women had a lower physical and mental HRQoL, but these differences had disappeared by T3 (Table 2).
3.3 HRQoL predictors in females and males at the beginning of cardiac rehabilitation
Table 3 shows that at T1, none of the factors predicted women’s symptoms, although their discomfort was negatively predicted by education and trait curiosity, and positively predicted by trait anxiety and the use of resignation as a coping strategy. In men, symptom severity was negatively associated with being professionally active before MI, and positively associated with current angina and trait anger. Being professionally active and being optimistic were negatively associated with discomfort, whereas the use of anxiolytics, trait anxiety and trait anger were positively associated with discomfort.
Table 3
Predictors of HRQoL at the beginning of cardiac rehabilitation in women and men – comprehensive models
|
Women
|
Men
|
Predictors
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Demographics
|
Education
|
-
|
− .21 (.03)
|
− .08 (ns)
|
-
|
Employment before MI
|
-
|
.02 (.ns)
|
− .18 (.02)
|
− .12 (.04)
|
Not included: age, marital status, place of residence
|
ΔR² = 0
|
ΔR² = .22
p = .001
|
ΔR² = .10
p = .001
|
ΔR² = .04
p = .009
|
Clinical characteristics
|
Angina before MI
|
-
|
-
|
.12 (ns)
|
-
|
Angina at T1
|
-
|
-
|
.21 (.003)
|
.04 (ns)
|
Anxiolytics
|
-
|
-
|
.08 (.ns)
|
.15 (.01)
|
Not included: type of MI, PTCA, complications, number of MI, betablockers, length of hospitalization, comorbidity, time gap between MI and rehabilitation, intensity of physical training
|
ΔR² = 0
|
ΔR² = 0
|
ΔR² = .13
p < .001
|
ΔR² = .10
p < .001
|
Personality factors
|
Curiosity-trait
|
-
|
− .33 (.003)
|
-
|
-
|
Anxiety-trait
|
-
|
.31 (.02)
|
-
|
.33 (.001)
|
Anger-trait
|
-
|
.20 (ns)
|
.31 (.001)
|
.23 (.008)
|
Optimism
|
-
|
-
|
-
|
− .19 (.003)
|
|
ΔR² = 0
|
ΔR² = .36
p < .01
|
ΔR² = .09
p < .001
|
ΔR² = .37
p < .001
|
Coping strategies
|
Resignation at T1
|
-
|
.19 (.04)
|
-
|
− .01 (.ns)
|
Religion at T1
|
-
|
-
|
.10 (.ns)
|
.09 (.ns)
|
Substance use at T1
|
-
|
-
|
-
|
.06 (ns)
|
Not included: reinterpretation, humor, social support, problem solving
|
ΔR² = 0
|
ΔR² = .03
p = .04
|
ΔR² = .01
p = ns
|
ΔR² = .01
p = ns
|
Total
|
adjR² = 0
|
adjR² = .57
|
adjR² = .30
|
adjR² = .49
|
If ‘-‘, the predictor was not included in the given analysis; not included, not entered into the comprehensive model; ns, not significant |
|
Women
|
Men
|
Predictors
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Demographics
|
Education
|
-
|
− .27 (.04)
|
-
|
− .08 (ns)
|
Employment before MI
|
-
|
-
|
− .17 (.02)
|
− .08 (ns)
|
Not included: age, marital status, place of residence
|
ΔR ²= 0
|
ΔR² = .10
p = .02
|
ΔR² = .08
p < .001
|
ΔR ²= .08
p = .002
|
Clinical characteristics
|
Angina before MI
|
-
|
-
|
.19 (.009)
|
-
|
comorbidity
|
-
|
-
|
.09 (ns)
|
-
|
Angina at T1
|
-
|
-
|
.31 (.001)
|
0.07 (ns)
|
Anxiolytics
|
-
|
.25 (.04)
|
.03 (ns)
|
.21 (.001)
|
Intensity of physical training
|
|
|
− .10 (ns)
|
-
|
Not included: type of MI, complications, PTCA, number of MI, betablockers, length of hospitalization, time gap between MI and rehabilitation
|
ΔR² = 0
|
ΔR² = .06
p = .06
|
ΔR² = .24
p < .001
|
ΔR² = .12
p < .001
|
Personality factors
|
Curiosity-trait
|
-
|
-
|
-
|
-
|
Anxiety-trait
|
-
|
-
|
-
|
.23 (.02)
|
Anger-trait
|
-
|
-
|
.26 (.001)
|
.17 (.06)
|
Optimism
|
-
|
-
|
-
|
− .23 (.001)
|
|
ΔR² = 0
|
ΔR² = 0
|
ΔR² = .07
p < .001
|
ΔR² = .26
p < .001
|
Coping strategies
|
Resignation at T1
|
-
|
.38 (.008)
|
-
|
-
|
Problem solving at T1
|
− .18 (ns)
|
|
|
-
|
Religion at T1
|
-
|
-
|
-
|
.18 (.ns)
|
Substance use at T1
|
-
|
-
|
-
|
.08 (ns)
|
Substance use at T2
|
− .26 (.08)
|
-
|
-
|
-
|
Resignation at T2
|
.29 (.051)
|
− .10 (ns)
|
-
|
-
|
Religion at T2
|
-
|
-
|
-
|
− .01 (ns)
|
Problem solving T2
|
-
|
-
|
− .08 (ns)
|
-
|
Not included: T1 and T2 reinterpretation, T1 and T2 humor, T1 and T2 social support
|
ΔR² = .12
p = .07
|
ΔR² = .12
p = .02
|
ΔR² = .006
p = ns
|
ΔR² = .03
p = ns
|
Total
|
adjR² =.07
|
adjR² = .22
|
adjR² = .36
|
adjR² = .45
|
3.4 HRQoL predictors in females and males at the end of cardiac rehabilitation
Table 4 shows that at T2, none of the factors investigated predicted women’s symptoms, although there were marginal negative relationships between symptoms and substance use, and positive with resignation. Education continued to be a negative predictor of discomfort, whilst the use of anxiolytics and resignation at T1 were positive predictors of discomfort. In men, the predictors of symptoms were, as at T1, pre-MI employment, current angina and trait anger; there was also a delayed effect of duration of angina before MI. The predictors of discomfort in men were the same as at T1, except that the effect of professional activity had disappeared.
|
Women
|
Men
|
Predictors
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Demographics
|
Education
|
-
|
-.27 (.04)
|
-
|
-.08 (ns)
|
Employment before MI
|
-
|
-
|
-.17 (.02)
|
-.08 (ns)
|
Not included: age, marital status, place of residence
|
ΔR ²= 0
|
ΔR² = .10
p = .02
|
ΔR² = .08
p < .001
|
ΔR ²= .08
p = .002
|
Clinical characteristics
|
Angina before MI
|
-
|
-
|
.19 (.009)
|
-
|
comorbidity
|
-
|
-
|
.09 (ns)
|
-
|
Angina at T1
|
-
|
-
|
.31 (.001)
|
0.07 (ns)
|
Anxiolytics
|
-
|
.25 (.04)
|
.03 (ns)
|
.21 (.001)
|
Intensity of physical training
|
|
|
-.10 (ns)
|
-
|
Not included: type of MI, complications, PTCA, number of MI, betablockers, length of hospitalization, time gap between MI and rehabilitation
|
ΔR² = 0
|
ΔR² = .06
p = .06
|
ΔR² = .24
p < .001
|
ΔR² = .12
p < .001
|
Personality factors
|
Curiosity-trait
|
-
|
-
|
-
|
-
|
Anxiety-trait
|
-
|
-
|
-
|
.23 (.02)
|
Anger-trait
|
-
|
-
|
.26 (.001)
|
.17 (.06)
|
Optimism
|
-
|
-
|
-
|
-.23 (.001)
|
|
ΔR² = 0
|
ΔR² = 0
|
ΔR² = .07
p < .001
|
ΔR² = .26
p < .001
|
Coping strategies
|
Resignation at T1
|
-
|
.38 (.008)
|
-
|
-
|
Problem solving at T1
|
-.18 (ns)
|
|
|
-
|
Religion at T1
|
-
|
-
|
-
|
.18 (.ns)
|
Substance use at T1
|
-
|
-
|
-
|
.08 (ns)
|
Substance use at T2
|
-.26 (.08)
|
-
|
-
|
-
|
Resignation at T2
|
.29 (.051)
|
-.10 (ns)
|
-
|
-
|
Religion at T2
|
-
|
-
|
-
|
-.01 (ns)
|
Problem solving T2
|
-
|
-
|
-.08 (ns)
|
-
|
Not included: T1 and T2 reinterpretation, T1 and T2 humor, T1 and T2 social support
|
ΔR² = .12
p = .07
|
ΔR² = .12
p = .02
|
ΔR² = .006
p = ns
|
ΔR² = .03
p = ns
|
Total
|
adjR² =.07
|
adjR² = .22
|
adjR² = .36
|
adjR² = .45
|
Table 4 Predictors of HRQoL at the end of cardiac rehabilitation in women and men – comprehensive models
If ‘-‘, the predictor was not entered into a block; not included, not entered into the comprehensive model; ns, not significant
3.5 HRQoL predictors in females and males one year after cardiac rehabilitation
Table 5 shows that at T3, the only predictor of women’s symptoms was angina at T3, although there was also a marginal association with professional activity at T3. Discomfort was positively predicted by use of beta-blockers at T1, angina at T3 and substance use at T2. In men, the strongest predictor of symptoms was angina at T3. The use of problem-solving at T1 was a negative predictor of discomfort at T3 (a delayed effect), whereas resignation at T3 was a positive predictor of discomfort.
|
Women
|
Men
|
Predictors
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Symptoms
β(p)
|
Discomfort
β(p)
|
Demographics
|
Marital status
|
.09 (ns)
|
-
|
-
|
-
|
Education
|
-
|
-
|
-.13 (ns)
|
-.16 (ns)
|
Employment before MI
|
-
|
-
|
-.16 (.07)
|
-
|
Employment at T3
|
-.25 (.06)
|
|
-
|
-.13 (ns)
|
Not included: age, place of residence
|
ΔR² = .22
p = .007
|
ΔR² = 0
|
ΔR² = .17
p < .001
|
ΔR² = .12
p = .002
|
Clinical characteristics
|
Betablockers
|
-
|
.29 (.04)
|
-
|
-
|
Comorbidity
|
.15 (ns)
|
-
|
-
|
-
|
Angina at T1
|
-
|
.22 (ns)
|
-
|
-
|
Anxiolytics
|
-
|
-
|
.11 (ns)
|
.10 (ns)
|
Rehospitalisation
|
|
-
|
.17 (.07)
|
-
|
Angina at T3
|
.49 (.001)
|
.36 (.01)
|
.35 (.001)
|
-
|
Not included: type of MI, PTCA, complications, number of MI, length of hospitalization, time gap between MI and rehabilitation, angina before MI, intensity of physical training, new MI
|
ΔR² = .27
p < .001
|
ΔR² = .28
p = .006
|
ΔR² = .21
p < .001
|
ΔR² = .04
p = .04
|
Personality factors
|
Anxiety-trait
|
-
|
-
|
-
|
-
|
Optimism
|
-
|
-
|
-.06 (ns)
|
-.08 (ns)
|
Not included: anger-trait, curiosity-trait
|
ΔR² = 0
|
ΔR² = 0
|
ΔR² = .02
p = ns
|
ΔR² = .02
p = ns
|
Coping strategies
|
Social suport at T1
|
-
|
-
|
-
|
.18 (ns)
|
Resignation at T1
|
-
|
-
|
.10 (ns)
|
-
|
Problem solving at T1
|
-
|
-
|
-.17 (ns)
|
-.30 (.02)
|
Substance use at T2
|
-
|
.31 (.03)
|
-
|
-
|
Resignation at T2
|
-
|
-
|
.06 (ns)
|
.11 (ns)
|
Humour at T2
|
-
|
-
|
-.16 (.08)
|
-
|
Problem solving at T2
|
-
|
-
|
.06 (ns)
|
-
|
Resignation at T3
|
-
|
-
|
.09 (ns)
|
.22 (.03)
|
Not included: T1, T2, T3 reinterpretation, T1, T2, T3 religion, T1 and T3 humor, T2 and T3 social support, T1 and T3 substance use, T3 problem solving
|
ΔR² = 0
|
ΔR² = .09
p = .03
|
ΔR² = .05
p = ns
|
ΔR² = .11
p = .01
|
Total
|
adjR² =.44
|
adjR² = .30
|
adjR² = .37
|
adjR² = .23
|
Table 5 Predictors of HRQoL one year after completion of cardiac rehabilitation in women and men – comprehensive models
If ‘-‘, the predictor was not entered into a block; not included, not entered into the comprehensive model; ns, not significant
3.6 HRQoL predictors after controlling for HRQoL at earlier timepoints
Given that there were gender differences in HRQoL at T1 and T2, we checked whether predictors of T2 and T3 HRQoL remained significant after controlling for HRQoL at the previous timepoints.
Using this approach there were two predictors of women’s symptoms at T2: substance use at T2 (β = –0.24, p = 0.01) and resignation at T2 (β = 0.20, p = 0.04). Education, use of anxiolytics and resignation at T1 no longer predicted T2 discomfort. Symptoms and discomfort at T1 explained an additional 59% and 54% of the variance in T2 symptoms and discomfort, respectively (p <0.001 for each). In men, T2 symptoms were still predicted by pre-MI angina and angina at T1, but pre-MI employment and trait anger were no longer significant predictors. Use of anxiolytics and optimism still predicted T2 discomfort, but trait anxiety did not. Symptoms and discomfort at T1 explained the additional 51% and 50% of the variance in T2 symptoms and discomfort, respectively (p <0.001 for each).
In women, T3 outcomes were very similar to those reported in the previous section. Adding symptoms at T1 and T2 did not increase significantly the amount of variance in T3 symptoms (6%, p = 0.29; symptoms at T1 β = 0.33, p = 0.09; symptoms at T2 β= 0.08, p = 0.69). Substance use at T2 and angina at T3 still predicted women’s T3 discomfort, but use of beta-blockers did not. Adding discomfort at T1 and T2 did not increase significantly the amount of variance in T3 discomfort (10%, p = 0.08; discomfort at T1 β = 0.03, p = 0.93; discomfort at T2 β = 0.03, p = 0.39). In men T3 outcomes were also very similar to those reported previously. Adding symptoms at T1 and T2 increased the amount of variance in T3 symptoms (14%, p <0.001), however symptoms at T1 and T2 were not significant predictors (β = 0.33, p = 0.09 and β = 0.08, p = 0.69, respectively). Problem-solving at T1 and resignation at T3 still predicted men’s T3 discomfort. Adding discomfort at T1 and T2 increased the amount of variance in T3 discomfort (11%, p = 0.002), however discomfort at T1 and T2 were not significant predictors (β = 0.09, p = 0.48 and β = 0.23, p = 0.07, respectively).