Participants
All patients who have had an elective, subacute or acute PCI procedure for ≥1 coronary occlusion participated in this study. They were recruited from the cardiology ward of the TweeSteden hospital (Tilburg). Participants were excluded if they were under the age of 18, had cognitive impairment, poor linguistic competence, dementia or other aspects that could have a negative impact on self-report. Participants with life threatening comorbidity (e.g. terminal illness) were also excluded.
Procedure
Researchers from the participating hospital selected potential participants from the patients file, if they have had a PCI procedure, and invited them to participate in this study. Potential participants had to indicate whether or not …show more content…
This approach uses four steps in establishing mediation [30, 36]: (I) The causal variable (X) should correlate with the outcome variable (Y) to establish that there may be a mediation effect (path c in Figure 1); (II) The causal variable should correlate with the mediator (M) by treating the mediator as an outcome variable (path a in Figure 1); (III) The mediator should affect the outcome variable. Though, the mediator and outcome variable may be correlated because they are both affected by the causal variable. Thus, the causal variable should be controlled in determining the effect of the mediator on the outcome (path b and c’ in Figure 1), and (IV) Complete mediation is present when the effect of the predictor on the criterion decreases to zero with the inclusion of the mediator. When the effect of the predictor on the criterion decreases, but is not equal to zero, partial mediation had occurred [37]. The paths (c, a, b, and c’) were estimated by using linear regression. After the establishment of either complete mediation or partial mediation ‘the Sobel test’ is used. The Sobel test was used supplementary to the Causal Steps Approach and not as a replacement. It uses unstandardized coefficients and standard errors for testing the null hypothesis that the ‘true’ mediation effect is zero, with the p-value adopt from the standard normal distribution. The Sobel test will give information about the size and significance of the mediation
All patients who have had an elective, subacute or acute PCI procedure for ≥1 coronary occlusion participated in this study. They were recruited from the cardiology ward of the TweeSteden hospital (Tilburg). Participants were excluded if they were under the age of 18, had cognitive impairment, poor linguistic competence, dementia or other aspects that could have a negative impact on self-report. Participants with life threatening comorbidity (e.g. terminal illness) were also excluded.
Procedure
Researchers from the participating hospital selected potential participants from the patients file, if they have had a PCI procedure, and invited them to participate in this study. Potential participants had to indicate whether or not …show more content…
This approach uses four steps in establishing mediation [30, 36]: (I) The causal variable (X) should correlate with the outcome variable (Y) to establish that there may be a mediation effect (path c in Figure 1); (II) The causal variable should correlate with the mediator (M) by treating the mediator as an outcome variable (path a in Figure 1); (III) The mediator should affect the outcome variable. Though, the mediator and outcome variable may be correlated because they are both affected by the causal variable. Thus, the causal variable should be controlled in determining the effect of the mediator on the outcome (path b and c’ in Figure 1), and (IV) Complete mediation is present when the effect of the predictor on the criterion decreases to zero with the inclusion of the mediator. When the effect of the predictor on the criterion decreases, but is not equal to zero, partial mediation had occurred [37]. The paths (c, a, b, and c’) were estimated by using linear regression. After the establishment of either complete mediation or partial mediation ‘the Sobel test’ is used. The Sobel test was used supplementary to the Causal Steps Approach and not as a replacement. It uses unstandardized coefficients and standard errors for testing the null hypothesis that the ‘true’ mediation effect is zero, with the p-value adopt from the standard normal distribution. The Sobel test will give information about the size and significance of the mediation