The authors regret that in our study, which assessed perceived levels of stress, resilience, and coping strategies related to COVID-19 in a sample of community dwelling older adults, the total score of the perceived stress scale (PSS) was computed incorrectly. The score presented did not consider that some of the items in the questionnaire should have been reversed before creating the overall sum. We report here a revised version of all the analyses, figures and tables that included the PSS variable. The analyses using the corrected PSS measure revealed similar results as reported in the original report and hence, our main findings and conclusions were not affected by the error and remain unchanged. The authors would like to apologize for any inconvenience caused.
The mean perceived stress was 15.5 ± 7.8 (SD), ranging from 0 to 37. The relationship between stress and demographics remained the same. Stress was inversely related to age (r = −0.25, p = 0.003). Women reported significantly more stress (mean 16.8 ± 8.2) than men (mean 13.8 ± 6.9, t = 2.3, p = 0.03). No relationship was found between stress and education. The PSS and the additional COVID-19 stress total scores remained significantly correlated with each other (r = 0.62, p < 0.001).
Bivariate correlations for all variables including the new PSS are presented in Fig. 4R. Similar as our previous report, resilience was significantly related to more frequent use of behavioral disengagement, self-blame, substance use, venting, and instrumental support. However, the relationship between resilience and self-distraction disappeared. Two additional significant relationships were observed; in that decreased stress was related to increased acceptance as well as decreased denial.
Fig. 4R.
Recalculation of the bivariate correlations between stress, resilience and coping strategies. The numbers and circles in the correlation matrix demonstrates significant r-values. The color-coded scale depicts the strength of the correlation with a darker color representing higher significance. Blue represents positive relationships and red are negative relationships.
Bivariate correlations for all CDC recommended strategies and the new PSS measure are presented in Supplemental Fig. 2R. As previously reported, higher resilience was strongly related to lower stress. We also observed an additional weak negative relationship between PSS and the coping strategy ‘I've been exercising regularly’, such that increased use of this coping behavior was related to decreased stress.
Supplemental Fig. 2R.
Recalculation of the bivariate correlations between stress, resilience and coping strategies recommended by the CDC. Note that the legends on the top have been replaced by numbers. The numbers and circles in the correlation matrix demonstrate significant r-values. The color-coded scale depicts the strength of the correlation with a darker color representing higher significance. Blue represents positive relationships and red are negative relationships.
The hierarchical multiple regression analyses for stress revealed similar results as previously reported in the paper. That is, in step 1, resilience and coping strategies accounted for 60% of the variance in stress (Table 2R), of which much was due to the independent main effect of resilience. Similar as reported in our paper, we observed significant independent main effects for substance use, self-blame and self-distraction. However, the previously significant main effect of instrumental support changed to trending (p = 0.052). We also observed additional significant main effects of denial and positive reframing.
Table 2R.
Revised Hierarchical Multiple regression analysis results of main (step 1) and interaction effects (step 2).
| Step 1 — Main Effects | ||
|---|---|---|
| Variables | ß (p) | R2 |
| 0.602*** | ||
| Resilience | −0.546 (<0.001) *** | |
| Acceptance | −0.004 (0.956) | |
| Active Coping | −0.031 (0.693) | |
| Behavioral Disengagement | 0.017 (0.798) | |
| Denial | 0.125 (0.039) * | |
| Emotional Support | −0.141 (0.0497) * | |
| Humor | 0.035 (0.587) | |
| Instrumental Support | 0.143 (0.0520) # | |
| Planning | 0.08 (0.286) | |
| Positive Reframing | −0.151 (0.04) * | |
| Religion | 0.038 (0.531) | |
| Self-Blame | 0.213 (0.002) ** | |
| Self-Distraction | 0.183 (0.014) * | |
| Substance Use | 0.194 (0.003) ** | |
| Venting | 0.071 (0.257) | |
| Step 2 – Interactions Between Resilience and Coping Strategies | |||
|---|---|---|---|
| Variables | ß | R2 | Δ R2 |
| 0.667*** | 0.065 | ||
| Resilience x Acceptance | 0.024 (0.965) | ||
| Resilience x Active Coping | 0.216 (0.699) | ||
| Resilience x Behavioral Disengagement | −0.841 (0.028) * | ||
| Resilience x Denial | 0.171 (0.629) | ||
| Resilience x Emotional Support | −0.934 (0.056) # | ||
| Resilience x Humor | −0.595 (0.179) | ||
| Resilience x Instrumental Support | 0.62 (0.179) | ||
| Resilience x Planning | −1.005 (0.084) # | ||
| Resilience x Positive Reframing | 0.281 (0.613) | ||
| Resilience x Religion | 0.095 (0.797) | ||
| Resilience x Self-Blame | 0.786 (0.027) * | ||
| Resilience x Self-Distraction | 0.295 (0.544) | ||
| Resilience x Substance Use | 0.128 (0.62) | ||
| Resilience x Venting | −0.58 (0.122) | ||
Note. Beta values represent standardized values at final model. # trending, *p < 0.05, ***p < 0.001.
The model at Step 2 including the interactions between resilience and each coping strategy remained statistically significant, F (15, 125) = 12.61, p < 0.001 (Table 2R). The interaction terms between resilience and coping strategies explained a significant proportion of the variance in stress, ΔR2 = 0.065, ΔF (29, 111) = 7.656, p < 0.001. Same as in our previous report, we observed a significant interaction effect between resilience and self-blame (ß = 0.328, Std.ß = 0.786, t = 2.241, p = 0.027; Fig. 5BR). Simple slope analysis demonstrated that at low levels of resilience (-1SD) the relationship between self-blame and stress was non-significant (ß = 0.61 ± 1, p = 0.54). At mean and high levels of resilience (+1SD), the relationships between self-blame and stress were positive and significant (ß = 2.56, p < 0.001 and ß = 4.52, p < 0.001 respectively).
Fig. 5.
AR and BR. The relationship between stress and planning (A) and self-blame (B) at low (-1SD), mean, and high (+1SD) levels of resilience. Responses are labeled as 0 = I haven't been doing this at all; 1 = I've been doing this a little bit; 2 = I've been doing this a medium amount; 3 = I've been doing this a lot.
The interaction between resilience and planning was trending (ß = −0.225 ± 0.129, Std.ß = −1.01, p = 0.084). However, same as previously reported, the simple slope analysis revealed that the high resilience group did not demonstrated a significant slope (p = 0.74), while the mean group was trending (ß = 1.04 ± 0.57, p = 0.07), and the low resilient group was significant (ß = 2.38 ± 1.02, p = 0.02; Fig. 5AR).
Finally, an additional interaction effect was observed between resilience and behavioral disengagement (ß = −0.53 ± 0.24, Std.ß = −0.841, p = 0.028; see Additional figure provided). However, the simple slope analysis revealed trending and non-significant relationships for each resilience group. That is, at low levels of resilience (-1SD) a trending positive relationship between behavioral disengagement and stress was observed (ß = 2.46, p = 0.08). For the mean group, a non-significant relationship was observed (ß = −0.68 ± 1.19, p = 0.57). At high levels of resilience (+1SD) a trending negative relationship was observed between behavioral disengagement and stress (ß = -.3.81, p = 0.09). This last interaction should be interpreted with caution, due to the trending findings in the post hoc simple slope analysis. Nevertheless, we believe that it adds further evidence, in line with the goodness-of-fit hypothesis, to the idea that high resilience dampens the effect on behaviors that are maladaptive.
Additional figure.
The relationship between stress and Behavioral Disengagement at low (-1SD), mean, and high (+1SD) levels of resilience. Responses are labeled as 0 = I haven't been doing this at all; 1 = I've been doing this a little bit; 2 = I've been doing this a medium amount; 3 = I've been doing this a lot.