Participant characteristics
A total of 244 patients consented to the study of which 158 (64.7%) returned the Decision Self-Efficacy Scale at baseline. Of these, 73 patients were treated using SABR, and 85 had VATS. We did not find any baseline difference between patients who completed the DSE (n = 158) and those who didn’t (N = 86) in terms of age (p:0.17), gender (p = 0.34) and PS > 1(p = 0.23). The baseline clinical characteristics of the participants included in this study are in listed in Table 1.
Table 1 Baseline demographic and clinical characteristics of participants N = 158.
Variable | All patients with DSE (n = 158) | SABR (n = 73) | Surgery (n = 85) |
Treatment (surgery, %) | 85 (53.8) | | |
Gender (male, %) | 69 (43.6) | 26 (35.6) | 43 (50.5) |
Age (years, SD) | 72.4 (8.6) | 74.5 (9.3) | 70.5 (7.5) |
Comorbidity (yes,%) | 135 (85.4) | 67 (91.7) | 68 (80) |
FEV1% (SD) | 83.5 (25.1) | 75.5 (27.4) | 89.2 (21.8) |
DLCO% (SD) | 77.6 (22.2) | 69.6 (22.8) | 83.5 (19.9) |
Current Smokers (n, %) | 34 (22.6%) | 19 (27.1) | 15 (18.7) |
PS 0–1 (n, %) | 104 (62.4) | 39 (53.4) | 15 (17.6) |
DSE (Mean, SD) | 81.7 (23) | 79.5 (23) | 83.6 (22.9) |
Results are expressed as mean and standard deviation (SD) for numeric variables and as count and percentages for categorical variables. FEV1: forced expiratory volume in one second; diffusing capacity of the lung for carbon monoxide (DLCO) expressed in percentage of predicted value, Karnofsky performance score (PS).
Mean value of the DSE was 81.7 (SD 23). In the Surgical group the mean score was 83.6 (SD 22.9) and in the SABR one was 79.5 (SD 23). DSE is the main score representing the overall efficacy in making the decision. Patients treated with SABR were older, with more comorbidities, lower FEV1 and DLCO values and higher PS. These differences were, however expected as the SABR treatment was indicated to those patients who were not physiologically fit for surgery.
Principal Components Analysis
A PCA analysis was conducted on the 11 items with oblique rotation (direct oblimin) as it was expected that the factors would not be independent. The Kaiser-Meyer-Olkin (KMO) measure verified sampling adequacy for the analysis (KMO = .91) well above the minimum criterion of .50, in addition all KMO values for individual items were ≥ .88 and Bartlett’s test of sphericity was also significant at p < .001. An initial analysis was run to obtain eigen values for each factor in the data. Two factors had Eigen values over Kaisers criterion of 1 and explained 81.2% of the variance. The screen plot depicted two inflections confirming Eigen values over 1. Table 2 shows the results of the PCA, suggesting that factor 1 (Items 5–11) represents overcoming barriers and factor 2 (1–4) represents information seeking.
Table 2
Summary of PCA results for the Decision self-efficacy scale (N = 158)
Item on the DSE scale | Overcoming barriers | Information seeking |
| Factor 1 | Factor 2 |
5 Ask questions without feeling dumb | 1.01 | |
6. Express my concerns about each choice | 0.954 | |
7. Ask for advice | 0.83 | |
8. Figure out the choice that best suits me | 0.726 | |
9. Handle unwanted pressure from others in making my choice | 0.911 | |
10. Let the clinic team know what’s best for me | 0.754 | |
11. Delay my decision if I feel I need more time | 0.68 | |
1. Get the facts about medication choices available to me | | 0.999 |
2. Get the facts about the benefits of each choice | | 0.98 |
3. Get the facts about the benefits and risks of each choice | | 0.848 |
4. Understand the information enough to be able to make a choice | | 0.631 |
Eigen values | 7.91 | 1.02 |
% of variance | 71.96 | 9.13 |
Construct validity
The decision self-efficacy scale performed well in terms of psychometrics in this sample (Table 3): we calculated a value of 0.96 for Cronbach’s alpha for the total DSE score, 0.94 for Factor1 and 0.95 for Factor2. The overall amount of missing data was < 5%, totalling just 1.5% of the dataset. However, a notable ceiling effects were apparent for the total DSE, Factor1 and Factor2, with > 15% of participants obtaining the maximum score (Table 3).
Table 3
Descriptive statistics, floor and ceiling effects and internal reliability for the DSE, F1 and F2
| N | Mean | (SD) | Floor effect (%min score) | Ceiling effect (%max score) |
DSE | 158 | 81.7 | 23 | 0.63% | 32.9% |
Factor1 | 158 | 82.7 | 23.1 | 1.27% | 41.1% |
Factor2 | 158 | 80.7 | 25.1 | 1.27% | 43% |
The items correlated significantly at p = 0.001. A determinant value of 2.09E-006 (above the necessary value of 0.00001) revealed that the level of collinearity would not be detrimental to the analysis therefore, no items were removed. The correlation matrix is provided in the Additional File 1.
Known-group differences
Group comparisons revealed no significant mean differences between the two treatment groups in terms of overall self-efficacy score (DSE): SABR 79.5, Surgery 83.6 (p = 0.09). There were no statistical differences between the two groups for each of the eleven items either (Table 4).
Table 4
Total DSE and F1, F2 and DSE individual questions scores in SABR and Surgical groups
Item | Surgery (n = 85) Mean (SD) | SABR (n = 73) Mean (SD) | p value |
DSE total score | 83.6 (22.9) | 79.5 (23.1) | 0.09 |
*Factor1 | 84.3 (24.1) | 80.6 (23.8) | 0.13 |
**Factor2 | 83.2 (23) | 78.8(26.2) | 0.19 |
1. Get the facts about medication choices available to me | 85.2 (23.5) | 80.4(26.1) | 0.13 |
2. Get the facts about the benefits of each choice | 83.5 (26) | 79.1 (26.0) | 0.1 |
3. Get the facts about the benefits and risks of each choice | 84.1 (25.8) | 80.1(27.3) | 0.21 |
4. Understand the information enough to be able to make a choice | 84.4 (26.1) | 82.8 (24.6) | 0.33 |
5. Ask questions without feeling dumb | 84.4 (30.8) | 79.7 (30.8) | 0.34 |
6. Express my concerns about each choice | 83.5 (26.8) | 78 (30) | 0.16 |
7. Ask for advice | 87 (26.8) | 82.8(26.3) | 0.11 |
8. Figure out the choice that best suits me | 84.1 (23.7) | 79.1 (28.8) | 0.35 |
9. Handle unwanted pressure from others in making my choice | 80.8 (29.2) | 77.3 (30.6) | 0.51 |
10. Let the clinic team know what’s best for me | 83.8 (22) | 77.7 (31) | 0.5 |
11. Delay my decision if I feel I need more time | 79.1 (31.5) | 76.7 (31.5) | 0.88 |
* Factor 1 (Items 5–11) represents overcoming barriers and **factor 2 (1–4) represents information seeking.
Patients with the Karnofsky performance score (PS) > 1, less fit, reported less self-efficacy in making their decision during the preoperative period. Indeed, patients with PS > 1 have a DSE mean of 73.8 (SD 26) compared to patients with a PS 0–1 who have a DSE mean of 85.8 (SD 20.3 p = 0.0024). Furthermore, these patients were different on both subscales identified in the factor analysis: F1 (PS > 1:76.1 V s PS 0–1:85.9 p = 0.0052) and F2 (PS > 1:71.6 V s PS 0–1:85.4 p < .001)
No statistically significant differences between DSE scores for men and women were evident (p = 0.37). Male patients had a mean DSE value of 84.0 (SD 21.3) and female of 79.9 (SD 24.2). No difference between these groups was found by Factor1 and Factor2, p = 0.41 and p = 0.62 respectively.
Similarly, when comparing DSE among younger and older people (using the cut-off above and below the median value of 72 years) no statistically significant differences were found (p = 0.4). In particular, older people > 72 years had a DSE mean of 82.5 (SD 23.7) and younger people < 72 years 81 (SD 22.6). No difference between these groups was found by Factor1 and Factor2, p = 0.21 and p = 0.85 respectively.