Skip to main content
. 2020 Mar 31;19:43. doi: 10.1186/s12904-020-00542-z

Table 3.

Predictors of Physically Active Behavior and Motivation for Physical activity

Dependent variable Significant predictors beta Wald p
Physically Active Behavior (R2 = 0.483; f = 0.97)a 1. Motivation for PA 1.044 7.76 0.005
2. Duration since metastatic cancer (months) 0.026 5.68 0.017

Independent variables in the regression models: age, sex, ECOG ≥2, number of comorbidities, tumor type (breast cancer vs. others), PA before cancer diagnosis, Interest in exercise program and knowledge about PA and QoL

Quality of regression model: p < 0.001; χ2 = 48.4

Motivation for PA (R2 = 0.685; f = 1.47)a 1. Fatigue b -2.301 7.03 .008
2. Clinically relevant depression c -1.390 4.25 .039
3. Knowledge about PA and QoL .929 9.20 .002
4. PA before cancer diagnosis .688 7.89 .005
5. Interest in exercise program .635 6.97 .008

Independent variables in the regression models: age, sex, ECOG-Performance status ≥2, number of comorbidities, tumor type (breast cancer vs. others)

Quality of regression model: p < 0.001; χ2 = 87.9

Abbreviations: ECOG Eastern Co-operative Oncology Group performance index, PA physical activity, QoL quality of life

a R2 = Explained variance; f = Effect size according to Cohen;

b FACT-F Score ≤ 34: cut-off point for diagnosis of cancer related fatigue [32]

c PHQ8-Score ≥ 10 points to clinically relevant depression [35]