Table 2. Correlations between biopsychosocial factors and morbidity indices and subjective disease burden. Partial correlation coefficients adjusted for morbidity indices.

	Correlation coefficient1	Partial correlation coefficients2
Independent variable	Unadjusted	Adjusted for Quan morbidity index	Adjusted for CDS
Physical functioning	- 0.512*	-0.467	-0.456
Compound effects: conditions interfere3	- 0.406*	-0.389	-0.396
Self-efficacy	- 0.378*	-0.389	-0.379
Financial constraints	- 0.365*	-0.345	-0.335
Overwhelmed by single condition	- 0.332*	-0.318	-0.313
Depression screen	- 0.286*	0.279	0.296
Medication adherence	- 0.247*	-0.278	-0.280
Compound effects: medications interfere3	- 0.222*	-0.204	-0.197
Knowledge about conditions	- 0.211*	-0.213	-0.219
Patient-provider communication	- 0.202*	-0.213	-0.208
Social activity	- 0.181**	-0.168	-0.164
Medication knowledge	- 0.159**	-0.166	-0.161
Health literacy	- 0.131ˆ	-0.104	-0.116
Compound effects: medications help3	0.055	0.032	0.021
Compound effects: conditions help3	0.003	0.002	-0.008
Quan morbidity index	0.226*	---	---
Chronic Disease Score	0.263*	---	---
Self-report list of conditions	0.807*	---	---

¹* p < 0.001; ** p < 0.01; ˆ p < 0.05 for association between independent variable and disease burden.

²Adjusted for Quan morbidity index or CDS. The significance of the difference between the unadjusted total and the partial correlation coefficients was tested using 1,000 bootstrapping samples. None of the differences was found to be significantly different from zero, indicating that adjusting for either of these two co-morbidity measures does not change the relationship between the independent variable and burden.

³We defined ‘compound effects of conditions’ as a report of treatments or symptoms for different conditions either interfering with or helping each other. (E.g. ‘my arthritis makes it hard for me to exercise to treat my diabetes.’)