Risk factors for poor health-related quality of life in cancer survivors with multiple chronic conditions: exploring the role of treatment burden as a mediator

Figures & data

Table 1 Descriptive and medical characteristics of study subjects (N=91)

Variable		n (%)
Age, median (range)	65.0 years (31–92 years)
Gender	Female	54 (59)
	Male	37 (41)
Race	White	74 (81)
	Black/AA	8 (9)
	Asian	6 (7)
	Native American	2 (2)
	Unknown	1 (1)
Ethnicity	Hispanic	2 (2)
Marital status	Married/partnered	57 (63)
	Not married	33 (36)
	Missing	1 (1)
Education status	College educated	67 (74)
	No more than HS	23 (26)
	Missing	1 (1)
Occupation status	Retired	39 (43)
	Employed (full time or part time)	34 (37)
	On disability	12 (13)
	Homemaker or full-time student	4 (4)
	Missing	2 (2)
Primary cancer type	Breast	19 (21)
	Genitourinary	17 (19)
	Malignant melanoma	12 (13)
	Gastrointestinal	12 (13)
	Hematologic	11 (12)
	Gynecologic	10 (11)
	Other	10 (11)
Cancer treatmenta	Surgery	57 (63)
	Radiotherapy	24 (26)
	Chemotherapy	22 (24)
	Observation	3 (3)
	Other	12 (13)
Total number of diagnosed conditions (including cancer)	Median	5 conditions
	Range	2–11 conditions
	2–3 conditions	16 (18)
	4 conditions	21 (23)
	5+ conditions	54 (59)
Types of conditions comorbid to cancer	Hypertension	49 (54)
	Arthritis	47 (52)
	Hyperlipidemia	46 (51)
	Diabetes (type 1 or 2)	42 (46)
	Low back pain	42 (46)
	Vision problems	34 (37)
	Depression	21 (23)
	Coronary artery disease	19 (21)
	Cardiac arrhythmia	18 (20)
	Anxiety	17 (19)
	Hearing problems	15 (17)
	Chronic kidney disease	14 (15)
	COPD	11 (12)
	Osteoporosis	11 (12)
	Congestive heart failure	6 (7)
	Substance abuse	5 (6)
	Headache	5 (6)
	Psoriasis	3 (3)
	Crohn’s disease	2 (2)
	Hepatitis	2 (2)

Note:

a Accumulated percentage exceeds 100% as combinations of more than one cancer treatment occurred (eg, surgery + radiotherapy, surgery + chemotherapy, radiotherapy + chemotherapy).

Abbreviations: AA, African-American; HS, high school.

Table 2 Pearson product-moment correlations of baseline predictors (ie, risk factors), treatment burden, and 6-month health-related quality of life

Predictors	Treatment burden (baseline)		Health-related quality of life (6 months)
	PETS Role-social activity limitationsa	PETS Physical/mental exhaustiona	PROMIS Global physicalb	PROMIS Global mentalb
Number of conditions	0.26*	0.18	−0.40***	−0.13
Education status (high school or less, college)	−0.12	−0.29**	0.17	0.12
Number of different physicians seen in past 6 months (1, 2, 3, 4, 5+)	0.45***	0.34**	−0.32**	−0.23*
Mental health diagnosisc (no, yes)	0.30**	0.41***	−0.31**	−0.36**
Self-management routine (no, yes)	−0.23*	−0.32**	0.13	0.31**
Health literacy	−0.20	−0.41***	0.26*	0.21
Self-efficacy	−0.45***	−0.46***	0.57***	0.57***
Social support	−0.12	−0.32**	0.23*	0.41***
Treatment burden
PETS role-social activity limitations	−	0.73***	−0.56***	−0.47***
PETS physical/mental exhaustion	0.73***	−	−0.56***	−0.56***

Notes:

a Higher score indicates more perceived burden.

b Higher score indicates better health.

c Diagnosis of depression, anxiety, or both.

*** P<0.001;

** P<0.01;

* P<0.05.

Abbreviations: PETS, Patient Experience with Treatment and Self-management; PROMIS, Patient-Reported Outcomes Measurement Information System.

Table 3 Multiple linear regressions of treatment burden scales onto risk factors

	Model 1: PETS role-social activity limitationsa
Predictors	B (SE)	95% CI	β	P-value (significance)
Number of diagnosed conditions	0.79 (1.19)	−1.57, 3.16	0.07	NS
Number of different physicians seen in past 6 months (1, 2, 3, 4, 5+)	6.72 (1.87)	2.99, 10.44	0.35	0.001
Mental health diagnosis (no, yes)b	3.19 (5.70)	−8.16, 14.53	0.06	NS
Self-management routine (no, yes)	−7.77 (6.24)	−20.20, 4.66	−0.12	NS
Self-efficacy	−1.32 (0.46)	−2.23, −0.41	−0.31	0.005
Model F and R^2	F (5, 76)=8.5, P<0.001; R^2=0.36
	Model 2: PETS physical/mental exhaustiona
Predictors	B (SE)	95% CI	β	P-value (significance)
Education status (HS or less, college)	−7.49 (5.17)	−17.80, 2.82	−0.13	NS
Number of different physicians seen in past 6 months (1, 2, 3, 4, 5+)	2.77 (1.71)	−0.64, 6.18	0.15	NS
Mental health diagnosis (no, yes)b	8.61 (5.29)	−1.93, 19.14	0.16	NS
Self-management routine (no, yes)	−12.95 (5.60)	−24.11, −1.80	−0.21	0.02
Health literacy	−6.38 (2.53)	−11.41, −1.34	−0.25	0.01
Self-efficacy	−0.90 (0.93)	−1.75, −0.04	−0.22	0.04
Social support	−0.53 (0.53)	−1.59, 0.53	−0.10	NS
Model F and R^2	F (7, 73)=8.4, P<0.001; R^2=0.45

Notes:

a Higher score indicates more perceived burden.

b Diagnosis of depression, anxiety, or both. B, unstandardized regression coefficient; β, standardized regression coefficient.

Abbreviations: HS, high school; NS, not significant; PETS, Patient Experience with Treatment and Self-management; SE, standard error.

Table 4 Multiple linear regressions of 6-month health-related quality-of-life outcomes onto baseline risk factors

	Model 3: PROMIS global physical healtha
Predictors	B (SE)	95% CI	β	P-value (significance)
Number of diagnosed conditions	−0.46 (0.42)	−1.30, 0.38	−0.12	NS
Number of different physicians seen in past 6 months (1, 2, 3, 4, 5+)	−1.20 (0.66)	−2.52, 0.12	−0.18	NS
Mental health diagnosis (no, yes)b	−0.28 (1.96)	−4.18, 3.63	−0.02	NS
Health literacy	1.64 (1.04)	−0.43, 3.72	0.16	NS
Self-efficacy	0.67 (0.16)	0.36, 0.98	0.49	<0.001
Social support	−0.02 (0.20)	−0.42, 0.38	−0.01	NS
Model F and R^2	F (6, 66)=8.4, P<0.001; R^2=0.43
	Model 4: PROMIS global mental healtha
Predictors	B (SE)	95% CI	β	P-value (significance)
Number of different physicians seen in past 6 months (1, 2, 3, 4, 5+)	−0.59 (0.69)	−1.96, 0.79	−0.09	NS
Mental health diagnosis (no, yes)b	−3.75 (2.03)	−7.81, 0.32	−0.19	NS
Self-management routine (no, yes)	4.13 (2.27)	−0.41, 8.66	0.18	NS
Self-efficacy	0.49 (0.17)	0.16, 0.83	0.34	0.005
Social support	0.34 (0.21)	−0.08, 0.75	0.17	NS
Model F and R^2	F (5, 66)=9.1, P<0.001; R^2=0.41

Notes:

a Higher score indicates better health.

b Diagnosis of depression, anxiety, or both. B, unstandardized regression coefficient; β, standardized regression coefficient.

Abbreviations: NS, not significant; PROMIS, Patient-Reported Outcomes Measurement Information System; SE, standard error.

Figure 1 Hypothesized mediational pathway.

Table 5 Mediation of the relationship between self-efficacy and 6-month health-related quality-of-life outcomes by overall treatment burden

	Outcome: PROMIS global physical healtha
Model 1: direct effect of self-efficacy	B (SE)	95% CI	β	P-value (significance)
Self-efficacy	0.68 (0.14)	0.41, 0.95	0.50	<0.001
Model F and R^2	F (2, 73)=21.1, P<0.001; R^2=0.37
Model 2: indirect effect of self-efficacy	B (SE)	95% CI	β	P-value (significance)
Self-efficacy	0.43 (0.14)	0.15, 0.70	0.32	0.003
Overall treatment burdenb	−3.39 (0.88)	−5.14, −1.65	−0.40	<0.001
Model F and R^2	F (3, 70)=20.4, P<0.001; R^2=0.47
	Outcome: PROMIS global mental healtha
Model 1: direct effect of self-efficacy	B (SE)	95% CI	β	P-value (significance)
Self-efficacy	0.88 (0.15)	0.58, 1.19	0.60	<0.001
Model F and R^2	F (2, 73)=17.9, P<0.001; R^2=0.33
Model 2: indirect effect of self-efficacy	B (SE)	95% CI	β	P-value (significance)
Self-efficacy	0.67 (0.16)	0.35, 0.99	0.46	<0.001
Overall treatment burdenb	−2.80 (0.99)	−4.78, −0.82	−0.30	0.006
Model F and R^2	F (3, 70)=14.7, P<0.001; R^2=0.39

Notes:

a Higher score indicates better health.

b Higher overall treatment burden factor score indicates more perceived burden. All estimates are based on multiple linear regression models statistically adjusting for number of diagnosed conditions. B, unstandardized regression coefficient; β, standardized regression coefficient.

Abbreviations: PROMIS, Patient-Reported Outcomes Measurement Information System; SE, standard error.