Cancer-Related Distress in Young Adults Compared to Middle-Aged and Senior Adults

Abstract

Purpose:

Little is known about cancer-related distress during young adulthood. Results from the few studies that have directly assessed this age group have indicated that young adults (YAs) may be at greater risk of developing psychosocial difficulties due to their unique challenges of coping with cancer. This study's objective was to investigate cancer-related distress in YAs compared to older adults.

Methods:

This retrospective cross-sectional study compared the distress level of YAs (18–39 years old) with that of middle-aged (40–64 years old) and senior adults (65–90 years old) using the Distress Thermometer (DT) and associated Problem List (PL). Factors that may be associated with distress by age group were examined, including demographics, cancer type, and PL items endorsed.

Results:

YAs had higher cancer-related distress than senior adults but similar distress levels to middle-aged adults. Findings from distress comparisons across demographics, cancer types, and PL items endorsed suggest that YAs and middle-aged adults had similar distress patterns when compared to senior adults, who had the lowest DT scores. Multivariable analyses indicated age-related risk factors for high distress, including gynecologic cancers for YAs; divorced, single, or unemployed statuses for middle-aged adults; and being of Hispanic ethnicity for senior adults. Female gender and practical, emotional, and physical problems were associated with distress for all age groups.

Conclusion:

There is a differential impact of cancer by age. It is important to screen for cancer-related distress, paying attention to risk factors by age to determine age-appropriate supportive care needs.

Cancer-related distress has been defined as an unpleasant emotional experience, ranging from natural feelings of sadness and fear to clinically significant symptoms of depression, anxiety, or other psychological, social, or spiritual problems that may interfere with the ability to cope with cancer and its treatment.¹ While distress is often a natural reaction to a cancer diagnosis, research has estimated that 13–43% of cancer patients experience elevated levels of distress.^2,3 Factors associated with distress levels include cancer type and stage, types of distress constructs assessed (i.e., global distress, anxiety, depression), and personal variables such as gender and age.^2,3

While studies of adults with cancer have suggested that younger age is a risk factor for higher levels of distress,^2–4 few studies have directly assessed cancer-related distress in young adults (YAs), defined here as those aged 18–39 years old. Limited research indicates that YAs may be at increased risk for developing psychosocial difficulties due to their unique cancer burdens during this developmental period marked by increasing autonomy, identity formation, career and family planning, and intimate relationships.^5,6 In a sample of 335 YAs 18–39 years old who had completed cancer treatment, 56.5% endorsed clinically relevant levels of elevated distress; the highest distress was among those who endorsed greater global impact of their cancer or were 13–24 months posttreatment (compared to 0–12 or 25–60 months posttreatment), especially those who had an interruption in education or employment.⁷ Roberts and colleagues found that YAs 22–35 years old did not endorse significant psychosocial distress compared to normative data, but did describe concerns about their health, future, and life goal changes, and worry that their children may get cancer.⁸ Similarly, there were no significant differences in levels of depression and anxiety in a Portuguese sample of 36 YA cancer patients and survivors 20–38 years old compared to a similar-aged control group; however, those on active treatment endorsed significantly lower scores on the quality of life domains of role, cognitive, and social functioning, as well as lower personal growth compared to the controls.⁹

Given the dearth of studies on cancer-related distress in YAs, our study objective was to increase understanding of cancer-related distress in younger compared to older adults. This retrospective study compared distress levels of YAs aged 18–39 years old with older adults (middle-aged adults 40–65 years old and senior adults 66–90 years old) using the National Comprehensive Cancer Network (NCCN) Distress Thermometer (DT).^1,10

We hypothesized that YAs would have significantly higher cancer-related distress than would older adults and that YAs with cancer would endorse a different pattern of stressors on the DT compared to older adults. Our study aims were to, using the DT and its associated Problem List (PL), (1) evaluate difference in the distress levels of YAs compared to older adults following consultation for cancer treatment; (2) examine how distress levels are affected by ethnicity, gender, marital status, employment status, and cancer type by age; and (3) describe the relationship between the DT score and PL items endorsed.

Methods

This was a retrospective cross-sectional chart review study. Inclusion criteria were 18–90 years old when the DT was administered, ability to read the DT in English, diagnosis of new or recurrent cancer, and completion of the DT and associated PL between May 1, 2008, and May 1, 2011. This data was routinely obtained during the initial or second outpatient visit as part of standard clinical practice at a large Midwestern academic cancer center affiliated with a private nonprofit medical college that serves as a local and regional cancer center. This study was approved by the joint Institutional Review Board of the medical college and cancer center with a waiver of informed consent.

Distress Thermometer

The DT^1,10 is a self-report measure used to evaluate cancer-related distress in adults. It is a one-page self-administered measure depicting a vertical thermometer with 0–10 on it, with 10 being the highest level of distress. Patients were instructed to circle the number that describes how much distress they experienced during the past week. A critical cutoff score of ≥4 was chosen, as that cutoff has been demonstrated to yield optimal sensitivity and specificity when compared to critical cutoff scores of well-established psychological symptom measures.^1,11 The DT also has a PL of 50 items that are divided into the Problems categories of Practical, Family, Emotional, Spiritual/Religious, and Physical. Patients check a “yes/no” box for all items that describe their symptoms during the past week. The DT is a valid and reliable screening measure for assessing distress in adult cancer populations, including the detection of depression and anxiety (sensitivity: 80.9% and 77.1%, and specificity: 60.2% and 66.1%, respectively).¹²

Procedure

After the patient completed the DT in the outpatient clinic, the nurse reviewed the DT score and PL items endorsed. Patients with a score of ≥4 on the DT were offered a referral to an appropriate provider, depending on the PL items identified. An administrative assistant then entered the data into a database. The database data were linked using the patient's medical record number to demographic and medical data collected and stored by the institution. Data extracted for this study included DT score, PL items endorsed, cancer type, age at DT administration, gender, ethnicity, and marital and employment statuses.

Statistical analyses

Descriptive statistics were calculated across the YA and two older adult groups. The ordinal DT score was compared across age groups using a Kruskal–Wallis test and a Mann–Whitney test for pairwise comparisons. Within each age group, the DT score was compared with the covariates by univariable and multivariable regression analysis using SAS 9.2 (SAS Institute, Cary, NC). The false discovery rate (FDR) was used to adjust for multiple comparisons.

Results

Sample characteristics

An initial sample of 5926 eligible patients completed the DT. One patient with orbital cancer was excluded because it was the only diagnosis of that kind, resulting in a final sample of 5925 (Table 1), of which 11.3% were YAs, 54.6% were middle-aged adults, and 34.1% were senior adults. Compared to the older age groups, YAs had a significantly higher proportion of females; lower proportions of married and divorced patients and a higher proportion of single patients; a lower proportion of Caucasians and higher proportions of African Americans and Hispanics; and a higher proportion of unemployed patients. YAs had significantly lower proportions of prostate/urologic, hepatobiliary, thoracic, and colorectal cancers and significantly higher proportions of blood/lymphatic/bone marrow transplant (BMT), neurologic, gynecologic, endocrine, and bone/connective tissue cancers compared to the older age groups (Table 1). YAs had significantly lower proportions of breast cancer compared to middle-aged but not senior adults.

Table 1.

Demographic and Cancer Type Variables for Young Adults and Older Adult Groups

				Pairwise test p-values^a
N=5925	YA, 18–39 years old (n=668)	Middle-aged, 40–64 years old (n=3238)	Senior, 65–90 years old (n=2019)	YA vs. M	YA vs. S	M vs. S
Age at study
Mean (SD)	31.1 (6.1)	54.3 (6.6)	73.7 (6.1)	n/a	n/a	n/a
Median	32.0	54.8	72.8
Frequency (%)
Gender				<0.0001	<0.0001	<0.0001
Male	276 (41.3%)	1614 (49.8%)	1133 (56.1%)
Female	392 (58.7%)	1624 (50.2%)	886 (43.9%)
Marital status
Married	289 (43.3%)	2154 (66.5%)	1337 (66.2%)	<0.0001	<0.0001	<0.0001
Single	341 (51.0%)	636 (19.6%)	165 (8.2%)
Divorced	32 (4.8%)	343 (10.6%)	159 (7.9%)
Widowed	1 (0.2%)	67 (2.1%)	332 (16.4%)
Separated	2 (0.3%)	25 (0.8%)	9 (0.4%)
Unknown	3 (0.4%)	13 (0.4%)	17 (0.8%)
Ethnicity				<0.0001	<0.0001	<0.0001
Caucasian	503 (75.3%)	2628 (81.2%)	1768 (87.6%)
African American	101 (15.1%)	435 (13.4%)	167 (8.3%)
Hispanic	31 (4.6%)	62 (1.9%)	25 (1.2%)
Other	33 (4.9%)	113 (3.5%)	59 (2.9%)
Employment status				<0.0001	<0.0001	<0.0001
Employed	402 (60.2%)	1942 (60%)	216 (10.7%)
Retired	0 (0%)	372 (11.5%)	1736 (86%)
Unemployed	260 (38.9%)	866 (26.7%)	44 (2.2%)
Unknown	6 (0.9%)	58 (1.8%)	23 (1.1%)
Cancer type
Prostate/urologic	72 (10.8%)	670 (20.7%)	579 (28.7%)	<0.0001	<0.0001	<0.0001
Blood/lymphatic/BMT	153 (22.9%)	553 (17.1%)	359 (17.8%)	0.0004	0.004	0.51
Head and neck	99 (14.8%)	434 (13.4%)	263 (13%)	0.33	0.24	0.70
Breast	59 (8.8%)	483 (14.9%)	146 (7.2%)	<0.0001	0.18	<0.0001
Neurologic	113 (16.9%)	246 (7.6%)	108 (5.4%)	<0.0001	<0.0001	0.002
Hepatobiliary^b	18 (2.7%)	173 (5.3%)	140 (6.9%)	0.004	<0.0001	0.02
Gynecologic	50 (7.5%)	170 (5.2%)	76 (3.8%)	0.02	<0.0001	0.01
Thoracic	5 (0.8%)	143 (4.4%)	147 (7.3%)	<0.0001	<0.0001	<0.0001
Colorectal	16 (2.4%)	137 (4.2%)	64 (3.2%)	0.03	0.31	0.05
Skin	21 (3.1%)	84 (2.6%)	71 (3.5%)	0.42	0.64	0.05
Endocrine^c	33 (4.9%)	83 (2.6%)	28 (1.4%)	0.001	<0.0001	0.004
Bone/connective tissue	29 (4.3%)	62 (1.9%)	38 (1.9%)	0.0002	0.0004	0.93

One eye orbital case was excluded. Significant values noted in bold.

Chi-square tests computed across the age groups.

Gastrointestinal, pancreatic, and liver were combined into hepatobiliary.

Endocrine was added.

BMT, bone marrow transplant; M, middle-aged adult; n/a, not applicable; S, senior adult; SD, standard deviation; YA, young adult.

Comparison of DT scores across age groups

There was no significant difference in DT scores when comparing the YAs and the combined older age group. The mean DT score for the YAs and the combined older age group was 3.6 (standard error [SE]=0.11) and 3.5 (SE=0.04), respectively (p=0.38). However, when comparing YAs separately to middle-aged and senior adults, the YAs had a similar mean DT score (mean=3.6, SE=0.11) to middle-aged adults (mean=3.8, SE=0.05, p=0.12), but both age groups had significantly more distress than did senior adults (mean=3.0, SE=2.8, p<0.0001). The percentage with DT score above the critical cutoff (≥4) was similar for the YA and middle-aged adult groups (47% vs. 49.7%), with significantly less in the senior adult group (39%).

Between-group comparisons of DT scores across age groups by demographics and cancer types

Cancer type

Comparing overall DT scores by age and cancer type (Table 2) indicated a significantly higher level of distress reported by both YAs and middle-aged adults compared to senior adults for those diagnosed with prostate/urologic, breast, colorectal, and gynecologic cancers. Middle-aged adults also reported a significantly higher level of distress compared to the senior adults for blood/lymphatic/BMT, head and neck, skin, and bone/connective tissue cancers, as well as neurologic and thoracic cancers (both marginally significant, FDR). The only significant difference in distress between the YAs (lower distress) and middle-aged adults was a marginal one (FDR) in skin cancer.

Table 2.

Distress Thermometer Scores by Age Group for Demographic and Cancer Type Variables

	YA, 18–39 years old (n=668)			Middle-aged, 40–64 years old (n=3238)			Senior, 65–90 years old (n=2019)			Pairwise test p-values^a
N=5925	n	Mean	SE	n	Mean	SE	n	Mean	SE	YA vs. M	YA vs. S	M vs. S
Gender
Male	276	3.1	0.17	1614	3.4	0.07	1133	2.7	0.08	0.13	0.08	<0.0001
Female	392	4.0	0.15	1624	4.3	0.07	886	3.5	0.10	0.10	0.0006	<0.0001
Marital status^b
Married	289	3.8	0.17	2154	3.6	0.06	1337	2.9	0.07	0.22	<0.0001	<0.0001
Single	341	3.5	0.16	636	4.2	0.12	165	3.2	0.21	0.002	0.26	0.0003
Divorced	32	3.9	0.52	343	4.4	0.17	159	3.5	0.25	0.35	0.35	0.001
Ethnicity
Caucasian	503	3.4	0.12	2628	3.8	0.06	1768	3.0	0.06	0.03	0.0004	<0.0001
African American	101	4.2	0.32	435	4.3	0.15	167	3.3	0.24	0.62	0.02	0.0002
Hispanic	31	5.0	0.56	62	3.7	0.41	25	5.0	0.70	0.07	0.99	0.14
Other	33	3.8	0.59	113	3.4	0.30	59	3.2	0.36	0.56	0.61	0.94
Employment status^c
Employed	402	3.7	0.14	1942	3.5	0.06	216	3.1	0.19	0.42	0.008	0.01
Retired	0			372	3.8	0.15	1736	3.0	0.07	n/a	n/a	<0.0001
Unemployed	260	3.6	0.18	866	4.6	0.11	44	3.2	0.47	<0.0001	0.26	0.003
Cancer diagnoses
Prostate/urologic	72	3.75	0.35	670	3.54	0.11	579	2.93	0.11	0.57	0.02	<0.0001
Blood/lymphatic/BMT	153	3.15	0.22	553	3.32	0.13	359	2.72	0.14	0.71	0.08	0.004
Head and neck	99	3.51	0.28	434	3.98	0.14	263	3.34	0.19	0.17	0.48	0.004
Breast	59	4.14	0.36	483	4.35	0.13	146	3.36	0.24	0.62	0.05	0.0002
Neurologic	113	3.23	0.27	246	3.43	0.19	108	2.68	0.25	0.59	0.16	0.03
Hepatobiliary^d	18	3.89	0.63	173	3.97	0.23	140	3.64	0.23	0.98	0.62	0.39
Gynecologic	50	5.56	0.40	170	4.67	0.23	76	3.63	0.29	0.08	0.0003	0.008
Thoracic	5	3.40	0.93	143	4.19	0.23	147	3.44	0.21	0.40	0.90	0.03
Colorectal	16	5.56	0.75	137	4.07	0.26	64	2.50	0.30	0.07	0.0004	0.0008
Skin	21	2.71	0.53	84	4.26	0.33	71	2.75	0.32	0.04	0.85	0.001
Endocrine^e	33	3.09	0.49	83	3.63	0.34	28	2.64	0.54	0.45	0.46	0.14
Bone/connective tissue	29	3.83	0.58	62	4.31	0.37	38	2.87	0.48	0.40	0.13	0.01

One eye orbital case was excluded. Significant values noted in bold.

Mann–Whitney tests computed across the age groups.

Widowed, separated, and unknown categories (total n=469) excluded from analyses.

Unknown employment status (n=87) excluded from analyses.

Gastrointestinal, pancreatic, and liver were combined into hepatobiliary.

Endocrine was added.

BMT, bone marrow transplant; M, middle-aged adult; n/a, not applicable; S, senior adult; SE, standard error; YA, young adult.

Gender

Both YA and middle-aged adult females (which did not differ from each other) had significantly higher levels of distress than senior adult females. Middle-aged adult males had significantly higher levels of distress than senior adult males, but YA males' DT scores did not differ from middle-aged or senior adult males (Table 2).

Ethnicity

Caucasian YAs had significantly lower DT scores than Caucasian middle-aged adults (Table 2), but both had significantly higher distress than Caucasian senior adults. African American YAs and middle-aged adults both had significantly higher DT scores compared to African American senior adults, but not compared to each other. Of note, Hispanic YAs and senior adults had the highest mean DT scores (5.0) compared to all other ethnic categories in all of the age groups.

Marital status

Both married YAs and middle-aged adults reported significantly higher distress than married senior adults. Single middle-aged adults reported significantly higher distress than single YAs and senior adults. Divorced middle-aged adults had significantly higher distress than divorced senior adults (Table 2).

Employment status

Both employed YAs and middle-aged adults had significantly more distress than employed senior adults, but did not differ from each other (Table 2), while unemployed middle-aged adults reported significantly higher distress compared to unemployed YAs and senior adults. Retired middle-aged adults had higher DT scores than retired senior adults; there were no retired YAs.

Relationship between DT score and categories of items endorsed on the PL by age

Results of DT score comparisons of specific categories of PL items by age are in Table 3. In the Practical Problems category, there were no significant differences in DT scores among the different problem items between YAs and either middle-aged or senior adults. Middle-aged adults who endorsed problems with housing or transportation reported significantly higher DT scores than senior adults who endorsed these problems, while insurance/financial problems were marginal (FDR). Within the Family Problems category, both YAs and middle-aged adults who endorsed problems with their partner had significantly higher distress levels than senior adults who endorsed this item; however, YAs and middle-aged adults did not differ in DT scores. In the Emotional Problems category, both YAs and middle-aged adults who endorsed problems with depression, fears, nervousness, sadness, worry, and loss of interest in usual activities had significantly higher distress than senior adults who endorsed each of these items; again, YAs and middle-aged adults did not differ in DT scores. Middle-aged adults who endorsed the single-item Spiritual/Religious Concerns category reported significantly higher DT scores than senior adults who endorsed this category. Within the category of Physical Problems, both YAs and middle-aged adults who endorsed problems with breathing, changes in urination, constipation, eating, fatigue, feeling swollen, getting around, memory/concentration, dry nose/congestion, pain, sexual function, dry/itchy skin, sleeping, and tingling in hands/feet had significantly higher levels of distress than senior adults who endorsed each of these items. Additionally, middle-aged adults who endorsed problems with appearance, bathing/dressing, diarrhea, fevers, indigestion, and nausea had higher DT scores than senior adults who endorsed these items. There were no significant differences in DT scores between YAs and middle-aged adults in in the Physical Problems category.

Table 3.

Relationship Between Distress Thermometer Scores and Problem List Items Endorsed By Age Group

	YA, 18–39 years old (n=668)			Middle-aged, 40–64 years old (n=3238)			Senior, 65–90 years old (n=2019)			Pairwise test p-values^a
Problem list category and items (N=5925)	n	Mean	SE	n	Mean	SE	n	Mean	SE	YA vs. M	YA vs. S	M vs. S
Practical Problems
Child care	31	5.55	0.50	44	5.89	0.45	11	5.00	0.81	0.64	0.35	0.43
Housing	42	5.90	0.45	135	6.37	0.23	44	5.16	0.39	0.33	0.18	0.007
Insurance/financial	142	4.93	0.24	594	5.18	0.12	155	4.62	0.22	0.30	0.40	0.03
Transportation	53	5.26	0.42	227	5.67	0.18	111	4.41	0.27	0.39	0.09	<0.0001
Work/school	132	4.93	0.25	366	5.02	0.15	33	5.00	0.50	0.70	0.95	0.91
Family Problems
Dealing with children	51	5.63	0.34	176	5.81	0.21	67	5.16	0.36	0.51	0.42	0.12
Dealing with partner	46	6.04	0.38	191	5.53	0.21	73	4.49	0.33	0.34	0.005	0.008
Emotional Problems
Depression	114	6.44	0.23	672	6.06	0.11	293	5.14	0.17	0.26	<0.0001	<0.0001
Fears	220	5.73	0.18	1204	5.57	0.08	475	5.04	0.12	0.48	0.001	0.0002
Nervousness	287	5.22	0.16	1307	5.36	0.08	606	4.69	0.11	0.42	0.007	<0.0001
Sadness	170	6.12	0.20	885	5.87	0.09	353	5.04	0.15	0.36	<0.0001	<0.0001
Worry	325	5.26	0.14	1689	5.19	0.07	820	4.52	0.09	0.79	<0.0001	<0.0001
Lost interest in usual activities	95	6.23	0.27	485	6.04	0.13	244	5.16	0.19	0.63	0.003	0.0002
Spiritual/Religious Concerns	17	6.06	0.82	126	6.40	0.25	53	5.15	0.40	0.77	0.26	0.007
Physical Problems
Appearance	92	5.03	0.29	317	5.58	0.16	130	4.82	0.24	0.07	0.48	0.008
Bathing/dressing	34	5.18	0.51	162	5.35	0.24	134	4.45	0.25	0.71	0.26	0.01
Breathing	41	5.51	0.43	336	4.93	0.15	255	3.95	0.18	0.19	0.001	<0.0001
Changes in urination	41	5.22	0.40	307	5.05	0.17	307	3.77	0.16	0.81	0.002	<0.0001
Constipation	78	5.10	0.34	441	4.84	0.14	312	4.13	0.16	0.54	0.02	0.002
Diarrhea	71	4.65	0.34	310	4.84	0.17	215	4.00	0.19	0.58	0.10	0.001
Eating	111	5.80	0.24	555	5.28	0.12	331	4.51	0.15	0.13	<0.0001	<0.0001
Fatigue	243	4.82	0.17	1264	4.69	0.08	788	3.96	0.10	0.55	<0.0001	<0.0001
Feeling swollen	94	5.01	0.28	502	5.04	0.13	263	4.04	0.17	0.88	0.004	<0.0001
Fevers	28	4.86	0.53	124	5.44	0.27	37	4.03	0.46	0.32	0.23	0.01
Getting around	62	5.53	0.38	376	5.31	0.15	317	4.50	0.16	0.60	0.01	0.0002
Indigestion	59	5.02	0.37	348	5.13	0.16	205	4.46	0.20	0.76	0.17	0.008
Memory/concentration	121	5.15	0.26	637	5.31	0.11	347	4.18	0.15	0.52	0.002	<0.0001
Mouth sores	31	5.00	0.56	196	4.71	0.21	116	4.01	0.25	0.61	0.12	0.06
Nausea	96	5.19	0.30	369	5.36	0.15	161	4.78	0.24	0.57	0.32	0.04
Nose: dry/congestion	100	4.70	0.27	428	4.67	0.14	277	3.87	0.17	0.91	0.009	0.0005
Pain	210	5.12	0.19	1022	5.05	0.09	572	4.20	0.12	0.78	0.0001	<0.0001
Sexual function	31	6.00	0.46	263	5.32	0.19	118	3.73	0.26	0.27	<0.0001	<0.0001
Skin: dry/itchy	138	4.78	0.25	628	4.80	0.11	441	3.77	0.14	0.91	0.0005	<0.0001
Sleeping	210	5.27	0.19	1088	5.10	0.09	551	4.24	0.12	0.51	<0.0001	<0.0001
Tingling in hands/feet	113	4.90	0.29	646	4.80	0.12	389	3.66	0.15	0.78	0.0002	<0.0001

Significant values noted in bold.

Mann–Whitney tests computed across the age groups.

M, middle-aged adult; S, senior adult; SE, standard error; YA, young adult.

Within-age-group comparisons of DT scores

For cancer type, univariable analyses (odds ratios [ORs]; data not shown) within each age group indicated that having a gynecologic cancer for the YAs (OR=1.58, p=0.003); gynecologic (OR=1.22, p=0.01), blood/lymphatic/BMT (OR=0.84, p=0.0005), breast (OR=1.15, p=0.005), or prostate/urologic cancer (OR=0.90, p=0.03) for the middle-aged adults; and blood/lymphatic/BMT (OR=0.88, p=0.05) diagnosis for the senior adults was predictive of higher DT scores. Multivariable analyses results did not differ from the univariable analyses, with the exception of nonsignificant breast (OR=1.11, p=0.06) and marginally significant (FDR) gynecologic (OR=1.19, p=0.04) and prostate/urologic (OR=0.91, p=0.04) cancers for middle-aged adults and nonsignificant blood/lymphatic/BMT diagnosis for senior adults (OR=0.89, p=0.08).

For demographic variables, univariable analyses (data not shown) indicated that female gender was predictive of higher DT scores within each of the age groups. Within middle-aged adults, being divorced (OR=1.21, p=0.002), single (OR=1.15, p=0.003), or unemployed (OR=0.77, p<0.001) was predictive of higher DT scores. Within the senior adult group, Hispanics had higher DT scores (OR=1.84, p=0.02). Multivariable results did not differ from the univariable analyses for demographic variables, with the exception of marginally significant (FDR) divorced (OR=1.13, p=0.05) and single (OR=1.10, p=0.05) statuses for the middle-aged adults.

The unadjusted within-age-group comparisons showed the same significant relationships among the five PL categories and DT scores in each of the age groups (Table 4). However, when each age group was adjusted for significant (p<0.05) demographics and cancer types from the univariable analyses, YAs only showed a significant relationship with the Practical, Emotional, and Physical Problems categories. Middle-aged adults showed a significant relationship with all but the Family Problems category, while senior adults continued to show a significant relationship with distress across all five PL categories.

Table 4.

Within-Age-Group Comparisons of Distress Thermometer Scores and Problem List Categories Endorsed

		Unadjusted univariable models			Adjusted multivariable model
	n	Risk ratio	p-value	95% CI	Risk ratio	p-value	95% CI
Within YA group^a (n=668)
Practical Problems	242	1.61	<0.0001	1.36–1.89	1.18	0.001	1.08–1.29
Family Problems	77	1.73	<0.0001	1.35–2.22	1.11	0.07	0.99–1.25
Emotional Problems	399	2.83	<0.0001	2.44–3.28	2.39	<0.0001	2.13–2.68
Spiritual/Religious Concerns	17	1.71	0.04	1.02–2.85	1.11	0.31	0.91–1.36
Physical Problems	463	1.96	<0.0001	1.65–2.32	1.42	<0.0001	1.27–1.60
Within middle-aged adult group^b (n=3238)
Practical Problems	913	1.48	<0.0001	1.37–1.61	1.18	0.0001	1.08–1.28
Family Problems	310	1.49	<0.0001	1.32–1.68	1.11	0.10	0.98–1.26
Emotional Problems	2072	2.47	<0.0001	2.30–2.65	2.19	<0.0001	2.02–2.37
Spiritual/Religious Concerns	126	1.72	<0.0001	1.43–2.07	1.31	0.004	1.09–1.58
Physical Problems	2349	1.67	<0.0001	1.54–1.81	1.27	<0.0001	1.17–1.39
Within senior adult group^c (n=2019)
Practical Problems	254	1.57	<0.0001	1.35–1.83	1.20	0.001	1.11–1.29
Family Problems	132	1.59	<0.0001	1.30–1.95	1.21	0.001	1.10–1.32
Emotional Problems	1076	2.56	<0.0001	2.33–2.82	2.36	<0.0001	2.21–2.52
Spiritual/Religious Concerns	53	1.73	<0.0001	1.26–2.38	1.22	0.002	1.08–1.18
Physical Problems	1483	1.74	<0.0001	1.56–1.96	1.29	<0.0001	1.19–1.38

Significant values noted in bold.

Multivariable model was adjusted for gender and gynecological cancer type.

Multivariable model was adjusted for gender; marital status; employment status; and blood/lymphatic/BMT, breast, gynecological, and prostate/urologic cancer types.

Multivariable model was adjusted for gender, ethnicity, and blood/lymphatic/BMT cancer type.

BMT, bone marrow transplant; CI, confidence interval; YA, young adult.

Discussion

While the level of distress was similar when comparing YAs and older adults with cancer, differences in distress levels emerged when stratifying the older adults by middle-aged and senior adult age groups. YAs had higher cancer-related distress than senior adults, but similar distress levels when compared to middle-aged adults. Nearly half of the YAs and middle-aged adults had DT scores above the critical cutoff compared to 39% of the senior adults, indicating that fewer senior adult patients reported critical levels of distress compared to younger patients. Similarly, DT score comparisons for the demographic and cancer type variables and categories of PL items endorsed suggest that YAs and middle-aged adults had similar distress patterns compared to senior adults, who again had the lowest levels of distress among the age groups.

Our results are consistent with findings from previous outpatient oncology studies that younger age was related to more psychosocial problems or higher levels of distress in samples that included the YA to senior adult age range.^4,13 This age difference may not be cancer-specific but rather more suggestive of a life span pattern of distress, as indicated by a large-scale health surveillance survey study of a random representative sample of adults, which found that adult age groups under 65 years old (18–34, 35–49, and 50–64) were more likely to have elevated distress levels compared to those who were 65 years and older, even after statistically controlling for demographic, socioeconomic, chronic illness (excluding cancer), and other health-related variables.¹⁴

Further examination of the relationship between predictive demographic and cancer type factors and distress outcomes within each age group provides a better understanding of the differential age-related risk factors for distress. We found that YAs with gynecologic cancers were at increased risk for greater distress compared to other cancer types within this age group. Middle-aged adults with a blood/lymphatic/BMT diagnosis were at risk for elevated distress, and to a lesser degree those with gynecologic and prostate/urologic cancers compared to other cancer types within this age group. There were no significant differences among cancer types for senior adults in the multivariate analyses. Being diagnosed with cancers that have a higher likelihood of negatively impacting a patient's fertility, sexual function, and body image at a time in their lives when they may be more likely to be family planning, dating, or early in a committed relationship may be inherently more distressing for younger compared to older adults with these diagnoses. Research on sexual distress and aging indicates that younger adult females who report having sexual problems tend to experience higher distress than older adult females who identified having the same type of problems.^15,16

Our demographics results were consistent with previous research indicating female gender as a risk factor for greater distress in both cancer^13,17 and healthy populations¹⁸ across all age groups. We found that ethnicity did not appear to contribute to distress for the YAs and middle-aged adults, but Hispanic seniors were at risk for high distress. In our study, marital and employment statuses were associated with distress only for middle-aged adults; divorced or single middle-aged adults were at risk for elevated distress, suggesting that marriage is a protective factor for this age group. Being a single or divorced middle-aged adult receiving cancer treatment may be particularly stressful because a middle-aged adult may not expect to be without the support of a partner at an age when individuals are more likely to be married compared to YAs and senior adults.¹⁹ Consistent with our finding, results from the 1999–2002 National Health Interview Surveys administered to the general United States population indicated that widowed, divorced, or separated middle-aged adults were more likely to report significant psychological distress compared to married middle-aged adults.¹⁹ In a similar longitudinal survey study of healthy Canadian adults, not having a spouse was found to be a risk factor for psychological distress in all adult age groups except adults who were 80 years and older.¹⁸ Finally, we found that unemployment was associated with elevated DT scores for middle-aged adults with cancer. Unemployed YAs may be completing their education and/or feel more optimistic about finding employment after cancer recovery, whereas senior adults may be more prepared for life after retirement, both situations that are not typical in middle adulthood. While our study did not find unemployment to be a risk factor for YAs, a longitudinal study of adolescent and YA cancer patients 15–39 years old found that being uninvolved in work or school was associated with higher distress at 4, 6, and 12 months postdiagnosis.²⁰ Similarly, the previously mentioned Canadian study of healthy adults found unemployment to be a risk factor for psychological distress in men and women 18–59 years old compared to those over 70 or 80 years of age.¹⁸ Consideration of employment and school status is important when assessing cancer-related distress in both YAs and middle-aged adults.

After we adjusted for demographic and cancer type correlates within each age group, Practical, Emotional, and Physical PL items endorsed were related to distress outcomes for each age group. Spiritual/Religious Concerns were associated with elevated distress levels for middle-aged and senior adults, whereas Family Problems (dealing with children or partner) were related to distress outcomes only for senior adults. Within each age group, those that cited housing problems had the highest mean DT score among items listed under Practical Problems. For Emotional Problems, depressive (depression, sadness, lost interest in usual activities) compared to anxiety-based (fears, nervousness, worry) symptoms had the highest mean DT scores within each age group. The highest mean DT score among Physical Problems items was sexual function for YAs, but was appearance for middle-aged and senior adults.

Our study had some limitations. While the DT was completed during the first or second cancer center consultation, reviewed data did not delineate whether patients were being seen for a new diagnosis, relapse, or second opinion. However, data obtained from a separate hospital clinical database indicated that approximately two-thirds of patients seen at the cancer center during the study time period were newly diagnosed—compared with slightly over one-fourth seen for relapse and very few seen for a second opinion—indicating that the majority of patients were likely seen for a new diagnosis. Additionally, the number of patients who declined to complete the DT is unknown, thereby limiting the study's generalizability. A confound of the results was our inability to assess if patients attending college categorized themselves as “unemployed.” This study was retrospective and cross-sectional; a prospective, longitudinal study would improve understanding of the age-related distress trajectory across treatment and survivorship care. Recent findings indicate that YAs may be at greatest risk for psychosocial distress in the acute period following diagnosis and during the transition to after-treatment care, suggesting that treatment status may be a contributing factor.²⁰ Finally, this study evaluated the relationship between age-related distress and cancer type categories, making it difficult to generalize findings to specific diagnoses or other prognostic factors. Future research is necessary to further evaluate the relationship between distress outcomes and cancer type, disease staging, and treatment-related factors by age.

Conclusion

Our findings suggest that it may be more difficult for both YAs and middle-aged adults to cope with cancer-related stressors compared to senior adults, who may be more likely to anticipate having health problems at their age, leading to lower levels of distress. An alternative explanation can be drawn from life span research suggesting that distress may naturally decrease with age, independent of cancer or other chronic illnesses. Regardless, it is essential to screen for cancer-related distress, with attention to demographic and cancer type risk factors by age, to determine age-appropriate supportive care needs.

Footnotes

Acknowledgments

The authors thank Laura Weber for her assistance with data abstraction, as well as all of the clinical cancer center staff and providers who worked tirelessly to incorporate distress screening into clinical practice using the Distress Thermometer.

Author Disclosure Statement

No competing financial interests exist.

References

National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: distress management (version 2.2014). Accessed April 2, 2015, from: www.nccn.org/professionals/physician_gls/pdf/distress.pdf

Linden

, Vodermaier

, Mackenzie

, Grieg

. Anxiety and depression after cancer diagnosis: prevalence rates by cancer type, gender, and age. J Affect Disord., 2012; 141(2–3):343–51.

Zabora

, BrintzenhofeSzoc

, Curbow

, et al. The prevalence of psychological distress by cancer site. Psychooncology., 2001; 10(1):19–28.

Graves

, Arnold

, Love

, et al. Distress screening in a multidisciplinary lung cancer clinic: prevalence and predictors of clinically significant distress. Lung Cancer., 2007; 55(2):215–24.

Eiser

, Penn

, Katz

, Barr

. Psychosocial issues and quality of life. Semin Oncol., 2009; 36(3):275–80.

Zebrack

, Hamilton

, Smith

. Psychosocial outcomes and service use among young adults with cancer. Semin Oncol., 2009; 36(5):468–77.

Yanez

, Garcia

, Victorson

, Salsman

. Distress among young adult cancer survivors: a cohort study. Support Care Cancer., 2013; 21(9):2403–8.

Roberts

, Severinson

, Carraway

, et al. Life changes and problems experienced by young adults with cancer. J Psychosoc Oncol., 1997; 15(1):15–25.

Monteiro

, Torres

, Morgadinho

, Pereira

. Psychosocial outcomes in young adults with cancer: emotional distress, quality of life and personal growth. Arch Psychiatr Nurs., 2013; 27(6):299–305.

10.

Holland

. Update: NCCN practice guidelines for the management of psychosocial distress. Oncology., 1999; 13(11A):459–507.

11.

Jacobsen

, Donovan

, Trask

, et al. Screening for psychologic distress in ambulatory cancer patients. Cancer., 2005; 103(7):1494–502.

12.

Mitchell

. Pooled results from 38 analyses of the accuracy of distress thermometer and other ultra-short methods of detecting cancer-related mood disorders. J Clin Oncol., 2007; 25(29):4670–81.

13.

Giese-Davis

, Waller

, Carlson

, et al. Screening for distress, the 6th vital sign: common problems in cancer outpatients over one year in usual care: associations with marital status, sex, and age. BMC Cancer., 2012; 12:441.

14.

Chittleborough

, Winefield

, Gill

, et al. Age differences in associations between psychological distress and chronic conditions. Int J Public Health., 2011; 56:71–80.

15.

Bancroft

, Loftus

, Long

. Distress about sex: a national survey of women in heterosexual relationships. Arch Sex Behav., 2003; 32(3):193–208.

16.

Stephenson

, Meston

. The young and the restless? Age as a moderator of the association between sexual desire and sexual distress in women. J Sex Marital Ther., 2012; 38(5):445–57.

17.

Grassi

, Johansen

, Annunziata

, et al. Screening for distress in cancer patients: a multicenter, nationwide study in Italy. Cancer., 2013; 119(9):1714–21.

18.

Drapeau

, Marchand

, Forest

. Gender differences in the age-cohort distribution of psychological distress in Canadian adults. Findings from a national longitudinal survey. BMC Psychol., 2014; 2:25.

19.

Schoenborn

. Marital status and health: United States, 1999–2002 (Advance data from vital and health statistics; no. 351). Hyattsville, Maryland: National Center for Health Statistics; 2004.

20.

Kwak

, Zebrack

, Meeske

, et al. Trajectories of psychological distress in adolescent and young adult patients with cancer: a 1-year longitudinal study. J Clin Oncol., 2013; 31(17):2160–6.