Employment Status and Work Ability in Long-Term Young Adult Cancer Survivors

Abstract

Purpose:

In young adult cancer survivors (YACSs) to explore the rate of being nonemployed and having low work ability, and to identify factors associated with these two outcomes.

Methods:

All Norwegian YACSs (N = 3558) diagnosed at ages 19–39 years and treated between 1985 and 2009 for breast or colorectal cancer, leukemia, non-Hodgkin lymphoma, or melanoma, and alive in 2015, were mailed a questionnaire. The response rate was 42%. For treatment, a minimal surgery-only group (N = 198) was defined as reference group, and 1000 YACSs represented the local, systemic, and systemic plus other treatment groups. Work status was compared with normative data.

Results:

The sample included 63% females. Median age at survey was 49 years (range 27–65), and median time since first cancer diagnosis was 16 years (range 6–31). At survey, 25% (95% confidence interval [CI]: 22–27) of YACSs were nonemployed, and 38% (95% CI: 35–41) reported low work ability. The rate of being nonemployed was similar to normative data. More female YACSs held disability pension compared with normative data. In multivariable analyses, an increasing number of adverse effects (AEs), cardiovascular diseases, lower basic education, reduced level of self-rated health, and increased level of depression were significantly associated with both being nonemployed and having low work ability.

Conclusions:

In YACSs surveyed at median 49 years of age, 25% were nonemployed and 38% had low current work ability. An increased mean number of long-term AEs and several other health-related factors were significantly associated with both these outcomes. Health care providers responsible for YACSs should be attentive to such factors and work ability.

Introduction

Young adulthood (19–39 years) is a phase of life focused on finishing education, career development, establishment of paired civil status, and eventual parenthood. A diagnosis of cancer during these years may interfere significantly with these tasks. As Norwegian workers can be employed until they are 67 years, young adult cancer survivors (YACSs) putatively have many years of work life ahead after their cancer treatment. However, a recent review, providing a comprehensive analysis of what is known about YACSs and employment, reported lack of knowledge on their long-term employment status and their work ability.¹ Concerning work status, official statistics separate persons who are employed from those who are nonemployed and supported economically from various insurance schemes. Nonemployed persons can still retain a subjective work ability useful for performing nonpaid tasks for their families, organizations, or for society.^2,3

The relationship between long-term adverse effects (AEs) and employment was not mentioned in the review,¹ and AEs are strongly related to treatment modalities.^4,5 The Childhood, Adolescent and Young Adult Cancers Survey (CAYACS) is a Norwegian population-based, cross-sectional questionnaire-based study of childhood cancer survivors and YACSs at long term since diagnosis focusing primarily on their perceived needs of information and follow-up care. The study also collected work-related data used in the current substudy of YACSs aiming at the following. (1) Examining the prevalence of being nonemployed compared with normative data, comparing being nonemployed with being employed, and identifying variables significantly associated with being nonemployed. (2) Examining the prevalence of low current work ability, comparing those with low and high work ability, and identifying variables significantly associated with low work ability.

Patients and Methods

Since 1953, the Cancer Registry of Norway has systematically identified all new cancer cases in Norway. The registry contains data related to initial diagnosis, primary treatment, and survival of cancer patients. Included in the CAYAC study were all childhood cancer incidences except central nervous system malignancies, diagnosed at ages 0–18 years, and individuals ages 19–39 years if diagnosed with one of five preselected first lifetime cancers from 1985 to 2009 and alive at September 2, 2015. Only individuals in the latter group were relevant for this substudy. They had cancers common during young adult years, with good prognosis, and were not included in other studies at our research unit for long-term AEs (www.ous-research.no/longterm).

The initial YACS sample included individuals with breast cancer (stage I–III, N = 1257), malignant melanoma (N = 2902), colorectal cancer (N = 380), leukemia (N = 338), or non-Hodgkin lymphoma (N = 623). Due to their large number, a random sample of 33% of melanomas was included (N = 960). This resulted in a total sample of 3558 YACSs. In September 2015, this sample was mailed a questionnaire covering information and follow-up care needs, current mental and physical health, lifestyle, and work issues, using instruments with good psychometric properties. Nonresponders got a reminder 5 months later. Among those invited, 1488 YACSs (42%) responded: 597 with breast cancer, 155 with colorectal cancer, 253 with lymphoma, 157 with leukemia, and 326 with melanoma. Among respondents, we excluded YACSs who did not report data on oncological issues or had a new cancer diagnosis within 2 years before the survey (N = 162). Thus, 1326 YACSs were eligible for this substudy.

Primary outcome variables

Current employment status was dichotomized into “employed” (full- and part-time work and on sick leave) versus “nonemployed” (work assessment allowance, disability pension, or others, including students or homemakers). Current work ability compared with the lifetime best on a 10-point Numerical Rating Scale (NRS) from 0 (“Currently not able to do work”) to 10 (“Work ability as previous lifetime best”) from the Work Ability Index (WAI) instrument.^2,3 The scorings were dichotomized into high (score 8–10) and low (score 0–7) work ability.⁶ This one-item approach has been validated.⁷

Cancer-related variables

Information on cancer types and stages was retrieved from the Cancer Registry, while data on cancer treatment and relapses were self-reported. The sample was divided into four groups according to treatment intensity with minimal surgery as reference. That group consisted of 198 YACSs with melanoma, excluding 128 with metastases, new cancer, relapse, or extensive treatments of their melanomas. The local treatment group had major surgery and/or radiotherapy only (N = 206), the systemic treatment-only group received chemotherapy and/or endocrine treatment (N = 194), while the systemic and other treatment groups also had surgery and/or radiotherapy (N = 600). The total sample of the current study thereby consisted of 1198 YACSs.

Long-term AEs were self-reported (yes/no). A list of relevant AEs was compiled based on existing literature.^4,5 Thirteen AEs were included: hormonal changes, reduced fertility, lung problems, problems of dental health, cognition and hearing, muscular cramps, nerve pains, numbness of hands/feet, sexual problems, osteoporosis, lymphedema, and radiation injuries. The number of reported AEs was summarized as a continuous variable (Total AEs). Since chronic fatigue, anxiety, and depression were rated by instruments, these AEs were not included.

Metastases in YACSs with breast or colorectal cancers were subclassified as regional or distant.

Instruments

We included four standardized instruments: The Fatigue Questionnaire,^8,9 the Hospital Anxiety and Depression Scale, anxiety subscale,¹⁰ the Patient Health Questionnaire-9,¹¹ and self-rated health item of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30).¹² Details are described in Table 1.

Table 1.

Information About Instruments Used in This Study

Scale	Content/subscales	Scoring	Sum score
The FQ	Mental (4 items)	0 (Less than before/not at all) to 3 (very much/worse)	Mental 0–12
	Physical (7 items)		Physical 0–21
	Total (11 items)		Total 0–33
HADS	Anxiety (7 items)	0 (Not present) to 3 (highly present)	Anxiety 0–21
The PHQ-9	Depression (9 items)	0 (Not at all) to 3 (nearly every day)	Depression 0–27
The EORTC QLQ-C30	Self-rated general health (1 item)	1 (Very poor) to 7 (excellent) converted to 0–100	General health 0–100

EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; FQ, Fatigue Questionnaire; HADS, Hospital Anxiety and Depression Scale; PHQ-9, Patient Health Questionnaire-9.

Single items

Current paired relationship was categorized as present or absent. Level of basic education was dichotomized into low (≤12 years) and high (>12 years). Cardiovascular diseases included a diagnosis of myocardial infarction, angina pectoris, heart failure, stroke, diabetes, or hypertension given by a doctor at any time. Correspondingly, musculoskeletal diseases included a diagnosis of arthrosis, rheumatic diseases, or other chronic diseases in muscles and/or joints. Mental disorders included mental health conditions leading to help-seeking. Obesity concerned a body mass index ≥30, and smoking identified current daily smoking of any number of cigarettes, at survey.

Normative data

Relevant work-related norm data were retrieved from the home pages of Statistics Norway (www.ssb.no/en) and the Norwegian Labor and Welfare Service (www.nav.no/en). Norwegian official registries are collected by agencies of the government, and since all individuals living in Norway have a unique personal number, they are used for cross-checking across registries.

Statistical analyses

Group comparisons of continuous variables were performed with independent sample t-tests, and if skewed distributions, nonparametric tests were used. Group comparisons of categorical variables were performed with chi-square tests. Internal consistencies of instruments were examined with Cronbach's coefficient alpha, which were 0.83 or higher for all scales within instruments.

Due to the majority of female YACSs in our sample, we adjusted relevant between-group comparisons for sex using multivariable linear and logistic regression analyses. Associations between independent variables and nonemployed status (employed as reference) and low current work ability (high as reference) were examined with univariate and multivariable logistic regression analyses. The strength of the associations was given as odds ratios with 95% confidence intervals (95% CIs) as appropriate. Due to multicollinearity with work ability, employment status was not included in the multivariable analysis of work ability (Table 5), and for the same reason, anxiety was omitted in the multivariable analyses of Tables 3 and 5. The significant p value was set as <0.05, and all tests were two sided. The software applied was IBM SPSS Statistics version 25 for PC (IBM Corporation, Armonk, NY).

Employment status

Among 1189 YACSs, 897 (75%, 95% CI: 73–78) were employed and 292 (25%, 95% CI: 22–27) were nonemployed. Among male YACSs, 83% were employed compared with 81% among males ages 25–66 years in the Norwegian population. For females, the corresponding numbers were 73% and 76%, respectively. Among YACSs, 10% of the males and 19% of females held disability pension, while the corresponding numbers for those aged 25–65 years in the Norwegian population were 11% and 13%, respectively. Nonemployed YACSs had significantly higher mean age at survey and longer mean time since first cancer diagnosis than those employed. Compared with employed YACSs, a significantly smaller proportion of nonemployed YACSs had received minimal treatment, while significantly more were treated with systemic and other treatments (Table 2).

Nonemployed YACSs also reported a significantly higher mean number of AEs, and a significantly higher proportion of them were females and YACSs with lower level of basic education compared with employed YACSs. The nonemployed YACSs also reported significantly higher rates of comorbid cardiovascular and musculoskeletal diseases as well as mental disorders, and had higher prevalence of obesity and daily smoking. Higher mean levels of total fatigue, anxiety, and depression, and lower levels of general health were also characteristic for the nonemployed compared with the employed YACS (Table 2). In the multivariable analysis, longer time since first cancer diagnosis, increasing number of AEs, female sex, low basic education, comorbid cardiovascular disease, decreased general health, and increased level of depression remained significantly associated with YACSs being nonemployed (Table 3).

Table 2.

Characteristics of Employed and Nonemployed Young Adult Cancer Survivors at Survey

Variables	Employed (N = 897)	Nonemployed (N = 292)	p	Total (N = 1189)
Age at first diagnosis, mean (SD)	33.2 (5.7)	32.9 (5.1)	0.69^a	33.0 (5.3)
Age at survey, mean (SD)	49.3 (7.3)	51.1 (9.0)	0.001 ^a	49.7 (7.8)
Time since diagnosis, mean (SD)	15.9 (6.7)	17.3 (6.9)	<0.00 ^a	16.8 (6.8)
Types of cancer, N (%)			0.06
Melanomas	164 (18)	34 (12)	0.008	198 (17)
Breast	369 (41)	121 (41)	0.95	490 (41)
Colorectal	109 (12)	39 (13)	0.61	148 (12)
Non-Hodgkin lymphomas	154 (17)	65 (22)	0.06	219 (19)
Leukemias	102 (11)	33 (11)	1.00	135 (11)
Treatment groups, N (%)			0.02
Minimal surgery	164 (18)	34 (11)	0.008	198 (17)
Local treatment	161 (18)	44 (15)	0.28	205 (17)
Systemic treatment only	140 (16)	52 (18)	0.41	192 (16)
Systemic+others	433 (48)	162 (56)	0.04	595 (50)
Relapse, N (%)	106 (12)	44 (15)	0.12	150 (13)
Metastases ever (breast and colorectal cancers), N (%), N = 638	266 (42)	90 (14)	<0.001	356 (56)
No. of cancers, N (%)			0.17
One cancer only	834 (93)	264 (90)		1098 (92)
Two or more cancers	63 (7)	28 (10)		91 (2)
No. of late adverse effects, mean (SD)	1.8 (2.1)	3.0 (2.7)	<0.001 ^a,b	2.1 (2.3)
Sex, N (%)			<0.001
Females	631 (70)	236 (81)		867 (73)
Males	266 (30)	56 (19)		322 (27)
In paired relationship, N (%)	709 (79)	217 (74)	0.09	926 (78)
Level of basic education, N (%)			<0.001 ^a
High (>12 years)	576 (64)	122 (42)		698 (59)
Low (≤12 years)	319 (36)	170 (58)		489 (41)
Work status, N (%)			NA
Full-time job	689 (77)	0 (0)		689 (58)
Part-time job	182 (20)	0 (0)		182 (15)
On sick leave	24 (3)	0 (0)		24 (2)
Disability pension	0 (0)	198 (68)		198 (17)
Work assessment allowance	0 (0)	41 (14)		41 (3)
Other statuses	2 (0)	53 (18)		55 (5)
Work ability index, mean (SD)	8.3 (1.8)	3.9 (2.9)	<0.001 ^a	7.4 (2.7)
Comorbidities, N (%)
Cardiovascular diseases	199 (22)	126 (43)	<0.001 ^a	325 (27)
Musculoskeletal diseases	432 (48)	191 (65)	<0.001 ^a	623 (52)
Mental disorders	213 (24)	118 (40)	<0.001 ^a	331 (28)
General health, mean (SD)	77.7 (20.6)	55.3 (24.6)	<0.001 ^a	72.3 (22.6)
Total fatigue, mean (SD)	12.3 (4.2)	16.3 (6.3)	<0.001 ^a	13.2 (5.1)
Anxiety, mean (SD)	4.2 (3.5)	6.1 (4.2)	<0.001 ^a,b	4.7 (3.7)
Depression, mean (SD)	4.4 (4.1)	8.3 (5.3)	<0.001 ^a,b	5.3 (4.7)
Obesity (BMI ≥30), N (%)	129 (15)	64 (23)	0.001 ^a	193 (16)
Daily smoking, N (%)	84 (9)	52 (18)	<0.001 ^a	136 (11)

Adjusted for sex.

Nonparametric test.

Bold values indicate p < 0.05.

BMI, body mass index; NA, not applicable; SD, standard deviation.

Table 3.

Univariate and Multivariable Logistic Regression Analyses of Young Adult Cancer Survivors Being Nonemployed (N = 292) and Employed (N = 897) as Reference

Variables	Univariate analyses			Multivariable analysis
Variables	OR	95% CI	p	OR	95% CI	p
Age at survey	1.03	1.01–1.05	0.001	1.01	0.98–1.05	0.40
Time since diagnosis	1.03	1.01–1.05	0.002	1.05	1.01–1.09	0.02
Treatment groups	—		0.02			0.74
Minimal surgery	1.00	—	—	1.00	—
Local treatment	1.31	0.80–2.16	0.29	0.84	0.46–1.53
Systemic treatment only	1.78	1.09–2.90	0.02	1.13	0.61–2.08
Systemic+others	1.79	1.19–2.71	0.005	0.90	0.52–1.56
No. of adverse effects	1.21	1.16–1.26	<0.001	1.12	1.04–1.21	0.004
Females (males reference)	1.77	1.28–2.46	0.001	1.90	1.27–2.84	0.002
Low basic education	2.52	1.92–3.30	<0.001	2.11	1.53–2.92	<0.001
Comorbidities
Cardiovascular diseases	2.66	2.01–3.52	<0.001	1.85	1.31–2.63	0.001
Musculoskeletal diseases	2.04	1.55–2.68	<0.00	0.95	0.68–1.34	0.78
Mental disorders	2.18	1.65–2.88	<0.001	1.30	0.91–1.85	0.15
General health	0.96	0.95–0.97	<0.001	0.98	0.97–0.98	<0.001
Total fatigue	1.16	1.13–1.19	<0.001	1.04	0.99–1.08	0.06
Anxiety	1.13	1.10–1.17	<0.001	MC^a
Depression	1.18	1.15–1.22	<0.001	1.06	1.01–1.11	0.02
Daily smoking	2.10	1.44–3.05	<0.001	1.22	0.77–1.93	0.41
Obesity	1.70	1.21–2.37	0.002	1.14	0.76–1.72	0.53

MC: not included in multivariate analysis due to multicollinearity with depression.

Bold values indicate p < 0.05.

95% CI, 95% confidence interval; OR, odds ratio.

Ethics

The project was approved by the Regional Committee for Medical and Health Research Ethics of South-East Norway (No. 2016/953). All participants provided informed consents.

Results

The sample consisted of 323 (27%) males and 875 females (63%). We present our findings adjusted for sex (Tables 1, 2, and 4), but these adjustments hardly differed from the nonadjusted p values. Median age was 35 years (range 19–39) at first cancer diagnosis, 49 years (range 27–65) at survey, and median time since first diagnosis was 16 years (range 6–31). Cancer and treatment data are displayed in Table 2. Among metastases in breast and colorectal cancers, 97% were regional and 3% were distant.

Current work ability

Among the 1140 YACSs reporting on work ability, 706 (62%, 95% CI: 59–65) had high, while 434 (38%, 95% CI: 35–41) had low current work ability. The work ability score in the total sample was 7.4 (standard deviation [SD] 2.7), 8.3 (SD 1.8) among those employed and 3.9 (SD 1.9; p < 0.001) in those not employed. Ninety-six percent of YACSs definitely had work experience, while that is unknown for the 4% who were students or homemakers. Significantly fewer YACSs with melanoma and significantly more with breast cancer and non-Hodgkin lymphomas had low work ability. YACSs with low work ability had also less often received minimal and local treatment, but significantly more often systemic and other treatments, and had more relapses and more than one cancer compared with the high work ability group (Table 3).

Those in the low work ability group had a significantly higher mean number of AEs, and were more often females and YACSs with a lower level of basic education compared with the group with high work ability. YACSs in the low work ability group also had a significantly higher rate of comorbid cardiovascular and musculoskeletal diseases as well as mental disorders, and higher rates of current obesity and daily smoking. Those with low work ability also reported significantly higher mean levels of total fatigue, anxiety, and depression, and lower level of general health, and in that group a significantly lower proportion held full-time jobs, while higher proportions held part-time jobs, were on sick leave, work assessment allowance, and on disability pension compared with those with high work ability (Table 4). In the multivariable analysis, systemic treatment both alone and combined with other treatments, increased mean number of AEs, comorbid cardiovascular disease, low basic education, decreased level of general health, and increased levels of total fatigue and depression remained significantly associated with low current work ability (Table 5).

Table 4.

Characteristics of Young Adult Cancer Survivors with High (N = 706) and Low Current (N = 434) Work Ability

Variables	High work ability (N = 706)	Low work ability (N = 434)	p
Age at first diagnosis, mean (SD)	32.6 (7.6)	33.3 (5.4)	0.07^a
Age at survey, mean (SD)	49.4 (7.6)	49.7 (8.0)	0.53^a
Time since diagnosis, mean (SD)	16.3 (6.7)	15.8 (6.7)	0.48^a
Types of cancer, N (%)			<0.001
Melanomas	157 (22)	33 (8)	<0.001
Breast	267 (38)	205 (47)	0.002
Colorectal	94 (13)	46 (10)	0.19
Non-Hodgkin lymphomas	113 (16)	94 (22)	0.018
Leukemias	75 (11)	56 (13)	0.25
Treatment groups, N (%)			<0.001
Minimal surgery	157 (22)	33 (7)	<0.001
Local treatment	135 (19)	60 (14)	0.023
Systemic treatment only	109 (15)	73 (17)	0.56
Systemic+others	305 (43)	268 (62)	<0.001
Relapse, N (%)	73 (10)	73 (17)	0.001
Metastases ever (breast and colorectal cancers), N (%), N = 638	196 (31)	143 (22)	0.004
No. of cancers, N (%)
One cancer only	664 (94)	387 (89)	0.003
Two or more cancers	42 (6)	47 (11)	0.003
No. of late adverse effects, mean (SD)	1.5 (1.9)	3.1 (2.5)	<0.001 ^a,b
Sex, N (%)			0.001
Females	487 (69)	339 (78)
Males	219 (31)	95 (22)
In paired relationship, N (%)	556 (78)	330 (76)	0.28
Level of basic education, N (%)			<0.001 ^a
High (>12 years)	464 (66)	213 (49)
Low (≤12 years)	241 (34)	221 (51)
Work status, N (%)			<0.001
Full-time job	583 (83)	104 (24)	<0.001
Part-time job	81 (11)	100 (23)	<0.001
On sick leave	5 (1)	19 (4)	<0.001
Disability pension	6 (1)	153 (35)	<0.001
Work assessment allowance	2 (0)	37 (9)	<0.001
Other statuses	28 (4)	22 (5)	0.38
Work ability index, mean (SD)	9.1 (0.9)	4.5 (2.3)	<0.001 ^a
Comorbidity, N (%)
Cardiovascular diseases	151 (21)	150 (35)	<0.001 ^a
Musculoskeletal diseases	307 (44)	285 (67)	<0.001 ^a
Mental disorders	142 (20)	171 (39)	<0.001 ^a
General health, mean (SD)	83.1 (17.2)	56.1 (22.7)	<0.001 ^a
Total fatigue, mean (SD)	11.2 (3.4)	16.4 (5.6)	<0.001 ^a
Anxiety, mean (SD)	3.7 (3.1)	6.2 (4.1)	<0.001 ^a,b
Depression, mean (SD)	3.5 (3.3)	8.3 (5.1)	<0.001 ^a,b
Obesity (BMI ≥30), N (%)	97 (14)	87 (21)	0.003 ^a
Daily smoking, N (%)	62 (9)	63 (15)	0.006 ^a

Adjusted for sex.

Nonparametric test.

Bold values indicate p < 0.05.

Table 5.

Univariate and Multivariable Logistic Regression Analyses of Young Adult Cancer Survivors Low (N = 434) and High Current Work Ability (N = 706) as Reference

Variables	Univariate analyses			Multivariable analysis
Variables	OR	95% CI	p	OR	95% CI	p
Age at first diagnosis	1.03	1.00–1.05	0.02	1.02	0.99–1.06	0.14
Treatment groups						0.06
Minimal surgery	1.00	—	—	1.0	—	—
Local treatment	2.11	1.30–3.43	0.002	1.31	0.69–2.47	0.41
Systemic treatment only	3.19	1.98–5.14	<0.001	2.07	1.08–3.95	0.03
Systemic+others	4.18	2.78–6.30	<0.001	1.98	1.11–3.52	0.02
No. of adverse effects	1.37	1.30–1.45	<0.001	1.13	1.04–1.23	0.003
Females (males reference)	1.61	1.22–2.12	0.001	1.23	0.82–1.84	0.31
Low basic education	2.00	1.57–2.55	<0.001	1.87	1.34–2.61	<0.001
Comorbidities
Cardiovascular diseases	1.94	1.49–2.54	<0.001	1.52	1.05–2.20	0.025
Musculoskeletal diseases	2.49	1.94–3.18	<0.001	1.20	0.85–1.68	0.30
Mental disorders	2.58	1.98–3.37	<0.001	1.20	0.83–1.75	0.33
General health	0.94	0.93–0.95	<0.001	0.96	0.95–0.97	<0.001
Total fatigue	1.32	1.27–1.37	<0.001	1.13	1.08–1.18	<0.001
Anxiety	1.21	1.17–1.26	<0.001	MC^a
Depression	1.33	1.20–1.38	<0.001	1.10	1.04–1.15	<0.001
Obesity	1.59	1.16–2.19	0.004	1.06	0.69–1.63	0.78
Daily smoking	1.76	1.21.–2.56	0.003	0.87	0.52–1.45	0.59

MC: not included in multivariate analysis due to multicollinearity with PHQ depression.

Bold values indicate p < 0.05.

Discussion

At survey, 25% of YACSs were nonemployed at a median age of 49 and 16 years since their first cancer diagnosis. Correspondingly, 38% of YACSs reported low current work ability, so the overlap between nonemployment and low work ability was not complete. Low basic education, increased mean number of AEs, comorbid cardiovascular disease, decreased general health, and increased level of depression were associated with both nonemployment and low work ability. These variables therefore are of obvious relevance for YACSs work life situation at long term. While basic education belongs to the past, the other four variables are currently relevant and should be in focus for targeted interventions by the health care providers responsible for the follow-up of YACSs.

To our knowledge, the significant associations between increased number of AEs and nonemployment and low work ability are new findings among YACSs, underlining the clinical relevance of checking YACSs for AEs at follow-up visits. Note that cardiovascular diseases, mental health conditions, and chronic fatigue were omitted from our list of AEs since they were covered by instruments. If these variables had been included among the AEs, the impact of AEs on work outcomes would have been even stronger. We separated the different treatment regimens into four treatment groups of increasing treatment burden with minimal surgery as a reference. Heavier treatment burden was significantly associated with both nonemployment and low current work ability in univariate analyses. In contrast, minimal surgery was negatively associated with poor work outcomes. More intensive treatments remained significantly associated with low work ability in the multivariable analysis. Treatment intensity thus seems to have a long-lasting effect on work outcomes. However, more intensive treatments also imply more AEs.^4,5

We found that the negative effects of AEs on work performance were statistically stronger than those associated with cancer types and treatment intensity. Clinically, YACSs with heavier treatment burden should be particularly examined for AEs at long-term follow-up. Regional metastases of YACSs with breast or colorectal cancers were more frequent in the employed and high work ability groups. This finding indicates that such metastases probably are of less relevance in YACSs concerning work performance. Several studies have shown that long-term YACSs are at increased risk for cardiovascular diseases.^13,14 We have documented the relevance of such comorbidities for the work status and work ability of YACSs. Compared with cancer-free controls, YACSs have poorer general health status¹⁵ and more mental health conditions.¹⁶ We demonstrated that such health status and present mental health conditions (increased level of depression and anxiety, and help-seeking for mental problems) were associated with nonemployment and low work ability in YACSs. Health care providers responsible for YACSs at long term should be aware of the negative work consequences regarding these health issues.

Being female YACSs was significantly associated with nonemployment, but not with low work ability status in multivariable analyses. Among cancer survivors in general, females are at greater risk for unemployment or early disability pension compared with men.^17,18 An explanation for our result could be that a proportion of female YACSs has too low work ability for official work life, but still do a lot of unpaid work within family, charity, or as volunteers. Increased prevalence of disability pension among female YACSs compared with the norm (19% vs. 13%) could also be part of the explanation for this finding. These results point to female YACSs being more vulnerable to exclusion from work life than males. The employment rates of YACSs of both sexes were similar to those found in the general Norwegian population. This finding is in contrast to studies that have reported higher rates of nonemployment among cancer survivors in general.¹⁷

One explanation may be the low rates of nonemployment in Norway for several decades compared with most other countries. Being employed or not is dependent on national legal and economic regulations concerning the labor market, prevalence of unemployment among young adults, and criteria for getting social support. The rate of disability pension among male YACSs was similar to the population and significantly lower than for female YACSs, who had a higher rate than the population. This finding is in line with results from previous Scandinavian studies.¹⁸ While employment status is categorized objectively, current work ability is a subjective measure compared with lifetime best. The advantage of assessing this outcome is its independence of employment status, making it useful for YACSs who are nonemployed but still able to do useful work. One problem with current work ability is the comparison with “lifetime best,” which could be years ago, indicating problems of recall and eventual response shift.⁷ A systematic review of work ability in cancer survivors found low work ability to be significantly associated with increased fatigue, anxiety, and depression as observed in our study.¹⁹ A Nordic study also demonstrated a significant association between low work ability and comorbidities as also found by us.²⁰

There is a lack of population-based normative data concerning current work ability, which represents a weakness of studies of that work outcome. Lifestyle factors, such as obesity and daily smoking, could have been established before the primary cancer diagnosis or later on. Even though not significant in multivariable analyses, they showed statistically significant associations with both work outcomes in the univariate models. Therefore, modifications of unhealthy lifestyle in YACSs could be relevant for improvement of current work ability and work status, and additionally have protective effects concerning relapses and second cancers. In their recent extensive review concerning YACSs and employment, Stone et al.¹ reported that low level of education was an important factor negatively associated with poorer work outcomes, confirmed in our study. Like them, we observed that aspects of mental health conditions (anxiety and depression) were significantly associated with poorer work outcomes. Our findings also supported their conclusion that: “the related studies found that survivors become equal to healthy controls overall in achieving a successful career” although we only examined employment rates.

Strengths and limitations

Patient identification by the Cancer Registry precluded any selection bias concerning the YACSs addressed by us. Use of established instruments for key variables is considered a major strength. The same could be said for our use of two different work outcomes, one categorical and objective and the other dimensional and subjective. Our response rate was 42%, which is quite common in population-based questionnaire surveys without any rewards for responding. The reliance of self-report of treatment is a limitation. Another is that we do not know if cancer was the specific reason for nonemployment or low work ability. The cross-sectional design of our substudy precludes drawing causal conclusions. Another limitation concerns our lack of data on family income, which could be relevant for work status of YACSs. Taken together, we do not consider these limitations to weaken our main findings.

Footnotes

Acknowledgments

The project has received funding from the Norwegian Research Council (Grant No. 218312), the Norwegian Cancer Association (Grant No. 45480), and from the Legacies of the Norwegian Radium Hospital (Grant No. 335007).

Author Disclosure Statement

No competing financial interests exist.

References

Stone

, Ganz

, Pavlish

, Robbins

. Young adult cancer survivors and work: a systematic review. J Cancer Surviv. 2017; 11(6):765–81.

Tuomi

, Ilmarinen

, Jakhola

, et al. Work ability index. 2nd ed. Helsinki: Finnish Institute of Occupational Health; 2006.

van den Berg

TIJ

, Elders

LAM

, de Zwart

BCH

, Burdorf

. The effects if work-related and individual factors on the Work Ability Index: a systematic review. Occup Environ Med. 2009; 66(4):211–20.

Ganz

. Survivorship: adult cancer survivors. Prim Care. 2009; 36(4):721–41.

Kiserud

, Dahl

, Fosså

. Cancer survivorship in adults. Recent Results Cancer Res. 2018; 210:123–43.

Dahl

, Cvancarova

, Dahl

, et al. Work ability inprostate cancer survivors after radical prostatectomy. Scan J Urol. 2016; 50(2):116–22.

Ahlstrom

, Grimby-Ekman

, Hagberg

, Dellve

. The work ability index and single-item question: associations with sick leave, symptoms, and health—a prospective study of women on long-term sick leave. Scand J Work Environ Health. 2010; 36(5):404–12.

Chalder

, Berelowitz

, Pawlikowska

, et al. Development of a fatigue scale. J Psychosom Res. 1993; 37(2):147–53.

Loge

, Ekeberg

, Kaasa

. Fatigue in the general Norwegian population: normative data and associations. J Psychosom Res. 1998; 45(1):53–65.

10.

Bjelland

, Dahl

, Haug

, Neckelmann

. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res. 2002; 52(2):69–77.

11.

Kroenke

, Spitzer

, Williams

JBW

. The PHQ-9: validity of a brief depression severity measure. J Intern Med. 2001; 16(9):606–13.

12.

Aaronson

, Ahmedzai

, Bergman

, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993; 85(5):365–76.

13.

Rugbjerg

, Mellemkjaer

, Boice

, et al. Cardiovascular disease in survivors of adolescent and young adult cancer: a Danish cohort study, 1943–2009. J Natl Cancer Inst. 2014; 106(6):dju110.

14.

Chao

, Xu

, Bhatia

, et al. Cardiovascular disease risk profiles in survivors of adolescent and young adult (AYA) cancer: the Kaiser Permanente AYA Cancer Survivors Study. J Clin Oncol. 2016; 34(14):1626–33.

15.

Kaul

, Veeranki

, Rodriguez

, Kuo

. Cigarette smoking, comorbidity, and general health among survivors of adolescent and young adult cancer. Cancer. 2016; 122(18):895–905.

16.

Kaul

, Avila

, Mutambudzi

, et al. Mental distress and health care use among survivors of adolescent and young adult cancer: a cross-sectional analysis of the National Health Interview Survey. Cancer. 2017; 123(5):869–78.

17.

de Boer

, Taskila

, Ojajärvi

, et al. Cancer survivors and unemployment: a meta-analysis and meta-regression. JAMA. 2009; 301(7):753–62.

18.

Carlsen

, Oksbjerg Dalton

, Frederiksen

, et al. Cancer and the risk for taking early retirement pension: a Danish cohort study. Scand J Public Health. 2008; 36(2):117–25.

19.

von Bonsdorff

, Rantanen

, Törmäkangas

, et al. Midlife work ability and mobility limitation in old age among non-disability and disability retirees–a prospective study. BMC Public Health. 2016; 16:154.

20.

Mehnert

, Barth

, Gaspar

, et al. Predictors of early retirement after cancer rehabilitation—a longitudinal study. Eur J Cancer Care. 2017; 26(5). DOI: 10.1111/ecc 12528