Abstract
Objectives:
To examine the association between arthritis diagnosis and educational and employment participation among young adults, and to determine whether findings differ by self-rated health.
Methods:
Data from the National Health Interview Survey, years 2009 to 2014, were combined and analyzed. Our sample was restricted to those aged 18-29 years either diagnosed with arthritis (n = 1,888) or not (n = 35,051). Prevalence and correlates of employment and educational participation in the past week were compared by arthritis status. Demographic characteristics, social role participation restrictions, health factors, and health system use variables were included as covariates. Self-rated health and arthritis diagnosis interaction was also examined within models. Weighted proportions, univariate, and multivariate associations were calculated to examine the association between arthritis and educational and employment participation.
Results:
Arthritis respondents were more likely to be female, married, report more social participation restrictions, fair/poor health, and functional limitations than those without arthritis. In multivariate models, arthritis was significantly associated with slightly higher employment participation (PR = 1.08; 95%CI 1.03-1.14), but not with educational participation (PR = 0.75; 95%CI 0.56-1.02). Stratified analyses showed that among those reporting higher self-rated health, arthritis diagnosis was associated with higher employment participation (PR = 1.09; 95%CI 1.03-1.14).
Conclusion:
For young adults with arthritis there may be need to design interventions that focus both on improving self-rated health and promoting labor market participation.
Keywords: Young adulthood, rheumatic disease diagnosis, employment, education, transition to adulthood
INTRODUCTION
In the United States, as in many other industrialized countries, arthritis is significantly related to work disability and challenges with educational attainment (1-4). While arthritis is diagnosed across the life course (5), research examining career experiences has focused on middle- and older-aged adults (6), and few studies have examined the educational and employment participation of young adults with arthritis.
Young adulthood is a unique time of life where a person may increasingly participate in social roles (e.g., marriage, parenting, and full-time employment) that mark a transition from the dependency of childhood to independence of adulthood (7). Of particular importance, involvement in educational and paid and unpaid work roles in young adulthood can provide opportunities for vocational skills building that determine a person’s career trajectory (7). Challenges with education and employment in young adulthood can have potential adverse economic consequences that can extend across the working life course including persistent unemployment, reduced long-term earnings, and deterioration of skills (8, 9). Other research finds that young adults are considered a vulnerable segment of the labor market and are more susceptible to shifts in economic conditions. These studies suggest that young adults are more likely to report unemployment, job strain, and periods of idleness (e.g. in neither education nor employment) when compared with middle-aged adults at the later phases of their career (8-10). An arthritis diagnosis may add to the challenges faced by young adults in their early work experiences.
Recently, demographic trends indicate that young people are extending the time spent in transitional roles as a way to delay entering adulthood (7, 11). Delayed transitions to adulthood may provide opportunities for young adults to gain the competencies and experience necessary to compete in a knowledge-based economy (7). Young adults with chronic conditions, including those diagnosed with arthritis, also experience delayed transitions associated with interruptions to education and employment stemming from the time needed for treatment regimens or to recover from medical interventions (4, 12). The extended time spent in the young adult life phase could provide those with arthritis an opportunity to catch up to their peers without a health condition, and attain a similar level of educational and employment participation.
To date, a majority of research on the relationship between arthritis and work experiences has focused on middle- and older adults. These studies show that those living with arthritis are more likely to report challenges sustaining employment (3, 13), limitations performing workplace activities, and productivity losses when compared with non-arthritis peers (1, 14, 15). Little is known about the experiences of people diagnosed with arthritis at the early career phase. In a recent literature review of primarily clinic-based studies of young people with arthritis, results showed mixed findings with some studies suggesting that young adults with arthritis are less likely to be employed when compared to their peers without a health condition and others indicating similar or higher employment participation (6, 16, 17).
One consistent research finding among existing clinic-based studies of young adults with arthritis is that those reporting poorer health outcomes, including lower self-rated health status and greater pain, fatigue, activity limitations and disease activity, were less likely to participate in education and employment when compared to those reporting more favorable health outcomes (6, 12, 18-21). These studies suggest that difficulties with the management and control of symptoms associated with arthritis may play a significant role in determining involvement in transitional roles. At the population-level, however, no studies have examined whether educational and employment participation differs among young adults with and without an arthritis diagnosis based on health status.
We use a national population-based data source to examine the association between arthritis diagnosis and participation in employment and education in the young adult life phase. We also examine whether employment and educational participation differ when compared by higher and lower self-rated health status among those with and without arthritis. Self-rated health was chosen because of its previously established association with arthritis symptoms (e.g., pain, fatigue and activity limitations) and ability to capture the perceived health status of arthritis and non-arthritis samples at a population level (22-24). We hypothesize that an arthritis diagnosis in young adulthood will be associated with a lower prevalence of educational and employment participation when compared with their peers without arthritis. We also hypothesize that the relationship between arthritis diagnosis and lower prevalence of educational and employment participation will remain consistent when comparing young adults by self-rated health status.
METHODS
Sample
Data from the 2009-2014 US National Health Interview Survey (NHIS) were combined and analyzed (25). NHIS is an ongoing health survey of all ages and uses a complex design to select a sample representing the US civilian, non-institutionalized population. Data are collected through in-home interviews conducted by trained interviewers (25). Our study focused on young adults (18-29 years) with (n = 1,888) and without (n = 35,051) an arthritis diagnosis who had records in the sample adult core and person files (25). In the NHIS, arthritis is defined using the question “Have you ever been told by a doctor or other health professional that you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?”
Outcome measures
Both outcomes of interest, participation in employment and education, were determined by asking respondents the question “What is your correct working status in the last week?” Those not employed were asked about the main reason for not working. Based on responses, participants were uniquely classified as employed, student, unemployed, disabled, or other (e.g., retired, homemaker). Those employed were considered to participate in employment, and students were considered to participate in education.
Covariates
Covariates were included in the analyses because of their association with employment and educational outcomes documented in previous literature (6).
Demographic variables:
Sex, race/ethnicity (i.e., non-Hispanic white, non-Hispanic black, Hispanic, and non-Hispanic other/multiple race), marital status (yes/no married or living as married), and highest level of educational attainment were analyzed.
Social participation restriction:
We defined restricted social participation as those who reported “very difficult” or “can’t do at all” to at least one of two questions asking about the ability to participate in social activities outside the home (e.g., go shopping, to the movies, or sporting events; attend clubs, parties, meetings, or visit friends) (26).
Health factors:
We examined self-rated health by asking the participant “In general, would you say that your health is...” response options included poor, fair, good, very good and excellent. Consistent with previous studies, participants reporting good/very good/excellent health were clustered into one group, labeled as “higher self-rated health”. Those reporting fair/poor health were grouped together as “lower self-rated health” (23, 27).
Other health factors examined included the number self-reported doctor-diagnosed comorbid conditions from a list of nine (i.e., hypertension, heart diseases, stroke, diabetes, asthma, cancer, weak or failing kidneys, hepatitis, and chronic obstructive pulmonary disease), smoking status and body mass index (calculated as kg/m2 from self-reported weighted and height: underweight/normal [≤24.9], overweight [25.0-29.9], and obese [≥30]). We defined functional limitation as those who answered “very difficult” or “can’t do at all” to one or more of eight activities (i.e., push/pull large objects, walk ¼ mile, stand for 2 hours, sit for 2 hours, stoop/bend/kneel, reach over one’s head, grasp small objects, climb stairs). We defined serious psychological distress as those scoring ≥ 13 out of 30 using a validated six item scale for population surveillance (28).
Healthcare service utilization:
We defined health insurance as covered/not covered based on self-report, and also number of doctor visits in the past year (i.e., 0-1, 2-5, or ≥6 visits).
Analysis
We tested the study hypotheses using descriptive, univariate, and multivariate analyses. To generate nationally representative population estimates, we applied sampling weights created by National Center for Health Statistics in all analyses. We used SAS software Version 9, which includes complex survey design procedures (29), and SUDAAN (30).
Weighted proportions with 95% confidence intervals (95% CIs) for the outcome measures and all study covariates were generated. Non-overlapping 95% CIs were used to conservatively indicate a statistically significant difference between the two groups. Weighted proportions were used to compare the relationship between arthritis and educational and employment participation by self-rated health status to identify potential interaction effects.
Next, we evaluated the association between arthritis and participation in education or employment in two ways using prevalence ratios (PR) with 95% CIs calculated from predicted margins. We first generated univariate models for each outcome with arthritis status as the exposure. Then, we controlled for study covariates in separate multivariate logistic regression models. The logistic models examined education and employment as their primary outcome variables. Not being involved in education or employment was the reference variable in each model, respectively. We used backwards elimination as the strategy for variable selection in the models ending when all variables were under the α ≤ 0.10 level. We conducted an additional multivariate logistic model, stratified by higher and lower self-rated health in models where the interaction between self-rated health and arthritis diagnosis was significant.
RESULTS
A description and comparison of young adults with and without arthritis is provided in Table 1. Mirroring population patterns in older age groups, young adults with arthritis were more often female (58.5% versus 49.7%) and married/living as if married (48.0% versus 34.9%), but reported similar educational attainment when compared with their peers without arthritis. Although generally low, those with arthritis indicated restricted social participation compared to those without arthritis (5.5% versus 0.8%). Arthritis diagnosis in young adulthood was associated with a higher prevalence of self-rated fair/poor health (19.6% versus 4.2%), obesity (37.1% versus 19.9%), one or more comorbid conditions (41.9% versus 15.6%), one or more functional limitations (23.2% versus 2.7%), serious psychological distress (9.7% versus 2.4%), and current smoking (35.4% versus 20.2%) when compared to those without arthritis. Arthritis diagnosis was also associated with a higher prevalence of reporting more than six doctor visits in the past year and a slightly higher prevalence of health insurance coverage when compared to those without arthritis.
Table 1:
Unweighted and weighted distributions of sample characteristics of young adults (18-29 years), with and without doctor-diagnosed arthritis, 2009-2014 National Health Interview Survey.
| Doctor diagnosed arthritis | Without doctor diagnosed arthritis | |||||||
|---|---|---|---|---|---|---|---|---|
| n unweighted | N weighted | % | 95% CI | n unweighted | N weighted | % | 95% CI | |
| Total | 1,188 | 1,677,361 | 35,051 | 48,907,368 | ||||
| Demographic factors | ||||||||
| Sex | ||||||||
| Male | 431 | 696,061 | 41.5 | 37.9 – 45.1 | 16,294 | 24,623,047 | 50.3 | 49.6 - 51.1 |
| Female | 757 | 981,301 | 58.5 | 54.9 – 62.1 | 18,757 | 24,284,321 | 49.7 | 48.9 – 50.4 |
| Race/ethnicity | ||||||||
| Non-Hispanic white | 783 | 1,196,321 | 71.3 | 67.8 – 74.8 | 18,030 | 29,064,761 | 59.4 | 58.3 – 60.5 |
| Non-Hispanic black | 193 | 237,818 | 14.2 | 11.5 – 16.9 | 5561 | 6,884,343 | 14.1 | 13.4 – 14.7 |
| Hispanic | 155 | 172,671 | 10.3 | 8.1 – 12.5 | 8398 | 9,909,058 | 20.3 | 19.4 – 21.1 |
| Non-Hispanic other | 57 | 70,551 | 4.2 | 2.8 – 5.7 | 3062 | 3,049,206 | 6.2 | 5.8 – 6.7 |
| Married or living as married | ||||||||
| Yes | 507 | 805,219 | 48.0 | 44.4 – 51.7 | 11579 | 17,053,316 | 34.9 | 34.0 – 35.7 |
| No | 681 | 872,143 | 52.0 | 48.3 – 55.6 | 23472 | 31,854,052 | 65.1 | 64.3 – 66.0 |
| Education | ||||||||
| Less than high school | 141 | 220,266 | 13.1 | 10.4 – 15.9 | 4,761 | 6,855,100 | 14.1 | 13.4 – 14.7 |
| High school or equivalent | 325 | 502,632 | 30.0 | 26.4 – 33.5 | 8813 | 13,017,924 | 26.7 | 25.9 – 27.5 |
| Some college/associate degree | 375 | 491,074 | 29.3 | 26.0 – 32.6 | 10233 | 14,773,700 | 30.3 | 29.4 – 31.2 |
| College graduate or above | 347 | 463,389 | 27.6 | 24.5 - 30.7 | 11162 | 14,130,640 | 29.0 | 28.1 - 29.8 |
| Social participation | ||||||||
| Restricted | 64 | 91,620 | 5.5 | 3.9 – 7.0 | 253 | 412,615 | 0.8 | 0.7 – 1.0 |
| Not restricted | 1124 | 1,585,741 | 94.5 | 93.0 – 96.1 | 34798 | 48,494,753 | 99.2 | 99.0 – 99.3 |
| Health factors | ||||||||
| Self-rated health | ||||||||
| Higher (Excellent/very good/good) | 955 | 1,347,317 | 80.4 | 77.5 – 83.3 | 33,406 | 46,856,500 | 95.8 | 95.6 – 96.1 |
| Lower (Fair/poor) | 232 | 329,058 | 19.6 | 16.7 – 22.5 | 1637 | 2,044,389 | 4.2 | 3.9 – 4.4 |
| Body mass index (kg/m2) | ||||||||
| Under- or normal weight (< 25.0) | 421 | 597,674 | 36.4 | 32.8 – 39.9 | 17577 | 24,826,442 | 52.0 | 51.3 – 52.8 |
| Over weight (25.0 – 29.9) | 324 | 436,335 | 26.6 | 23.4 – 29.8 | 9729 | 13,384,453 | 28.0 | 27.3 – 28.8 |
| Obese (> 30.0) | 417 | 608,878 | 37.1 | 33.4 – 40.7 | 6838 | 9,515,613 | 19.9 | 19.4 – 20.5 |
| Number of comorbid conditions | ||||||||
| None | 723 | 975,134 | 58.1 | 54.2 – 62.0 | 29,353 | 41,311,557 | 84.5 | 84.0 – 84.9 |
| One | 333 | 504,712 | 30.1 | 26.6 – 33.6 | 4,944 | 6,601,553 | 13.5 | 13.1 – 13.9 |
| Two | 97 | 152,899 | 9.1 | 6.9 – 11.3 | 653 | 864,474 | 1.8 | 1.6 – 1.9 |
| Three or more | 35 | 44,617 | † | † | 101 | 129,784 | 0.3 | 0.2 – 0.3 |
| Specific functional limitation | ||||||||
| One or more limitations | 269 | 388,581 | 23.2 | 20.0 – 26.3 | 932 | 1,322,984 | 2.7 | 2.5 – 2.9 |
| No limitations | 919 | 1,288,780 | 76.8 | 73.7 – 80.0 | 34,119 | 47,584,383 | 97.3 | 97.1 – 97.5 |
| Serious psychological distress | ||||||||
| Yes | 121 | 159,352 | 9.7 | 7.5 – 11.9 | 844 | 1,136,164 | 2.4 | 2.1 – 2.6 |
| No | 1049 | 1,486,491 | 90.3 | 88.1 – 92.5 | 33,718 | 47,105,060 | 97.6 | 97.4 – 97.9 |
| Smoking status | ||||||||
| Current smoker | 414 | 593,724 | 35.4 | 31.7 – 39.1 | 6,979 | 9,861,292 | 20.2 | 19.6 – 20.9 |
| Former smoker | 131 | 170,619 | 10.2 | 8.0 – 12.3 | 2,877 | 3,934,608 | 8.1 | 7.7 – 8.5 |
| Never smoker | 642 | 910,975 | 54.4 | 50.6 – 58.2 | 25,072 | 34,928,750 | 71.7 | 70.9 – 72.5 |
| Health service use variables | ||||||||
| Number of doctor visits | ||||||||
| None or one | 303 | 406,213 | 24.5 | 21.5 – 27.5 | 17,376 | 24,280,888 | 50.3 | 49.5 – 51.0 |
| Two to five | 435 | 602,271 | 36.3 | 32.8 – 39.9 | 11,865 | 16,778,885 | 34.7 | 34.0 – 35.5 |
| More than six | 437 | 649,848 | 39.2 | 35.5 – 42.8 | 5,347 | 7,230,485 | 15.0 | 14.4 – 15.5 |
| Health insurance | ||||||||
| Not covered | 268 | 372,755 | 22.4 | 19.5 – 25.3 | 9,541 | 12,979,574 | 26.7 | 26.0 – 27.5 |
| Covered | 914 | 1,293,578 | 77.6 | 74.7 – 80.5 | 25,278 | 35,552,165 | 73.3 | 72.5 – 74.0 |
| Work status in the past week | ||||||||
| Employed (employment participation) | 754 | 1,050,704 | 63.1 | 59.4 – 66.9 | 23,648 | 32,292,033 | 66.2 | 65.4 – 67.0 |
| Student (educational participation) | 73 | 105,076 | 6.3 | 4.3 – 8.3 | 4,116 | 6,236,096 | 12.8 | 12.0 – 13.5 |
| Unemployed | 141 | 200,292 | 12.0 | 9.4 – 14.6 | 4,165 | 6,354,462 | 13.0 | 12.5 – 13.6 |
| Disabled | 122 | 181,509 | 10.9 | 8.7 – 13.2 | 636 | 1,016,536 | 2.1 | 1.8 – 2.3 |
| Other (homemaker/retired) | 91 | 126,921 | 7.6 | 5.8 – 9.5 | 2,380 | 2,876,077 | 5.9 | 5.5 – 6.2 |
Notes: † = Estimate is potentially unreliable with a relative standard error between 20–30 and not presented in this table.
Univariate analyses indicated that young adults with arthritis reported a similar prevalence of employment participation when compared with their peers without arthritis (63.1% versus 66.2%; univariate PR = 0.95; 95%CI = 0.90 - 1.01). Educational participation among respondents with arthritis was less than half that of their peers without arthritis (6.3% versus 12.8%; univariate PR = 0.49; 95% CI = 0.36 - 0.68) (Table 1).
Among those with higher self-rated health, young adults with arthritis reported similar employment participation when compared with those without arthritis (70.3% versus 67.1%) (Table 2). Among those with lower self-rated health, young adults with arthritis had lower employment participation when compared with those without arthritis (33.1% versus 46.1%). The prevalence estimate for educational participation among those with either higher or lower self-rated health and arthritis were unreliable (relative standard error [RSE] > 30%). Accordingly, for the educational participation model, the self-rated health-arthritis diagnosis interaction was not presented in Table 2 or carried forward in subsequent multivariate analysis.
Table 2.
Unweighted and weighted distributions of employment participation by arthritis and self-rated health status among young adults (18-29 years).
| Employment Participation | ||||||||
|---|---|---|---|---|---|---|---|---|
| Higher self-rated health‡ | Lower self-rated health‡‡ | |||||||
| n unweighted | N weighted* | %† | 95% CI | n unweighted | N weighted* | %† | 95% CI | |
| Arthritis | 670 | 943,186 | 70.3 | 66.5 – 73.9 | 83 | 106,530 | 33.1 | 25.9 – 41.1 |
| No arthritis | 22,838 | 31,351,164 | 67.1 | 66.2 – 67.9 | 804 | 936,502 | 46.1 | 43.1 – 49.1 |
Notes: * = single year estimate; † = Estimated rates for employed (numerator) out the of the specified group row (denominator) and may not sum up to 100%; ‡ = higher self-rated health includes those who reported that their health was excellent, very good or good ; ‡‡ = lower self-rated health included those reporting that their health was fair or poor.
Table 3 presents findings from the multivariate model examining the association between arthritis diagnosis and educational participation when controlling for demographic, social role participation restriction, health factors and health service use variables. Although findings were statistically similar, the prevalence of educational participation among young adults with arthritis was slightly lower than their peers without any health condition (PR = 0.75; 95%CI 0.56-1.02).
Table 3:
Prevalence ratios from a multivariable model examining the association between arthritis and educational participation.
| Educational participation | |||
|---|---|---|---|
| PR | SE | 95% CI | |
| Arthritis diagnosis | |||
| Arthritis diagnosis vs. no arthritis diagnosis | 0.75 | 0.11 | 0.56 – 1.02 |
| Demographic factors | |||
| Sex | |||
| Male vs. female | 1.05 | 0.05 | 0.96 – 1.15 |
| Race/ethnicity | |||
| Non-Hispanic black vs. non-Hispanic white | 0.97 | 0.06 | 0.87 – 1.09 |
| Hispanic vs. non-Hispanic white | 1.06 | 0.05 | 0.96 – 1.17 |
| Non-Hispanic other vs. non-Hispanic white | 1.72 | 0.09 | 1.55 – 1.90 |
| Marital status | |||
| Not married vs. married or living as if married | 4.40 | 0.28 | 3.88 – 5.00 |
| Social participation restrictions | |||
| Not restricted vs. restricted | 1.99 | 0.56 | 1.13 – 3.47 |
| Health factors | |||
| Self-rated health | |||
| Higher vs. lower | 1.09 | 0.12 | 0.89 – 1.35 |
| Body mass index (kg/m2) | |||
| Underweight/normal weight vs. overweight | 1.29 | 0.06 | 1.17 – 1.41 |
| Underweight/normal weight vs. obese | 1.64 | 0.09 | 1.47 – 1.83 |
| Smoking status | |||
| Never smoked vs. current smoker | 2.53 | 0.18 | 2.20 – 2.92 |
| Never smoked vs. former smoker | 1.68 | 0.16 | 1.39 – 2.02 |
| Health service use variables | |||
| Number of doctor visits | |||
| 2 -5 visits vs. ≤ 1 | 1.10 | 0.05 | 1.01 – 1.20 |
| ≥ 5 visits vs. ≤ 1 | 1.01 | 0.06 | 0.89 – 1.14 |
| Health insured | |||
| Covered vs. not covered | 1.68 | 0.10 | 1.50 – 1.89 |
Notes: PR = Prevalence ratio; SE = Standard error; CI = 95% confidence interval; * = Education was not considered as a candidate variable for the current student model.
As shown in Table 4, controlling for the covariates that met model-building criteria, an arthritis diagnosis in young adulthood was associated a slightly increased likelihood of employment participation (PR = 1.08; 95% CI 1.03-1.14). Of note, the interaction between arthritis diagnosis and self-rated health was significantly associated with employment participation (PR = 1.09; 95%CI 1.04-1.15) and indicated a need for a multivariate model stratified by self-rated health. Results of the stratified multivariable model comparing results by higher and lower self-rated health are displayed in Table 5. Among young adults reporting higher self-rated health, an arthritis diagnosis was associated with 9% higher prevalence of employment participation when compared to those without arthritis (PR = 1.09; 95%CI 1.03 – 1.14). Among those reporting lower self-rated health, an arthritis diagnosis was associated with a similar prevalence of employment participation when compared to peers without arthritis (PR = 0.86; 95%CI 0.69-1.07).
Table 4:
Prevalence ratios from a multivariable model examining the association between arthritis and employment participation.
| Employment Participation | |||
|---|---|---|---|
| PR | SE | 95% CI | |
| Arthritis diagnosis | |||
| Arthritis diagnosis vs. no arthritis diagnosis | 1.08 | 0.03 | 1.03 – 1.14 |
| Demographic factors | |||
| Sex | |||
| Male vs. female | 1.09 | 0.01 | 1.07 – 1.12 |
| Race/ethnicity | |||
| Non-Hispanic black vs. non-Hispanic white | 0.89 | 0.01 | 0.87 – 0.92 |
| Hispanic vs. non-Hispanic white | 0.98 | 0.01 | 0.95 – 1.01 |
| Non-Hispanic other vs. non-Hispanic white | 0.78 | 0.02 | 0.74 – 0.81 |
| Education | |||
| College and above vs. less than high school | 1.68 | 0.04 | 1.61 – 1.76 |
| College and above vs. high school or equivalent | 1.29 | 0.02 | 1.26 – 1.33 |
| College and above vs. some college | 1.26 | 0.02 | 1.22 – 1.29 |
| Marital status | |||
| Not married vs. married or living as if married | 0.89 | 0.01 | 0.87 – 0.91 |
| Social participation | |||
| Not restricted vs. restricted | 1.79 | 0.20 | 1.43 – 2.24 |
| Health factors | |||
| Self-rated health | |||
| Excellent/very good/good vs. fair/poor | 1.13 | 0.03 | 1.07 – 1.19 |
| Body mass index (kg/m2) | |||
| Underweight/normal weight vs. overweight | 0.93 | 0.01 | 0.91 – 0.95 |
| Underweight/normal weight vs. obese | 0.94 | 0.01 | 0.92 – 0.96 |
| Smoking status | |||
| Never smoked vs. current smoker | 0.96 | 0.01 | 0.93 – 0.98 |
| Never smoked vs. former smoker | 0.92 | 0.01 | 0.89 – 0.95 |
| Serious psychological distress | |||
| Not psychologically distressed vs. psychologically distressed | 1.19 | 0.05 | 1.10 – 1.29 |
| Specific functional limitations | |||
| No functional limitations vs. one or more limitations | 1.24 | 0.05 | 1.15 – 1.34 |
| Health service use variables | |||
| Number of doctor visits | |||
| 2 -5 visits vs. ≤ 1 | 0.97 | 0.01 | 0.95 – 1.00 |
| ≥ 5 visits vs. ≤ 1 | 0.86 | 0.02 | 0.83 – 0.90 |
| Health insured | |||
| Covered vs. not covered | 1.04 | 0.01 | 1.02 – 1.07 |
|
Arthritis diagnosis x self-rated health
|
1.09 | 0.03 | 1.04 – 1.15 |
Notes: PR = Prevalence ratio; SE = Standard error; CI = 95% confidence interval
Table 5.
Prevalence ratio for stratified multivariable model examining the association between arthritis diagnosis and employment participation, by self-rated health.
| Employment Participation | ||||||||
|---|---|---|---|---|---|---|---|---|
| Higher self-rated health‡ | Lower self-rated health‡‡ | |||||||
| N | PR | 95% CI | SE | N | PR | 95% CI | SE | |
| Arthritis diagnosis vs. no arthritis diagnosis† | 32,848 | 1.09 | 1.03 – 1.14 | 0.03 | 1,726 | 0.86 | 0.69 – 1.07 | 0.09 |
Note: † = Model controlling for sex, ethnicity, marital status, education, social participation restriction, comorbidities, body mass index, smoking status, serious psychological distress, specific functional limitations, number of doctor visits, health insurance; ‡ = higher self-rated health includes those who reported that their health was excellent, very good or good ; ‡‡ = lower self-rated health included those reporting that their health was fair or poor.
DISCUSSION
This study was the first to conduct a population-level examination of participation in education and employment of young adults diagnosed with arthritis. While statistically similar, young adults with arthritis reported a slightly lower prevalence of educational involvement in relation to their non-arthritis counterparts. Interestingly, arthritis diagnosis was associated with an increased likelihood of participating in employment. Notably, the significant association between arthritis and a higher likelihood of employment was only reported among young adults also reporting higher self-rated health. Findings suggest the need to design interventions that address both self-rated health and the promotion of labor market participation. Results also provide a foundation for future research to understand the educational and employment experiences of young adults with arthritis.
Our study points to a potential association between arthritis diagnosis and a lower prevalence of participation in education. While statistically similar at the multivariate level, young adults with arthritis reported a slightly lower prevalence of educational participation. Findings may demonstrate a possible disadvantage young adults with arthritis possess when establishing their careers. US demographic data show that the number of young adults enrolled in post-secondary school has steadily increased over the last two decades to provide specialized competences and training required for higher paying and technical jobs (31). For young adults, an arthritis diagnosis may mean few opportunities to obtain educational resources to compete against peers for higher skilled jobs.
A stratified design enabled us to examine the association between arthritis and employment participation when compared by self-rated health. Stratified analysis also enabled us to account for potential suppression effects in the multivariate model. In contrast to our hypothesis, among young adults reporting higher self-rated health, an arthritis diagnosis was associated with significantly higher employment prevalence. Our research provides support for several previously conducted clinical studies of young adults, which also show that arthritis diagnosis can be associated with an increased likelihood of employment when compared with those without arthritis (6, 16, 17, 32). Findings also indicate the importance of self-rated health in young adulthood and its association with vocational outcomes. At the same time, recent research highlights the challenges associated with delivery of health services to young people with arthritis including health service gaps during the pediatric to adult care transition (33, 34), ongoing disease activity (18), poor mental health (12, 18, 19), and lower perceived social support (21, 33, 35). Greater insights are needed to determine the types of interventions that should be delivered to young adults with arthritis to improve both health and employment outcomes. To compliment the findings from the current study, there is also a need to understand at-work experiences of young adult diagnosed with arthritis and their ability to stay at-work. A recent study of young adults with arthritis found that close to 40% of those working reported absenteeism, job disruptions, or a perceived productivity loss (36). Among young adults with arthritis, it is not clear what role work productivity losses may play in sustaining employment over a longer period of time, highlighting a critical knowledge gap.
The association between arthritis and higher prevalence of employment participation could also indicate an early transition into the labor market among those with higher self-rated health. Young adults with arthritis may have more of a need for income and/or extended medical benefits associated with paid work to independently manage their condition and participate in other valued social roles (e.g., living alone, supporting a spouse or family) (7, 11). An early employment transition could mean that young adults with arthritis are forfeiting pre-vocational roles that are important to long-term career development such as graduate school, unpaid internships, or volunteering.
There are several study limitations to acknowledge. First, we utilize a self-reported question in the NHIS to identify young people with arthritis diagnosis. While this method does not substitute clinical diagnosis, it is considered a valid case-finding question for public health surveillance (37). Our data are cross-sectional, and causation cannot be ascertained. Additionally, we offer a snapshot of educational and employment experience in the past week, and may not capture unpredictable interruptions to education and employment that may be associated with episodic arthritis symptoms over a longer period (15). Study strengths include the use of a large and rich source of data for education, employment, and various covariates. The NHIS also enabled us to ascertain a large representative sample of young adults with arthritis and conduct a population-based study focused on the nation.
Young adulthood is a time of life where a person typically establishes their career. By utilizing population-level data of young adults with and without arthritis this study was the first to compare participation in educational and employment roles. While few significant differences existed, results did highlight higher employment prevalence among young adults with arthritis who also indicated higher self-rated health. Findings point to the importance of sustaining self-rated health in young adulthood to labor market involvement. Overall, our study provides important insights on early career experiences and contributes to a growing body of scholarship that aims at minimizing work disability across the life course for people living with arthritis.
SIGNIFICANCE AND INNOVATIONS.
At the population level, no studies have examined the association between arthritis and educational and employment participation during the young adult life phase.
Among young adults, arthritis was associated with greater employment participation, but only among those indicating greater self-rated health.
Findings point to the need to plan interventions that may focus on improving both self-rated health and promoting labor market participation of younger populations with arthritis.
Acknowledgment:
We would like to acknowledge Dr. Charles Helmick for his thoughtful comments on the manuscript.
Footnotes
Conflicts of Interest: None to report
Disclosure: Findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
REFERENCES
- 1.Allaire S, Wolfe F, Niu J, Lavalley M, Michaud K. Work disability and its economic effect on 55–64-year-old adults with rheumatoid arthritis. Arthritis Care & Research. 2005;53(4):603–8. [DOI] [PubMed] [Google Scholar]
- 2.Gignac MAM, Cao X, Lacaille D, Anis AH, Badley EM. Arthritis-related work transitions: A prospective analysis of reported productivity losses, work changes, and leaving the labor force. Arthritis Care & Research. 2008;59(12):1805–13. [DOI] [PubMed] [Google Scholar]
- 3.Badley EM, Wang PP. The contribution of arthritis and arthritis disability to nonparticipation in the labor force: a Canadian example. The Journal of Rheumatology. 2001;28(5):1077–82. [PubMed] [Google Scholar]
- 4.Moorthy LN, Peterson MG, Hassett A, Baratelli M, Lehman TJ. Impact of lupus on school attendance and performance. Lupus. 2010;19(5):620–7. Epub 2010 Jan 11. [DOI] [PubMed] [Google Scholar]
- 5.Centers for Disease Control and Prevention. Arthritis: Addressing the nation’s most common cause of disability at a glance 2015. 2015. [cited 2015 November 30th]; Available from: http://www.cdc.gov/chronicdisease/resources/publications/aag/arthritis.htm
- 6.Jetha A The impact of arthritis on the early employment experiences of young adults: A literature review. Disability and health journal. 2014. [DOI] [PubMed] [Google Scholar]
- 7.Arnett JJ. Emerging adulthood: A theory of development from the late teens through the twenties. American psychologist. 2000;55(5):469. [PubMed] [Google Scholar]
- 8.Marshall K Youth neither enrolled nor employed. Ottawa, ON: Statistics Canada; 2012. [Google Scholar]
- 9.Scarpetta S, Sonnet A, Manfredi T. Rising youth unemployment during the crisis. 2010. [Google Scholar]
- 10.Bradley H, Devadason R. Fractured transitions: Young adults' pathways into contemporary labour markets. Sociology. 2008;42(1):119–36. [Google Scholar]
- 11.Arnett JJ. Emerging adulthood: The winding road from late teens through the twenties. New York, NY.: Oxford University Press; 2004. [Google Scholar]
- 12.Packham JC, Hall MA. Long-term follow-up of 246 adults with juvenile idiopathic arthritis: education and employment. Rheumatology (Oxford). 2002;41(12):1436–9. [DOI] [PubMed] [Google Scholar]
- 13.Arthritis Yelin E.. The cumulative impact of a common chronic condition. Arthritis and rheumatism. 1992;35(5):489–97. [PubMed] [Google Scholar]
- 14.Verstappen S, Bijlsma J, Verkleij H, Buskens E, Blaauw A, Ter Borg E, et al. Overview of work disability in rheumatoid arthritis patients as observed in cross-sectional and longitudinal surveys. Arthritis Care & Research. 2004;51(3):488–97. [DOI] [PubMed] [Google Scholar]
- 15.Gignac MAM, Badley EM, Lacaille D, Cott CC, Adam P, Anis AH. Managing arthritis and employment: Making arthritis-related work changes as a means of adaptation. Arthritis Care & Research. 2004;51(6):909–16. [DOI] [PubMed] [Google Scholar]
- 16.Gerhardt CA, McGoron KD, Vannatta K, McNamara KA, Taylor J, Passo M, et al. Educational and occupational outcomes among young adults with juvenile idiopathic arthritis. Arthritis Care & Research. 2008;59(10):1385–91. [DOI] [PubMed] [Google Scholar]
- 17.Arkela-Kautiainen M, Haapasaari J, Kautiainen H, Vilkkumaa I, Mälkiä E, Leirisalo-Repo M. Favourable social functioning and health related quality of life of patients with JIA in early adulthood. Annals of the rheumatic diseases. 2005;64(6):875–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Oen KG, Cheang M. Epidemiology of chronic arthritis in childhood. Seminars in Arthritis and Rheumatism. 1996;26(3):575–91. [DOI] [PubMed] [Google Scholar]
- 19.Foster HE, Marshall N, Myers A, Dunkley P, Griffiths ID. Outcome in adults with juvenile idiopathic arthritis: a quality of life study. Arthritis Rheum. 2003;48(3):767–75. [DOI] [PubMed] [Google Scholar]
- 20.Malviya A, Rushton SP, Foster HE, Ferris CM, Muthumayandi K, Deehan DJ. Exploring the relationships between adult juvenile idiopathic arthritis and employment. Arthritis & Rheumatism. 2012:n/a-n/a. [DOI] [PubMed] [Google Scholar]
- 21.Jetha A, Badley E, Beaton D, Fortin PR, Shiff NJ, Rosenberg AM, et al. Transitioning to employment with a rheumatic disease: the role of independence, overprotection, and social support. The Journal of rheumatology. 2014;41(12):2386–94. [DOI] [PubMed] [Google Scholar]
- 22.Perruccio AV, Badley EM, Hogg-Johnson S, Davis AM. Characterizing self-rated health during a period of changing health status. Social science & medicine. 2010;71(9):1636–43. [DOI] [PubMed] [Google Scholar]
- 23.Perruccio AV, Power JD, Badley EM. Arthritis onset and worsening self-rated health: A longitudinal evaluation of the role of pain and activity limitations. Arthritis Care & Research. 2005;53(4):571–7. [DOI] [PubMed] [Google Scholar]
- 24.Hootman JM, Cheng WY. Psychological distress and fair/poor health among adults with arthritis: state-specific prevalence and correlates of general health status, United States, 2007. International journal of public health. 2009;54(1):75–83. [DOI] [PubMed] [Google Scholar]
- 25.Centers for Disease Control and Prevention. National Health Interview Survey. 2015. [cited 2015 November 30]; Available from: http://www.cdc.gov/nchs/nhis.htm
- 26.Theis K, Murphy L, Hootman J, Wilkie R. Social Participation Restriction Among US Adults With Arthritis: A Population-Based Study Using the International Classification of Functioning, Disability and Health. Arthritis care & research. 2013;65(7):1059–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kaplan M, McFarland B, Newsom J, Huguet N. Spending more, feeling worse: medical care expenditures and self rated health. Journal of epidemiology and community health. 2004;58(6):529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kessler RC, Barker PR, Colpe LJ, Epstein JF, Gfroerer JC, Hiripi E, et al. Screening for serious mental illness in the general population. Archives of general psychiatry. 2003;60(2):184–9. [DOI] [PubMed] [Google Scholar]
- 29.SAS Institute Inc. SAS 9.3. Version 9.3 ed. Cary, NC: SAS Institute Inc. ; 2015. [Google Scholar]
- 30.RTI International. SUDAAN. 2015. [cited 2015 November 30]; Available from: http://www.rti.org/sudaan/
- 31.Arnett JJ. Suffering, selfish, slackers? Myths and reality about emerging adults. Journal of Youth and Adolescence. 2007;36(1):23–9. [Google Scholar]
- 32.Yelin EH. Musculoskeletal conditions and employment. Arthritis & Rheumatism. 1995;8(4):311–7. [DOI] [PubMed] [Google Scholar]
- 33.McDonagh JE. Transition of care from paediatric to adult rheumatology. Archives of disease in childhood. 2007;92(9):802–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tucker LB, Cabral DA. Transition of the adolescent patient with rheumatic disease: issues to consider. Rheumatic Disease Clinics of North America. 2007;33(3):661–72. [DOI] [PubMed] [Google Scholar]
- 35.Packham J, Hall M. Long-term follow-up of 246 adults with juvenile idiopathic arthritis: social function, relationships and sexual activity. Rheumatology. 2002;41(12):1440–3. [DOI] [PubMed] [Google Scholar]
- 36.Jetha A, Badley E, Beaton D, Fortin PR, Shiff NJ, Gignac MA. Unpacking early work experiences of young adults with rheumatic disease: An examination of absenteeism, job disruptions and productivity loss. Arthritis care & research. 2015. [DOI] [PubMed] [Google Scholar]
- 37.Sacks JJ, Harrold LR, Helmick CG, Gurwitz JH, Emani S, Yood RA. Validation of a surveillance case definition for arthritis. The Journal of rheumatology. 2005;32(2):340–7. [PubMed] [Google Scholar]