“Should I tell my employer and coworkers I have arthritis?” A longitudinal examination of self-disclosure in the work place

Authors

  • Monique A.M. Gignac,

    Corresponding author
    1. Arthritis Community Research & Evaluation Unit, and Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
    • Arthritis Community Research & Evaluation Unit, Toronto Western Research Institute, Main Pavilion 10-316, 399 Bathurst Street, Toronto, Ontario, M5T 2S8 Canada
    Search for more papers by this author
  • Xingshan Cao

    1. Arthritis Community Research and Evaluation Unit, University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author

Abstract

Objective

To examine arthritis self-disclosure at work, factors associated with disclosure, and prospective relationships of self-disclosure and work place support with changes to work place interactions, work transitions, and work place stress.

Methods

Using a structured questionnaire, participants with osteoarthritis or inflammatory arthritis were interviewed at 4 time points, 18 months apart. At time 1, all participants (n = 490; 381 women, 109 men) were employed. Of the entire sample, 71% were retained throughout the study. Respondents were recruited using community advertising and from rheumatology and rehabilitation clinics. Self-disclosure and perceived support from managers and coworkers was assessed, as well as demographic, illness, work-context, and psychological variables. Generalized estimating equations modeled associations of disclosure and support on changes at work (e.g., job disruptions, work place stress).

Results

At each time point, 70.6–76.6% of participants had self-disclosed arthritis to their manager and 85.2–88.1% had told a coworker. Intraindividual variability in disclosure was considerable. Factors associated with self-disclosure were often inconsistent over time, with the exception of variables assessing the need to self-disclose (e.g., activity limitations) and perceived coworker support. Self-disclosure was not associated with changes to work. However, coworker support was related to fewer job disruptions, help with work tasks, and being less likely to reduce hours. Perceived managerial support was associated with less work place stress.

Conclusion

Greater awareness is needed about issues related to self-disclosing arthritis at work. This study emphasizes the importance of a supportive work place, especially supportive coworkers, in decisions to discuss arthritis at work and in changes to work that might enable people to remain employed.

INTRODUCTION

A consistent body of research highlights the negative impact of arthritis on employment, with many individuals reporting difficulties with work activities or even having to give up their jobs (1–11). As a result, there has been increased attention to the need for work place–focused interventions that would enable individuals with arthritis to remain employed (12–18). However, the ultimate success of these interventions may relate, in part, to whether individuals are willing to tell employers and coworkers about their disease. To date, there have been few studies of self-disclosure of arthritis in the work place (19–21). The research that does exist, across a range of health conditions, has shown that individuals are often reluctant to tell others about their health problems and that they do not always self-disclose (19, 22–31). In the current research, we examined whether individuals had told managers and coworkers about their disease, and examined factors associated with self-disclosure. We also prospectively examined the association of self-disclosure and perceived work place support with subsequent changes to work, including work place interactions, work transitions related to productivity and giving up employment, and perceived arthritis-related work stress.

Self-disclosure of personal information involves some risk. Individuals with chronic disabling conditions may report concerns related to experiencing stigma (i.e., negative or damaging attitudes) that can be manifested in the form of lost credibility or status with employers, lost support or opportunities (e.g., for promotion), and even loss of one's job (19, 30). However, benefits of self-disclosure may include support from others and a reaffirmation of one's work identity (32–34). Self-disclosure may also be necessary in order to access instrumental support such as work place accommodations, benefits, and treatment, or to receive legal protection (19, 29, 35, 36).

Research on self-disclosure at work has been mainly cross-sectional or involving small samples in qualitative studies. It has focused primarily on individuals with health conditions that are sometimes viewed as stigmatizing, such as epilepsy, mental health, and human immunodeficiency virus/acquired immunodeficiency syndrome, and has emphasized the reactions of employers more than coworkers. Moreover, factors associated with self-disclosure have not always been consistent. For example, some studies have found that younger individuals and women are more likely to report disclosing health issues at work, but other studies have not (26, 31). Other research has found that those working in public sector jobs are more likely to disclose, whereas those in academia are the least likely to tell others (19).

Perceived need is consistently reported as a significant factor related to self-disclosure. Specifically, individuals who believe that their health problems might affect their job performance, who require accommodations or health management at work (e.g., diabetes mellitus testing), or who need to frequently attend medical appointments are more likely to disclose health conditions (19, 26–29, 37). Also consistently associated with disclosing is the perception of colleagues and organizational policies as supportive (19, 20, 26, 28, 29, 22, 38).

Studies of the potential work place consequences of self-disclosure remain rare. What research exists has found that disclosing is associated with practical support, such as a reduced workload, a decreased work pace, and changes to the work environment (39, 40). A cross-sectional analysis of nearly 500 individuals with osteoarthritis (OA) or inflammatory arthritis found that, compared with individuals who had not discussed arthritis with their employer, those who had self-disclosed were more likely to report help from colleagues with job tasks and to make work place modifications (34). However, individuals self-disclosing also reported more positive work place perceptions. It may be that a positive work environment, not disclosure, was responsible for changes at work. Prospective research is needed to disentangle the roles of self-disclosure and perceived support in changes to work.

This study longitudinally examined self-disclosure of arthritis to coworkers and supervisors. The research asked to what extent individuals self-disclosed their condition, and how consistent or variable this decision and the factors associated with it were. Unlike studies examining potentially stigmatizing health conditions, we expected individuals with arthritis to be relatively forthcoming about their disease. However, arthritis symptoms can fluctuate, and research has found that many people with arthritis make a variety of work transitions over the course of their work life (e.g., change jobs) (11, 22). This could result in variability in the decision to self-disclose to others, as well as in the factors associated with disclosure. In general, however, self-disclosure is hypothesized to be more likely among those with greater need (e.g., pain, fatigue, work place activity limitations). In keeping with previous studies, psychosocial factors such as perceived coworker and managerial support were also expected to be related to greater self-disclosure.

We also asked whether self-disclosing arthritis matters. Specifically, about what work place outcomes, if any, are associated with disclosure. We focused on arthritis disclosure and perceived work place support separately to disentangle their associations with subsequent changes at work. Work outcomes include changes to work place interactions (e.g., fewer job disruptions, more help from coworkers); fewer work transitions like absenteeism, reduced work hours, changing jobs, or leaving the labor force; and decreased reports of work place stress.

SUBJECTS AND METHODS

Subjects.

Individuals with OA or inflammatory arthritis residing in southwestern Ontario, Canada were interviewed 4 times (T1–T4), each 18 months apart. The sample was purposive to ensure diversity across occupations and that individuals receiving fewer health care services were not systematically excluded. Respondents were recruited largely from community advertisements, with additional respondents from rheumatology and rehabilitation clinics and The Arthritis Society, Ontario Division. Eligibility criteria were a reported physician diagnosis of OA or inflammatory arthritis, arthritis duration of ≥1 year, paid employment at baseline (T1), no comorbid conditions causing physical disability (e.g., multiple sclerosis), and fluency in English.

Procedure.

An in-depth, structured questionnaire lasting ∼2 hours was administered at the participant's home or a location of their choice. Ethical approval was received and informed written consent was obtained at each interview.

Measures.

Demographics, arthritis type and duration, and joints affected.

Data on age and sex were collected. Respondents provided the type(s) of arthritis diagnosed by their physician and the time since their diagnosis. Arthritis type was coded in 3 categories: inflammatory arthritis (e.g., rheumatoid arthritis, psoriatic arthritis), OA, or both. A homunculus displaying major joints affected by arthritis was used to calculate the total number of joints affected (41).

Symptom severity.

Pain and fatigue were measured over the preceding month. Work place research suggested that the chronicity of severe pain was more relevant to employment than pain severity alone without reference to its frequency (11, 42, 43). Pain was assessed with the question “How often did you have severe pain from your arthritis?” and fatigue with the question “How often have you felt fatigue as a result of your arthritis?” Responses were on a 5-point Likert-type scale (where 1 = no days and 5 = all days).

Employment status, years with current employer, and occupation.

Participants were asked if they were employed full or part time and the number of years they had worked for their current employer. Occupation was classified using the Human Resources Development Canada National Occupation Classification Matrix 2001 (44). Occupations were collapsed into 4 groups: 1) business, finance, and administration; 2) health, science, and arts; 3) sales and service; and 4) trades, transportation, and equipment operation.

Work place activity limitations.

The 11-item Work Place Activity Limitations Scale (WALS) was used to ask about arthritis-related employment activity limitations (22, 34). Items include getting to, from, and around the work place; sitting/standing for long periods; lifting; reaching; and the schedule and pace of the work. Responses were on a 4-point scale (where 0 = no difficulty and 3 = not able to do). Participants indicating that an activity was not applicable to their job were assigned a score of 0 (no difficulty) for that activity. Responses were summed, with total scores ranging from 0–33. The internal consistency of the WALS using Cronbach's alpha was 0.78 at T1 and 0.81 at each of T2, T3, and T4.

Work place help from others.

For each item of the WALS, participants were asked whether they had received help from others at work to manage difficulties with work tasks. Scores across the 11 items were summed from 0–11.

Work transitions and interactions: absenteeism, job disruptions, reduced work hours, changed jobs, left the labor force.

Respondents were asked if they had been absent from work, related to arthritis, in the previous 6 months (yes/no). Job disruptions were assessed with 10 items. Examples include work time lost because of arriving late/leaving early; being unable to attend meetings, take on extra responsibilities, pursue job promotions, or work the shift/schedule desired; and work interruptions of ≥20 minutes (11). Respondents indicated whether job disruptions related to arthritis occurred in the previous 6 months (yes/no); scores were summed from 0–10. Reduced work hours were assessed with the question “Because of your arthritis, have you changed the number of hours you work in an average week since the last interview?” (yes/no). Respondents were also asked if they had changed jobs (“Since your last interview, have you changed your occupation/job as a result of your arthritis?” [yes/no]). Leaving the labor force was assessed with a question on employment status. Categories for employment status were full time, part time, leave of absence/short-term disability, long-term disability, unemployed (i.e., looking for work), not employed (i.e., not looking for work), retired, homemaker, and other. Respondents working full or part time were considered employed. Other changes in employment status were categorized as left the labor force.

Work stress.

The 15-item Chronic Illness Job Strain Scale was included at T3 and T4, and measured perceptions of stress related to working with arthritis (42). Responses were on a 5-point scale (where 1 = not at all stressful and 5 = extremely stressful). Scores were summed and ranged from 15–75. Examples include, “To what extent is uncertainty about how you will feel at work from day to day stressful?” and “To what extent does the impact of arthritis on your ability to use your skills and training make employment stressful (i.e., not being able to use your training)?” Cronbach's alpha for the scale items was 0.95 at both T3 and T4.

Pain catastrophizing.

A 13-item scale measured perceptions of pain, including pain rumination, magnification, and helplessness (45, 46). Responses were on a 5-point scale (where 0 = not at all and 4 = all the time). Cronbach's coefficient alpha for the scale at T1, T2, T3, and T4 was 0.94, 0.93, 0.95, and 0.94, respectively.

Perceived managerial and coworker support.

Drawing on previous job support research (47, 48), 4 items measured perceived managerial support and 8 items measured perceived coworker support. Items were on a 5-point scale (where 1 = strongly disagree and 5 = strongly agree). Sample items include, “My manager does things to help employees balance their work and their personal life” and “I have coworkers who would support me through tough personal times.” Cronbach's coefficient alpha for the managerial support scale was 0.87 at T1 and 0.86 at T2, T3, and T4. Cronbach's coefficient alpha for the coworker support scale at T1, T2, T3, and T4 was 0.87, 0.89, 0.90, and 0.91, respectively.

Arthritis self-disclosure.

Respondents were asked, “Does your employer/supervisor know about your arthritis?” (yes/no). They were also asked, “How many, if any, of your coworkers know about your arthritis?” Responses were coded as 1 = none, 2 = a few, 3 = approximately half, 4 = most, and 5 = all. To facilitate analyses, we combined categories 2 and 3 (labeled “less than half”) and 4 and 5 (labeled “more than half”).

Statistical analysis.

We used frequencies, means, and SDs to assess sample characteristics and self-disclosure. Self-employed respondents were excluded from analyses examining managerial self-disclosure but were retained in analyses of coworker disclosure if they indicated the presence of these individuals. Analyses of variance and chi-square tests examined bivariate associations of coworker and managerial self-disclosure with demographic, health, and work place factors. These associations were examined at each time point to test whether they were consistent or variable over time.

To prospectively examine whether self-disclosure and perceived support were associated with subsequent changes in the work place, the data were structured into a person-wave format in which each participant had as many records as time points in which he or she participated. Because these analyses focused on potential consequences of disclosure and support at a later time point, they included only participants with ≥2 time points of data. A fixed-effect longitudinal model was specified. The model used generalized estimating equations (GEEs), a method of parameter estimation for repeated observations that is useful when dependent measures are discrete, and that adjusts for the autocorrelations within individuals across time points (49, 50). Specifically, a fixed-effect Poisson model with GEEs was specified to test whether self-disclosure and perceived support were associated with fewer job disruptions and more help from others. A binomial distribution was assumed for a fixed-effect GEE model on work transitions (e.g., reducing work hours), and a normal distribution was assumed for a fixed-effect GEE model on work stress. The relationships between self-disclosure and perceived support (independent variables) with subsequent work place changes (dependent variables) were examined, controlling for age, sex, arthritis type, disease duration, joints affected, pain, fatigue, work place activity limitations, occupation, full-/part-time work, years with employer, and employment changes measured in an earlier wave. Except for sex, control variables could vary across time.

RESULTS

At time 1, participants were, on average, 51 years old, and three-quarters of them (77.8%) were women (Table 1). Over half of the participants (55.8%) had OA, 34.7% reported inflammatory arthritis, and 9.5% reported both OA and inflammatory arthritis. The time since diagnosis varied widely (mean 9.3 years). One-third (33.2%) of the sample reported severe pain and half reported fatigue most or all days. Of the participants, 82% were employed full time, with 70% working in business/administration or health/science/teaching. Respondents reported a range of work place activity limitations and stress and pain catastrophizing perceptions, with the average respondent reporting some difficulty with work place activities, mild to moderate stress, and a low frequency of pain catastrophizing cognitions. Most participants reported agreeing with items assessing coworker support, with slightly less agreement for items measuring managerial support. Of the participants, 71% were retained in the sample over 4.5 years: n = 490 at T1, and n = 349 at T4. Demographic and occupational characteristics remained similar over time and have been reported in detail previously (11).

Table 1. Sample characteristics of participants at time 1*
 Value
  • *

    Values are the number (percentage) unless otherwise indicated. WALS = Workplace Activity Limitations Scale; CIJSS = Chronic Illness Job Strain Scale; PCS = Pain Catastrophizing Scale.

  • Based on time 3 data, and administered at time 3 and time 4 only.

Age, mean ± SD years51.1 ± 9.3
Sex 
 Male101 (22.2)
 Female354 (77.8)
Diagnosis 
 Inflammatory arthritis158 (34.7)
 Osteoarthritis254 (55.8)
 Both43 (9.5)
Duration of arthritis, mean ± SD years9.3 ± 8.8
No. of joints affected, mean ± SD8.4 ± 4.7
Pain 
 1 (no days)84 (18.6)
 2 (a few days)108 (23.9)
 3 (some days)110 (24.3)
 4 (most days)99 (21.9)
 5 (all days)51 (11.3)
Fatigue 
 1 (no days)70 (15.5)
 2 (a few days)70 (15.5)
 3 (some days)87 (19.3)
 4 (most days)112 (24.8)
 5 (all days)113 (25.0)
Employment status 
 Full time373 (82.0)
 Part time82 (18.0)
Occupation 
 Administration, management159 (35.0)
 Health, science159 (35.0)
 Sales, services86 (18.9)
 Trade, transportation50 (11.0)
Time with current employer, mean ± SD years10.5 ± 9.1
WALS, mean ± SD score (0–33)6.4 ± 4.4
CIJSS, mean ± SD score (15–75)34.4 ± 14.7
PCS, mean ± SD score (0–52)16.7 ± 12.7
Perceived support, mean ± SD 
 Coworker (8–40)30.8 ± 5.8
 Managerial (4–20)14.1 ± 4.1

At each time, approximately three-quarters of participants had disclosed their arthritis to their supervisor (70.6–76.6%) (Table 2). Approximately half of participants reported telling more than half of their coworkers about their arthritis (49–54.9%). Only 9.6–13.6% of respondents did not tell anyone at work about their condition. Differences in the number of participants at each time point made it difficult to examine consistency in self-disclosure. However, the data suggest considerable intraindividual changes in disclosure. Specifically, only ∼15.7% of participants who at T1 reported that they had not disclosed arthritis to their manager consistently reported not disclosing at other times. The remainder of the participants reported self-disclosing their arthritis at a later time point or were lost to followup. Among those telling their supervisor about their arthritis at T1, approximately one-fifth (18.7%) later reported not self-disclosing.

Table 2. Participants who self-disclosed their arthritis to employers and coworkers at time 1 to time 4*
 Time 1Time 2Time 3Time 4
  • *

    Self-employed respondents were excluded from analyses examining managerial self-disclosure but were retained in analyses of coworker disclosure if they indicated the presence of these individuals.

  • Out of 429 eligible participants at T1, 318 at T2, 265 at T3, and 219 at T4.

Self-disclosure to employer, no.440325269224
 No, %24.623.429.429.0
 Yes, %75.576.670.671.0
Self-disclosure to coworkers, no.455335276235
 None, %11.913.414.913.6
 Less than half, %39.131.635.937.0
 More than half, %49.054.949.349.4
No self-disclosure to either employer or coworkers, no. (%)41 (9.6)37 (11.6)36 (13.6)27 (12.3)

Examination of demographic, health, work, and psychosocial factors associated with self-disclosing arthritis to managers also revealed little consistency over time, with some exceptions (Table 3). In keeping with our hypotheses, greater perceived support from coworkers was associated with disclosing arthritis to one's manager at T1–T3. Need also related to managerial self-disclosure, with greater work place activity limitations and disease duration significantly associated with or approaching significance with disclosure at all time points, and more joints affected significantly associated with or approaching significance at T1, T2, and T4. Pain catastrophizing (e.g., pain rumination, magnification) was also significantly associated with disclosing or approached significance at T1, T2, and T4.

Table 3. Bivariate analyses of health and work place variables associated with self-disclosure of arthritis to managers and coworkers at time 1 through time 4*
 Self-disclosure of arthritis to managerSelf-disclosure of arthritis to coworkers
NoYesNo coworkersLess than halfMore than half
  • *

    Values are the mean ± SD unless otherwise indicated. Variables examined: age, sex, diagnosis, arthritis duration, pain, fatigue, joints affected, occupation, years with current employer, work place activity limitations, pain catastrophizing, perceived coworker support, and perceived managerial support. OA = osteoarthritis; RA = rheumatoid arthritis; WALS = Workplace Activity Limitations Scale; PCS = Pain Catastrophizing Scale.

  • Significant (P < 0.05). (Other values are trends [P < 0.10]. Nonsignificant findings [P > 0.10] are omitted to facilitate interpretation of the results.)

Time 1
 Diagnosis, %     
  OA only30.669.414.642.542.9
  RA or both RA/OA16.983.18.534.856.7
 Arthritis duration, years7.1 ± 7.19.8 ± 9.36.6 ± 6.28.3 ± 8.210.6 ± 9.6
 Joints affected6.7 ± 3.88.9 ± 4.96.9 ± 3.87.8 ± 4.59.0 ± 4.9
 Fatigue2.9 ± 1.43.4 ± 1.42.9 ± 1.43.1 ± 1.43.5 ± 1.4
 WALS5.1 ± 3.97.0 ± 4.55.4 ± 4.06.1 ± 4.37.1 ± 4.4
 PCS14.9 ± 12.118.0 ± 12.9
 Perceived managerial support13.0 ± 3.914.4 ± 4.113.4 ± 4.213.5 ± 4.014.6 ± 4.2
 Perceived coworker support28.7 ± 5.231.4 ± 5.828.5 ± 5.629.6 ± 5.532.3 ± 5.6
 Occupation, %     
  Administration19.081.012.034.653.5
  Health, science29.270.88.247.844.0
  Sales, services20.279.814.032.653.5
  Trade, transportation33.366.718.038.044.0
Time 2     
 Arthritis duration, years9.0 ± 8.411.1 ± 9.08.7 ± 7.49.6 ± 7.511.9 ± 9.6
 Joints affected6.1 ± 3.97.6 ± 4.55.6 ± 3.66.8 ± 4.27.9 ± 4.6
 WALS5.4 ± 4.56.5 ± 4.35.3 ± 4.05.7 ± 3.87.0 ± 4.9
 PCS12.8 ± 11.716.6 ± 12.213.4 ± 12.613.3 ± 10.817.4 ± 12.4
 Perceived coworker support29.9 ± 5.831.9 ± 5.727.6 ± 6.131.2 ± 5.232.7 ± 5.6
Time 3     
 Arthritis duration, years9.9 ± 7.513.2 ± 9.39.0 ± 7.011.1 ± 7.814.4 ± 9.8
 Joints affected7.2 ± 4.56.7 ± 4.18.5 ± 5.2
 Fatigue2.6 ± 1.43.0 ± 1.32.9 ± 1.42.6 ± 1.33.1 ± 1.3
 WALS5.7 ± 4.26.7 ± 4.45.3 ± 3.85.6 ± 4.07.5 ± 4.6
 Occupation, %     
  Administration13.731.654.7
  Health, science9.445.844.9
  Sales, services26.932.740.4
  Trade, transportation18.213.668.2
 Perceived coworker support28.6 ± 5.532.6 ± 4.828.4 ± 6.230.9 ± 5.132.8 ± 4.8
Time 4     
 Arthritis duration, years10.5 ± 5.715.3 ± 9.911.6 ± 7.112.1 ± 7.516.6 ± 10.2
 Joints affected6.8 ± 4.48.1 ± 4.9   
 Fatigue2.6 ± 1.22.9 ± 1.32.7 ± 1.52.7 ± 1.33.1 ± 1.3
 WALS5.9 ± 4.17.9 ± 4.85.8 ± 4.76.5 ± 4.68.5 ± 4.5
 PCS10.3 ± 10.213.0 ± 11.2   
 Occupation, %     
  Administration22.277.8
  Health, science28.471.6
  Sales, services46.353.7
  Trade, transportation23.876.2
 Perceived managerial support14.0 ± 3.514.9 ± 3.8   
 Perceived coworker support  30.0 ± 4.332.2 ± 5.332.5 ± 5.6

There was greater consistency in factors associated with self-disclosing arthritis to coworkers. Need factors such as work place activity limitations and disease duration were significantly related to disclosure at all time points. Having more affected joints and greater fatigue was significantly associated with or approached significance with self-disclosure at 3 of the 4 time points. Perceived coworker support was also important and was significantly related to or approached significance at all time points. Type of arthritis and occupation were inconsistently associated with disclosure (significant at 1 time point only). Age and sex were not significantly related to arthritis disclosure.

We examined whether arthritis self-disclosure or perceived work place support were related to changes at work by testing the longitudinal association of each of these variables with subsequent alterations to work place interactions, work transitions, and work place stress (Table 4). Analyses controlled for age, sex, arthritis type, disease duration, number of joints affected, pain, fatigue, work place activity limitations, occupation, job status, and years with employer. Arthritis self-disclosure to managers and coworkers was not associated with subsequent changes at work across the different time points. However, perceptions of support were related to work place changes. Controlling for other variables, individuals who perceived their managers as more supportive reported less work place stress related to their arthritis (P < 0.02). Perceptions of greater coworker support were associated with subsequent changes to the work environment in the form of fewer job disruptions (P < 0.03), more work place help from others (P < 0.01), and fewer reports of reducing work hours (P < 0.02).

Table 4. Longitudinal analyses using generalized estimating equations (GEEs) examining the association of arthritis self-disclosure and perceived support with subsequent changes to work place interactions, work transitions, and work place stress*
Dependent variablesPrevious disclosure to employerPrevious disclosure to coworkersPrevious perceived managerial supportPrevious perceived coworker support
More than halfLess than half
  • *

    Values are the b (95% sample regression coefficients). Analyses controlled for age, sex, arthritis type, disease duration, joints affected, pain, fatigue, Workplace Activity Limitations Scale score, occupation, job status, and years with employer.

  • Also controlled for managerial and coworker support.

  • Also controlled for arthritis self-disclosure to manager and coworkers.

  • §

    Poisson distribution was assumed.

  • Significant (P < 0.05).

  • #

    Binary distribution was assumed.

  • **

    Normal distribution was assumed.

Work place interactions§     
 Job disruption0.11 (−0.29, 0.50)0.09 (−0.39, 0.57)−0.06 (−0.49, 0.37)0.00 (−0.03, 0.03)−0.02 (−0.05, 0.00)
 Help from others0.16 (−0.19, 0.51)−0.28 (−0.73, 0.18)−0.27 (−0.72, 0.19)0.01 (−0.02, 0.03)0.02 (0.00, 0.04)
Work transitions#     
 Absenteeism0.08 (−0.48, 0.64)0.12 (−0.62, 0.87)0.19 (−0.48, 0.86)−0.01 (−0.06, 0.04)0.01 (−0.03, 0.05)
 Reduced work hours−0.2 (−0.86, 0.45)0.17 (−0.73, 1.08)0.21 (−0.59, 1.01)0.04 (−0.03, 0.11)−0.05 (−0.08, −0.01)
 Changed jobs−0.28 (−1.08, 0.52)0.34 (−0.69, 1.36)−0.36 (−1.25, 0.53)−0.01 (−0.09, 0.08)−0.02 (−0.08, 0.03)
 Left labor force0.04 (−0.67, 0.74)0.31 (−0.72, 1.35)0.57 (−0.37, 1.51)0.00 (−0.06, 0.06)−0.01 (−0.06, 0.03)
Work place stress**0.95 (−2.43, 4.32)−2.05 (−6.72, 2.61)−0.69 (−4.69, 3.32)−0.34 (−0.63, −0.04)0.08 (−0.16, 0.33)

DISCUSSION

Research examining self-disclosure of arthritis at work is rare. However, if arthritis interventions to reduce work disability are to be successful, greater awareness is needed about the issues and sensitivities related to discussing arthritis. This is particularly important given research findings that people with chronic illnesses sometimes experience stigma or discriminatory work practices resulting from self-disclosure of health problems. This study found that, across 4.5 years, many individuals had told their managers and coworkers about their arthritis, unlike findings from studies of individuals with more stigmatizing health conditions.

A greater percentage of participants had discussed arthritis with coworkers than with managers. At the same time, there was considerable variability in self-disclosure to employers. Among those who, at the outset of the study, reported they had not told their manager about their disease, most later reported discussing arthritis. Approximately one-fifth of participants who initially reported disclosing to their manager later reported not disclosing. Although it is not possible to “un-tell” someone about one's health, variability in disclosing is likely due to changes in managers over time and is consistent with research indicating that people with arthritis make numerous work transitions during their career, including changing jobs (11). Variability of self-disclosure within an individual highlights the need for additional research to understand when and to whom individuals self-disclose and the factors related to the timing of telling others. It also underscores that arthritis self-disclosure is not a single event in the work lives of people with arthritis, but reflects a series of decisions that individuals make throughout the course of their disease. Currently, we know little about this decision-making process.

Variability in self-disclosure also underscores the complexity surrounding the reasons why people disclose. Previous research found that perceived need and a supportive environment where individuals feel that it is safe to disclose were critical factors. This study also found evidence for the important role of need and support. Having more joints affected, fatigue, and greater activity limitations at work were relatively consistently associated with disclosure. However, there was inconsistency in the role of other health and work place factors, including disease and job type. For example, on one occasion those with OA were less likely to self-disclose than those with inflammatory arthritis. This may reflect differences not only in disease severity but also in treatment or the perceived seriousness of the disease. Research finds that those with OA often normalize their condition, seeing it as a part of aging and believing that little can be done to manage their symptoms (43). As such, these individuals may feel there is little to be gained by self-disclosing arthritis at work. Individuals in sales and service jobs often work variable hours and have less access to benefits. A more precarious job environment with little in the way of benefits may influence their decision to not self-disclose. These factors need to be explored in future studies that ask participants about their reasons for and against disclosing arthritis to others.

Interestingly, it was perceived coworker support, not supervisor support, which was more likely to be related to self-disclosing, regardless of whether managers or coworkers were examined. The reasons for this are not clear. However, coworkers are integral to the day-to-day work lives of individuals and their cooperation may be needed to implement many of the work place changes that would help people with arthritis manage their disease at work. It may be that individuals with arthritis see little point in disclosing to managers without the support of their coworkers. It also may be that supportive coworkers encourage individuals with arthritis to make their needs known to managers. The importance of coworker relationships is emphasized in conceptual models of work stress and in research finding that coworker problems are among those that can make work the most stressful (51, 52). The findings of this research emphasize the need for additional understanding of the role of coworkers in self-disclosure. These individuals may be integral to the success of arthritis work place interventions.

This study is one of the few examining longitudinal associations of self-disclosure on work-related outcomes. The findings reveal that self-disclosure to managers or coworkers, in and of itself, made little difference to the work lives of people with arthritis. As noted earlier, this underscores the complexity of self-disclosure and that it is not adequately captured as a single event (disclosed: yes or no). More important to work place outcomes was whether the work environment was supportive. As mentioned earlier, perceived coworker support was particularly important in understanding subsequent changes to work that were made to manage arthritis and that may ultimately help people with arthritis remain employed. Coworker support was associated with fewer job disruptions such as taking extended breaks, being unable to work the schedule desired, or being unable to take on extra work. It was associated with greater assistance at work with job tasks. Participants reporting greater coworker support also were less likely to subsequently reduce their work hours. However, coworker support was not associated with reduced absenteeism, changing jobs, or giving up employment. It is likely that coworker support enables individuals to alter or redistribute specific work tasks. Other factors such as disease severity, work place policies, and job demands likely play an important role in influencing absenteeism, changing jobs, and giving up employment.

Managerial support was not related to changes in outcomes, except reduced stress, which also may benefit people with arthritis in remaining employed. Additional research is needed to replicate these findings and to expand on the work outcomes studied to include perceived stigma and even potentially biased work place practices. As noted earlier, there was considerable intraindividual and interindividual variability in self-disclosure to managers. Reasons underlying this variability need to be captured in future studies.

Several limitations to this research have already been discussed, including the importance of illuminating decisions around the reasons for and timing of arthritis self-disclosure. Research also needs to replicate our findings and examine the short- and long-term consequences of disclosure. To do this, researchers should examine diverse samples of people with different types of arthritis and collect more detailed information about their diagnosis and the contexts within which they work. This includes examining health treatment regimens that might interfere with work scheduling, as well as work place benefits and policies that might encourage individuals to self-disclose in order to make accommodations at work.

Despite these limitations, this study highlights the variability and complexity in the decisions of individuals with arthritis to tell others at work about their disease. The study findings emphasize the importance of a supportive work place, especially supportive coworkers, in decisions to discuss arthritis at work and in changes to work that might enable people with arthritis to remain employed.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Gignac had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Gignac.

Acquisition of data. Gignac.

Analysis and interpretation of data. Gignac, Cao.

Ancillary