Abstract
The workplace is an important setting for improving population psychological health. This study aimed to identify priority industries and populations in Australia with highest adverse effects of psychological distress. The study included 5834 workers aged 18 to 64 years who participated in the 2020 to 2021 National Health Survey. Logistic regression analyses were used to estimate the prevalence of psychological distress by industry, age group, and gender. Productivity losses were analyzed using negative binomial regression. Increased odds of experiencing distress were found for younger workers 18 to 29 years (odds ratio [OR] = 2. 45) and female workers (OR = 1.52). Industry of employment did not impact distress once age and gender were accounted for (P = .956). Being highly/very highly distressed resulted in a mean of 21.56 more distress-related loss days per year (0.78 vs 22.34) compared with low/moderate distress. Targeted and tailored workplace intervention programs for these groups are required to build a healthy and productive future workforce.
Keywords: Australia, distress, female, industry, mental health, work, young workers
What We Already Know
The prevalence of psychological distress has been trending upward for some time.
Both work and non-work factors are associated with poor mental health outcomes.
Workplace mental health programs can effectively improve worker mental health.
What This Article Adds
The prevalence of psychological distress was higher in younger workers and female workers in Australia; however, there was no significant variation in distress by industry.
The workplace productivity impacts of psychological distress are substantial, with distressed workers reporting an additional 21 days absent per year.
Workplace mental health programs should be targeted and tailored to young workers and female workers.
Introduction
The scale of psychological distress in the modern day is unprecedented. More than 40% of Australians aged 16 to 85 have experienced a mental disorder at some time in their life. 1 In the workplace, psychological conditions account for nearly 10% of serious injury claims, 2 and cost workplaces approximately AUD$17 billion annually, mostly due to reduced productivity. 3
Against this backdrop, evidence suggests that mental health programs delivered in the workplace can improve worker well-being. 4 With two thirds of Australian adults employed, and workers spending around 50% of their waking hours at work, the workplace is an obvious setting for action to deliver population-level mental health benefits, and there are growing calls for policy-makers to focus on the workplace for opportunities to deliver improved mental health. 5
Improving workforce mental health requires policies and workplace interventions tailored to the specific needs of priority groups. Identifying where distressed workers are in the workforce landscape can guide policy-makers in prioritizing sectors and developing tailored interventions. Furthermore, understanding the associated productivity losses can foster business interest in addressing mental health.
While a body of research in Australia has suggested high-priority sub-groups for psychological health at work are females and younger workers, 6 there has also been recent evidence to suggest that industry patterns may exist in the prevalence of psychological distress. 7 Using National Health Survey (NHS) data from 2020 to 2021, when the COVID-19 pandemic was impacting on the population, this research aimed to provide an updated insight into the relationship between distress and industry, age, and gender, and estimate absenteeism related to this distress.
Study Aims
The primary aim of this study was to investigate workforce psychological distress in Australia during 2020 to 2021 by key covariates: industry, age, and gender. Second, we aimed to estimate absenteeism related to this distress to inform business and policy-makers with respect to impact on industries and resource allocation.
Methods
Design and Participants
The NHS is collected by the Australian Bureau of Statistics (ABS) at 3-yearly intervals. Usually collected face-to-face, the 2020 to 2021 survey was collected during the COVID-19 pandemic, necessitating a change to online data collection. The 2020 to 2021 NHS had a lower response rate (35%) than previous NHS cycles. A comprehensive description of the study participants, survey, and sampling design for the 2020 to 2021 NHS can be found elsewhere. 8 Data collection took place between August 2020 and June 2021. Survey scope excludes overseas visitors, very remote areas, and those living in non-private dwellings. The final sample consisted of 13 281 fully responding persons.
National Health Survey unit record data were accessed in the secure ABS DataLab environment under an agreement between Universities Australia and the ABS. Ethics approval was granted by the Macquarie University Medicine and Health Sciences Ethics subcommittee (approval ID: 13039).
Measures
Psychological distress
Psychological distress was measured using the Kessler Psychological Distress Scale (K10). 9 The ten-item Likert scale measures psychological distress, particularly symptoms of anxiety and depression, in the previous 4 weeks. Total scores range from 10 to 50, and were categorized as low (10-15), moderate (16-21), high (22-29), and very high (30-50), in accordance with ABS category parameters. 8 Total scores of 20 or more are indicative of the presence of mental disorder(s) 10 and need for treatment. 11 In line with previous studies,12 -14 Kessler Psychological Distress Scale scores of high or very high (22 and higher) were combined and used to indicate psychological distress in this study.
Analysis was restricted to working adults aged 18 to 64 years as K10 data were only collected from persons aged 18 or over, and the interpretation of the K10 for those aged over 65 requires specific consideration. 15
Productivity
Respondents were asked the number of days they were unable to work, study, or carry out day-to-day activities because of feelings in the last 4 weeks. This measure is an inclusion in the extended K10 Plus. 16 Consistent with previous research, these days are assumed to impact on work for this population given the sample is restricted to the working population.3,17
Socio-demographic variables and work-related variables
Industry of occupation was coded for participants’ main job according to the Australian and New Zealand Standard Industrial Classification, 2016. 18 Sex was reported as per collected in the survey: male or female. Age was categorized into three groups: 18 to 19, 30 to 49, and 50 to 64 years. Socio-economic status was estimated using the Socio-Economic Indicators For Areas (SEIFA) produced by the ABS and transformed into quintiles. Personal weekly income (gross) was categorized as quintiles. Usual hours of work per week were categorized into seven groups: 1 to 15, 16 to 24, 25 to 34, 35 to 39, 40, 41 to 49, and 50 hours or over.
Statistical Analysis
Replicate weighting variables in the ABS-provided CURF (confidentialized unit record file) were used throughout to account for the complex sample design. The standard error (SE) of estimated prevalence rates and confidence intervals (CIs) around odds ratios (ORs) (derived from regression models) were estimated using the delete-one jackknife technique. 19 All analyses were performed using STATA 17.
Psychological distress
To examine the prevalence of psychological distress (K10 score ≥ 22) across the study variables, we present weighted proportions with a 95% CI.
The key covariates of interest were industry, sex (male/female), and age group. Univariate logistic regression was used to assess their association with distress. To determine the independent association of the key covariates, multivariate analysis was conducted adjusting for each other and the potential confounders with known or suspected associations with distress that were available in the NHS and could potentially be beneficial for the purpose of targeting intervention strategies (state/territory of residence, income, socio-economic quintile, and usual hours of work). In the analysis of industry, the industry with the largest number of workers, health care and social assistance, was used as the reference category. Wald tests were performed (using STATA’s testparm command) to estimate whether industry was collectively significant in each model. Records with missing data for income (n = 26) were excluded from multivariate analysis.
Distress-related productivity impacts
Weighted means for number of distress-related work loss and work cutback days were tabulated by industry, K10 distress category, gender, age group, state, income, socio-economic quintile, and usual hours of work. Annual loss days were reported for ease of interpretation (annual days = 4-week rate × 12, to allow for 48 working weeks and 4 weeks of annual leave) in line with previous research. 3
The relationship between the number of distress-related work loss days and the study variables was analyzed using negative binomial regression, expressed as an OR. Univariate regressions were used to analyze the association between work loss days and the key covariates of interest: industry, sex (male/female), and age. To determine the independent association of the key covariates, multivariate analysis was conducted adjusting for each other and potential confounders (state, income, socio-economic quintile, and usual hours of work).
The association between distress and distress-related loss days was analyzed in a univariate regression only and excluded from the multivariable regressions to avoid the impact of collinearity on estimated coefficients.
Total distress-related work loss days
Estimates of the employed labor force by industry, age, and sex were sourced from the ABS Labor force survey 20 using ABS-provided sample weights to ensure conformity to the estimated population by age and sex. Monthly time series was extracted for the period 2020 to 2021. Distress-related loss days and associated 95% CIs were calculated as workforce size × mean workforce loss days.
Results
The data set consisted of 5834 employed individuals aged 18 to 65 years. Kessler Psychological Distress Scale score was unavailable for one individual, sex was not stated in six cases, and a further 312 cases were removed due to inadequately described industry, leaving a final sample of 5515 individuals (2644 males [47.9%] and 2871 females [52.1%]).
Table 1 shows that 18.2% of all workers experienced psychological distress. The highest rate of distress was found in the administrative and support services workforce, where more than a quarter (25.9%) of workers reported distress, while the lowest rate was found in mining (8.5%). More than one in five female workers experienced distress (21.9%), compared with one in seven males (14.8%). More than a quarter of the 18- to 29-year-old workforce experienced distress, with rates of distress reducing with increasing age. Higher rates of distress were found in Victorian workers compared with other states. Study weights largely correlated with pre-study weights.
Table 1.
Distress Status by Industry and Relevant Covariates for Working Australians Aged 18 to 64 Years.
| High/very high distress | |||
|---|---|---|---|
| N (%) | % | 95% CI | |
| Total sample | 5515 | 18.20 | [16.62, 19.89] |
| Industry | |||
| Agriculture, forestry, and fishing | 87 (1.6) | 13.38 | [5.34, 29.71] |
| Mining | 98 (1.8) | 8.46 | [4.01, 17.00] |
| Manufacturing | 273 (5.0) | 13.75 | [9.27, 19.94] |
| Electricity, gas, water, and waste services | 77 (1.4) | 13.05 | [4.89, 30.46] |
| Construction | 360 (6.5) | 16.87 | [11.73, 23.66] |
| Wholesale trade | 145 (2.6) | 19.74 | [11.03, 32.81] |
| Retail trade | 414 (7.5) | 19.07 | [14.93, 24.03] |
| Accommodation and food services | 195 (3.5) | 22.39 | [15.24, 31.65] |
| Transport, postal, and warehousing | 225 (4.1) | 17.76 | [11.35, 26.70] |
| Information media and telecommunications | 88 (1.6) | 20.70 | [9.16, 40.32] |
| Financial and insurance services | 216 (3.9) | 23.08 | [16.71, 30.98] |
| Rental, hiring, and real estate services | 80 (1.5) | 19.38 | [10.34, 33.38] |
| Professional, scientific, and technical services | 614 (11.1) | 15.32 | [11.76, 19.72] |
| Administrative and support services | 165 (3.0) | 25.89 | [16.46, 38.26] |
| Public administration and safety | 692 (12.5) | 13.59 | [9.62, 18.84] |
| Education and training | 651 (11.8) | 20.62 | [15.91, 26.30] |
| Health care and social assistance | 868 (15.7) | 18.89 | [14.89, 23.65] |
| Arts and recreation services | 89 (1.6) | 20.35 | [10.55, 35.64] |
| Other services | 178 (3.2) | 19.55 | [11.44, 31.38] |
| Gender | |||
| Male | 2644 (47.9) | 14.84 | [12.80, 17.15] |
| Female | 2871 (52.1) | 21.92 | [19.56, 24.47] |
| Age group (years) | |||
| 18 to 29 | 835 (15.1) | 26.43 | [21.80, 31.63] |
| 30 to 49 | 2821 (51.2) | 17.63 | [15.76, 19.66] |
| 50 to 64 | 1859 (33.7) | 12.40 | [10.46, 14.63] |
| State | |||
| New South Wales | 1038 (18.8) | 16.80 | [14.16, 19.82] |
| Victoria | 870 (15.8) | 22.38 | [19.09, 26.05] |
| Queensland | 706 (12.8) | 17.45 | [13.91, 21.66] |
| South Australia | 647 (11.7) | 15.42 | [11.77, 19.96] |
| Western Australia | 769 (13.9) | 14.81 | [11.26, 19.23] |
| Tasmania | 579 (10.5) | 18.86 | [14.40, 24.32] |
| Northern Territory | 294 (5.3) | 19.14 | [13.71, 26.08] |
| Australian Capital Territory | 612 (11.1) | 19.46 | [16.12, 23.30] |
| Income quintile a | |||
| 1 (lowest) | 408 (7.4) | 20.7 | [14.7, 28.3] |
| 2 | 421 (7.6) | 21.5 | [15.3, 29.4] |
| 3 | 1033 (18.7) | 20.1 | [16.9, 23.8] |
| 4 | 1652 (30.0) | 19.6 | [16.7, 22.9] |
| 5 (highest) | 2001 (36.3) | 13.6 | [11.2, 16.5] |
| SEIFA quintile | |||
| 1 (most disadvantaged) | 632 (11.5) | 22.35 | [17.01, 28.77] |
| 2 | 914 (16.6) | 14.21 | [11.20, 17.86] |
| 3 | 1129 (20.5) | 18.48 | [15.30, 22.15] |
| 4 | 1366 (24.8) | 18.74 | [15.47, 22.52] |
| 5 (least disadvantaged) | 1474 (26.7) | 18.27 | [15.45, 21.48] |
| Usual hours of work | |||
| 1 to 15 hours | 469 (8.5) | 28.05 | [21.93, 35.11] |
| 16 to 24 hours | 572 (10.4) | 17.33 | [13.2 1, 22.40] |
| 25 to 34 hours | 720 (13.1) | 20.26 | [16.21, 25.03] |
| 35 to 39 hours | 1442 (26.1) | 16.13 | [13.57, 19.07] |
| 40 hours | 1134 (20.6) | 16.24 | [13.53, 19.38] |
| 41 to 49 hours | 485 (8.8) | 20.12 | [14.07, 27.92] |
| 50 hours and over | 693 (12.6) | 15.82 | [11.97, 20.62] |
Missing data (n = 26) were distributed evenly across quintiles.
Industry was not significantly associated with distress in logistic regression models (Table 2). The univariate regression for industry found working in mining was associated with 60% lower odds of being distressed (OR = 0.40, 95% CI [0.16, 0.96]) than the reference group, health care and social assistance. However, this association was no longer evident after adjustment for age and gender (OR = 0.50 95% CI [0.18, 1.35]), and all covariates (OR = 0.59 95% CI [0.21, 1.67]).
Table 2.
Association Between Psychological Distress (Combined High/Very High) and Industry, Gender and Age: Odds Ratios From Logistic Regression Models.
| Univariate regressions | Model 2: multivariate—industry, gender, and age | Model 3: Multivariate—all covariates | |||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | |
| Industry | .746 a | .918 a | .956 a | ||||||
| Agriculture, forestry, and fishing | 0.66 | [0.20, 2.18] | .492 | 0.90 | [0.26, 3.09] | .861 | 0.82 | [0.22, 3.08] | .767 |
| Mining | 0.40 | [0.16, 0.96] | .040 | 0.50 | [0.18, 1.35] | .168 | 0.59 | [0.21, 1.67] | .319 |
| Manufacturing | 0.68 | [0.40, 1.17] | .160 | 0.83 | [0.48, 1.45] | .507 | 0.84 | [0.48, 1.47] | .531 |
| Electricity, gas, water, and waste services | 0.64 | [0.19, 2.24] | .483 | 0.91 | [0.25, 3.29] | .880 | 1.03 | [0.28, 3.83] | .958 |
| Construction | 0.87 | [0.51, 1.50] | .615 | 1.10 | [0.60, 2.03] | .757 | 1.12 | [0.58, 2.13] | .737 |
| Wholesale trade | 1.06 | [0.50, 2.21] | .882 | 1.40 | [0.65, 3.01] | .384 | 1.42 | [0.64, 3.14] | .388 |
| Retail trade | 1.01 | [0.67, 1.53] | .954 | 0.96 | [0.63, 1.46] | .843 | 0.87 | [0.56, 1.36] | .535 |
| Accommodation and food services | 1.24 | [0.73, 2.09] | .416 | 1.13 | [0.67, 1.89] | .647 | 1.06 | [0.61, 1.83] | .832 |
| Transport, postal, and warehousing | 0.93 | [0.55, 1.56] | .774 | 1.18 | [0.70, 2.00] | .522 | 1.21 | [0.69, 2.09] | .499 |
| Information media and telecommunications | 1.12 | [0.38, 3.33] | .835 | 1.22 | [0.43, 3.41] | .706 | 1.36 | [0.48, 3.86] | .560 |
| Financial and insurance services | 1.29 | [0.79, 2.11] | .307 | 1.40 | [0.86, 2.28] | .167 | 1.50 | [0.91, 2.48] | .112 |
| Rental, hiring, and real estate services | 1.03 | [0.46, 2.34] | .938 | 1.05 | [0.47, 2.34] | .909 | 1.14 | [0.52, 2.50] | .737 |
| Professional, scientific, and technical services | 0.78 | [0.51, 1.18] | .230 | 0.87 | [0.57, 1.34] | .532 | 0.93 | [0.59, 1.46] | .742 |
| Administrative and support services | 1.50 | [0.76, 2.96] | .236 | 1.46 | [0.71, 3.01] | .304 | 1.47 | [0.71, 1.80] | .293 |
| Public administration and safety | 0.68 | [0.44, 1.04] | .073 | 0.79 | [0.51, 1.23] | .294 | 0.90 | [0.56, 1.43] | .644 |
| Education and training | 1.12 | [0.72, 1.72] | .615 | 1.14 | [0.74, 1.76] | .542 | 1.13 | [0.71, 1.80] | .599 |
| Health care and social assistance b | 1.00 | 1.00 | 1.00 | ||||||
| Arts and recreation services | 1.10 | [0.48, 2.49] | .822 | 1.12 | [0.49, 2.55] | .787 | 1.01 | [0.41, 2.47] | .982 |
| Other services | 1.04 | [0.51, 2.15] | .906 | 1.16 | [0.56, 2.40] | .689 | 1.03 | [0.48, 2.23] | .929 |
| Gender | |||||||||
| Male b | 1.00 | 1.00 | 1.00 | ||||||
| Female | 1.61 | [1.28, 2.02] | <.001 | 1.58 | [1.22, 2.07] | .001 | 1.52 | [1.16, 2.01] | .003 |
| Age group (years) | <.001 c | <.001 c | <.001 c | ||||||
| 18 to 29 | 2.54 | [1.81, 3.55] | <.001 | 2.48 | [1.77, 3.49] | <.001 | 2.45 | [1.75, 3.43] | <.001 |
| 30 to 49 | 1.51 | [1.20, 1.90] | <.001 | 1.50 | [1.20, 1.90] | .001 | 1.55 | [1.23, 1.94] | <.001 |
| 50 to 64 b | 1.00 | 1.00 | 1.00 | ||||||
Overall parameter estimate for industry (from Wald test).
Reference category.
Overall P for trend.
Being female was significantly associated with distress in all models. The odds of experiencing distress were found to be 61% higher for female than male workers in a univariate regression (OR = 1.61, 95% CI [1.28, 2.02]). Adjusting for age and industry slightly attenuated these odds (OR = 1.58, 95% CI [1.22, 2.07]). After adjustment for all covariates, the odds of experiencing distress were found to be 52% higher for female than male workers (OR = 1.52, 95% CI [1.16, 2.01]).
Younger age group was associated with higher psychological distress in univariate and multivariate regressions. After adjusting for all covariates (model 3), compared with older workers 50 to 64 years, workers aged 18 to 29 years had 154% higher odds of being distressed while workers aged 50 to 64 years had 51% higher odds of being distressed (OR = 2.54, 95% CI [1.81, 3.55] and OR = 1.51, 95% CI [1.20, 1.90], respectively).
Workers reported a mean of 4.70 days lost to distress per year. Being highly or very highly distressed resulted in a mean of 21.56 more distress-related loss days each year than being low-moderately distressed (0.78 vs 22.34 days). Workers in agriculture, forestry, and fishing had the fewest distress-related loss days per month (0.42 days per year, 95% CI [–0.21, 1.06]), followed by workers in mining (1.83 days, 95% CI [0.05, 3.61]). Males reported fewer distress-related work loss days than females (3.56 vs 5.97 days), while workers in the youngest age group had the highest loss days (8.73 days per year), with loss days decreasing as the age group increased (Table 3).
Table 3.
Distress-Related Work Loss Days by Industry and Relevant Covariates for Working Australians Aged 18 to 64 Years.
| Annual loss days | 95% CI | |
|---|---|---|
| Total sample | 4.70 | [3.71, 5.70] |
| Industry | ||
| Agriculture, forestry, and fishing | 0.42 | [–0.21, 1.06] |
| Mining | 1.83 | [0.05, 3.61] |
| Manufacturing | 3.05 | [1.09, 5.01] |
| Electricity, gas, water, and waste services | 13.43 | [–10.29, 37.16] |
| Construction | 5.21 | [–0.75, 11.17] |
| Wholesale trade | 2.48 | [0.30, 4.66] |
| Retail trade | 5.99 | [1.52, 10.46] |
| Accommodation and food services | 8.54 | [4.60, 12.48] |
| Transport, postal, and warehousing | 2.94 | [–0.09, 5.97] |
| Information media and telecommunications | 2.24 | [0.25, 4.22] |
| Financial and insurance services | 3.51 | [0.83, 6.20] |
| Rental, hiring, and real estate services | 3.34 | [0.34, 6.34] |
| Professional, scientific, and technical services | 2.96 | [1.06, 4.86] |
| Administrative and support services | 7.29 | [–0.02, 14.59] |
| Public administration and safety | 3.09 | [1.15, 4.73] |
| Education and training | 6.81 | [3.80, 9.81] |
| Health care and social assistance | 5.44 | [2.58, 8.31] |
| Arts and recreation services | 4.19 | [–0.77, 9.15] |
| Other services | 3.74 | [1.02, 6.46] |
| Gender | ||
| Male | 3.56 | [2.35, 4.77] |
| Female | 5.97 | [4.50, 7.44] |
| Age group (years) | ||
| 18 to 29 | 8.73 | [5.43, 12.03] |
| 30 to 49 | 3.32 | [2.36, 4.28] |
| 50 to 64 | 3.74 | [2.00, 5.49] |
| Distress category | ||
| Low/moderate distress | 0.78 | [0.59, 0.97] |
| High/very high distress | 22.34 | [17.20, 27.48] |
| State | ||
| New South Wales | 3.96 | [2.18, 5.74] |
| Victoria | 5.73 | [3.67, 7.79] |
| Queensland | 4.58 | [2.01, 7.15] |
| South Australia | 4.88 | [2.42, 7.34] |
| Western Australia | 4.70 | [2.17, 7.22] |
| Tasmania | 5.19 | [2.25, 8.13] |
| Northern Territory | 4.08 | [0.32, 7.83] |
| Australian Capital Territory | 3.35 | [2.08, 4.63] |
| Income quintile | ||
| 1 (lowest) | 6.96 | [2.40, 11.04] |
| 2 | 6.72 | [3.36, 10.08] |
| 3 | 6.48 | [3.60, 9.36] |
| 4 | 4.92 | [2.88, 6.96] |
| 5 (highest) | 2.16 | [1.56, 2.76] |
| SEIFA quintile | ||
| 1 (most disadvantaged) | 6.97 | [3.50, 10.44] |
| 2 | 4.06 | [1.45, 6.68] |
| 3 | 5.71 | [3.72, 7.69] |
| 4 | 4.10 | [2.40, 5.79] |
| 5 (least disadvantaged) | 3.64 | [2.12, 5.17] |
| Usual hours of work | ||
| 1 to 15 hours | 8.71 | [5.81, 11.61] |
| 16 to 24 hours | 6.95 | [2.96, 10.94] |
| 25 to 34 hours | 4.28 | [2.32, 6.23] |
| 35 to 39 hours | 3.15 | [1.74, 4.56] |
| 40 hours | 5.18 | [2.13, 8.23] |
| 41 to 49 hours | 6.42 | [2.56, 10.28] |
| 50 hours and over | 1.57 | [0.28, 2.86] |
The univariate regression for work loss days by distress category showed high/very high distressed workers reported 28 times more distress-related loss days than workers with low/moderate distress (OR = 28.68, 95% CI [20.28, 40.56], P < .001).
In regression models for distress-related work loss days (Table 4), workers in the agriculture, forestry, and fishing industry were found to have had significantly fewer distress-related work loss days than the reference industry, health in univariate analyses, and this difference was maintained when controlling for age and gender (model 2), and all other covariates (model 3) (OR = 0.08, 95% CI [0.02, 0.30]). Workers in the information, media, and telecommunications industry had significantly lower distress-related loss days when age and gender were controlled for (model 2), however, this difference was no longer significant with the addition of all other covariates (model 3).
Table 4.
Association Between Number of Distress-Related Work Loss Days and Industry, Gender, and Age: Odds Ratios From Negative Binomial Regression Models.
| Univariate | Model 2: multivariate—industry, gender, and age | Model 3: multivariate—all covariates | |||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | |
| Industry | .046 | .067 | .067 a | ||||||
| Agriculture, forestry, and fishing | 0.08 | [0.02, 0.38] | .002 | 0.10 | [0.02, 0.45] | .003 | 0.08 | [0.02, 0.30] | <.001 |
| Mining | 0.34 | [0.10, 1.14] | .078 | 0.55 | [0.14, 2.10] | .375 | 1.03 | [0.22, 4.80] | .969 |
| Manufacturing | 0.56 | [0.25, 1.27] | .161 | 0.62 | [0.26, 1.46] | .267 | 0.59 | [0.25, 1.42] | .238 |
| Electricity, gas, water, and waste services | 2.47 | [0.06, 103.36] | .630 | 4.22 | [0.06, 301.20] | .502 | 3.49 | [0.19, 65.71] | .398 |
| Construction | 0.96 | [0.21, 4.47] | .956 | 1.12 | [0.28, 4.53] | .867 | 1.03 | [0.26, 4.09] | .962 |
| Wholesale trade | 0.46 | [0.15, 1.41] | .169 | 0.60 | [0.18, 1.97] | .392 | 0.93 | [0.24, 3.67] | .916 |
| Retail trade | 1.10 | [0.42, 2.89] | .843 | 0.96 | [0.36, 2.55] | .932 | 1.09 | [0.30, 3.90] | .894 |
| Accommodation and food services | 1.57 | [0.70, 3.49] | .264 | 1.09 | [0.49, 2.45] | .823 | 1.20 | [0.54, 2.63] | .652 |
| Transport, postal, and warehousing | 0.54 | [0.14, 2.10] | .369 | 0.77 | [0.20, 3.05] | .707 | 0.89 | [0.20, 4.00] | .873 |
| Information media and telecommunications | 0.41 | [0.13, 1.28] | .123 | 0.33 | [0.12, 0.89] | .028 | 0.47 | [0.16, 1.35] | .157 |
| Financial and insurance services | 0.65 | [0.22, 1.37] | .427 | 0.56 | [0.23, 1.33] | .184 | 0.92 | [0.42, 2.03] | .832 |
| Rental, hiring, and real estate services | 0.61 | [0.19, 1.93] | .398 | 0.72 | [0.22, 2.31] | .572 | 0.74 | [0.22, 2.49] | .626 |
| Professional, scientific, and technical services | 0.54 | [0.22, 1.37] | .193 | 0.61 | [0.25, 1.46] | .261 | 0.87 | [0.36, 2.13] | .757 |
| Administrative and support services | 1.34 | [0.36, 5.05] | .661 | 1.26 | [0.34, 4.66] | .722 | 1.05 | [0.26, 4.31] | .945 |
| Public administration and safety | 0.57 | [0.28, 1.14] | .111 | 0.66 | [0.30, 1.47] | 0.304 | 1.15 | [0.49, 2.72] | .744 |
| Education and training | 1.25 | [0.61, 2.56] | .535 | 1.13 | [0.51, 2.46] | .763 | 1.58 | [0.75, 3.34] | .229 |
| Health care and social assistance b | 1.00 | 1.00 | 1.00 | ||||||
| Arts and recreation services | 0.77 | [0.17, 3.56] | .734 | 0.59 | [0.17, 2.04] | .394 | 0.46 | [0.13, 1.64] | .226 |
| Other services | 0.69 | [0.27, 1.72] | .417 | 0.64 | [0.24, 1.72] | .369 | 0.86 | [0.31, 2.36] | .768 |
| Gender | |||||||||
| Male b | 1.00 | 1.00 | 1.00 | ||||||
| Female | 1.68 | [1.12, 2.52] | .014 | 1.64 | [1.10, 2.43] | .015 | 1.63 | [1.01, 2.63] | .047 |
| Age group (years) | .002 | .012 | .117 c | ||||||
| 18 to 29 | 2.33 | [1.24, 4.38] | .009 | 2.48 | [1.26, 4.86] | .009 | 2.07 | [0.94, 4.57] | .070 |
| 30 to 49 | 0.89 | [0.51, 1.54] | 0.668 | 1.03 | [0.60, 1.76] | 0.918 | 1.04 | [0.59, 1.83] | .878 |
| 50 to 64 b | 1.00 | 1.00 | 1.00 | ||||||
Overall parameter estimate for industry.
Reference category.
Overall P for trend.
Significantly higher loss days were associated with female gender in all models (OR = 1.63, 95% CI [1.01, 2.63]). Compared with older workers (50-64 years), workers aged 18 to 29 years had higher distress-related loss days in all models, although the relationship was no longer statistically significant in model 3 (OR = 2.07, 95% CI [0.94, 4.57], P = .07).
Table 5 presents estimated workforce loss days due to distress for 2020 to 2021 by industry, sex, and age. Total days lost due to distress were estimated at 56 million. The largest workforces made the biggest contributions to distress-related loss days (health care and social assistance, education, and training and retail), except in the case of the accommodation and food industry, which experienced an estimated 5.75 million distress-related loss days in a workforce of only 673 000 workers. The female workforce lost 33 million days due to distress, compared with 22 million days for the male workforce. Younger workers aged 18 to 29 accounted for the greatest number of distress-related loss days of all age groups (26 million), despite having the smallest workforce.
Table 5.
Workforce Size and Estimated Total Distress-Related Work Loss Days for 2020 to 2021, by Industry, Age, and Gender for Working Australians Aged 18 to 64 Years.
| Workforce size
a
N (‘000) |
Estimated total workforce loss days for 2020 to 2021 (‘000,000) b | 95% CI (‘000,000) c | |
|---|---|---|---|
| Total sample | 11 933 | 56.09 | [44.21, 68.02] |
| Industry | |||
| Agriculture, forestry, and fishing | 260 | 0.11 | [–0.05, 0.28] |
| Mining | 259 | 0.47 | [0.01, 0.93] |
| Manufacturing | 824 | 2.51 | [0.90, 4.13] |
| Electricity, gas, water, and waste | 149 | 2.01 | [–1.53, 5.54] |
| Construction | 1104 | 5.76 | [–0.83, 12.33] |
| Wholesale trade | 355 | 0.88 | [0.11, 1.65] |
| Retail trade | 1146 | 6.87 | [1.74, 11.99] |
| Accommodation and food services | 673 | 5.75 | [3.10, 8.40] |
| Transport, postal, and warehousing | 606 | 1.78 | [–0.05, 3.62] |
| Information media and telecommunications | 183 | 0.41 | [0.05, 0.77] |
| Financial and insurance services | 460 | 1.62 | [0.38, 2.85] |
| Rental, hiring, and real estate services | 189 | 0.63 | [0.06, 1.20] |
| Professional, scientific, and technical services | 1108 | 3.28 | [1.17, 5.38 |
| Administrative and support services | 389 | 2.84 | [–0.01, 5.68] |
| Public administration and safety | 835 | 2.58 | [0.96, 3.95] |
| Education and training | 1051 | 7.16 | [3.99, 10.31] |
| Health care and social assistance | 1711 | 9.31 | [4.41, 14.22] |
| Arts and recreation services | 209 | 0.88 | [–0.16, 1.91] |
| Other services | 457 | 1.71 | [0.47, 3.95] |
| Gender | |||
| Male | 6266 | 22.31 | [14.73, 29.89] |
| Female | 5667 | 33.83 | [25.50, 42.16] |
| Age group (years) | |||
| 18 to 29 | 2989 | 26.09 | [16.23, 35.96] |
| 30 to 49 | 5746 | 19.08 | [13.56, 24.59] |
| 50 to 64 | 3198 | 11.96 | [6.40, 17.56] |
Discussion
Our study found over one sixth (18.2%) of employed Australians experienced psychological distress in 2020 to 2021 at a level indicative of requiring treatment. This is notably higher than the 11.4% distress rate observed among working population of Australians aged 18 to 64 in 2017 to 2018. 21 The gender-specific distress rates in 2020 to 2021 (14.8% for males and 21.9% for females) also exceeded those found in an analysis of the 2017 to 2018 NHS (9.7% for males and 13.5% for females). 7 While direct comparisons are constrained by methodological differences, these findings align with prior research indicating a rising trend in psychological distress in the population over the last 20 years,22,23 a trend likely exacerbated by the COVID-19 pandemic. 24
Gender continues to be a significant factor in workforce distress. Females consistently report higher distress rates, 22 and there is evidence that the pandemic has disproportionately worsened females’ mental health. 25 Our findings corroborate this trend, showing a nearly 50% higher distress rate among females compared with males in 2020 to 2021, versus a 39% higher rate in 2017 to 2018. 7 Given our focus on employed adults, we hypothesize that the elevated distress rates among females partly reflect the challenges of balancing work, often from home, with periods of home schooling implemented in Australia at various points over this time. Despite these pandemic-related anomalies, the persistently high distress rates among females are concerning, particularly as females continue to shoulder the majority of caring responsibilities in the post-pandemic era, amidst changing employment conditions, 5 including remote work and expectations of round-the-clock digital availability. These factors underscore the need for industry to provide support to workers with caring responsibilities to ensure their mental health is maintained.
Young workers remain a high-priority sub-group in the workforce, with over a quarter of 18- to 29-year-old workers experiencing distress in 2020 to 2021. While some evidence suggests that the pandemic disproportionately increased distress among younger people compared with older people,25,26 the persistent pattern of high distress rates for young workers calls for sustained attention and focus. Young workers are exposed more often to psychosocial risk factors such as workplace conflict, low job control, bullying, and precarious employment arrangements. 27 The rise of the gig economy and heightened job insecurity suggests young workers will continue to face insecure, low quality, precarious jobs, leaving them particularly vulnerable to poorer mental health outcomes. 28 Policy-makers and business should explicitly focus on addressing youth-specific mental health risk factors, and further, on designing workplaces that positively contribute to the mental health of young workers.
Interestingly, industry of employment did not appear to impact distress during 2020/2021, contrasting with the 2017 to 2018 period. 7 This lack of variation by industry may be attributed to the widespread increase in distress from a non-work stressor (COVID-19); however, it is likely the nationwide income support payments made available during the pandemic also contributed. Research comparing US state policies and their impact on distress during the pandemic showed that supportive social policies can effectively weaken the link between income shocks and worsening mental health. 29 In Australia, workers from all industries were eligible for these payments, providing a buffer for some of the stress associated with working during the pandemic.
Our study found substantial productivity losses associated with distress, including an estimated 56 million loss days over the period. Workers experiencing distress were found to lose an additional 21.58 days of work per year compared with non-distressed workers. These losses are felt more acutely in the younger workforce and among female workers. Despite a smaller workforce, distress among younger workers aged 18 to 29 years accounted for 26 million loss days—the largest loss of all age groups, reinforcing their priority status for intervention support. Collectively, these findings extrapolate to extensive economic losses attributable to distress.
A limitation of this study was the cross-sectional design that precluded us from attributing a causal relationship between the variables investigated and distress. However, our aim was to describe workforce sub-group differences to allow for the targeting of interventions toward high-priority groups. Similarly, we did not include a range of other variables potentially associated with distress, such as marital status and education, as our focus was on variables that could define subgroups to which policy interventions could be targeted.
The transition to online data collection for the 2020 to 2021 NHS brings benefits and limitations. While online collection of K10 has been shown to increase accuracy due to reduced social desirability bias, 30 the change makes comparisons with previous data sets difficult to interpret. Given the consistent findings by others of an increase in distress during the pandemic,1,24 it is likely the observed increase in distress reflects both a genuine rise in distress in the population, and improved accuracy in reporting, although the contribution of each is unknown. The lower response rate for 2020 to 2021 NHS is an additional limitation and this may have impacted on our ability to detect differences in distress by industry.
Finally, we acknowledge that the measure of productivity loss used in this research (self-reported distress-related loss days) has limitations, including potential recall bias. However, its short 4-week recall period, ease of use and distress specificity means this measure is widely used in national surveys. 3 In addition, our calculations for total workforce loss days (for industry, sex, and age) use total employed persons and as such does not account for part-time worker variations across sub-groups.
This study’s major strengths include its use of data from a nationally representative survey of all employed adults, including the self-employed, and that productivity estimates are restricted to those that are distress related, allowing for a more precise estimate of the additional impact of distress on workforce productivity.
Conclusion
These findings highlighted the enduring vulnerability to psychological distress experienced by female workers and younger workers, and the resulting lost productivity. Workplace programs that aim to contribute positively to mental health should be developed with and for female workers and younger workers that take their specific needs into account. Such programs, if implemented successfully, could see both substantial improvements in the well-being of the workforce along with significant productivity gains.
While industry was not found to impact on the distress of workers in 2020 to 2021, there is a need to remain vigilant to the shifting workplace landscape, including but not limited to technological changes and pandemics, and to continue to monitor and identify emerging priority groups with respect to worker mental health.
Supplemental Material
Supplemental material, sj-docx-1-aph-10.1177_10105395241306477 for Workforce Psychological Distress and Absenteeism in Australia: The Correlates of Industry, Age, and Gender by Kristy Burns, Louise A. Ellis, Abilio De Almeida Neto and Janaki Amin in Asia Pacific Journal of Public Health
Acknowledgments
The authors are grateful to the Australian Bureau of Statistics for access to the National Health Survey and Labour Force Survey data sets via DataLab.
Footnotes
Author Contributions: Kristy Burns and Janaki Amin conceived the ideas and analyzed the data. The first draft of the manuscript was written by Kristy Burns and all authors contributed to the final manuscript.
Availability of Data and Materials: The unit-record survey data are available to researchers, in accordance with ABS data access procedures and policies. More information is available at the ABS website: https://www.abs.gov.au/websitedbs/d3310114.nsf/home/microdata+entry+page.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval: The study was conducted according to the guidelines of the Declaration of Helsinki. The conduct of ABS surveys was approved under the Census and Statistics Act 1905. All accessed data were de-identified. Ethical approval for the current analysis of ABS data was granted by the Macquarie University Medicine and Health Sciences Ethics subcommittee (approval ID: 13039).
ORCID iD: Kristy Burns 
https://orcid.org/0000-0002-3060-4812
Supplemental Material: Supplemental material for this article is available online.
References
- 1. Australian Bureau of Statistics. National study of mental health and wellbeing 2020-22. Published 2022. Accessed February 1, 2024. https://www.abs.gov.au/statistics/health/mental-health/national-study-mental-health-and-wellbeing/2020-21
- 2. Safe Work Australia. Key WHS statistics Australia 2022. Published 2023. Accessed February 1, 2024. https://www.safeworkaustralia.gov.au/sites/default/files/2023-01/key_whs_stats_2022_17jan2023.pdf
- 3. Australian Productivity Commission. Mental health. Report no. 95. Published 2020. Accessed February 14, 2024. https://www.pc.gov.au/inquiries/completed/mental-health/report/mental-health.pdf
- 4. Rugulies R, Aust B, Greiner BA, et al. Work-related causes of mental health conditions and interventions for their improvement in workplaces. Lancet. 2023;402(10410):1368-1381. [DOI] [PubMed] [Google Scholar]
- 5. Frank J, Mustard C, Smith P, et al. Work as a social determinant of health in high-income countries: past, present, and future. Lancet. 2023;402(10410):1357-1367. doi: 10.1016/s0140-6736(23)00871-1 [DOI] [PubMed] [Google Scholar]
- 6. LaMontagne AD, D’Souza RM, Shann CB. Socio-demographic and work setting correlates of poor mental health in a population sample of working Victorians: application in evidence-based intervention priority setting. Int J Ment Health Promot. 2012;14(2):109-122. doi: 10.1080/14623730.2012.703048 [DOI] [Google Scholar]
- 7. Burns K, Schroeder EA, Fung T, Ellis LA, Amin J. Industry differences in psychological distress and distress-related productivity loss: a cross-sectional study of Australian workers. J Occup Health. 2023;65(1):e12428. doi: 10.1002/1348-9585.12428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Australian Bureau of Statistics. 2020-21 National Health Survey: first results methodology. Published 2022. Accessed December 10, 2023. https://www.abs.gov.au/methodologies/national-health-survey-first-results-methodology/2020-21
- 9. Kessler RC, Andrews G, Colpe LJ, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32(6):959-976. [DOI] [PubMed] [Google Scholar]
- 10. Andrews G, Slade T. Interpreting scores on the Kessler Psychological Distress Scale (K10). Aust N Z J Public Health. 2001;25(6):494-497. [DOI] [PubMed] [Google Scholar]
- 11. Lace JW, Greif TR, McGrath A, et al. Investigating the factor structure of the K10 and identifying cutoff scores denoting nonspecific psychological distress and need for treatment. Ment Health Prev. 2019;13:100-106. [Google Scholar]
- 12. Brijnath B, Antoniades J, Temple J. Psychological distress among migrant groups in Australia: results from the 2015 National Health Survey. Soc Psychiatry Psychiatr Epidemiol. 2020;55(4):467-475. [DOI] [PubMed] [Google Scholar]
- 13. Temple JB, Kelaher M. Is disability exclusion associated with psychological distress? Australian evidence from a national cross-sectional survey. BMJ Open. 2018;8(5):e020829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Anderson TM, Sunderland M, Andrews G, Titov N, Dear BF, Sachdev PS. The 10-item Kessler psychological distress scale (K10) as a screening instrument in older individuals. Am J Geriatr Psychiatry. 2013;21(7):596-606. [DOI] [PubMed] [Google Scholar]
- 15. Trollor JN, Anderson TM, Sachdev PS, Brodaty H, Andrews G. Age shall not weary them: mental health in the middle-aged and the elderly. Aust N Z J Psychiatry. 2007;41(7):581-589. [DOI] [PubMed] [Google Scholar]
- 16. Australian Department of Health. Primary mental health care minimum data set: scoring the Kessler-10 Plus. Published 2020. Accessed February 21, 2024. https://pmhc-mds.com/doc/pmhc-scoring-k10p.pdf
- 17. Bryan ML, Bryce AM, Roberts J. Presenteeism in the UK: effects of physical and mental health on worker productivity. Sheffield Economic Research paper series 2020005. Published 2020. Accessed 21 January, 2024. https://ideas.repec.org/p/shf/wpaper/2020005.html
- 18. Australian Bureau of Statistics. Australian and New Zealand standard industrial classification 2006 (ABS Cat No. 1292.0). Published 2006. Accessed 1 February, 2024. https://www.abs.gov.au/statistics/classifications/australian-and-new-zealand-standard-industrial-classification-anzsic/latest-release.
- 19. Birrell CL, Steel DG, Batterham MJ, Arya A. How to use replicate weights in health survey analysis using the National Nutrition and Physical Activity Survey as an example. Public Health Nutr. 2019;22(18):3315-3326. doi: 10.1017/S1368980019001927 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Australian Bureau of Statistics. Labour force, Australia, detailed (Cat. 6291.0.55.003). Published 2021. Accessed April 30, 2023. https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia/latest-release
- 21. Australian Bureau of Statistics. Kessler 5 Category (Psychological Distress) for Working Adults Aged 18-64 Years (2017-18) [National Health Survey, Tablebuilder]. National Health Survey, 2017-18. Australian Bureau of Statistics; 2018. Accessed July 15, 2023. tablebuilder.abs.gov.au/webapi/jsf/tableView/tableView.xhtml [Google Scholar]
- 22. Enticott J, Dawadi S, Shawyer F, et al. Mental health in Australia: psychological distress reported in six consecutive cross-sectional national surveys from 2001 to 2018. Front Psychiatry. 2022;13:815904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Butterworth P, Watson N, Wooden M. Trends in the prevalence of psychological distress over time: comparing results from longitudinal and repeated cross-sectional surveys. Front Psychiatry. 2020;11:595696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Butterworth P, Schurer S, Trinh TA, Vera-Toscano E, Wooden M. Effect of lockdown on mental health in Australia: evidence from a natural experiment analysing a longitudinal probability sample survey. Lancet Public Health. 2022;7(5):e427-e436. doi: 10.1016/s2468-2667(22)00082-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Pierce M, Hope H, Ford T, et al. Mental health before and during the COVID-19 pandemic: a longitudinal probability sample survey of the UK population. Lancet Psychiatry. 2020;7(10):883-892. doi: 10.1016/s2215-0366(20)30308-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Ellis LA, Dammery G, Wells L, et al. Psychological distress and digital health service use during COVID-19: a national Australian cross-sectional survey. Front Psychiatry. 2022;13:1028384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Milner A, Krnjacki L, LaMontagne AD. Psychosocial job quality and mental health among young workers: a fixed-effects regression analysis using 13 waves of annual data. Scand J Work Environ Health. 2017;43(1):50-58. [DOI] [PubMed] [Google Scholar]
- 28. Arena A, Petrie K, Deady M, Glozier N, Morris R, Harvey S. Has the mental health of working Australians changed over the last 20 years? modern work: how changes to the way we work are impacting Australians’ mental health. Published 2021. Accessed January 21, 2024. https://www.blackdoginstitute.org.au/workplace-mental-health-white-paper/
- 29. Donnelly R, Farina MP. How do state policies shape experiences of household income shocks and mental health during the COVID-19 pandemic? Soc Sci Med. 2021;269:113557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Klein JW, Tyler-Parker G, Bastian B. Measuring psychological distress among Australians using an online survey. Aust J Psychol. 2020;72(3):276-282. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-aph-10.1177_10105395241306477 for Workforce Psychological Distress and Absenteeism in Australia: The Correlates of Industry, Age, and Gender by Kristy Burns, Louise A. Ellis, Abilio De Almeida Neto and Janaki Amin in Asia Pacific Journal of Public Health