Association of Physical Health Indices With Antidepressant and/or Anxiolytic Use in Former US Coal Miners

Abstract

Objective:

Most health research on coal miners has focused primarily on cardiorespiratory health and injuries, despite the presence of significant risk factors for poor mental health in the industry, including illness and pain, traumatic experiences, and socioeconomic stressors.

Methods:

Clinical and occupational data from coal miners examined in a West Virginia clinic between 2004 and 2015 were used to examine associations between physical indicators of health and antidepressant/anxiolytic medication use (proxy for depression/anxiety) through multivariable models.

Results:

Antidepressant/anxiolytic use was prevalent (35%) in this population. Lung function impairment, hypertension, cardiovascular disease, chronic pain, and trauma significantly increased the odds of antidepressant/anxiolytic medication use.

Conclusions:

This study reveals a high burden of depression and/or anxiety among US coal miners and underscores the importance of evaluating this workforce for their mental health, which has largely been ignored to date.

Coal miners have significant risk factors for poor mental health outcomes, including chronic illness and pain, primary and secondary trauma, and socioeconomic stressors.^1–6 Although the majority of occupational health research on coal miners to date has focused on cardiorespiratory health and physical trauma, there is a paucity of systematic data collection regarding the mental health of these workers, including psychological, social, and emotional health and well-being. To date, there are only two studies of mental health symptomatology in US coal miners.^4,7 One study of 2808 predominantly former miners seen in a southwestern Virginia Black Lung Clinic found symptoms of depression (37%), anxiety (39%), and post-traumatic stress disorder (PTSD) (26%) were common. Hypoxemia, as defined by current supplemental oxygen use, was significantly associated with higher rates of anxiety and depression symptoms.⁴ A recent random-digit dialing study of coal miners in Appalachia found similarly high rates of depression (23%), anxiety (16%), and PTSD (25%). Clinically significant depression and anxiety were more frequently reported by coal miners than non-miner participants.⁷

Chronic obstructive pulmonary disease (COPD) and other chronic lung diseases have been associated with depression and anxiety in the general population.^8–14 The risk of developing depression is significantly higher among those with COPD, and that risk increases with COPD severity.¹¹ Conversely, depression and anxiety have been shown to increase morbidity from heart disease, COPD, and stroke.¹⁵ Coal miners are known to have high rates of obstructive lung disease and lung function impairment.^3,6,16–20 Recent data from federally funded US Black lung clinics demonstrated that the age-standardized prevalence of airflow obstruction among former miners ≥45 years of age was 18.9% overall, including 12.2% among never smokers. The highest burden of disease was found in Central Appalachian (Kentucky, Virginia, and West Virginia) miners.^6,21 Depression is also prevalent, as high as 20%, among subjects with cardiovascular disease (CVD).²² Several studies have found that depression increases the risk of developing CVD, and CVD increases the risk of depressive symptoms; thus, there is a bidirectional causal relationship between symptoms of depression and CVD through a combination of behavioral and biological mechanisms.^22–25

Even though reports of fatal and nonfatal injury have declined over the last 40 years,^26,27 coal miners remain at increased risk of work-related injury compared with other mining sectors²⁸ and other industries.²⁹ Acute injury and subacute chronic injury often lead to chronic pain, which is highly prevalent among coal miners.¹ Chronic pain and depression are related and highly comorbid, and appear with increased prevalence in affected populations.³⁰

Portions of the work presented in this paper come from the published thesis of DeVaughn.³¹ The aim of this study was to estimate the prevalence of antidepressant and anxiolytic medication use and their associations with multiple indices of physical health in former US coal miners.

METHODS

Study Population

This is a retrospective cross-sectional study of physical and mental health among a population of former coal miners evaluated at a clinic specializing in disability evaluations for coal mine dust lung disease (CMDLD) in southern West Virginia between 2004 and 2015. De-identified demographic, clinical, and occupational information about these miners was obtained from structured clinical notes and test results. Demographic data included age, race/ethnicity, and sex. Data from occupational history narratives were extracted and included total years of coal mine employment and mining occupations. Occupations were classified as high risk for coal mine dust exposure based on the US MSHA “designated occupations” list.³² Most miners (81%) reported between three and six occupations in their mining careers.

The disability evaluations included pre-bronchodilator spirometry, diffusing capacity of the lungs for carbon monoxide (D_LCO) testing, and chest radiograph. Percent-predicted (PP) values and lower limits of normal (LLN) for pre-bronchodilator forced expiratory volume in 1 second (FEV₁), forced vital capacity (FVC), and D_LCO were calculated using the Global Lung Function Initiative reference equations.³³ FEV₁ impairment was defined as FEV₁ < LLN. The classification for severity of FEV₁ impairment, as described by Pellegrino et al,³⁴ was used to characterize FEV₁ of <70% predicted as “moderate to severe” impairment. Patterns of spirometric abnormalities were defined as follows:

1. Restrictive: FVC < LLN, and FEV₁/FVC ≥ LLN

2. Obstructive: FVC ≥ LLN and FEV₁/FVC < LLN

3. Mixed obstructive and restrictive (“Mixed”): FVC < LLN and FEV₁/FVC < LLN

D_LCO impairment was defined as D_LCO < LLN.

Chest radiographs were classified according to the International Labour Office (ILO) system by B Readers certified by the US National Institute for Occupational Safety and Health.³⁵ Simple coal workers’ pneumoconiosis (CWP) was defined as the presence of small opacities with long-axis diameter of ≤1 cm, a profusion of 1/0 or greater, and an absence of large opacities. The presence of large opacities ≥1 cm in long-axis diameter was characterized as progressive massive fibrosis (PMF).³⁵ Those miners with complete evaluations, chest radiograph, spirometry, and D_LCO, were included in this cross-sectional study. The most recent examination results were used for those miners with multiple examinations.

Outcome Classification: Antidepressant and Anxiolytic Medication Use

All miners reported a list of their current medications to the examining physician as part of their medical history. This often included antidepressants and/or anxiolytic medications. The examining physician was evaluating miners for CMDLD and therefore is not the treating physician responsible for prescribing the medications reported by the miner. A study physician coded all unique medications listed in the data set with up to three primary indications (eg, hypertension, angina, depression, diabetes) and pharmacological categories (eg, analgesic, corticosteroid, opioid) based on clinic guidelines.³⁶ All medications with a treatment indication of depression and/or anxiety were included in the outcome definition (Table 1). No attempt was made to classify these medications into separate antidepressant and anxiolytic categories given that many of these drugs are used to treat both conditions. Clinical diagnoses of depression and/or anxiety were not available in the data set.

TABLE 1.

Distribution of Pharmacological Classes of all Medications Used in the Outcome Definition of Those Taking Antidepressant and/or Anxiolytic Medications Among a Population of Former Coal Miners Seen in a West Virginia Black Lung Clinic between 2004 and 2015 (n = 3207)

Pharmacological Drug Class	N	%
Tricyclic antidepressant	116	3.62
SSRI/SNRI/SARI	684	21.33
Benzodiazepine	304	9.48
Antihistamine	57	1.78

SSRI/SNRI/SARI, serotonin and norepinephrine reuptake inhibitors/selective serotonin reuptake inhibitors/serotonin antagonist and reuptake inhibitors.

Covariates

Body mass index (BMI) was calculated from the height and weight of the miners and was analyzed as a continuous and categorical (underweight BMI <18.5 kg/m², normal BMI 18.5–24.9 kg/m², overweight BMI 25–29.9 kg/m², and obese BMI ≥30 kg/m²) variable. Comorbidities including hypertension, CVD, cancer, PTSD, hearing loss, chronic pain, sleep problems, and sleep apnea were determined based on self-reports of these conditions. CVD included listed diagnoses of cardiovascular disease, atherosclerotic heart disease, atherosclerosis, congestive heart failure, coronary artery bypass graft, myocardial infarction, and myocardial ischemia. Cancer included any listed instance of cancer in the dataset. Using the medication classification process discussed above, all reported instances of current prescribed opioid medication use were classified and used as a proxy for severe pain.

Statistical Analysis

The distribution of demographic, occupational, and physical health indices was examined across the population and by outcome status. t Tests were used to examine difference in group means for continuous variables, and chi-squared tests were used to examine group differences for categorical variables to evaluate bivariate relationships in covariates of age, sex, smoking status, coal mining tenure, BMI, spirometry, diffusion capacity, chest radiography, prescribed opioid use, PTSD history, musculoskeletal pain, hearing loss, hypertension, and CVD by outcome (anxiolytic and/or antidepressant medication usage) status.

Associations between indices of physical health and antidepressant and/or anxiolytic medication use were assessed through multivariable logistic regression models while controlling for demographic and occupational factors. Variables were retained in the final multivariable model if they were significant independent predictors of the outcome and/or confounders of our primary variables of interest. Adjusted odds ratios (ORs) and 95% confidence intervals (CI) were derived from multivariable logistic regression models. All statistical analyses were performed in SAS V.9.4.³⁷

RESULTS

Study Population

The overall source population included 5463 miners. The miners in the study were nearly exclusively male (99.5%) and non-Hispanic White (98.3%), with ages ranging from 30 to 80 years (mean [SD], 62.5 [8.7] years) at the time of clinical examination. Never smokers comprised 32.7% of the population study, and ever smokers had a mean 26.1 pack-years of cigarette smoking as of the date of examination. The mean coal mining tenure was 27 years (SD, 8.5), and the majority (93%) worked in mining jobs with high risk of dust exposure (ie, designated occupations). The analysis was restricted to those with complete pulmonary testing—ILO chest radiograph interpretations, spirometry, and D_LCO testing (n = 3207, 59%). This subsample was similar to the overall study group in demographic and occupational characteristics, which did not differ substantially between the overall source population of these data and the analytic data sets (Table 2).

TABLE 2.

Comparison of Demographic and Occupational Characteristics in Overall Source Population and Analytical Study Population of Former Coal Miners Seen in a West Virginia Black Lung Clinic Between 2004 and 2015

	Overall Source Population (n = 5463)		Analytic Study population (N = 3207)
Characteristic	n	%	n	%
Age, mean (SD)	62.4 (9.0)	—	62.5	—
Sex
Male	5437	99.5	3193	99.6
Female	26	0.5	14	0.4
Race/Ethnicity
Non-Hispanic White	3865	70.7	3154	98
Non-Hispanic Black	65	1.2	53	2
Ever smoker	3677	67.3	2157	67.3
Pack-years, mean (SD)	26.8 (18.7)	—	25.9 (18.1)	—
Total coal mining tenure, mean (SD), yr	27.3 (8.7)	—	27.6	—
Reported use of antidepressant and/or anxiolytic medications	1885	34.5	1112	34.7

Population Differences by Antidepressant and/or Anxiolytic Medication Use Status

The prevalence of antidepressant and/or anxiolytic medication use among the analytic sample (n = 3207) was 35% (n = 1112). Benzodiazepine use was reported among 9.5% of the study population; 21.3% reported use of an SSRI, SNRI, or SARI, and 3.6% reported use of a tricyclic antidepressant (Supplemental Table 1, http://links.lww.com/JOM/C55). Those miners reporting antidepressant and/or anxiolytic medication use were slightly younger than those not taking these medications (mean age, 61.2 vs 63.3 years; P < 0.0001) (Table 3). They were more likely to be ever smokers and had a significantly greater number of pack-years (70.7%; mean pack-years, 27.3), compared with those ever smokers who were not taking these medications (65.4%; mean pack-years, 25.0). Total coal mine employment tenure was slightly lower for miners taking antidepressant and/or anxiolytic medication (mean [SD] tenure, 26.0 [8.3] years) compared with miners not taking these medications (mean [SD] tenure, 28.5 [8.7] years; P < 0.0001). There was no significant difference in the proportion of miners in designated high-dust occupations across medication usage status.

TABLE 3.

Demographic and Occupational Characteristics of Former Coal Miners Seen in a West Virginia Black Lung Clinic Between 2004 and 2015 (n = 3207) by Antidepressant and/or Anxiolytic Medication Use

	Miners Taking Antidepressant and/or Anxiolytic Medication (n = 1112)		Miners NOT Taking Antidepressant and/or Anxiolytic Medication (n = 2095)
Characteristic	n	%	n	%	P a
Age, mean (SD)	61.2 (8.4)	—	63.3 (8.5)	—	<0.0001
Sex					0.9347
Male	1107	99.6	2086	99.6
Female	5	0.4	9	0.4
Race/Ethnicity					0.0318
Non-Hispanic White	1101	99.1	2053	98.0
Non-Hispanic Black	11	0.9	42	2.0
Ever smoker	786	70.7	1371	65.4	0.0026
Pack-years, mean (SD)	27.3 (18.1)	—	25.0 (18.1)	—	0.0042
Coal mining start year, mean (SD)	1970 (9.0)	—	1968 (9.6)	—	<0.0001
Total coal mine employment, mean (SD), yr	26.0 (8.3)	—	28.5 (8.7)	—	<0.0001
Occupationsb
Hand loaders coal diggers general miners	681	61.2	1300	62.1	0.6526
Development work; rockman	645	58	1113	53.1	0.0083
Roof bolters and helpers	570	51.3	1010	48.2	0.1002
Shuttle car operators	573	51.5	1046	49.9	0.3883
Timberman and helpers	518	46.6	863	41.2	0.0034
Designated occupation	1043	93.8	1947	92.9	0.3564

^aSignificant differences in characteristics are based on P < 0.05.

^bFive most commonly listed occupations

Pulmonary Health and Antidepressant and/or Anxiolytic Medication Use

Miners taking antidepressant and/or anxiolytic medications had significantly worse lung function compared with those not taking these medications. These miners had lower mean FEV₁ PP values (80.2 vs 82.2; P = 0.004) and were significantly more likely to have an abnormal spirometric pattern (40.7% vs 36.9%; P = 0.036) than miners not taking these medications (Table 4). D_LCO PP was significantly lower among miners taking antidepressant and/or anxiolytic medications (mean, 89.3) compared with miners not taking these medications (mean, 92.7; P < 0.0001). Further, miners taking antidepressant and/or anxiolytic medications were significantly more likely to have abnormally low diffusion capacity tests (D_LCO < LLN) (22.8%) than those not taking these medications (19.3%; P = 0.023).

TABLE 4.

Pulmonary, Cardiovascular, and Other Health Indices of Former Coal Miners Seen in a West Virginia Black Lung Clinic Between 2004 and 2015 (n = 3207) by Antidepressant and/or Anxiolytic Drug Use

	Miners Taking Antidepressant and/or Anxiolytic Medication (n = 1112)		Miners NOT Taking Antidepressant and/or Anxiolytic Medication (n = 2095)
Characteristic	n	%	n	%	P a
Abnormal spirometric pattern	453	40.7	774	36.9	0.0355
Pattern of impairment					0.0088
None	659	59.3	1321	63.1
Restrictive	158	14.2	223	10.6
Obstructive	217	19.5	430	20.5
Mixed	78	7.0	121	5.8
FEV1 PP,b mean (SD)	80.2 (19.4)	—	82.2 (19.5)	—	0.0042
DLCO PP,c mean (SD)	89.3 (20.4)		92.7 (21.4)	—	<0.0001
DLCO < LLN	253	22.8	405	19.3	0.0225
Radiographic disease					0.0901
No CWP	617	51.3	1182	57.3
Small opacity CWP	420	42.5	725	35.1
PMF	65	6.2	156	7.6
PTSD history	67	6.0	30	1.4	<0.0001
Prescription opioid use	648	58.3	650	31.0	<0.0001
Antipsychotic drug use	56	5.0	26	1.2	<0.0001
Musculoskeletal Pain	1065	95.8	1950	93.1	0.0025
BMI category					0.0085
BMI < 18.5	11	0.99	9	0.43
18.5 ≤ BMI <25	137	12.3	227	10.9
25 ≤ BMI <30	341	30.7	748	35.7
BMI ≥30	622	56.0	1109	53.0
Hearing loss	841	75.6	1552	74.5	0.3376
Sleep apnea	197	17.7	239	10.3	<0.0001
Sleep problems	569	51.2	936	37.4	0.0005
Cardiovascular diseased	316	28.4	479	24.2	0.0005
Cancer	5	0.4	15	0.7	0.3618
Hypertension	714	64.2	1235	59.0	0.0037

^aSignificant differences in characteristics are based on P < 0.05.

^bForced expiratory volume in 1 second, percent predicted.

^cDiffusing capacity of the lungs for carbon monoxide, percent predicted.

^dCardiovascular disease includes the following self-reported conditions: atherosclerotic heart disease, atherosclerosis, cardiovascular disease, congestive heart failure, coronary artery bypass graft, myocardial infarction, and myocardial ischemia.

The prevalence of small opacity CWP in miners taking antidepressant and/or anxiolytic medications was 42.5% compared with the prevalence of miners not taking these medications at 35.1%. The prevalence of PMF in miners taking antidepressant and/or anxiolytic medications was 6.2% compared with the proportion of miners not taking these medications at 7.6%. These differences were not statistically significant (P = 0.09).

Other Health Findings

Among those taking antidepressant and/or anxiolytic medication compared with those not taking these medications, there were significantly higher proportions of self-reported PTSD (60% vs 1.4%; P < 0.0001), prescribed opioid medication usage (58.3% vs 31%; P < 0.0001), antipsychotic drug usage (5% vs 1.2%; P < 0.0001), reported musculoskeletal pain (95.8% vs 93.1%; P = 0.003), sleep apnea (17.7% vs 10.3%; P < 0.0001), sleep problems more generally (51.2% vs 37.4%; P = 0.001), cardiovascular disease (28.4% vs 24.2%; P = 0.001), and hypertension (64.2% vs 59%; P = 0.004). There was no significant difference in mean BMI or the proportion of miners reporting hearing loss or cancer between groups.

Multivariable Model Results

The primary pulmonary physiologic variables examined in this study were FEV_I1 PP and D_LCO PP. When assessed independently, a decrease in either FEV₁ PP or D_LCO PP resulted in significantly increased odds of antidepressant and/or anxiolytic medication usage, although the model containing D_LCO PP had a slightly better model fit than the FEV₁ PP model. However, when both predictors were assessed simultaneously, the effect of FEV₁ PP was attenuated and no longer significant, whereas D_LCO PP remained strongly associated with medication use. Therefore, the final model of the relationship between physical health and medication usage in this study included the D_LCO PP, along with reported age, coal mine employment years, smoking status, opioid use, PTSD history, cardiovascular disease, and hypertension (Table 5). Importantly, findings show that as D_LCO PP decreased, the odds of a miner taking antidepressant and/or anxiolytic medications increased significantly (OR per 10-unit change, 1.07; 95% CI, 1.03–1.11). Other important predictors of medication usage included cardiovascular diseases (OR, 1.28; 95% CI, 1.07–1.54), hypertension (OR, 1.29; 95% CI, 1.09–1.52), opioid use (OR, 2.84; 95% CI, 2.43–3.33), and PTSD history (OR, 4.67; 95% CI, 2.99–7.44). The results for age and coal mine employment years show that as each increase, the odds of miners taking antidepressant and/or anxiolytic medications decreased significantly. The number of pack-years for smoking miners was a slightly significant predictor of antidepressant and/or anxiolytic medication usage.

TABLE 5.

Associations Between Antidepressant and/or Anxiolytic Use and Selected Physical Health Indices Among Former Coal Miners (n = 3207)

Variable	Adjusted OR (95% CI)a
Age	0.98 (0.97–0.99)
Coal mine employment, yr	0.98 (0.97–0.99)
Cigarette pack-years	1.01 (1.00–1.01)
Diffusing capacity percent predicted (10-unit decrease)b	1.07 (1.03–1.11)
Prescribed opioid use	2.84 (2.43–3.33)
PTSD history	4.67 (2.99–7.44)
Cardiovascular diseasec	1.28 (1.07–1.54)
Hypertension	1.29 (1.09–1.52)

^aAdjusted odds ratios (ORs) and 95% confidence intervals (CI) are derived from a multivariate logistic regression model that includes all covariates listed in the table.

^bOdds ratio reflects a 10-unit decrease in D_LCO percent predicted.

^cCardiovascular disease includes the following self-reported conditions: atherosclerotic heart disease, atherosclerosis, cardiovascular disease, congestive heart failure, coronary artery bypass graft, myocardial infarction, and myocardial ischemia.

DISCUSSION

In this study of former US coal miners with disability evaluations at a Black lung clinic in West Virginia between 2004 and 2015, we found significant relationships between indicators of physical health and mental health. Impaired lung diffusing capacity, chronic pain (indicated by prescribed opioid use), self-reported PTSD, hypertension, and cardiovascular disease were all significantly associated with increased odds of taking antidepressants and/or anxiolytic medications. In particular, we found that as diffusion capacity decreases, the odds of antidepressant and/or anxiolytic medication use increased, which is consistent with previous literature on mental health and the negative relationship with respiratory health outcomes in patients with sarcoidosis³⁸ and interstitial lung disease.³⁹

Coal miners suffer excess morbidity and mortality from occupational injuries and coal mine dust lung diseases, including CWP and COPD.^16,40 The prevalence of CMDLD and in particular CWP and its most severe form progressive massive fibrosis (PMF) have been increasing among active and former coal miners most dramatically in the central Appalachian states of Kentucky, Virginia, and West Virginia, since the mid-1990s.^41–43 Despite the regulation of dust levels in US coal mines, it is estimated that 20% of long-tenured, active miners in the central Appalachian region have radiographic evidence of CWP.⁴³ Given the reciprocal relationship between chronic respiratory disease and comorbid mental illness, an aggressive approach to the identification and treatment miner’s mental health is clearly indicated.

The prevalence of antidepressant and/or anxiolytic use in this population of former coal miners was 35%. Benzodiazepines, the most common class of medications prescribed for anxiety, were used by approximately 10% of our study population, which is higher than the prevalence of prescribed benzodiazepine use among the US adult male population (7%) in 2015–2016.⁴⁴ However, despite the observed high rates of antidepressant and anxiolytic use in our study, our results may reflect an underestimate the prevalence of depression and/or anxiety in the study population, as data from the overall US population demonstrate that only one-third of American adults experiencing severe depressive symptoms were using an antidepressant,⁴⁵ and only 44% of adults 50 years and older with any mental illness were using prescription medications to help with their mental health.⁴⁶ The prevalence of antidepressant and/or anxiolytic medication use in this study is similar to the prevalence of depressive symptoms (37%) and anxiety (39%) among the Central Appalachian coal miners seen in a similar Black lung clinic setting. The Appalachian region had a higher prevalence of depression among Medicare recipients (17%) compared with the US overall (15%) in 2012.⁴⁷ Reported PTSD was the strongest predictor of antidepressant and/or anxiolytic medication use among former miners in this study, consistent with studies demonstrating that PTSD is highly comorbid with depression.⁴⁸ Musculoskeletal pain and prescribed opioid use were both highly prevalent and highly associated with increased odds of antidepressant and/or anxiolytic medication use in this population. This finding is consistent with an Australian study that found an association between physical pain and mental distress among coal mine workers.⁴⁹ In a recently published analysis on our study population, Friedman et al¹ found that nearly all miners reported a history of traumatic injury and chronic pain, and that chronic pain of the head/neck/spine was most strongly associated with opioid use. The rate of reported prescription opioid use among this cohort is higher than the opioid dispensing rate among US men 55 years and older from a similar time period (32% in 2008 and 25% in 2018),⁵⁰ reflecting the high burden of chronic pain in our study population.

We found no association between the severity of radiographic CWP and antidepressant and/or anxiolytic medication usage. However, previous investigations have found that, although lung function is significantly worse among those with PMF and small opacity CWP than those without radiographic disease, lung function impairment and obstructive lung disease are common even among those without radiographic disease.^6,51 Our findings are also consistent with a prior study of mental health among coal miners evaluated in a southwestern Virginia Black lung clinic, which also found no association between radiographic disease and symptoms of depression or anxiety based on reported medication usage.⁴ Findings of increased odds of antidepressant and/or anxiolytic medication use in former miners with hypertension and CVD are consistent with previous literature on the relationship between depression and depressive symptoms among those with CVD^22,23,25 and anxiety and anxiety symptoms among those with hypertension in the general population.^16,40

This study has several limitations. Information on antidepressant and/or anxiolytic medication use was assessed at the same time as self-reported history of injury, concurrent mental health conditions (eg, PTSD), cardiovascular health, and pulmonary health, and therefore, this study was unable to establish temporality in our analysis of physical and mental health indicators. Further, as there were no diagnostic data on mental health conditions in this population, we relied on self-reported use of medications that are typically prescribed for depression and/or anxiety. Some antidepressant and anxiolytic medications may be used for other health conditions (eg, smoking cessation or insomnia),⁵² although we believe the effect of this potential misclassification on the outcome to be minimal, as these medications are primarily prescribed to treat symptoms of depression and/or anxiety.^45,53 This study makes the assumption that reported antidepressant and/or anxiolytic medication use indicates ongoing, or long-term use, as previous literature of antidepressant use among US adults would suggest that the majority (60%) of those reporting use of these medications has been taking them for over2 years, with 14% reporting use for over 10 years.⁴⁵ However, we were unable to quantify the duration or dose of reported medication use for anxiety, depression, and/or pain relief. Finally, many measures of physical health were self-reported, which may have led to misclassification of other predictor variables in the present study.

Our study had a number of strengths. Our study of a large and detailed clinical data set on coal miners was demographically and occupationally representative of the population of coal miners served by federally funded Black lung clinics across the United States.⁶ There were numerous indices of physical health, including objective results of pulmonary function tests and chest radiograph classifications, to assess in relationship to mental health. The examinations were also systematically assessed by the same physician in the same clinic, resulting in consistency in the interview and evaluation of the miners. Furthermore, because miners were visiting this clinic for pulmonary health evaluations, selection bias based on the mental health status of the miners is likely minimal.

CONCLUSION

The findings of this study suggest a high burden of depression and/or anxiety among US coal miners and underscores the importance of evaluating this workforce for their mental health, which has largely been ignored to date. We also found consistent significant associations between physical health indices—including lung function, CVD, hypertension, and pain—and antidepressant and/or anxiolytic medication use. Coal miners are at significantly increased risk for poor mental health outcomes given their excess morbidity of coal mine dust lung diseases and chronic pain from traumatic injuries, exposure to trauma, and economic stressors. Our results demonstrate that Black lung clinics, which treat miners for their pulmonary health, are a potential point of evaluation and intervention for improving the mental health of coal miners. There is also an opportunity for significant engagement with employers, occupational health clinicians, researchers, and the government to prioritize funding in this area and address risk factors, develop early intervention programs for mental health, and reduce risk factors for poor mental health related to work in coal mining.

Supplementary Material

Supplemental digital contents are available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.joem.org).

LEARNING OUTCOMES.

• We found significant relationships between indicators of physical health and mental health among former coal miners with disability evaluations at a Black lung clinic in West Virginia between 2004 and 2015.

• Impaired lung-diffusing capacity, chronic pain (indicated by prescribed opioid use), self-reported post-traumatic stress disorder, hypertension, and cardiovascular disease were all significantly associated with increased odds of taking antidepressants and/or anxiolytic medications.