Asthma and Serious Psychological Distress: Prevalence and Risk Factors Among US Adults, 2001-2007 FREE TO VIEW

In 2007, more than 17 million adults in the United States reported current asthma.¹ Because no cure for asthma exists, effective control requires routine access to medication, medical care, appropriate self-management, and environmental precautions to avoid triggers that make symptoms worse.² This regimen, however, can be particularly challenging for the one of 17 adults diagnosed with a serious mental illness in the United States.³ The link between respiratory health and mental health has been well documented.^4-15 Among adults with asthma in clinical and community-based settings, rates of anxiety and depression twice those of the US population have been identified^4,5 and associated with increased severity of airway obstruction,⁶ functional impairment,^7-9 risk of hospitalization,⁸ and a decreased ability to manage symptoms.^9-15

In the last decade, nationally representative data from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System¹⁶ have been used to describe findings about asthma and mental health to the general US population.^17-21 Studies have found that adults with asthma have significantly more physically and mentally unhealthy days,^17,21 days with activity limitations,^18,19 a higher prevalence of frequent mental distress,²⁰ lower health-related quality of life (HRQOL),^18,20,21 and are more likely to report smoking, physical inactivity, and low income.^19-21 An association between current depression, HRQOL, and poor asthma management has also been documented.²¹ Taken together, previous investigations suggest adults with asthma are at increased risk for poor mental health; however, it is worth noting that the use of single years of data^17-21 or single-item outcome measures^17-20 may limit the reliability and validity of published estimates.²² Moreover, many results have yet to be replicated with other surveys or representative samples.

The National Health Interview Survey (NHIS) is a multipurpose survey that provides data on mental health and HRQOL. This study expanded on previous findings by combining 7 years of data from the NHIS (2001-2007) and using validated outcome measures for mental health and HRQOL. The objective of this study was twofold: (1) to estimate the prevalence of serious psychological distress (SPD) among US adults with current asthma, and (2) to examine the odds of SPD among adults with and without asthma by demographic, behavioral, and health-related factors.

The NHIS is an annual household survey of a sample of the civilian, noninstitutionalized population of the United States conducted by the National Center for Health Statistics.²³ We used data collected from 2001 to 2007, which provided detailed health and demographic information on all adult members (aged 18 years or older) of sampled households. Earlier years were not included because specific health data were unavailable. From 2001 to 2007, a total of 205,644 adults completed the interview, representing an average response rate of 82.5%. We restricted analyses to adult respondents who provided complete data on lifetime and current asthma history, psychological distress, and other covariates.

The NHIS included two questions on asthma: “Have you ever been told by a doctor or other health professional that you had asthma?” and “Do you still have asthma?” We classified respondents as having current asthma if they answered “yes” to both questions.

We classified race/ethnicity into the following categories: non-Hispanic white, non-Hispanic black, Hispanic, and non-Hispanic other. Because there were so few respondents in other racial/ethnic categories (eg, American Indian/Alaska Native and Native Hawaiian or Other Pacific Islander), we aggregated these records in the non-Hispanic other category. Additional demographic covariates included age group (18-34 years, 35-64 years, 65 years and older), education level (some high school, high school graduate, some college, college graduate), marital status (never married, married, widowed/separated/divorced), and employment status (currently working, retired, not working). The family income to federal poverty level (FPL) ratio²⁴ was calculated by dividing family income by the appropriate poverty threshold for each year, adjusting for family size. For the 7-year study period, values for the family income to FPL ratio were not available for 20.5% of the population surveyed; missing values were imputed based on procedures specified by National Center for Health Statistics and described elsewhere.²⁵ We categorized the income to FPL as ≤ 0.99, 1.00 to 1.99, 2.00 to 3.99, or ≥ 4.00.

Health-related behaviors and circumstances (smoking,^26,27 alcohol use,²⁸ and BMI²⁹) and chronic health conditions (hypertension, coronary heart disease, heart disease, stroke, emphysema, cancer, diabetes, chronic bronchitis, and arthritis^30,31) were included in the analysis based on their documented association with asthma or psychological distress. Smoking status and alcohol use were categorized as “current,” “former,” or “never.” Respondents were considered normal or underweight if their BMI²³ was ≤ 24.99, overweight if their BMI was 25.00 to 29.99, and obese if their BMI was ≥ 30.00. The number of chronic conditions other than asthma was calculated by respondents answering “yes” to having ever been told by a health professional that they had the corresponding chronic condition, based on a list provided by NHIS.²³ Respondents who reported they had “borderline diabetes” were coded as having diabetes.

The Kessler 6 (K6) Nonspecific Psychological Distress Scale³² is a six-item self-report measure included in the Sample Adult Core portion of the 2001-2007 NHIS.²³ Respondents were asked to report how often they felt “sad,” “nervous,” “restless,” “hopeless,” “worthless,” or that “everything was an effort” in the past 30 days. Responses were scored on a five-point Likert scale, ranging from 0 meaning “none of the time” to 4 meaning “all of the time.” A sum score from 0 to 24 was produced, corresponding to the level of distress, and a standard cutoff score ≥ 13 was considered indicative of “serious mental illness accompanied by moderate to severe psychological distress.”^32-34 The K6 has been used widely to screen for Diagnostic and Statistical Manual-IV mood and anxiety disorders in the general population^34,35 and has demonstrated excellent internal consistency (Cronbach α = 0.89) across subsamples.³⁵

The Health and Activities Limitation Index (HALex)³⁶ incorporates information about perceived health and activity limitations into a composite quality-of-life score.³⁷ General health was measured with the question: “Would you say your health is excellent, very good, good, fair, or poor?” Activity limitations were assessed based on the questions: “Because of a physical, mental, or emotional problem: (a) Do you need the help of other persons with personal care needs? (b) Do you need the help of other persons in handling routine needs? (c) Are you unable to work at a job or business? (d) Are you limited in the kind or amount of work you do? and (e) Are you limited in any way?” Responses were inserted into a classification matrix and scored, ranging from 0.1 (most dysfunctional) to 1.0 (healthiest).³⁶ The HALex has demonstrated usefulness as a HRQOL measure among adults with chronic diseases in clinical and national studies, particularly in examining outcomes by health-related factors such as smoking and obesity.^36-39

The study sample was weighted to produce national prevalence estimates for adults aged 18 years or older, and weights were divided by seven to adjust for multiple years. We used SAS version 9.1⁴⁰ PROC SURVEY FREQ (SAS Institute, Inc.; Cary, NC) for univariate analyses, PROC SURVEY MEANS to calculate sample averages and 95% CI, and SUDAAN version 10⁴¹ (RTI International; Research Triangle Park, NC) to calculate prevalence estimates, adjusted odds ratios (AOR), and 95% CIs, after adjusting for imputed income data and the complex sample design of NHIS. Average annual HALex and K6 scores were plotted and compared graphically. Multivariate logistic models were developed to adjust for all study variables and control for confounding.⁴² Unless otherwise noted, only differences significant at the 0.05 level are discussed.

Of the 205,644 adults who completed the survey, we excluded 18,906 respondents with incomplete or missing data. The final analysis was based on a total of 186,738 (91.2%) records. From 2001 to 2007, the average annual prevalence of current asthma was 7.0% (data not shown). Compared with adults without asthma, a greater proportion of adults with asthma were female, non-Hispanic black, had a family income below the FPL, were not working, not married, current smokers, and obese (Table 1). The prevalence of SPD among adults with asthma (7.5%) was more than double that of the overall US population (3.0%) and adults without asthma (2.6%). The prevalence of reporting three or more health-related comorbidities was significantly higher in adults with asthma (21.5%) than adults without asthma (8.4%).

A simple plot comparing average annual HALex and K6 scores suggests a negative association between HRQOL and SPD, with decreasing values for the HALex corresponding to increasing scores for the K6, regardless of asthma status (Fig 1). Adults with asthma had consistently worse indicators of HRQOL and SPD. The average annual HALex score for adults with asthma (0.754, 95% CI, 0.749-0.760) were significantly lower than score for adults without asthma (0.857, 95% CI, 0.856-0.859).

Table 2 shows the average annual prevalence and odds of reporting SPD in adults with and without asthma. Among adults with asthma, the prevalence of SPD was highest among the following subgroups: women aged 35 to 64 years, Hispanics, adults with some high school education, those below the FPL, previously married, and not working. The prevalence of SPD was four times higher (16.7%) among adults not working compared with other categories of employment status, and five times higher (16.0%) among adults with income below the FPL compared with those in the highest income category. The proportion of SPD was higher among adults with asthma who report current smoking, obesity, former alcohol use, or having three or more comorbid health conditions. A multivariate logistic model, including all study variables, showed all risk factors, except BMI and marital status, were significantly associated with SPD in adults with asthma. Controlling for all covariates, adults with asthma who were not working (AOR, 2.5; 95% CI, 2.0-3.1), current smokers (AOR, 2.4; 95% CI, 1.9-2.9), and had three or more comorbid conditions (AOR, 3.9; 95% CI, 2.8-5.3) had significantly higher odds of SPD. Respondents with asthma reporting Hispanic ethnicity (AOR, 1.4; 95% CI, 1.1-1.8) had significantly higher odds of SPD compared with those of Hispanic ethnicity without asthma (AOR, 1.0; 95% CI, 0.9-1.1).

Table 3 shows the adjusted odds of self-reported fair/poor health and measures of activity limitation stratified by asthma and SPD status. Overall, adults with either asthma, SPD, or asthma and SPD were significantly more likely to report worse HRQOL than adults with neither condition. Adults with asthma had significantly higher odds of reporting fair/poor health and various functional limitations compared with adults without asthma or SPD. Adults with asthma and SPD had significantly worse HRQOL for every component of the HALex than adults with asthma alone. Adults with SPD and adults with both asthma and SPD had 4.6 (95% CI, 4.2-5.0) and 6.6 (95% CI, 5.4-8.0) times the odds of reporting fair/poor health compared with adults with neither condition.

Our findings confirm that SPD is a prevalent condition among adults with asthma in the United States. We found an inverse dose-response relationship between SPD and HRQOL for all adults, regardless of asthma status, although adults with asthma tended to report worse mental health and quality of life. Overall, a similar pattern of risk factors was found to predict both the independent occurrence of SPD and the co-occurrence of asthma and SPD. These results support those of previous investigations.^17-21 Our use of several years of nationally representative data in combination with validated measures, such as the K6 and HALex, contributes to previous findings by producing more reliable estimates of SPD among adults with and without asthma and describing its relationship with HRQOL.

From 2001 to 2007, we found that the prevalence of SPD was 2.5 times higher among adults with asthma compared with those without asthma (7.5% vs 2.6%). Our estimate falls within the range (5.6%-9.0%) reported by other studies of mental health among adults with chronic illnesses, such as diabetes⁴³ and arthritis,⁴⁴ and is similar to emerging evidence from states that have estimated SPD using the Behavioral Risk Factor Surveillance System.^45,46 In contrast, other prevalence estimates have ranged from 3.8% for generalized anxiety disorder to 19.4% for depression among adults with current asthma,²¹ 21.3% for major depression among adults with asthma/chronic bronchitis/emphysema,⁴⁶ and 22.2% for panic attacks.⁴⁷ Although our study used the optimal K6 cutoff for assessing SPD in the general population, more sensitive cutoffs have been proposed to identify mild to moderate cases of specific disorders (ie, depression).³⁴ Further analysis of validated mental health measures in annual health surveys may improve our understanding of the observed variation in prevalence and facilitate public health surveillance for mental health in the United States.

Reduced quality of life has been associated with both the prevalence¹⁷ and onset of asthma,⁴⁸ yet we found that HRQOL decreased as SPD increased for all persons, regardless of asthma status. Persons with SPD and no asthma also reported substantially worse HRQOL and more work-related limitations compared with those with asthma but no SPD. Similar findings have been reported indicating that experiencing psychological distress may have a greater impact on quality of life than asthma alone.⁴⁹ The National Heart, Lung, and Blood Institute has long supported tracking mental health functioning as one dimension of HRQOL for persons with asthma,² and our results highlight the adverse impact that any level of psychological distress may have on health, well-being, and the ability to manage a chronic disease like asthma.

The understanding of the relationship between respiratory and mental health described in this report is complicated by several factors. For example, we found a similar pattern of risk factors predicted both SPD and the co-occurrence of SPD and asthma. Persons with asthma also reported other chronic conditions more frequently, compared with those without asthma. However, psychological distress can result from the presence of any chronic illness⁵⁰ or stress associated with sociodemographic characteristics.⁵¹ Furthermore, it may be difficult for persons to differentiate between asthma symptoms and respiratory symptoms caused by anxiety.⁵² In addition, some asthma medications (eg, β agonists) can cause symptoms similar to those of panic disorders.⁵³ Other studies suggest⁵³ that a large proportion of adults with asthma could benefit from the implementation of collaborative-care models^54,55 that integrate mental health screening and services in primary care settings. Other models of health care coordination, such as self-management and proactive integrated care, have been shown to improve quality of life and outcomes in respiratory disease.⁵⁶

This study was subject to at least the following limitations. First, responses to the NHIS are not confirmed by medical records or follow-up with health-care providers, so inaccurate reporting of asthma or mental health symptoms may result in misclassification during analysis. Second, responses may be subject to recall bias resulting in the underestimation of study outcomes. Third, the NHIS is a cross-sectional survey; thus, we cannot determine whether the onset of asthma or other factors preceded the occurrence of psychological distress. Finally, our findings are limited by the possibility that persons experiencing SPD are less likely to participate in surveys; hence, our findings might underestimate the actual occurrence of mental illness.

This study demonstrates the importance of SPD as a marker of mental health among adults with asthma and emphasizes the necessity of identifying and treating psychologic distress for every adult. SPD is a powerful predictor of quality of life and has a synergistic effect with asthma on health status. Further insight is needed to identify determinants for change in health behaviors and outcomes in asthma care. Resulting interventions should be developed, implemented, and evaluated to reduce SPD among persons with asthma and other chronic diseases.

Author contributions: Mr Oraka, Dr King, and Dr Callahan had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Mr Oraka: contributed to study concept and design, analysis and interpretation of data, drafting of the manuscript, statistical analysis, and study supervision.

Dr King: contributed to study concept and design and drafting of the manuscript.

Dr Callahan: contributed to critical revision of the manuscript for important intellectual content.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.