Sarcoidosis severity and socioeconomic status

Identification of patients

Adult patients (aged ≥18 yrs) were identified and entered into a sarcoidosis registry through computerized hospital ambulatory billing records for sarcoidosis-related International Classification of Diseases, adapted codes at Georgetown University Medical Center (GUMC), a private university hospital. Additional patients were identified from sarcoidosis clinic records at DC General Hospital (DCGH), a municipal hospital. While GUMC serves a socioeconomically diverse population, more patients are White and insured. DCGH serves a predominant minority (mostly Black), low-income population.

Medical charts from patients seen at GUMC and DCGH from July 1, 1992–December 31, 1996 were abstracted using a Sarcoidosis Registry Form including sociodemographic, insurance, sarcoidosis laboratory and chest radiograph data, and clinical symptoms. Seventeen per cent of medical records were abstracted twice by different abstractors. Of the 20 records abstracted, there were four discrepancies, all errors of omission.

Inclusion criteria

The registry population included all patients, aged≥18 yrs, with confirmed or suspected sarcoidosis (based on clinical diagnosis, with or without tissue biopsy) seen at least once. Patients who were mentally incapacitated (mentally retarded or psychotic), incarcerated, or non-English speaking, were excluded from further study.

Measures

A standardized telephone questionnaire was developed and pretested. This Sarcoidosis Telephone Survey Form assessed current health status, SES and demographic characteristics, and insurance status. General health status was assessed using questions taken from the Health Status Questionnaire Short-Form 36 18. Quality of life was assessed using physical and social function questions from the Medical Outcomes Study Short-Form General Health Survey 19. Data on use of services, education, income, and health insurance were obtained using National Health Interview Survey questions. Dyspnoea was assessed by the Modified Medical Research Dyspnoea Scale 20.

Data analysis

Most analyses consisted of descriptive statistics. Where applicable, data analyses included cross-tabulations, using the nonparametric contingency coefficient, to determine the degree and significance of an association between several variables related to sociodemographic, health status, and medical services usage. The contingency coefficient, based on Chi-squared, is a nonparametric test for use with nominal level variables, and expressed as a p-value significance level. p-Values <0.05 were considered statistically significant.

Study population

Of the 282 sarcoidosis registry patients, 63% were from GUMC. Most patients (78%) were seen at a pulmonary or sarcoidosis clinic. The remaining patients were seen at GUMC speciality clinics: Rheumatology (5%), Ophthalmology (14%), General Internal Medicine or Family Practice (2%), and Dermatology (1%).

Telephone survey population comparisons

Of the 282 registry patients, 256 were eligible for the phone survey. One-hundred and fifteen of these did not have a current identifiable phone number, leaving 141 eligible patients with an identifiable current phone; 22% refused participation, leaving a total interviewed population of 110.

The interviewed, telephone-accessible, eligible registry patients were compared by demographic, insurance, or symptom characteristics (table 1⇓). There were no significant differences for either comparison group for the 15 demographic, employment insurance, or sarcoidosis symptom characteristics; the interviewed and locatable groups were nearly identical, but fewer males and DCGH patients were interviewed than represented in the group of eligible registry patients. When compared by hospital and clinic of origin, interviewed patients differed significantly, with fewer DCGH clinic patients. Interviewed patients had a mean age of 46 yrs and were predominantly Black, female, seen at GUMC, and privately insured. The most common sarcoidosis symptoms were cough and shortness of breath, recorded for about one-quarter of the patients; other symptoms were uncommonly reported.

Interview group characteristics

Few patients had less than a high school education (19%). Household income was <US$20,000 for 30%, all of whom were Black, but 41% of patients had incomes >US$50,000. Twenty-seven per cent had no or public insurance (Medicaid); only 7% had no insurance. Black patients were significantly more likely to have less education, lower income, and no insurance. Most patients, however, had more than a high school education, private insurance (including Medicare), and incomes >US$20,000. Table 2⇓ reports self-perceived general health status of sarcoidosis patients in two categories derived from a five-point scale: Excellent to Poor (where responses do not add up to 100% on any table, information was not available; percentages are for known responses). About 70% of patients reported being in excellent or good health, with no significant differences by demographic characteristics. Fair-to-poor general health status was significantly related to less education, income, and having no or public insurance.

Sarcoidosis severity and socioeconomic status insurance indicators

The relationship of sarcoidosis severity to SES and insurance status is shown in table 3⇓. The interview ascertained subjective measures of sarcoidosis severity-related symptoms in the past month: dyspnoea, chest pain, cough, fever, weight loss, skin rash, uveitis or other eye problems, and fatigue. Objective severity measures were derived from chart abstracts: radiographic stage of chest disease and pulmonary function tests.

Severity of dyspnoea was significantly associated with level of education, lower income, and having no or public insurance. While nearly half of patients reported no dyspnoea, this was more likely in higher SES and privately insured patients. The number of self-reported sarcoidosis-related symptoms was unrelated to education, income, or insurance status, but the highest income patients were more likely to report no symptoms or fewer symptoms, if symptomatic.

Some patients were identified in nonpulmonary speciality clinics and others were referred from outside the hospital; radiographic and pulmonary function data were available in about one-half of the patients. For those with radiographs, more extensive lung involvement was significantly associated with lower income but not with education and insurance. Patients with incomes <US$20,000 were 3.5 times as likely to have Stage 3–4 disease than patients with incomes >US$50,000. Patients with the highest incomes were more likely (36% versus 7%) to have Stage 1 disease than patients with the lowest incomes. Pulmonary function test results were, in general, not associated with differences by income or insurance, although those less educated were significantly more likely to have moderate or severe restriction in vital capacity (40% versus 11%).

Table 4⇓ reports sarcoidosis symptom change in the past year. Few patients reported disease worsening; only by race was there a significant change in a year, with more Black than White patients reporting worsening (15% versus 8%), and more Black patients reporting improvement (45% versus 17%). No other indicator was associated with changed sarcoidosis status.

Work and activity limitations due to sarcoidosis

Patients were asked about activity limitations in the past 4 weeks due to physical or emotional disability (such as feeling depressed or anxious) in number, amount, and kind of activities. In addition, they were asked if their physical or emotional state had affected social activities.

There were consistent, significant associations among income, insurance status, and the amount and kind of limitations in the activities sarcoidosis patients could do because of both physical and emotional disability (table 5⇓). In contrast, education was inconsistently and marginally associated with activity limitations. Lower income patients were more likely to be limited in both the amount and kind of activity they could do because of physical or emotional health. Insurance was associated with each type of work limitation due to physical disability and to amount and kind due to emotional disability, with those not or publicly insured consistently worse off. Respondents reported limitations in normal social activities due to sarcoidosis on a five-point scale ranging from “not at all” to “extremely”, with the poorest and less well insured respondents significantly more likely to report social activities.

Implications

The findings have several implications. Those with no or public insurance were more likely to have activity limitations, to report limitations in both kind and amount of activity, and to report dyspnoea. Pulmonary impairment signified by dyspnoea could compromise employability and access to work-associated private insurance, particularly for manual workers. Sarcoidosis patients with functional impairment and reduced income could lose insurance or become eligible for Medicaid, public welfare insurance available for those with medical need. The poor may delay medical care or not receive Medicaid until impaired. Diagnosis may occur when disease is more advanced. Not all physicians will see Medicaid patients, thus restricting access to care. SES and insurance are not associated with worsening of sarcoidosis health status in a year, suggesting that sarcoidosis control is possible despite insurance and income restrictions. Consistent with that, there were no differences in physician visits or medication use by sarcoidosis severity. Associations of income and insurance with disease severity and consequent limitations may be mediated through delayed presentation of diagnosis and care. Delayed presentation may lead to greater physical and social impairment.

The functional impairment and possible financial consequences of sarcoidosis have not previously been documented. However, a study of sarcoidosis patients' quality of life showed fatigue and loss of energy to be major aspects of the disease for those who were symptomatic 23. Sarcoidosis patients, particularly those fatigued, were also more likely to have depressive symptoms but not psychologically impaired health. Documentation of physical and social impairment for sarcoidosis patients adds to the significance of this disease. As sarcoidosis is a disease of young adulthood, those with impairment, particularly pulmonary impairment, are affected at a time when impairment could restrict household income when income needs are highest. The USA has recently revised its welfare system, decreasing the time people can remain on welfare. This change may particularly compromise young adults impaired by diseases such as sarcoidosis. Public and private work retirement systems depend upon years of work and income; disease effects may have lifelong consequences. Furthermore, sarcoidosis is not classified as an occupationally-related disease, so affected adults are not eligible for worker's compensation.

The confounding of race by SES measures is particularly complex for sarcoidosis, as there may be a genetic basis for sarcoidosis among some Blacks 21, but if SES and insurance affect disease severity, these are factors amenable to social policy, unlike race 22.

Confirmation and elaboration of this study's findings in a study on a larger population of Black and White patients seen at the same stage of disease is desirable. Such a study would identify the relative significance of race, SES, and insurance in disease severity at presentation and over time. This may be accomplished by analysis of the findings of A Case Control Etiology of Sarcoidosis Study (ACCESS) 24, in which 718 newly diagnosed patients were identified at the time of disease presentation and followed for 2 yrs.

While sarcoidosis is uncommon, it may be more significant in its consequences than many more common disorders. The aetiology of the disease remains unknown; there is no curative therapy. Because of its age of onset and form of disability, the activities of young adults (child rearing and work) are particularly compromised. The significance of many diseases is measured by mortality, an uncommon consequence of sarcoidosis, but measures of the functional and social consequences of diseases and time of onset are less dramatic, but important health measures 25. These findings suggest that better understanding of the aetiology and treatment of sarcoidosis is a worthy health investment.