Sarcoidosis: social predictors of severity at presentation

Social predictors including demographic characteristics, socioeconomic status (SES), and barriers to care affect health status 1, 2. SES is used to classify people by health status, but SES-related health status may result from social opportunity and health behaviour. Both may be affected by barriers to healthcare, particularly financial limitations on access to medical services. Low SES and barriers to healthcare are associated with a worse disease course and outcomes for some degenerative, pulmonary, infectious, and malignant diseases. Sarcoidosis occurs more frequently and has a more severe disease course among Blacks than Whites 3. Blacks are more likely to be of low SES and to be insured less well. Cross-sectional studies of ambulatory sarcoidosis patients showed education, income and/or public health insurance to be related to disease severity 4. Among Black patients, municipal hospital ambulatory sarcoidosis patients had more severe illness than university hospital sarcoidosis patients 5. By contrast, a study of hospitalised sarcoidosis patients in South Carolina showed no associations of disease severity with measures of income and education 6.

Nothing is known about the associations among severity of sarcoidosis at presentation (time of diagnosis), SES and barriers to care. A Case-Control Etiologic Study of Sarcoidosis (ACCESS) was a prospective, multicentre, case-control study of sarcoidosis of patients enrolled within 6 months of diagnosis 7, providing an opportunity to test the association of SES and demographic measures and barriers to care with disease severity at presentation.

Methods

Results

Discussion

Measures of income and other barriers to care are associated with sarcoidosis severity at initial presentation in descriptive analysis consistent with observations with severity and other diseases 1, 2. Low income and type of insurance were highly correlated with Black race, female sex, and barriers to care in descriptive analysis. Having low income and no or other public insurance may cumulatively discourage seeking medical care. Blacks are more likely to be unemployed or have low income jobs without insurance if employed 15. ACCESS blacks were more likely to have family income of <US$20,000 (37 versus 9% of Whites) and to have other public insurance (16 as compared with 3% of Whites). Most other public insurance is Medicaid, health insurance for those impoverished, which pays providers less, restricts access to physicians, particularly specialists, by whom ACCESS cases were diagnosed 16. The categorical nature of Medicaid eligibility is more problematic for young adult males, in contrast to females who may be eligible if pregnant or caring for children.

Even more restrictive than Medicaid is no insurance. Black males were those most likely to be uninsured (20 versus 8% for Black females and 3% for all Whites). The relatively low number of Black males in the study (79 Black males versus 163 White males) may reflect these income and insurance barriers to care. When Black males and their families are comparably insured as in a health maintenance organisation, the rate of Black males with sarcoidosis is only slightly lower than that of Black females 17.

Barriers to care linked to healthcare costs may further delay recognition of asymptomatic or mild sarcoidosis. Low income ACCESS cases were less likely to have a regular physician when compared with higher income cases (73 versus 98%). Other barriers to care i.e. delay in obtaining medicines, not seeing a physician when desired and reporting cost concerns about medical care, were also associated with sarcoidosis severity. For cases reporting delay in getting medicines and not calling a physician, cost was the most common reason given.

Physician visits require out-of-pocket expenses for related nonmedical expenses, such as transportation, babysitting, lost work income, as well as deductible and co-insurance expenditures. These costs discourage nonessential and routine care, which may account for lower rates of mild disease among those insured with low or intermediate income. However, once an individual presents to a physician, income and insurance do not delay sarcoidosis diagnosis, as shown in a substudy of the diagnostic pathway for ACCESS cases 18.

Logistical regression models support the relationship of sarcoidosis severity to SES and barriers to care. This relationship is seen in both the models of objective and subjective measures of severity. Income is consistently associated with severity, even when analysed with insurance and comorbidities/other symptoms in the same model. Among other barriers to care, delay in obtaining medication is the most consistent predictor of disease severity. A multivariate regression of a case-control study of sarcoidosis also showed difficulty obtaining a prescription and not having a regular physician to be associated with cases 19.

Regarding demographic measures, cases aged ≥40 yrs had more severe disease at presentation, by objective and subjective measures of diminished physical function and physical role limitations. Black race was associated with greater severity by objective, but not subjective, measures of disease at presentation, consistent with the possibility that Blacks are more likely to have financial barriers to care and delay seeking care. Black patients were less likely to have normal FVC % pred at presentation, and Black males were less likely to present with normal FVC % pred. The apparent severity of disease in Black males could be due to fewer Black males with asymptomatic and mild disease presenting early for diagnosis. ACCESS cases were significantly more likely to have private insurance and a regular physician than ACCESS telephone-area matched controls, suggesting people with better insurance and access to medical care are more likely diagnosed as cases.

By contrast to Black race, female sex was a predictor for all subjective, but no objective, measures. Perhaps, females experience sarcoidosis symptoms and limitations differently than males. Females in population surveys are more likely to report dyspnoea than males 20–21. Females, compared with males, with minor respiratory infections report more shortness of breath 22. In the National Health and Nutrition Examination Survey no. III (NHANES III), females had substantially lower FVC % pred than males 23. Others have reported using the SF-36 indices, that females with sarcoidosis were more likely than males to report poorer QoL 24. Differences in interpretation of symptoms by sex may affect subjective perception of disease severity; females with sarcoidosis were more likely to be depressed 24. Those depressed have a lowered threshold for discomfort, which may result in more negative responses to SF-36 questions onQoL 26–27. Similarly lower income is a predictor of sarcoidosis severity at presentation for multiple measures of severity; low FEV₁ % pred, dyspnoea, and all the SF-36 QoL indices. National cross-sectional studies using diverse QoL measures indicate that lower income is associated with worse QoL scores 28–29. When factors affecting QoL in low income populations are studied in a national study, the factors of employment status and activity limitations account for most of the variability in QoL. Other SES, demographic, and health insurance factors contribute little in additional explanation 29. Dyspnoea and reduced pulmonary function due to sarcoidosis could result in activity limitation and contribute to the lower SF-36, physical function, and rate limit scores for those of lower income. Nevertheless, in the current study's analysis, lower income remains as a separate predictive factor for severity even when numerous other demographic, disease symptoms, and social factors are considered. However, QoL is an important aspect of care not fully recognised by physicians in providing sarcoidosis care. A recent study of ambulatory sarcoidosis patients argues that QoL provides important measures of patient experience with sarcoidosis not captured by physiological measures. They conclude that physicians' assessment of severity, which is primarily influenced by physiological measures, does not sufficiently acknowledge QoL aspects of sarcoidosis important to patients 30.

Conclusions

The current study's findings indicate that income, insurance, and cost barriers to care are significantly associated with sarcoidosis severity at presentation, even after adjusting for race, sex, age, and other comorbidities/symptoms. Cost barriers to care may cause some patients, particularly those with asymptomatic or early sarcoidosis, to delay presentation. Cost barriers reflect social policies restricting financial access to medical care and drugs. Although improved public education about sarcoidosis may increase awareness of sarcoidosis and its early symptoms, cost barriers will detrimentally affect decisions to seek care and comply with therapeutic recommendations. Identifying patients earlier would permit earlier therapeutic intervention (when appropriate) to permit symptomatic relief or effect disease course 3, 31, 32. Reduced financial barriers to care are needed for earlier sarcoidosis detection and care.