Preventive care has the potential to reduce the incidence and severity of disease. Yet access to preventive care and treatment in the United States is uneven (1). Patients with substance dependence and low income may have reduced access to care, because both poverty and substance abuse are barriers to care (2). Substance abuse treatment programs have traditionally been separate from primary medical care, a situation exacerbated by the increase in "carved out" behavioral health care implemented by counties and states (3). This fragmentation in services may particularly affect low-income patients, who face other barriers to care, including lack of insurance, residence in areas where safety-net services are overburdened, and limited transportation.
Editor's note: This report is part of TRAININGrounds, an occasional series by, about, and for residents. The series editor is Joshua L. Roffman, M.D. Prospective authors should contact Dr. Roffman to discuss possible submissions to TRAININGrounds. Contact him at the Department of Psychiatry, Massachusetts General Hospital, 149 13th St., Bldg. 149, Rm. 2613, Charlestown, MA 02129 (e-mail: firstname.lastname@example.org).
Access to medical care is important. Available evidence suggests that chronic medical conditions are more prevalent among substance-dependent adults and that medical care may improve substance abuse treatment outcomes (4,5). Studies of community mental health centers have described these centers' limited ability to provide clients with medical referrals and to coordinate care (6,7). Furthermore, one survey among substance abuse treatment programs found that few had service linkages to medical care (8).
Adding a client-level perspective to this body of literature, our study describes access to care and use of preventive care services among adults in publicly funded treatment for substance dependence. Few studies have examined use of preventive medical services among patients with substance dependence (9). Given that few substance abuse treatment providers offer on-site medical care, a richer picture of clients' health care utilization and need for services can help inform programmatic changes in substance abuse treatment facilities to improve health and reduce morbidity.
For this study, we partnered with a Los Angeles County contracted provider of substance abuse treatment, Behavioral Health Services, Inc. (BHS), which operates 11 sites in the county and serves approximately 6,000 adult clients annually. BHS staff were concerned that clients had limited access to medical services. To explore this issue we conducted a cross-sectional survey among 254 outpatient and residential adult clients enrolled in treatment at seven BHS sites in Los Angeles County during June 2008.
We attempted to survey all clients at seven sites on predetermined dates in June 2008. Participants were recruited from reception areas, dining areas, and general communal areas of the sites. Clients eligible to participate were age 18 or older, fluent in English or Spanish, and not on probation for a federal offense. Trained, bilingual interviewers administered the survey.
Survey domains included health care access and receipt of recommended preventive care, self-reported health concerns and chronic conditions, measurement of body mass index (weight divided by height [kg/m2]), health status, and demographic information. Our main access-to-care measures included the following items: having a source of usual care and visiting a physician in the past 12 months. Preventive care measures were those identified by the U.S. Preventive Services Task Force as having good or fair evidence to support improved health outcomes. Measures included blood pressure screening; lipid screening; and cervical, breast, and colon cancer screenings. We measured each participant's height and weight with a portable calibrated scale and height rod and with the patient's shoes removed.
Descriptive statistics and bivariate and multivariate analyses were used to examine relationships among client characteristics and use of medical care. Senior staff from BHS and the RAND Institutional Review Board approved the study procedures.
Of 417 participants enrolled in treatment programs during the study period, 254 completed the survey (61%). Mean age was 38.4±10.5 years, and 62% were male (N=158) and 38% were female (N=96). On average, participants had been in treatment 89±80 days; two-thirds (N=163) were in residential treatment, and one-third (N=85) reported current homelessness. Forty-nine percent (N=125) reported beginning drug treatment within six months of a prison or jail release. Thirty-seven percent (N=93) were Latino, 27% (N=68) African American, 21% (N=53) white, 12% (N=30) of mixed race-ethnicity, and 4% (N=10) Native American. Forty-four percent (N=112) had less than a high school education. Sixty percent (N=152) had no medical insurance, and one-third of participants (N=84) reported their general health as fair or poor.
Although 63% (N=161) of participants reported having a usual source of care, 14% and 22% of that group reported that it was the emergency department or more than one source of care, respectively, suggesting that a smaller proportion of patients had a single source of non-emergency department care (Table 1). Lack of medical insurance was most often cited as the reason for no usual source of care. However, 70% of participants reported having a physician visit that did not involve the emergency department within the prior 12 months. Forty-seven percent (N=119) of participants visited an emergency department (16%) or doctor's office (30%) during the current substance abuse treatment episode.
Nearly all participants reported blood pressure screening, and over half (55%) reported lipid screening. Among women, half reported being screened for cervical cancer within the past year. Almost one-third of eligible women (31%) had received a mammogram within the prior 12 months. Approximately one-third of participants older than age 50 (37%) had received colon cancer screening. Forty-six percent of participants had been tested for HIV in the prior six months.
Fifty percent of clients surveyed reported one or more chronic illnesses, most commonly hypertension (22%), asthma (19%), and arthritis (19%). Forty-one percent were obese. Seventy-three percent of clients described one or more current health worries, and almost two-thirds of this group reported receiving no physician care for at least one current health concern. Arthritis, obesity, dental problems, and tobacco cessation were concerns most frequently cited.
In bivariate analyses, receipt of serum cholesterol screening, cervical and colon cancer screening, and HIV testing were each associated with both a usual source of care and medical insurance. There were no associations between a usual source of care or medical insurance and an emergency department or doctor's office visit during the current treatment episode. Participants with at least one chronic illness had greater odds of an emergency department or doctor's office visit during the current substance abuse treatment episode (odds ratio [OR]=3.10, 95% confidence interval [CI]=1.85–5.19) and of having a usual source of care (OR=2.19, CI=1.20–3.69). In multivariate analysis, use of a doctor's office or the emergency department during treatment was associated with one or more chronic illnesses (OR=2.37, CI=1.33–4.24) and current homelessness (OR=2.64, CI=1.43–4.86), after adjustment for gender, age, insurance, race-ethnicity, Alcohol Use Disorders Identification Test score for alcohol consumption (AUDIT-C), and usual source of care. Those with a usual source of care had greater odds (OR=2.59, CI=1.33–5.02) of receiving at least one preventive care screening (among serum cholesterol screening and screening for cervical, breast, and colon cancer) in multivariate analysis, after adjustment for gender, age, insurance, race-ethnicity, AUDIT-C score, current homelessness, and recent release from incarceration.
Most of the clients in our sample were uninsured. Despite this, one-half to two-thirds reported that they had accessed care or that they had a regular source of medical care in the community. There was evidence that almost half of clients utilized health care services during the current substance abuse treatment episode. Nonetheless, half of our participants did not have a usual source of care, receipt of most preventive screenings was low, and two-thirds of clients had current concerns related to their health. Although we cannot make direct comparisons with a matched sample who did not have substance dependence, available evidence suggests that even in insured populations having a usual source of care is infrequently reported and receipt of preventive care screenings is inadequate (10). For uninsured adults in particular, a usual source of care is associated with increased receipt of preventive care screenings (11) and, moreover, is important for those with chronic conditions to avoid disease complications and preventable hospitalizations (12).
Our finding that clients accessed medical care in the community may be the result of the primary care safety net system in Los Angeles County. One hypothesis suggested by our analysis is that presence of a chronic illness is associated with use of medical care while persons are in substance abuse treatment and with having a usual source of care. Studies of persons in treatment for substance abuse have demonstrated that health care use is driven by medical conditions (13) and that increases in medical care utilization occur early during episodes of intense treatment (14) for persons who are privately insured and for those who receive health care through the U.S. Department of Veterans Affairs.
Given our community partner's concern about clients' health care access, substance abuse treatment may represent a favorable time to help clients establish a regular source of medical care and increase receipt of preventive services. Moreover, clients are often motivated to improve their health upon entering treatment. For clients with a community source for usual care, the substance abuse treatment provider might assess whether clients have unmet health needs and assist clients to arrange appropriate care.
To our knowledge there are few local and national standards that guide substance abuse treatment programs' approach to facilitating and coordinating aspects of clients' care related to physical health. However, the Institute of Medicine has maintained in its Quality Chasm series that better coordinated and integrated physical and behavioral health care will result in improved outcomes for persons with mental illness, including substance dependence (15). Sustaining and building on clients' utilization of community medical care and promoting use of preventive services during treatment may meet some clients' health care needs and move us closer to the Institute of Medicine's goal.
Our study has several limitations. Because of the anonymity provided to clients, we did not collect demographic data from nonresponders, and they may have differed from respondents in important ways. Moreover, we relied on client self-report for most of our measures. Self-report is more reliable for recent events, such as use of an emergency department, involving chronic conditions and likely overestimates the use of medical care services. Our study was restricted to one drug treatment provider in Los Angeles County and was cross-sectional in design. However, the diversity in the seven sites chosen and the moderate sample size make the results applicable to other sites of our community partner and has generated meaningful hypotheses and possible programmatic actions for further exploration.
Funding for this study was provided by the Robert Wood Johnson Clinical Scholars Program.
The authors report no competing interests.