Clinical services used and types of health information sought by low-income Latino patients at a large urban, free clinic

The present study describes the clinical services used and the types of health information sought by Latino patients at a large urban, free clinic in Los Angeles County, California, USA. A subgroup data analysis of the 2005-2006 Patient Health Assessment Survey, originally administered as part of a clinic quality improvement program, was conducted using a subsample of low-income Latino patients (n = 301) who rely on the targeted clinic for usual care. The descriptive analysis explored health services utilization, including the types of health information sought by members of this clinic population and patient feedback about provider communication strategies that are conducive for mutual decision-making. Overall, healthy eating (61%) and chronic disease management (58%) were among the most frequently sought after health information by Latino patients at the clinic. Provider communication strategies, including explaining things in a way that the patient could understand and the use of non-medical language, were suggested by a majority of respondents (> 70%) as desirable ways to improve the medical encounter experience. Findings from this study may benefit quality improvement efforts in other community health clinics that provide free care to low-income Latino patients with similar backgrounds. © 2011 Californian Journal of Health Promotion. All rights reserved.


Introduction
Quality improvement efforts to enhance clinical services often require a broader understanding of patients' health services use and the type of health information they seek, especially among underserved populations or racial/ethnic subgroups who rely on free community health clinics for their usual care (Hardcastle et al., 2008;Schrop et al., 2006). In these populations, disease prevention and lifestyle behavioral modifications (e.g., counseling and brief intervention for diet, physical activity, weight management, smoking, and/or alcohol and drug use) remain some of the most effective and reliable strategies for reducing health risks and medical complications associated with chronic disease (Stead et al., 2008;McAlpine & Wilson, 2007;Peterson, 2007;Ma et al., 2004;Mellen et al., 2004). Unfortunately, limited access to care, lower educational attainment, poor emphasis on disease prevention, and significant communication inequalities have continued to contribute to worse health outcomes in these populations, as compared to their more affluent cohorts (Kontos et al., 2011;Heaton & Frede, 2006;Schrop et al., 2006).
Research evidence underscores the importance of communication and having a voice during the medical encounter to facilitate mutual decisionmaking (Tataw & Bazargan-Hejazi, 2010a). Data from the literature suggest that receiving appropriate health information and directives from physicians (Street, 1992) is associated with higher patient satisfaction (Tataw et al., 2010b;Galil et al., 2006) and greater compliance with treatment (Counsell et al., 1993), especially if the information is tailored specifically to reduce linguistic and cultural barriers. Kontos et al. (2011) have argued that racial/ethnic minorities as well as other groups from lower socioeconomic backgrounds often face significant communication barriers that negatively impact their access to and use of accurate health information for improving health. To date, much remains unknown about the patterns of clinical services use and/or the health information seeking behaviors of lower income, minority populations (Hardcastle et al., 2008;Schrop et al., 2006).
To help address this gap, the present study examined the types of clinical services used and the types of health information sought by lowincome Latino patients at a large, urban, free clinic in Los Angeles County, California, USA. The study also explored patient perceptions about communication strategies used by providers at this clinic.

Study design of the original patient health assessment survey
The Patient Health Assessment Survey (PAS) was a self-administered survey given to 475 patients who were systematically selected to participate in a quality improvement (QI) program at four primary care services delivery sites in a large urban, free clinic in the western region of Los Angeles County. This multi-site free clinic is among the largest, public-sector free clinics in the United States, with eight service locations spanning Venice, Santa Monica, Mar Vista, and Culver City. The clinic serves more than 24,400 men, women, children, teens, and seniors annually; 97% of them are low-income and nearly three-quarters are uninsured. More than 106,000 primary care, specialty care, mental health, dental, and health education visits occur each year. The PAS was conducted during the period December 2005 to January 2006. Although most of the activities of the QI program are continuous, the patient survey component is only repeated periodically. Patients were invited to participate in the 2005-2006 PAS if they met the following inclusion criteria: 1) aged 20 years and older and 2) had at least one prior visit to the clinic in the past 12 months. Participation in the survey was voluntary. Prospective respondents were approached in the waiting rooms of the clinic sites, typically during adult medicine clinic visits. Among patients who were approached and invited to participate, more than half (53%) consented to the survey and completed the questionnaire.
The self-administered PAS questionnaire contained 94-multiple-item questions that examined patient socio-demographics, personal health risk behaviors, the types of health information patients seek, the use of clinical services, and patient perceptions about the quality of care at the clinic. The questionnaire also evaluated patient satisfaction ratings of the care they received at the clinic during the preceding 12 months. The question items were adapted from several sources, including prior clinic patient health assessment surveys and other national as well as local population-based surveys such as the National Health Interview Survey (NHIS), the Behavioral Risk Factor Survey (BRFS), and the Los Angeles County Health Survey (LACHS). Details about these surveys, including their design, methods, instrument construction process, and question items, are described elsewhere (NHIS, http://www.cdc.gov/nchs/nhis.htm; BRFS, http://www.cdc.gov/brfss/; LACHS, http://publichealth.lacounty.gov/ha/hasurveyintr o.htm). The design and implementation of the 2005-2006 PAS were informed and guided by the literature on health education and framework on patient health empowerment (Hubley, 2006). The questionnaire was available in both English and Spanish.

The present study
The present study was a cross-sectional analysis of the original PAS dataset (n = 475). A subgroup analysis was conducted, focusing on data for 301 Latino patients who participated in Table 1 Percentage of survey respondents who were seeking and/or using the types of health information and clinical services listed (n=301) the original survey. Data analyses examined patients' self-reported demographic profiles, health risk behaviors, interests in the different types of health information available, and the use of clinical services at the clinic, including the Women, Infants, and Children (WIC) program, prenatal classes, diabetes self-care programs, and lifestyle-related counseling on diet, physical activity, cigarette/tobacco use, and/or alcohol and other drug use.

Key health services indicators
Key health services indicators examined in the analysis included the 14-item (Yes = 1, No = 0) question on health information seeking (HIS) or the types of health information respondents reported having interests in (Table 1); the 13item (Yes = 1, No = 0) question on clinical services used (CSU) ( Table 1); the 5-question battery assessing provider counseling on health habits (Table 1) Table 2).
Other health-related variables examined in the analysis included: patients' perceived physical (1-item) and mental (1-item) health status, each measured as excellent = 5, very good = 4, good = 3, fair = 2, and poor = 1; cigarette/tobacco use in the past 3 months, which was measured as daily or almost daily = 4, weekly = 3, monthly = 2, once or twice = 1, or not smoked cigarettes in the past 3 months = 0; and problem drinking, asking male respondents "how many times in the past 30 days have you had 5 or more drinks of beer, wine or liquor in a day?", or for women "4 or more drinks of beer, wine or liquor in a day?". Respondents were also asked if they have had or currently have any chronic medical conditions, such as diabetes, asthma, arthritis, hypertension, and/or depression (Yes = 1, No = 0). Demographic variables included age, gender, education, employment, country born, number of years lived in the United States, and language spoken at home. Provider did anything to make sure patient understood 2 Provider drew pictures during the visit 1

Data analysis
Descriptive and bivariate analyses were performed on key health services indicators used in the present study. Patient-reported frequencies of services use and suggestions for ways to improve clinical services were evaluated to generate frequencies for of each of the indicators (e.g., HIS, CSU, PCS, receipt of counseling).
Pearson correlation coefficients were calculated to explore the relationships between the indicators (e.g., PCS and CSU). The sum of the scores for the related items from each of these indicators was calculated. Higher values were interpreted as higher patient reports of services used and/or desired outcomes (i.e., patient preference of what they would like providers to do differently when communicating to them about care management plans).
A p-value of < 0.05 was considered to be statistically significant in the study analyses. All analyses were conducted using the SPSS Version 16.0 statistical software package (SPSS, Inc.; Chicago, IL). The study received exemption approval from the Institutional Review Boards at the Charles R. Drew University of Medicine and Science and the UCLA David Geffen School of Medicine.

Results
The majority of the study respondents (Latino patients) were female (79%), unmarried, and less than 30 years of age. Approximately 41% rated their overall physical health as poor/fair, compared to only 15% who rated their overall mental health as poor/fair. Twenty percent of the respondents reported excessive (daily or almost daily) cigarette/tobacco use; 21% reported problem use with alcohol (> 5 drinks per day for men and > 4 drinks per day for women in the past 30 days). Almost half (47%) indicated that their providers have told them they have at least one of the following chronic conditions: hypertension, heart disease, or diabetes.
Of the 14 health-related topics listed in HIS, healthy eating, maintenance of a healthy weight, and managing chronic diseases were among the top five most frequently sought after health information; collectively, at least 58% of the survey respondents reported having interest in these topics. Many respondents also expressed interest in stress management, depression care, and alternative and complementary medicine (e.g., chiropractic, acupuncture, yoga, herbal medicine).
Survey respondents frequently reported using general medical services; reproductive care (e.g., family planning, infant care, birth control/oral contraceptive use, etc); and a range of health education, counseling and other preventive services. Table 1 shows patient-reported use of  clinical services at the clinic for the period 2005-2006. Respondents provided feedback on what they thought providers have done well or could do differently to help facilitate better patientprovider communication and mutual decisionmaking. They reported or suggested an array of proactive communication strategies including explaining things in ways that could be understood and using non-medical language ( Table 2). In correlation analysis, these communication strategies were positively correlated with the use of health promotion services by these patients at the clinic (r = 0.41, p < .001). This relationship between PCS and CSU, however, is likely confounded by other contributing factors such as the type and nature of the health promotion services being offered, patients' co-morbidities, organizational structure/characteristics, and patient visit patterns.

Discussion
In the present study, we examined self-reported clinical services used and the types of health information sought by Latino patients at a large, urban, free clinic in Los Angeles County. The study yielded two interesting findings that may have clinical practice implications for this underserved population.
First, a substantial percentage of Latino patients who participated in the 2005-2006 PAS reported desiring more information about health promotion and self-care strategies for managing chronic disease. They were interested in such alternative treatment modalities as acupuncture, chiropractic, yoga, and stress management (Table 1). Together, these results suggest an unmet demand for self-care information in this population. Kontos and co-workers (2011) have previously studied similar instances of unmet demands or needs in minority populations, concluding that access barriers to quality health information can exacerbate disparities in health outcomes among disadvantaged populations. Communication inequalities such as these and others can perpetuate gaps in health knowledge and beliefs among lower income individuals (i.e., less empowered to seek out accurate information and to act), as compared to those from a more affluent background (Kontos et al., 2011). According to Johnson and colleagues (2004), the patient's voice can be strengthened through training to help address these communication inequalities.
Second, a substantial percentage of Latino patients who participated in the survey reported a preference for more proactive provider communication strategies (e.g., use of nonmedical language, simple explanations, more time spent explaining care management plans, etc.). As communication strategies are generally not costly nor labor-intensive to implement, institutional investments in ongoing provider training and in building a practice environment which supports the usage of these types of strategies for empowering patients and for facilitating mutual decision-making (e.g., allowing more time for providers to talk to their patients and for patients to ask questions) would seem reasonable and feasible to do (Greene & Yedidia, 2005;Chakravarthy et al., 2002). Evidence from organizational research suggests that if system-level changes are undertaken with the support of managers and staff, the adoption of these types of patient-centered communication strategies into routine clinical practice could face minimal institutional resistance and result in salutary impacts on patient health outcomes in the long-term (Greene & Yedidia, 2005;Chakravarthy et al., 2002; Committee on Quality of Health Care in America, 2001). Use of patient-centered communication is timely for patient populations such as the population described in the present study, as the reported rates of poor physical and mental health and risky behaviors, including smoking and problem drinking, are relatively high in this subgroup.

Study Limitations
The generalizability of the study findings is subject to at least three limitations. First, the study sample is not representative of the general population in Los Angeles County or of the insured Latino groups. Instead, the sample is more representative of uninsured or underinsured Latino patients who rely upon public sector clinics in Los Angeles County for selected clinical services. Second, self-report and selection bias may have overestimated or underestimated the respondents' interests and their reported use of clinical services. Third, the study design may be inadequate for evaluating other factors which may influence health information seeking behaviors, patient preferences and satisfaction, and quality of clinical services delivered. For example, the original survey did not specifically ask about health literacy or cultural tailoring of health education materials for Latino patients. These two factors, along with others, are likely to be just as important as health information availability or use of proactive provider communication strategies for enhancing quality of care in this low-income clinic population.