BACKGROUND. Twenty-three million Americans have limited English proficiency. Language barriers can have major adverse consequences in health care, but little is known about whether pharmacies provide adequate care to patients with limited English proficiency.
OBJECTIVES. We sought to evaluate pharmacies' ability to provide non–English-language prescription labels, information packets, and verbal communication, and assess pharmacies' satisfaction with communication with patients who have limited English proficiency.
METHODS. We used a cross-sectional, mixed-methods survey of Milwaukee County, Wisconsin, pharmacies. Survey questions addressed sociodemographic and language-service characteristics of pharmacies. A pharmacist or technician at each pharmacy was asked 45 questions by telephone, fax, or mail. The main outcome measures were the ability of pharmacies to provide non–English-language prescription labels, information packets, and verbal communication; and pharmacy satisfaction with communication with patients who have limited English proficiency.
RESULTS. Of 175 pharmacies, 73% responded. Forty-seven percent of the pharmacies never/only sometimes can print non–English-language prescription labels, 54% never/only sometimes can prepare non–English-language information packets, and 64% never/only sometimes can verbally communicate in non-English languages. Eleven percent use patients' family members/friends to interpret. Only 55% were satisfied with their communication with patients who have limited English proficiency. In multivariate analyses, community pharmacies had significantly lower odds of being able to verbally communicate in non-English languages, whereas pharmacies using telephone interpreting services had significantly higher odds. Pharmacies' suggestions for improving patient communication included continuing education, producing a chain-wide list of resources, hiring bilingual staff, using telephone interpreters, analyzing translation quality/accuracy of labels and information packets, and adding more languages to pharmacy software.
CONCLUSIONS. Approximately half of Milwaukee pharmacies never/only sometimes can provide non–English-language prescription labels or information packets, and approximately two thirds never/only sometimes can verbally communicate in non-English languages. One in 9 pharmacies that verbally communicate in non-English languages use patients' family members/friends to interpret. Almost half of the pharmacies are dissatisfied with their communication with patients who have limited English proficiency. Community pharmacies are less likely and pharmacies using telephone interpreting services are more likely to be able to verbally communicate in non-English languages. Study findings indicate that improvements in pharmacies' communication with patients who have limited English proficiency may result by increasing the quality and number of non-English languages in existing computer programs, hiring bilingual staff, and using telephone interpreting services when in-person interpreters are unavailable.
- health disparities
- language or communication barriers
- language services
- limited English proficiency
- health services accessibility
Nineteen percent of Americans (51.9 million) speak a non-English language (NEL) at home, and 8.6% (23.1 million) of Americans have limited English proficiency (LEP), which is defined as a self-rated ability to speak English less than very well.1 Between 1990 and 2000, the number of Americans who spoke a NEL at home grew by 15.1 million (a 47% increase), and the number with LEP grew by 7.3 million (a 53% increase).2 In the United States, 9.8 million children younger than 18 years old (18%) speak a NEL at home, 3.5 million of which have LEP.3 Census data indicate that these trends will continue to rise dramatically over the next 10 years.4
Language barriers can have a major adverse impact on access to health care and health status, health outcomes, patient satisfaction, communication, and patient safety.2,5–10 Language problems have also been shown to be a major access barrier to health care for Latino children and contribute to racial/ethnic disparities in children's health care.11–14 Nevertheless, very little is known about the impact of language barriers on the quality of care that patients with LEP receive in pharmacies. Only 3 published articles (to our knowledge) have addressed this issue,15–17 but previous work has not evaluated pharmacies' ability to provide prescription labels or information packets in diverse NELs, pharmacies' satisfaction with their ability to communicate with patients with LEP, or pharmacies' own suggestions for improving access to care at pharmacies for patients with LEP.
Our study objective was to evaluate the ability of pharmacies in a major metropolitan area to provide prescription labels, information packets, and verbal communication in NELs. Additional study aims included assessing pharmacies' satisfaction with their ability to communicate with patients with LEP, identifying pharmacies' suggestions for improving communication with patients with LEP, and examining features of model pharmacies that already provide high-quality communication with patients with LEP.
Study Design and Sample
A cross-sectional, mixed-methods survey was conducted in which attempts were made to interview a pharmacist or pharmacy technician at all Milwaukee County, Wisconsin, pharmacies, including those embedded within larger stores, such as in supermarkets or retail stores. The pharmacy database was compiled by using information from the greater Milwaukee area Yellow Pages; www.zapconnect.com, an Internet pharmacy information site; and rosters from the Pharmacy Society of Wisconsin. Surveys were completed by telephone if possible or by fax or mail if requested. Callbacks were made every 2 to 3 days until survey completion or refusal.
A non–community pharmacy was defined as any pharmacy that is clinic- or hospital-based or services a long-term care facility, with this designation made through self-identification (the specific choices available included community versus clinic, hospital, long-term care, and other). Chain- and urban-pharmacy designations were also made through self-identification, with the choices being chain versus nonchain and urban versus suburban, rural, or other. Bilingual staff were defined as pharmacy personnel, including pharmacists and pharmacy technicians, who identified themselves as bilingual. Information packets were defined as any patient-education materials that accompanied filled prescriptions, including information and instructions on how to take medications, possible adverse effects, drug interactions, and contraindications. Model pharmacies were defined as those that already provided high-quality communication with patients with LEP, as documented by a self-report that the pharmacy is able to communicate always or most of the time with patients with LEP. The LEP volume was defined as the percent of patients with LEP who were served, multiplied by the annual number of prescriptions filled by the pharmacy; this measure was constructed as a potential weight to adjust for the number of prescriptions filled but was not entered into multivariate models, because it was not significantly associated with any of the main study outcomes in bivariate analyses. All multivariate models, however, included adjustment for the number of prescriptions filled annually.
Data and Measures
The survey consisted of 45 Likert-scale (choices: never, some of the time, most of the time, or always), yes/no, and open-ended questions. Each survey took ∼10 minutes to complete. Survey questions addressed selected sociodemographic and language-service characteristics of the pharmacies. Questions were asked about the pharmacies' ability to provide NEL prescription labels and NEL information packets and to verbally communicate in NELs, with follow-up questions regarding how each of the previous processes were conducted (bilingual staff, computer programs, telephone interpreting services, or Internet interpreting services), as well as the names of specific programs or devices used. Additional questions were asked about the pharmacies' overall level of satisfaction with their ability to communicate with patients with LEP. Qualitative data on suggestions for improving pharmacies' ability to meet the needs of patients with LEP were collected by using open-ended questions. These data were analyzed by using customary qualitative analytic techniques, including margin coding and grounded theory. Themes were identified, and a relevant taxonomy was created. Finally, questions were asked to examine pharmacy characteristics associated with the various study outcomes. These characteristics included the number of years the pharmacy had been open; the annual number of prescriptions filled; the estimated proportion of patients with LEP, nonwhite patients, and patients who speak a NEL at home; and the estimated proportion of Medicaid patients. This study was approved by the institutional review board of the Children's Hospital of Wisconsin.
All statistical analyses were performed using SAS 9.1 software. Bivariate analyses of nominal variables were completed using the χ2 test or, for cases in which ≥25% of the cells had expected counts of <5, Fisher's exact test. Cochran-Mantel-Haenszel tests for nonzero correlation were used in bivariate comparisons of binary and ordinal variables. The Wilcoxon test was used for bivariate analyses that compared the medians of continuous variables. Categorical data are presented as sample size (percentage), and continuous data are presented as median (interquartile range [IQR]). A P value of <.05 was considered statistically significant.
Multivariate logistic regression was used for the following outcomes: ability to provide NEL prescription labels, ability to provide NEL information packets, ability to verbally communicate in NELs, and pharmacy's satisfaction with overall ability to communicate with patients with LEP. Ability-related outcomes were dichotomized as never/some of the time versus most of the time/always (with “some of the time” defined as multilingual staff, prescription labels, written materials, verbal communication, and/or all relevant languages available only intermittently on certain days or during certain business hours, according to the self-rated assessment by the interviewed staff member). The satisfaction outcome was dichotomized as satisfied/very satisfied versus very dissatisfied/dissatisfied/neither satisfied nor dissatisfied. All models were adjusted for pharmacy setting (community versus noncommunity) and type (chain versus nonchain), number of years the pharmacy has been open, number of prescriptions filled annually, and proportions of patients who have LEP, are nonwhite, and are covered by Medicaid. The potential interaction between community- and chain-pharmacy designations was also evaluated.
An additional multivariate model was run to examine the relationship between the use of telephone interpreting services and an ability to verbally communicate with patients with LEP. This analysis was limited to those who reported being able to communicate at least some of the time, because data on use of telephone interpreting services were not requested from those who reported never being able to communicate with patients with LEP. For this model, the dependent variable was ability to verbally communicate in NELs always/most of the time versus some of the time, and the independent variables were use of telephone interpreting services, pharmacy setting (community versus noncommunity) and type (chain versus nonchain), number of years the pharmacy has been open, number of prescriptions filled annually, and proportions of patients who have LEP, are nonwhite, and are covered by Medicaid.
Response Rate and Demographic Features
Using the Yellow Pages, an Internet pharmacy site (www.zapconnect.com), and the Pharmacy Society of Wisconsin rosters, 510 pharmacies were identified as potential survey participants (Fig 1). After duplicates were first excluded, 219 pharmacies remained, and calls were made to these pharmacies. After excluding disconnects, additional duplicates, and pharmacies that did not provide retail outpatient services, 175 pharmacies were contacted; 128 pharmacies responded to the survey, yielding a response rate of 73%. Eighty-three percent of the survey respondents were pharmacists; 79% of pharmacies were community pharmacies, 77% were chain pharmacies, and 72% were urban pharmacies (Table 1). The median percentage of pharmacy patients who speak a NEL at home was 5, and the median percentage of patients with LEP was 3. The median percentage of nonwhite patients was 35, and the median percentage of patients covered by Medicaid was 30. The median number of years that the pharmacies had been open was 9, and the median LEP volume was 1300.
Language-Service Characteristics of Pharmacies
When asked about ability to provide NEL prescription labels, 47% of the pharmacies reported that they can never or only sometimes prepare NEL prescription labels (Table 2). NEL prescription labels are most commonly prepared by using computer programs (88%), but approximately one quarter of the pharmacies use bilingual staff; telephone interpreters and other methods are used infrequently. The most common NEL prescription-label language provided by pharmacies was Spanish, followed by French, Russian, German, Vietnamese, Arabic, Chinese, Italian, Japanese, Korean, Polish, Portuguese, Tagalog, and Hmong. More than half of the pharmacies never or only sometimes can provide NEL information packets, including 15% who can never provide NEL packets. Among those capable of doing so, 95% use a computer program, and bilingual staff are used frequently. Approximately two thirds of the pharmacies never or only sometimes can verbally communicate in NELs, including 1 in 6 who never can verbally communicate in NELs. Among those able to verbally communicate in NELs, approximately two thirds use bilingual staff, one third use telephone interpretation services, and one third use other methods. Among pharmacies that are able to verbally communicate in NELs, 1 in 9 uses patients' family members or friends to interpret. Only approximately half of the pharmacies reported being satisfied or very satisfied with their communication with patients who have LEP.
Community pharmacies are significantly less likely to be able to verbally communicate in NELs compared with noncommunity pharmacies (29% vs 63%, respectively; P = .001; Table 3). Chain pharmacies are significantly more likely to be able to provide NEL prescription labels than nonchain pharmacies (58% vs 37%; P = .04; Table 3). There were no statistically significant differences between community and noncommunity pharmacies in the ability to provide NEL prescription labels or information packets or in satisfaction with the overall ability to communicate with patients with LEP. Chain and nonchain pharmacies did not significantly differ in the ability to provide NEL information packets or to verbally communicate in NELs or in overall satisfaction with ability to communicate with patients with LEP.
A higher median proportion of Medicaid-covered patients was associated with a significantly higher likelihood of a pharmacy's ability to provide NEL prescription labels (Table 4). Among pharmacies able to provide NEL prescription labels most of the time or always, the median proportion of Medicaid-covered patients was 38%, compared with 30% for those able to provide labels only some of the time or never. No statistically significant associations were found between pharmacies' number of years open, number of prescriptions filled annually, percent of patients who speak a NEL at home, percent of patients with LEP, percent of nonwhite patients, or volume of patients with LEP and the ability to provide NEL prescription labels, to provide NEL information packets, or to verbally communicate in NELs.
Use of telephone interpreting services was significantly associated with the ability to provide NEL information packets and verbally communicate in NELs (Table 5). Twelve percent of the pharmacies that were able to provide NEL information packets most of the time use telephone interpreting services, whereas none of those able to provide the packets some of the time or always used telephone interpreting services (P = .01). Only 20% of the pharmacies that reported being able to verbally communicate some of the time in NELs use telephone interpreting services, compared with 45% of those able to verbally communicate in NELs most of the time and 50% of those able to do so always (P = .01). In addition, this finding had a nonzero correlation (ie, a dose-response relationship), with use of telephone interpreting services directly correlated with ability to verbally communicate in NELs (P = .005). None of the pharmacy characteristics were found to be significantly associated with pharmacy satisfaction with the overall ability to communicate with LEP patients.
Community pharmacies were found to have significantly lower odds of being able to verbally communicate in NELs (odds ratio [OR]: 0.3; 95% confidence interval [CI]: 0.1–0.9), after adjusting for chain-pharmacy class, number of years open, number of prescriptions filled annually, percentage of patients with LEP, percentage of nonwhite patients, and percentage of Medicaid patients (Table 6). In the second multivariate model, limited only to pharmacies that can communicate verbally with patients with LEP at least some of the time, community pharmacies continued to have significantly lower adjusted odds of being able to verbally communicate in NELs (OR: 0.2; 95% CI: 0.05–0.8), and pharmacies that use telephone interpreting services were significantly more likely to be able to verbally communicate in NELs (OR: 4.2; 95% CI: 1.3–13.1), after adjusting for chain-pharmacy class, number of years open, number of prescriptions filled annually, percentage of patients with LEP, percentage of nonwhite patients, and percentage of Medicaid patients (Table 7). No interactions were documented between community- and chain-pharmacy designations.
No independent variable was found to be significantly associated with either of the remaining 2 outcome variables (ability to provide NEL prescription labels and ability to provide NEL information packets) in multivariate analyses.
Pharmacies' Suggestions for Improving Communication With Patients Who Have LEP
Pharmacies' suggestions for improving communication with patients who have LEP included continuing education classes, producing a chain-wide list of resources (when applicable to the practice setting), hiring bilingual staff members, using telephone interpretation services, analyzing the translation quality and/or accuracy of the written material (NEL prescription labels and information packets), and adding more languages to the pharmacy computer software (Table 8).
Model pharmacies shared a combination of NEL computer programs, bilingual staff, and use of telephone interpreting services when in-person interpreters were unavailable. For example, 1 study pharmacy reported that 3% of its patients have LEP. The pharmacy can print NEL prescription labels in 14 languages and prepares NEL information packets in 2 languages. They have bilingual staff who speak 4 NELs and regularly use telephone interpreting services when needed. Another study pharmacy reported 20% patients with LEP, NEL prescription labels in 13 languages, NEL information packets in 1 language, bilingual staff who speak 1 NEL, and regular use of telephone interpreting services. A third study pharmacy reported 40% patients with LEP. This pharmacy can print prescription labels and prepare information packets in 1 NEL. They have bilingual staff who speak 4 NELs and regularly use telephone interpreting services.
More than 4 billion prescriptions are written each year in the United States.18 Given that 8.6% of Americans have LEP,1 one can estimate that as many as 336 million prescriptions are written each year for patients with LEP. About half of the pharmacies surveyed in this study reported never or only some of the time being able to provide NEL prescription labels, suggesting that as many as half of the 336 million prescriptions given to patients with LEP in the United States could be written in a language that patients cannot fully understand, which poses potentially serious risks of medical errors and injury. Published studies document the danger of providing LEP patients with English-only prescription labels or instructions. Recent work detailed the case of a 10-month-old girl with iron-deficiency anemia who experienced a 12.5-fold overdose of iron and was hospitalized for iron intoxication after her parents (who had LEP) were given medication instructions and a prescription only in English. When asked in Spanish, the parents reported giving 15 mL of iron elixir based on the prescription label that read: “15 mg per 0.6 mL, 1.2 mL daily.”10 In another case, a 6-week-old infant was admitted for a barbiturate overdose caused by a 10-fold medication dosing error by a mother with LEP who did not understand the outpatient dosing instructions available only in English.19 To avoid such medical errors and injuries, pharmacies should routinely use computer translation software to provide NEL prescription labels for patients with LEP.
More than half of the pharmacies in this study could never or only sometimes provide NEL information packets to patients with LEP. This is the first study (to our knowledge) to document deficiencies in pharmacies' ability to provide NEL information packets. Pharmacy information packets contain information and instructions on how to take medications, as well as possible adverse effects, drug interactions, and contraindications—information that is essential for patient care and safety.20–23 Written information can be effective in improving patient adherence to regimens for antibiotic therapy.22 Information packets can serve as a useful addition to verbal counseling provided by the pharmacist, and patient satisfaction is higher when information packets are used.24 Lack of access to pharmacy information packets for patients with LEP may place them at greater risk for drug interactions and adverse drug effects. Additional research is needed to identify the adverse consequences experienced by patients with LEP who do not receive NEL pharmacy information packets.
Verbal Communication in Pharmacies
Almost two thirds of the pharmacies reported never or only sometimes being able to verbally communicate with patients with LEP. Verbal communication between patients and pharmacists allows patients to ask questions about medications and pharmacists to identify possible drug allergies, ensures that patients understand how to properly take and refill medications, and verifies that patients have the correct means of administering medications (including syringes, inhalers, droppers, and teaspoons). Previous studies have documented that verbal counseling by pharmacists improves patient outcomes and is associated with greater patient satisfaction.25–27 Access to trained interpreters or bilingual staff should be provided for patients with LEP in pharmacies, because studies indicate that optimal communication, patient satisfaction, and outcomes and the fewest interpreter errors occur when patients with LEP have access to trained professional interpreters or bilingual providers.5
Ad hoc Interpreter Use in Pharmacies
One in 9 pharmacies reported using patients' family members or friends to verbally communicate with patients with LEP. Studies document that use of such ad hoc interpreters (including family members, friends, untrained medical staff, and strangers from waiting rooms or the street) can be dangerous.2,5 Ad hoc interpreters are significantly more likely than professional interpreters to commit errors of potential or actual clinical consequences, with 77% of ad hoc interpreter errors on average having potential clinical consequences.2,5,28 Ad hoc interpreters also are unlikely to have adequate training on medical terminology and confidentiality, sometimes have priorities that conflict with patients, and may inhibit or preclude essential discussions regarding sensitive issues such as domestic violence, substance abuse, psychiatric illness, and sexually transmitted diseases.2,28,29 It is especially risky to have children interpret, because they are less likely to have full command of 2 languages and medical terminology (frequently making interpreter errors of clinical consequence), are especially avoidant of sensitive but important clinical issues, often are embarrassed by and ignore questions about menstruation, bowel movements, and other bodily functions,5 and may not adequately understand the legal responsibilities of parents to make treatment decisions and provide informed consent. Inadequate communication can have tragic consequences for patients.2 The use of ad hoc interpreters in pharmacies, therefore, should be discouraged and used only as a last resort.
Community Pharmacies and Communication With Patients With LEP
Community pharmacies were found to be particularly at risk for not being able to verbally communicate with patients with LEP. The reasons for this finding are not clear and could not be ascertained in this study. One can speculate that noncommunity pharmacies (which include those located in clinics, hospitals, and long-term care facilities) may have more ready access to medical interpreters or are part of larger institutions that provide more extensive language-access services. Additional research is needed to determine why community pharmacies are at a greater risk of not being able to verbally communicate with patients with LEP.
Certain study limitations should be noted. Only Milwaukee County pharmacies were surveyed, so the results may not necessarily apply to other major metropolitan areas or suburban or rural regions of the country. The findings may also be subject to nonresponse bias, because 27% of the eligible pharmacies did not respond. Given, however, that the nonresponders may be less likely to provide language services for patients with LEP, the study findings may underestimate the actual prevalence of the identified problems. The survey data collected did not allow us to determine whether bilingual staff had been systematically hired by certain pharmacies to meet the needs of their patients with LEP or the bilingual staff were an unintended and fortuitous happenstance of the hiring process. Finally, the focus of the survey and the analyses of data in aggregate did not permit us to conduct detailed identification and analysis of smaller nonchain pharmacies in ethnic neighborhoods that might have high-quality communication with patients with LEP (just as there frequently are grocery stores in ethnic communities that carry ethnic foods and have more bilingual/multilingual staff). We did, however, qualitatively search for and identify such pharmacies, some of which are highlighted in “Model Pharmacies” earlier and below.
Several pharmacies in this study reported exemplary efforts to ensure language access for patients with LEP. Computer translation software is frequently used by these model pharmacies to prepare NEL prescription labels and information packets. Although computer programs are used by 88% of study pharmacies to print NEL prescription labels and 95% to print NEL information packets, it is unclear whether the accuracy of such programs has been evaluated or validated adequately. A recent study that examined pharmacy translation software found that most pharmacies surveyed could provide prescription labels in Spanish; however, at 1 large chain pharmacy, the computer could not translate commonly used terms such as “dropper full” or “for 30 days.”15 The accuracy and validity of a translation program, therefore, should be established before the software is used. In terms of information packets, an efficient option would be for international pharmaceutical companies to make available to US pharmacies translated NEL information packets (already being used in other countries).
Model pharmacies frequently employ bilingual staff members who can verify the accuracy of NEL prescription labels and information packets. Bilingual pharmacists also verbally counsel patients with LEP, explain medication instructions, adverse effects, and contraindications, and answer patient questions. To increase the number of bilingual staff, pharmacies can target hiring pharmacists and pharmacy technicians from the surrounding community to ensure that the language needs of that area are being met. Another option is to have pharmacy schools incorporate population-appropriate NEL courses into their curricula.17 For example, Wake Forest University School of Medicine has added a medical Spanish class as a required course for students to graduate.30 Continuing education medical NEL courses could also be offered for staff members who wish to achieve NEL fluency.
Model pharmacies use telephone interpreting services when in-person interpreters are unavailable. Previous studies have shown that patients with LEP who have access to telephone interpreters are more satisfied than those who have ad hoc interpreters.5,31 Because telephone interpreting services can be awkward to use and do not allow for observation of nonverbal cues, telephone services should be used only when bilingual staff or medical interpreters are not available, such as in the case of rare language groups or cost limitations.13,32 More research is needed on the effectiveness of telephone interpreting services in the pharmacy setting.
Practice and Policy Implications
Data from the American Community Survey indicate that 16% of the population of Milwaukee County speak a language other than English at home and 7% have LEP.33 In contrast, in this study, Milwaukee County pharmacists reported a median of 5% of their patients speaking a language other than English at home and a median of 3% having LEP. This substantial gap between demographic prevalence and practitioner perception suggests that (1) pharmacists may underestimate the proportions of their patients who have LEP, (2) family members who pick up the prescriptions may be the English-proficient members of households, and pharmacists, therefore, may not have direct contact with many of their patients with LEP, or (3) patients with LEP get fewer prescriptions because they are more likely to have impaired access, no health insurance, or better health status. Either of the first 2 possibilities suggests that the problems documented in this study are more serious, because pharmacists are only aware of the “tip of the iceberg” of language barriers among their patients. For example, if the pharmacist is not aware that many patients have LEP, then he or she might not even bother to print labels in NELs or consider having translated information packets.
Physicians and other clinicians who write prescriptions should be aware that pharmacies often are unable to communicate well with their patients with LEP. We suggest, therefore, that a greater awareness among clinicians of such potential pharmacy language barriers for patients with LEP could enhance adherence and patient safety by focusing greater attention on improved communication regarding prescriptions before filling them. Improvements in pharmacy communication quality with patients who have LEP, however, might relieve clinicians from some of these burdens.
Both the federal government and individual states are major payers for prescription drugs through Medicaid and Medicare. States and the federal government, therefore, might consider stronger efforts to encourage pharmacies and prescription drug plans to improve services to patients with LEP. Pharmacies and prescription drug plans might want to consider their potential liability for harm to patients who misunderstand their prescriptions and educational materials because of language barriers. Governments, working with pharmacy benefit managers and pharmacies, could try to develop better standards of care for patients with LEP and improve access to bilingual/multilingual materials.
About half of the study pharmacies never or only sometimes can provide NEL prescription labels and information packets, and approximately two thirds never or only sometimes can verbally communicate in NELs. Almost half of the pharmacies are dissatisfied with their communication with patients who have LEP, and 1 in 9 pharmacies that verbally communicate in NELs use patients' family members or friends to interpret, which increases the risk of communication errors. Community pharmacies are less likely and pharmacies that use telephone interpreting services are more likely to be able to verbally communicate in NELs. The study findings suggest that many pharmacies may not provide adequate services to their patients with LEP, thereby limiting appropriate access to health care and increasing the risk of compromised patient safety. Improvements in pharmacies' communication with patients who have LEP may result by increasing the quality and number of NELs in existing computer translation programs used for prescription labels, hiring bilingual staff, and using telephone interpreting services when in-person interpreters are unavailable.
This study was funded through a grant from the National Heart, Lung, and Blood Institute.
We thank Hong Ji for statistical assistance with some of the analyses; Emmanuel Ngui for providing comments on an earlier manuscript draft; and Richard Hayney, RPh, and the Pharmacy Society of Wisconsin for supplying pharmacy rosters for the survey database and commenting on earlier versions of the manuscript. We owe special thanks to the 2 anonymous reviewers for their helpful comments.
- Accepted January 25, 2007.
- Address correspondence to Glenn Flores, MD, Division of General Pediatrics, Department of Pediatrics, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
This work was presented in part as a platform presentation at the annual meeting of the Pediatric Academic Societies; April 29, 2006; San Francisco, CA; the annual research meeting of AcademyHealth; June 27, 2006; Seattle, WA; and the annual meeting of the American Public Health Association; November 7, 2006; Boston, MA.
The National Heart, Lung, and Blood Institute had no role in the design or conduct of the study; the collection, management, analysis, or interpretation of the data; or the preparation, review, or approval of the manuscript.
- ↵US Census Bureau. Selected social characteristics in the United States: 2005. Available at: http://factfinder.census.gov/servlet/ADPTable?_bm=y&-geo_id=01000US&-qr_name=ACS_2005_EST_G00_DP2&-ds_name=ACS_2005_EST_G00_&-_lang=en&-_sse=on. Accessed September 21, 2006
- ↵US Census Bureau. QT-P17: ability to speak English—2000. Available at: http://factfinder.census.gov/servlet/QTTable?_bm=y&-qr_name=DEC_2000_SF4_U_QTP17&-geo_id=D&-ds_name=D&-_lang=en. Accessed July 28, 2006
- ↵US Census Bureau. P25-1130: population projections of the United States by age, sex, race, and Hispanic origin—1995 to 2050. Available at: www.census.gov/prod/1/pop/p25-1130/p251130.pdf. Accessed July 31, 2006
- ↵Flores G. The impact of medical interpreter services on the quality of health care: a systematic review. Med Care Res Rev.2005;62 :255– 299
- Siganga WW, Huynh TC. Barriers to the use of pharmacy services: the case of ethnic populations. J Am Pharm Assoc (Wash DC).1997;NS37 :335– 340
- Ku L, Flores G. Pay now or pay later: providing interpreter services in health care. Health Aff (Millwood).2005;24 :435– 444
- ↵Flores G. Language barrier. Available at: www.webmm.ahrq.gov/case.aspx?caseID=123. Accessed July 24, 2006
- ↵Ngui EM, Flores G. Satisfaction with care and ease of using health care services among parents of children with special health care needs: the roles of race/ethnicity, insurance, language, and adequacy of family-centered care. Pediatrics.2006;117 :1184– 1196
- ↵Flores G, Olson L, Tomany-Korman SC. Racial and ethnic disparities in early childhood health and health care. Pediatrics.2005;115(2) . Available at: www.pediatrics.org/cgi/content/full/115/2/e183
- Westberg SM, Sorensen TD. Pharmacy-related health disparities experienced by non-English-speaking patients: impact of pharmaceutical care. J Am Pharm Assoc (Wash DC).2005;45 :48– 54
- ↵Muzyk AJ, Muzyk TL, Barnett CW. Counseling Spanish-speaking patients: Atlanta pharmacists' cultural sensitivity, use of language-assistance services, and attitudes. J Am Pharm Assoc (Wash DC).2004;44 :366– 374
- ↵Aspden P, Wolcott J, Bootman JL, Cronenwett LR, eds. Preventing Medication Errors. Washington, DC: National Academics Press; 2007
- ↵Koren G, Barzilay Z, Greenwald M. Tenfold errors in administration of drug doses: a neglected iatrogenic disease in pediatrics. Pediatrics.1986;77 :848– 849
- ↵Koo MM, Krass I, Aslani P. Factors influencing consumer use of written drug information. Ann Pharmacother.2003;37 :259– 267
- ↵Buck, ML. Providing patients with written medication information. Ann Pharmacother.1998;32 :962– 969
- Singhal PK, Gupchup GV, Raisch DW, Schommer JC, Holdsworth MT. Impact of pharmacists' directive guidance behaviors on patient satisfaction. J Am Pharm Assoc (Wash DC).2002;42 :407– 412
- ↵Erickson SR, Kirking DM, Sandusky M. Michigan Medicaid recipients' perceptions of medication counseling as required by OBRA ′90. J Am Pharm Assoc (Wash DC).1998;38 :333– 338
- ↵Flores G, Laws MB, Mayo SJ, et al. Errors in medical interpretation and their potential clinical consequences in pediatric encounters. Pediatrics.2003;111 :6– 14
- ↵American Medical Association. Student request results in course in medical Spanish. Available at: www.ama-assn.org/amednews/2002/06/24/prsc0624.htm. Accessed July 25, 2006
- ↵US Census Bureau. Milwaukee County, Wisconsin: selected social characteristics in the United States—2005. Available at: http://factfinder.census.gov/servlet/ADPTable?_bm=y&-geo_id=05000US55079&-qr_name=ACS_2005_EST_G00_DP2&-ds_name=&-redoLog=false. Accessed December 18, 2006
- Copyright © 2007 by the American Academy of Pediatrics