Health care illiteracy: Implications for providers

Approximately 90 million adults in the United States are illiterate, a condition that frequently underlies low health literacy. Risk factors include poverty, incarceration, lack of education, immigration status, age, and race.

Zachary Hartsell, MPAS, PA-C

The author is a hospitalist PA in the Department of Internal Medicine, Mayo Clinic Hospital, Scottsdale, Ariz, and a member of the JAAPA editorial board. He has indicated no relationships to disclose relating to the content of this article.

   If you prefer to view this article in PDF form, click here.

One quarter of adults in the United States are functionally illiterate.¹ Although the US Department of Education conducted the National Adult Literacy Survey (NALS) in 1992, the problem of how this high rate of illiteracy affects health care still goes widely unnoticed. A report released last year by the Agency for Healthcare Research and Quality (AHRQ) emphasized that health care illiteracy remains prevalent and has still not been effectively addressed by health care providers.² Regulatory agencies and national organizations have noticed the problem, however. At AAPA’s 32nd Annual PA Conference in Las Vegas, Nevada, Julie Theriault, PA-C, president-elect of the Academy, listed health literacy as the theme of the 2004-2005 Host City Prevention Campaign.³

What is health literacy?

The Institute of Medicine (IOM) defines functional health literacy as “the ability to read, understand, and act on health information.”¹ This definition encompasses how the patient gathers and processes health information and how the health care provider presents information to the patient. Many PAs do not recognize the legal and financial implications of failing to identify low-literacy patients in their practice, and many are unaware of the low-cost solutions available for ensuring adequate care of low-literacy patients.

Low health literacy has several causes, including poor reading ability and learning disorders such as dyslexia. Sensory handicaps can also contribute. Cultural background can affect health-literacy status: Patients from other cultures may be unable to read or speak English, or they may be unable to read in their native language even though they can speak it. They may not understand American views about health. For example, a person born in a small village in sub-Saharan Africa will have different health views and beliefs than someone born in Iowa. Even foreign patients who can speak English may have difficulty figuring out some aspects of American health care and may not understand how to sign a consent form or fill out a Medicare application.

Further defining the problem

The NALS questioned approximately 26,000 people across the United States and used literacy levels to categorize them into five groups. The two lowest levels, NALS 1 (functionally illiterate) and NALS 2 (marginally literate), comprised 49.5% of the survey respondents. Based on these data, estimates are that 40 to 44 million people in the United States cannot understand the most basic written material.⁴

The NALS found that the mean reading level for the survey population was equivalent to an eighth-grade education and that risk factors for illiteracy were advanced age, low socioeconomic status, recent immigration, prior or current imprisonment, and poor education. In addition, 44% of people aged 65 or older were in the lowest reading levels (NALS 1 and 2).¹ Illiteracy was not confined to these groups. The largest group of illiterate people were white persons who were born in this country.⁴ Furthermore, 20% of those surveyed in NALS 1 had a high school diploma, although having a diploma did not correlate with reading level.⁵

Refining the problem Since the NALS of 1992, several studies have explored the scope of illiteracy from a health care perspective. A study of 500 indigent patients who presented to outpatient medical clinics found that the mean reported grade level completed was 10th grade, yet the actual mean literacy level tested was fifth grade and 58% of patients in the study read below that level.¹ Another study of 396 parents and caregivers of pediatric patients found that the mean self-reported reading level of caregivers was 11th grade, but they tested at only the 6th- to 8th-grade level. These results indicate that patients self-report their reading level four to five grade levels above their actual ability.^1,6 One study of 2,659 patients in public hospitals found that 21% to 31% of patients could not understand an appointment slip; 11% to 33% did not understand instructions written at a fourth-grade level; 24% to 58% could not understand the instruction “take pills on an empty stomach”; and 40% to 74% did not understand the hospitals’ general informed consent form.⁷ Finally, a study of 500 patients with diabetes and hypertension who were under medical treatment found that only 50% of those with low literacy levels understood the symptoms of hypoglycemia, compared with 97% of patients with adequate literacy levels.⁸

Difficulty processing health care information Low health literacy also affects people whose general literacy levels are adequate. For example, studies of mammography in elderly, well-educated women found that they had little overall comprehension of concepts of probability or risk reduction. In addition, 46% of these women had difficulty equating percentages with probability.⁹

Consequences of illiteracy

Several experts have projected the costs of health illiteracy to be as high as $73 billion annually.¹⁰ Not calculated in this estimate are the significant noneconomic costs, such as emotional distress, that cannot be measured in dollars. A study in Arizona of 400 Medicaid patients found that those with the lowest reading levels had medical costs almost 400% higher than those of patients with higher reading levels ($12,974 versus $2,969). The patients with low health-literacy levels tended to have more hospitalizations, fewer primary care physician visits, and poor adherence to physician recommendations.¹¹

Overall, these higher costs result in higher insurance and tax rates. A review of medical expenditures in low-literacy patients found that Medicare absorbs 39% of the increased costs, employers 17%, and Medicaid 14%. Up to 14% of the increased costs are written off by hospitals and insurers.¹²

Emotional costs Some noneconomic costs of illiteracy, such as patient shame, are well documented. Parikh and colleagues studied 202 random patients in a public hospital emergency department and found that 42% were identified as having low functional literacy.¹³ Furthermore, 28.7% had low functional illiteracy and also admitted to having difficulty reading.¹³ Of these patients, nearly 40% reported feeling ashamed of their reading status, and 67.2% had not told their spouse about their reading difficulties, 53% had never told their children, and a surprising 19% had never told anyone.¹³

Poor health status The question of whether literacy affects a person’s health status was raised in 1998, when death rates for chronic diseases, rates of communicable diseases, and injuries were all found to be inversely related to education level.¹⁴ Since these data were released, several other studies have attempted to define health in relation to literacy status. Examples of this correlation include a study that found that when they were compared with patients in higher-literacy groups, the patients who read at less than a fourth-grade level have the greatest illness-related dysfunction in their daily lives.¹ Another observational study of 408 diabetic patients with low health literacy found that they had the lowest rates of glycemic control and higher rates of retinopathy than did diabetic patients in general.⁸

This poorer comparative health status is likely due to one or more factors yet to be identified.¹¹ A study of 212 men at a prostate cancer clinic found that lower-literacy patients were more likely to present with advanced cancer and less likely to have received prior screenings.³ Although failure to receive preventive care is unlikely to be solely responsible for the poorer health of low-literacy patients, it does seem to be a factor.

Diagnosing health illiteracy

Low-literacy patients often do not want to reveal their poor reading skills or lack of understanding. When dissembling on the part of the patient is combined with the pressure on the provider to see patients in short visits, eliciting information about a patient’s literacy status can become difficult. Time and sensitivity constraints have kept literacy experts from recommending illiteracy screening for the general patient population. Some providers have attempted general screening of patients in their practice by using methods such as self-reported grade level. As Jackson and associates have shown, however, this assessment often does not correlate with patients’ actual reading level.⁶

Current recommendations direct practitioners to observe their patients and provide formal illiteracy testing for those with signs of low literacy. Such persons frequently take medications incorrectly. They may have someone read handouts or patient material to them, need to sound out words, or be confused by simple written instructions. Clinicians in certain high-risk practices—such as those caring for prisoners or elderly persons in clinics in underserved areas—should screen most patients. Providers should also take random literacy samples of patients in their practices so they get an idea of the overall literacy levels of their patient population.

Despite their limited use as general screening tools, several formats of health literacy testing are used in literacy research.

The Test of Functional Healthcare Literacy in Adults (TOFHLA) is considered the gold standard of health literacy testing. It measures both reading comprehension and numerical comprehension, but the length of time needed to administer this test is prohibitive. The word recognition test and the word comprehension test are both shorter than the TOFHLA, so these tests are often used in its place.¹⁵ Both tests were designed by educators to test overall literacy levels, and both have been shown to correspond well with a patient’s overall health literacy.¹

Other literacy tests include measures of word recognition. These are often first-line tests because the TOFHLA can take up to 30 minutes to administer. Popular word recognition tests include the Rapid Estimate of Adult Literacy in Medicine (REALM), the Wide Range Achievement Test (WRAT), and the Slossen Oral Reading Test (SORT). REALM is especially easy to use, so it is the most frequently used literacy test in clinical practice and research. REALM categorizes patients between the third grade and high school and that category represents their literacy (reading) level.¹⁶

REALM results correspond well with those of TOFHLA (correlation coefficient [r]=0.84).¹ Disadvantages include the test’s lack of availability in foreign languages, as well as its inability to measure skill in read-ing or understanding numbers. These limitations make REALM an incomplete tool for measuring overall health literacy, but its advantages make it an effective screening tool. Although REALM is available in Spanish, the Spanish language version has not been validated.¹

Other types of testing useful in assessing low-literacy patients include evaluations of written materials provided to patients that focus on the reading level of the material. Although detailed discussion of these techniques is beyond the scope of this paper, PAs should know that computer applications are available that can accurately measure the reading level of any given document. These results are useful in tailoring educational materials to patients with a known literacy level or for creating patient educational materials based on the “average” reading level of patients.   

Finally, vision should be considered, especially in practices with elderly patients who often have refractory impairments. Studies by pharmacists have shown that the visual acuity needed to read instructions on the average OTC medication label is 20/50 at 165 inches, while 20/20 vision was required to read the instructions on the labels of several brands.³

Implications for clinicians

In this era of increased vigilance about patient safety, health care illiteracy is poised to be a perpetually important issue to providers, and providers may be affected in several ways. Studies have shown that 10% of medication errors resulted from poor communication and that patients with low-literacy skills are five times more likely to misinterpret a prescription.¹⁰ Couple this data with the greater likelihood that low-literacy patients will use medications or require hospitalization or medical services, and the full implications of the problem are apparent.⁵

A report by the Health Promotion Council of Southeastern Pennsylvania notes that no provider in the United States has ever faced legal action directly related to care provided to a low-literacy patient.¹⁷ In several cases, however, regulatory action was taken against institutions for failure to provide adequate care to low-literacy patients. In addition, the report outlined several extreme situations where individual providers could theoretically be liable for malpractice. There are four major concerns related to cost and regulation that low-literacy patients present to health care providers:

1. Medical malpractice Several medical organizations, such as the American Medical Association (AMA) and the US Department of Health and Human Services (DHHS), have outlined recommendations to ensure that low-literacy patients understand their conditions and medical treatment. Many legal experts believe that such recommendations set a standard of care. Thus, providers who do not alter their practices to accommodate low-literacy patients could face a malpractice suit if low literacy is responsible for a poor outcome.

For a provider to be found guilty of medical malpractice, there must be an obligation to treat the patient (duty), a resulting injury to the patient, and a breach of this duty that resulted in an injury to the patient (proximate cause). The injured patient also must prove that the provider did not practice in accordance with the standards of care. Adamson and colleagues interviewed more than 2,000 patients in 107 practices and verified the link between low health literacy and increased malpractice claims.¹⁸ Their study found that physicians with the poorest communication skills had the highest rate of malpractice claims.¹⁸

2. Patients’ rights A second area of legal and regulatory concern in treating low-literacy patients focuses on the use of interpreters. Title VI of the Civil Rights Act of 1964 ensures that no person will be excluded from participation in federal assistance on the basis of color, creed, or national origin. This law is applicable to any agencies receiving Medicare or Medicaid reimbursement. All health care facilities receiving any type of federal reimbursement are required to provide interpreters at no cost to non-English speakers. Failure to do so can result in fines or, in severe cases, possible exclusion from the Medicare and Medicaid programs.

3. Regulation The third area to consider involves the scrutiny of regulatory agencies like the National Committee for Quality Assurance (NCQA) and the Joint Commission on Accreditation of Healthcare Organizations (JCAHO). Low patient literacy has taken on increasing importance to these groups in light of the connection between medical mistakes and low health literacy. Both organizations have listed medical mistakes as a primary focus of their reviews. Since 1996, JCAHO requires that health institutions review the educational and informational material given to patients to ensure that it is at the appropriate reading level for the patient population. Failure to provide appropriate and understandable information to patients could result in loss of accreditation.

4. Patient consent The final legal and regulatory concern involves providers who perform procedures on or obtain consent from low-literacy patients. Most consent forms for procedures are nearly impossible for low-literacy patients to comprehend.¹⁹ Even patients who can read the forms or understand the provider’s oral explanation often have difficulty processing the information. The report from the Health Promotion Council of Southeastern Pennsylvania does state that while no cases directly involve health literacy and failure to obtain informed consent, the provider has a legal and regulatory responsibility to ensure that the patient understands this process.¹⁷ The Pennsylvania report encourages providers to speak in common terms understandable to the general public, to change consent forms so that low-literacy patients can understand them, and to have the patient repeat back what has just been explained.¹⁷

Managing low-literacy patients

Despite the extent of low health literacy and the focus placed on the issue by national organizations, a consensus for the management of low-literacy patients does not exist. Several different approaches have been recommended (see Table 1).   

The AMA was the first to introduce recommendations for managing low-literacy patients in its 1998 policy statement from the House of Delegates. According to Policy H-160.931, the AMA “. . . encourages the development of literacy appropriate, culturally diverse health-related patient education materials for distribution in the outpatient and inpatient setting. . . .”⁴ The report also encourages “ . . . the development of undergraduate, graduate, and continuing medical education programs that train physicians to communicate with patients who have limited literacy skills.”¹²    

This policy was the basis for the AMA campaign focusing on health literacy. “Help Your Patients Understand” was launched in 2003 to educate health care providers about how to identify and respond to health illiteracy. The campaign encourages behaviors to improve communication, such as using plain, nonmedical language, limiting the amount of information given at each visit, and using the repeat-back method of learning. The repeat-back method involves presenting information to the patient, then having the patient explain or demonstrate the information back to the provider.²⁰ Other recommendations involve include creating and using patient-friendly written materials (fifth- to sixth-grade level) and using nonwritten material like videos and computer programs.²⁰    

A 2004 report by the IOM reviewed the problem of health literacy in America but provided few recommendations.¹⁴ An account released around the same time by the AHRQ summarizes current management of low-literacy patients. The authors found 29 articles that reviewed interventions affecting low-literacy outcomes and assessed the overall quality of the reports as “fair.”² The most interesting finding was that the medium of material mattered less than the level of the material presented in relation to the patients’ literacy level.²¹ That is, when material was presented in the same degree of difficulty in written and video formats, little difference was observed. Asking patients how they learn best should be the standard of care.²¹

One novel approach to improving patient outcomes is to create educational programs that are geared for low-literacy patients and cover specific diseases. One such educational program for those with heart failure provided easily understood instructions for monitoring weight and symptoms and explained how to determine when to seek help. After 3 months, the group had a 20% decrease in reported heart failure symptoms.²²

Supporting research into literacy issues is one way that all clinicians can make an impact. By devising easier, more effective testing and by obtaining a better understanding of the risk factors that lead to illiteracy, providers will be able to better identify low-literacy patients in the future. More research into effective learning may help to minimize the negative ramifications associated with low literacy in health care treatment.  

REFERENCES

© 2007 Haymarket Media, Inc. and the American Academy of Physician Assistants. All rights reserved.
Use of jaapa.com subject to License agreement. Please read our Disclaimer and Privacy policy.