A Recap of Plenary and Selected Breakout Sessions
IHA's 8th Annual Health Literacy Conference
Thursday, May 7, 2009
"Research in Health Literacy: Identifying the Gaps"
Rima Rudd, Sc.D, MSPH
Rima Rudd, Sc.D, MSPH, senior lecturer on society, human development, and health at the Harvard School of Public Health and a pioneer in health literacy, presented the opening keynote at the Institute for Healthcare Advancement's 8th Annual Health Literacy conference. Her goal was to help identify the field's next steps. "In understanding what health literacy is, the current status and focus of the field, and how it got there," she said, "we can gain a better sense of the work that still needs to be done."
Two solid premises underlie all work in health literacy, Dr. Rudd said, whether we are working in small local clinic or larger community initiatives. The first premise is the belief in the critical role that literacy plays in the acquisition of healthcare. In fact, Dr. Rudd pointed out, literacy is the foundation of all communication, and as such, it influences a person's ability to access and navigate through all aspects of our social environment, not just health information.
The second premise is that of social justice, said Dr. Rudd. People working in the field of health literacy are keenly aware that health disparities emerge from social inequalities. This recognition enables health literacy advocates to state very clearly and very firmly, said Dr. Rudd, "that literacy is a social inequality in this country." The differences in literacy that we see in our country, she said, have everything to do with differences in available resources, access barriers, biases, discriminatory housing and schooling, etc.
Dr. Rudd said research has established that there is a large - and widening - gap between the literacy demands of health systems and the actual literacy skills of the population. Healthcare's increased demand on literacy is the result partly of the burden of disease, partly of the increased complexity of our healthcare systems, and partly of the assumption we now make in the management of chronic disease that people must be full partners in health decisions. We are placing more and more literacy demands on patients; but the literacy skills of those patients cannot meet the demands.
We also are well aware of the consequences of the healthcare inequalities that exist between those with strong literacy skills and those with weaker skills, said Dr. Rudd. Well over 1500 peer-reviewed published articles testify to these disparities and the resulting disparities in health outcomes, which include differences in:
- rates of participation in screening and disease prevention
- participation in healthful activities
- basic understanding of causes and treatments of disease, and
- comprehension of medicines and regimens.
"In disease management we see very dramatic differentials in chronic disease management and success. We see it in disease management in general, we see it in morbidity, and we see it in well being and health status. And recently, we see it in mortality. People are dying unnecessarily because they are not able to manage their own health and well being," said Dr. Rudd.
Status of the Field
The field of health literacy has taken the first steps, said Dr. Rudd, but it needs to take more. It began by looking at the work done in the field of education to identify and locate the problem. "All of the adult literacy assessment surveys conducted by the field of education gave us evidence that the problem existed and that it might have health consequences," she explained. From there, health literacy advocates set out to make the argument that the findings were relevant not just to education but to public health, health education, medicine, nursing, and other areas in the health field as well. The problem, according to Dr. Rudd, is that, "you'll note, when you look at the literature, people are still making that argument," said Dr. Rudd. She called for the field to move on now to solving the problems it has uncovered.
The health literacy field claims to have measurements of patients' health literacy skills, she went on, and is now interested in measuring health consequences by the skill level of users, among various population groups, and within medical fields and specialties. However, Dr. Rudd cautions, in reality, patients' health literacy skills have not been adequately measured. "In the health field, we use very poor approximations of reading skills with a little bit of health content in it, and we call that health literacy measurement," she said. "It is not." However, despite the imperfections of the measurements, "those very weak tests offer some profound findings that there are differentials in health outcomes between people who can use materials with ease and those who cannot."
"As a field, we've been relatively good at locating half of the problem. We've been very good at making the argument within our own field of research and our own field of practice. We've measured the scope of the problem, though I want to point out the ways in which that measurement has been somewhat limited. We've measured the consequences; we have a very strong body of literature that shows us the consequences of lack of literacy skills and health outcomes." Then she went on to identify those areas where she thinks the field need to focus more efforts: "We've not been very good at identifying solutions; we're still experimenting. And we've not been as rigorous as we can in making sure we test out and do really good evaluative studies to make sure that we can identify the state of the art. The state of the art in medicine, for example, does not go much beyond the teach-back technique and questioning. So, there is a wide range in which we can make contributions."
A Divergence: Health Literacy and Education
Moving forward, health literacy needs to recognize the limits of its relationship with education, according to Dr. Rudd. She acknowledged that health literacy research has gained much from education and that it can still gain a lot of insight from adult educators into how to ameliorate the problem. "But the fields are also divergent," she cautioned. She noted that some people in the health literacy field are anxious to use pre- and post-tests to see if there are changes taking place. "This might be where we are making a mistake," she warned.
"We in the health field are not about improving literacy skills-this is not our area of expertise. We don't have time for that. We need to ameliorate the problem, we need to address the issues, we need to make changes in the face of the existing literacy skills," she said. To realize the changes that those in the health field want to see, the materials need to be made more comprehensible and the demands on literacy need to be made more overt. The field of education is not moving in the same direction, noted Dr. Rudd. Education does not want to lower the bar on the demand side; they want to improve and increase literacy skills. And while the health field should be supportive of that, she said, "We don't have time to wait for it to happen."
In order to change its demands on literacy, the field of health literacy needs to recognize the fact that health literacy is actually an array of skills. She called particular attention to the importance of the oral exchange and to the work currently being done with numeracy. "The oral exchange is critical to healthcare. That is where the diagnosis takes place, that is where the healing begins, that is where recommendations take place," she said. But the ability to find the words to describe an amorphous feeling located somewhere inside the body or within the emotional array that makes up a person and then communicate that feeling to someone else requires profound literacy skills, she explained, as does listening to someone else's speech and following the logic.
As for numeracy, Dr. Rudd encouraged attendees to continue exploring the topic. "We have paid attention to the fact that we ask people all the time to deal with numbers. They have to measure, they have to monitor, they have to use tools that have measurement scales-and the scales differ from one tool to another. They also need to deal with mathematical concepts, like risk and probability, that practitioners in the field take for granted but for which we have not done a good job of offering definitions or explanations," she said.
Another area that emerges again and again as an issue for health literacy is the definition of terms. "When you define health literacy as the skill set or capability of individuals," Dr. Rudd explained, "then the locus of responsibility seems to fall on the shoulders of that individual." Instead, she said, the health community must acknowledge the part it plays in the disconnect between patients' literacy skills and the demands placed on them by healthcare materials. To fail to do this leads to the kind of reinventing of the wheel that has been taking place in the field of health literacy, she said, because then patients become the focus of all measurement and they become the part of the equation that is pressed to change. "We know from national and international studies that the average adult in all industrialized nations struggles using currently available print materials to accomplish everyday tasks. We have solid evidence of that. How many times do we have to re-prove it?"
Not only are not enough measurements of the demand side being taken, not enough thought has gone into exactly what is being measured when they are taken, claimed Dr. Rudd. "We need to deconstruct health activities, understand the tasks we ask people to undertake, and link those tasks to literacy skills so that we can understand the demand side better," she said.
In addition, said Dr. Rudd, the health literacy field needs to examine the legal processes related to healthcare. She pointed to the practice of asking patients to sign consent forms after they have already been prepped for a procedure and the practice in some pharmacies of asking patients to sign the waiver for counseling on the medication without ever asking if they have questions. In each case, she said, the form is not being used to ensure that patients understand or have the opportunity to ask questions. Instead, it is being used as insurance for the institution seeking the signature, and the manner in which that signature is obtained is circumventing the intention of the law.
The effects of professional training have not been studied, she went on. There have been no good studies where the healthcare professional is the subject, where, for example, a group of doctors are trained in behaviors that support health literacy goals and then followed to see if they do a better job with patients than other professionals.
Dr. Rudd said it is time for the health literacy field to take the next step and undertake meaningful evaluation studies of healthcare professionals' communication skills. She suggested that training in communication should be a rigorous requirement for licensing and accreditation, and she called for advances in the use of assistive technologies. "We are so far behind the times. We are not using sound and videos and visuals the way that we can. And we know that it helps." She said that bureaucratic priorities of health institutions must be modified to better support the public. Health literacy advocates must look at the healthcare field's decision-making processes and the numeric demands that are being placed on patients and examine whether information is being presented in a way that enables patients to make decisions or use information, she said.
Dr. Rudd also noted that the health literacy field needs to move beyond doctor/patient interactions and look at prevention. "We know that people who have limited literacy do not engage in preventive activities the way that other members of the population do," she said. The healthcare arena, she said, needs to consider that perhaps it is not making it easy for individuals with low literacy to take advantage of what society has to offer in this area.
Health literacy advocates must expand the boundaries of their focus. "We need to become involved in developing policy initiatives at the state and national levels," said Dr. Rudd. She noted that Cynthia Baur of the CDC, who is drafting the action plan that is coming out of the Surgeon General's work on health literacy, is interested in the response of the health literacy field. In addition, she said, "We need to integrate health literacy into disaster management. There are very few voices looking at that connection."
* * *
"Causal Pathways Between Health Literacy and Health Disparities"
Michael K. Paasche-Orlow, MD
Disparities in health outcomes among various populations are well documented. The healthcare arena has the responsibility and the desire to address these inequalities, but it needs guidance in order to develop effective interventions. Michael K. Paasche-Orlow, M.D., is interested in health literacy and its association with the demographics that have been the central focus of discussions on health disparities. Dr. Paasche-Orlow explored the possible role of health literacy in eliminating certain disparities.
Ethnicity and race seem to be factors in the disparities found in a number of areas, Dr. Paasche-Orlow pointed out. The National Assessment of Adult Literacy (NAAL, 2003) showed that ethnicity and race are related to literacy, for example. While more than a third of adults in the United State scored at the Below Basic or Basic levels of the NAAL, lower scores were more common among certain groups, including ethnic and racial minorities. In addition, reports by the National Assessment of Educational Progress (NAEP) show a connection between ethnicity/race and disparities in educational achievement, he said.
In 2005, for example, the NAEP compared proficiency levels of black, Hispanic, and white 12th graders. It showed that the percentage of black and Hispanic students who scored below basic proficiency on tests of math and reading was twice the percentage of white students who scored below basic proficiency in those disciplines.
In medicine, the story is similar, he said. A study of the prevalence of marginal health literacy across 85 medical studies showed that racial and ethnic minorities, along with the elderly and people with lower levels of education, were more likely to have marginal health literacy skills. "It's the same story," he pointed out, "but using a different database."
There is ample literature establishing connections among health literacy, health outcomes, and race and ethnicity, said Dr. Paasche-Orlow. So the question of how disparities in health literacy might be connected to disparities in health outcomes among racial and ethnic minorities seems a logical one.
Dr. Paasche-Orlow reported that a study of 200 HIV patients showed lower adherence to medicine regimens among the African-American patients. In fact, he said, "African-Americans were about two and a half times more likely to show non-adherence. But if we corrected for literacy, the correlation went away and health literacy became the dominant, independent predictor of non-adherence. Those with lower levels of literacy were twice as likely to be non-adherent."
Findings like this, said Dr. Paasche-Orlow, open up possibilities for interventions. If the disparities in health outcomes are not about race, as the healthcare field has long assumed, but about literacy, we can do something about it, he explained.
On the other hand, failing to take disparities in literacy into account could lead to mistaken assumptions and unfortunate consequences, said Dr. Paasche-Orlow. He pointed to articles that conclude that African-Americans choose aggressive end-of-life care more often than whites and noted that they do not reveal how the information was presented to patients and do not control for other factors. To test if health literacy could explain the reported differences in care preferences, Dr. Paasche-Orlow surveyed patients regarding their preferences on care in the case of advanced dementia. Care options were described verbally, and then patients were shown a two-minute video of someone with advanced dementia. They were asked about their care preferences following the verbal explanations and again after watching the video. Following the verbal presentations, said Dr. Paasche-Orlow, African-Americans were more likely to opt for aggressive care. However, when the analysis was adjusted to account for health literacy, the effect of race went away. In both the unadjusted analysis and the adjusted analysis, he said, low health literacy was a strong predictor of choosing aggressive care. Interestingly, after the video, there were no differences in preferences for end-of-life care by level of health literacy or by race, he said.
"This is an important example," said Dr. Paasche-Orlow, "because if you thought that the important feature of the story was culture, you would have to ask what you could do to uphold and support the person's culture. But if you identify the undergirding phenomenon as literacy, then you ask, 'How can I empower the person with information and skills to overcome misunderstandings?'" The different assumptions about what is driving the disparities would lead you in completely different directions, he explained.
A Model for Research
Dr. Paasche-Orlow sees in these examples a lesson for health-related research. In situations with an apparent racial connection, he said, researchers need to ask if that connection can be nullified or changed if looked at from the perspective of health literacy. "This has only been done in about four or five papers to date," he said, though he predicted that the field will begin to see more analysis of how health literacy might start to explain disparities in health outcomes.
He presented a model for approaching research on the connection between health literacy and health outcomes and the role of race in that connection. To begin, he said, physical abilities (vision, hearing, verbal ability, memory, and reasoning), language, culture, social support, income, employment, and occupation should be set aside. This is necessary not because these things are unimportant, but because it allows the discussion to focus on the most direct paths between health literacy and outcomes, explained Dr. Paasche-Orlow.
The model looks at access and utilization of health care, the provider-patient interaction, and self care. It emphasizes that, in each of these areas, there are factors on both the side of the patient and on the side of the system/provider/environment so that "you can never ask a question about a person's skill without also asking about the complexity of the context," said Dr. Paasche-Orlow.
This is perhaps the model's biggest strength, but he said it is limited by the model's focus on the healthcare context. "This idea needs to explode onto the public health agenda as well," he claimed. To do that, he said, models that reflect the challenges of public health must be created.
What to Do
Research into disparities has to start measuring health literacy, insisted Dr. Paasche-Orlow. If the field of research that focuses on disparities were to incorporate measurements of literacy into its studies, the health literacy field could more effectively design interventions, he said. In the meantime, there are a number of things that can be done to address those health literacy disparities that have already been identified.
For example, it is time to "massively invest in and improve patient/physician education," he asserted. We know that what happens in healthcare can sometimes exacerbate underlying educational disparities, said Dr. Paasche-Orlow, yet, the total time spent on patient education upon discharge from the hospital averages eight minutes. "And if you look at that exchange, it is almost always passive," he added. In a study of 97 audiotaped provider-patient interchanges, only two showed providers engaging patients in conversation," he said. "And in both instances, the provider was a clinical pharmacist who was trying to assess the patient's understanding."
Instruction on dealing with disparities in patients' health literacy needs to be incorporated into the agenda of medical schools, continued Dr. Paasche-Orlow. Currently, he said, it is not addressed, so physicians are leaving patient education up to the nurses. But the nurses are already overwhelmed with responsibilities, he said. Consequently, effective patient education does not always happen.
Finally, said Dr. Paasche-Orlow, health literacy advocates need to explore the opportunities presented by health information technology (HIT). HIT offers access to and interface with such technology, but that needs to be carefully managed, he cautioned. As the field of HIT moves forward, care must be taken to ensure that the solutions can be used by people with limited literacy.
When dealing with disparities in health outcomes, things may not always be what they seem. According to Dr. Paasche-Orlow, disparities that seem to be based on race or ethnicity may actually be literacy-based. To develop effective interventions, he said, the field of health literacy must work to reveal the true underlying reasons for disparities. At the same time, it must look at ways to better educate both physicians and patients so that those disparities cease to be significant.
* * *
"From the IOM Roundtable: Translating Research into Practice"
George Isham, MD (for Ruth Parker)
In the course of any field of study's growth and evolution, opportunities for discussion are invaluable. The entire body of knowledge in a field benefits from researchers and practitioners discussing the issues inherent in the field. Health literacy is no exception. The Institute of Medicine (IOM) recognized the benefit that a forum for open discussion of the challenges faced by health literacy would provide and convened a Health Literacy Roundtable. Ruth Parker, M.D., a leading researcher in the field of health literacy, was slated to present information from that roundtable at the 2009 IHA Health Literacy Conference. Unfortunately, she took ill just before the conference, but George Isham, M.D., chairman of the IOM Health Literacy Roundtable and a conference speaker (see Friday Recap) was able to deliver her presentation on some of the important topics that have been raised by the IOM Health Literacy Roundtable.
Dr. Isham began by contrasting the goal and function of an IOM roundtable with that of the different committees. While the various IOM committees conduct studies and write reports intended to advise the government on issues in medicine, the function of an IOM roundtable is to advance the national agenda in its particular area of focus, said Dr. Isham. A roundtable does not reach conclusions or make recommendations; it is not supposed to. Instead, he said, it is meant to convene dialogue. IOM roundtables are supported by contributions from stakeholders rather than sponsored by the government or by private foundations, he said, and their deliberations are open to the public. IOM roundtables try to present open workshops and may publish workshop proceedings. They are, explained Dr. Isham, a mechanism to disseminate the discussion.
The IOM reviews roundtables every three years. Some continue for a long time while others are shorter lived. The roundtable on health literacy convened in 2006 and is beginning its second three-year term. Its mission is to advance the field of health literacy by translating research findings to practical strategies that can be implemented.
A Perfect Storm
One of the first issues raised by the IOM Roundtable on Health Literacy is the idea that health literacy encompasses more than healthcare itself. Roundtable members recognized that health systems, society, the education system, the media, business, and government are all components of health literacy and that these stakeholders need to become more a part of the conversation in the field than they previously have been, said Dr. Isham.
Once this larger view of health literacy is embraced, he continued, it becomes apparent that there is a perfect storm brewing in this country with respect to literacy. Shifting demographics, a changing job market, and a forecasted decline in literacy rates among the population make the need for practical approaches to address the problem of health literacy concerns increasingly urgent, said Dr. Isham.
In 2004, the IOM reported that 90 million Americans had inadequate health literacy, he said. We know that the consequences of low health literacy include a decrease in preventive services use and consequent delayed diagnoses, increased ER use and hospitalization, poorer physical and mental health, and increased mortality risk among those with low health literacy. Add to this a report from the Educational Testing Service (ETS) which predicts that, in the year 2030, we will see a 5% decline in average literacy among adults aged 16-65 years old as well as a 7% increase in the disparity of scores, and it becomes clear that all of the stakeholders in health literacy will be faced with increased challenges, said Dr. Isham.
Among the working-age population, he explained, such a decrease in general literacy will translate into increased unemployment; increased uninsured; increased prevalence of inadequate health literacy; and increased problems of disparity, quality, and costs. For those over 65, he warned, the prediction is even worse.
How to Prepare
So how do we prepare for this? Dr. Isham insisted that all stakeholders need to "think about who is most vulnerable and tailor interventions to aid those at high risk. We also need to set a long-term goal of increasing patient skills via education while also building organizational infrastructure now--because we can't wait until the last minute to deal with a crisis. We need to ensure patients' clarity and understanding of health tasks and focus on what patients need to know to do."
There is no guarantee that what we try will work, he said. But we are at a point where we know that if we do nothing, it will get worse.
The IOM Health Literacy Roundtable also discussed the fact that, in the quest to develop interventions that will alleviate the effects of low health literacy, both sides of the health literacy framework need to be addressed, said Dr. Isham. "We need to deal with the skills/ability side as well as the demands/complexity side of the health literacy equation," he explained, "yet the tools we contemplate bringing to bear on this issue tend to be on the side of the patient, and they focus on the patient/physician interaction." Work needs to be done on the demand side as well, he said.
He observed that, if our aim is to improve healthcare, looking at the office visit is of limited use. It is a limited portal through which to look at doctors, he said, and it is a difficult way to impact the self-advocacy skills of patients.
As an illustration, Dr. Isham called attention to the issue of labeling. Medicine labels, inserts, and guides, he pointed out, can be very confusing. Not a lot of effort has gone into making them readable, logical, and understandable. Then he presented an enhanced label prototype that was designed with the patient in mind. The reformatted label is an example of addressing the demand side of the labeling problem, he explained. Instead of expecting patients to increase their literacy skills in order to read a label, the suggested label format makes the label itself easier to read and understand.
This idea of addressing both sides of the health literacy equation was widely accepted among roundtable members, said Dr. Isham. There seems to be general agreement that health literacy occurs when skills/ability are aligned with demands/complexity, and there are some encouraging steps being taken to achieve that alignment.
On the skills/ability side, he said, definite steps have been identified that the IOM roundtable believes will help health literacy advocates translate research into practice. They want to see a United State Department of Health and Human Services (HHS) ongoing national data repository of health literacy skills/ability, said Dr. Isham. They would also like the HHS to make available a tract level data map of the entire country detailing skills/ability. In clinical practices, the roundtable members would like to promote the use of "universal precautions" rather than individual skill testing.
The roundtable also discussed possible next steps on the demands/complexity side of the health literacy problem, reported Dr. Isham. They would like to see the Centers for Medicare and Medicaid Services (CMS) develop meaningful metrics of "health literate" care and service providers. This initiative would emphasize the use of models and demonstrations, said Dr. Isham, and would include incentives for early adopters. The IOM roundtable also voiced the need for the National Institutes of Health and professional societies to define essential "need to know to do" items for health.
Perhaps the most important idea to be voiced at the IOM roundtable, said Dr. Isham, was that of incorporating health literacy into HHS's Healthy People 2020 objectives. There is a great desire, he said, to link health literacy to national efforts to measure health quality, disparities, and cost.
The overall message coming from the IOM roundtable on health literacy, said Dr. Isham, is the urgency of the need for practical ideas to improve overall health literacy in this country. We need more research in this field, he said, and we need to test rapidly what is going on. Health literacy advocates need to look at both the skills/ability and the demands/complexity sides of the health literacy problem and develop interventions that will move those two halves of the framework into greater alignment with each other. And, said Dr. Isham, we need to focus on the fact that the public does not necessarily need to be well versed in the issues and complexities of the health literacy discussion, but it does need to know what to do to stay healthy.
* * *
"A Breakthrough Health Literacy Program: Empowering Parents, Benefitting Children, Improving the Healthcare System"
Ariella Herman, Ph.D.
As the field of health literacy matures, it becomes increasingly important to move beyond research and measurement to the development of specific interventions that can be put into place. The field needs to identify some best practices that can be adopted, analyzed, and improved. Ariella Herman, Ph.D, presented a description of just such a program. As founder of the UCLA/Johnson & Johnson Health Care Institute, Dr. Herman has been fully immersed in addressing the consequences of health literacy for the past eight years.
Health literacy is a stronger predictor of a person's health than age, income, employment status, education level, or race, said Dr. Herman. But even for those with high literacy levels, disparities related to health literacy pose a problem. Approximately 90 million adult Americans cannot appropriately use health information, she said, and this leads to increased use of emergency services and hospitalization and decreased use of preventive services and compliance with medication regimens. This, in turn, costs the American economy at least $73 billion annually, she said.
In 2000, Dr. Herman asked Head Start agency directors to identify the major barriers to quality healthcare among the families they served. "The answer was the parents' lack of health literacy and the parents not coming to training when it was offered," said Dr. Herman. With the intention of addressing these concerns, she teamed with Johnson & Johnson to launch the UCLA/Johnson & Johnson Health Care Institute (HCI) in 2001 and to develop a training methodology for Head Start agencies to use.
In order to ensure that they were motivated to take on the HCI program, individual Head Start agencies were required to apply to participate. Once accepted, the director, health coordinator, social services coordinator, and a few staff members from each agency attended a two-day Train-the-Trainer workshop. The workshop went beyond teaching attendees how to teach; it helped each agency develop a health improvement project (HIP) specific to the challenges it faced, and it addressed topics like staff motivation and marketing.
HCI and Head Start distributed IHA's book What To Do When Your Child Gets Sick to low-income families. Using the HIP it had developed and presenting the information in the languages spoken by the families it served, each agency trained its families on how to use the book to make appropriate healthcare decisions for their children. Numerous approaches were used. One agency translated its training into four languages and projected the translated materials onto different screens, said Dr. Herman. The families had earpieces that allowed them to listen to the spoken information in their own languages. Another agency addressed the problem of quantitative literacy, or numeracy, by showing parents how to tell the difference between a teaspoon and a tablespoon, how to read a thermometer, etc. Each family was given the different measurement tools, so everyone left with a first aid kit. "Throughout the training, we never tried to change anyone's culture," said Dr. Herman. "We wanted to give them additional knowledge."
After the training, the Head Start agencies were required to conduct follow-up telephone calls with the families. "We found that we had to reinforce changes in behavior," said Dr. Herman. Through the calls, staff members answered any remaining questions, reiterated what had been taught, and tracked information on parent behavior. To date, this methodology has been used to train 14,000 families in seven languages and across 10 ethnicities via 55 programs in 38 states, said Dr. Herman.
Prior to the training, the Head Start families reported that the first place they turned when a child was ill was the doctor (68.79%) or the emergency room (4.39%). Only 4.7% of the families questioned said that they consulted the book first, according to Dr. Herman. After the training, however, the percentage of families who reported consulting the book first rose to 47.55% while the percentage of families who went straight to the doctor or emergency room dropped by more than half.
These results were echoed in other findings, said Dr. Herman.
- The number of doctor and emergency room visits for the five main childhood illnesses (cold, cough, ear infection, fever, and flu) dropped after families received the training.
- From 2001-2008, the average number of school days missed by children in the study dropped 29% and the number of work days missed by their parents dropped 42%.
- From 2001-2008, there was a 42% drop in doctor visits and a 58% drop in emergency room visits among the families studied.
Translating these results into dollars saved, said Dr. Herman, shows the economic impact of the training. The average annual savings per family tracked was $554.74. When you compare that to the $60 cost to train a single family-an amount that goes down as more families are trained-the benefit becomes clear, she said.
A number of agencies have elected to continue using the HCI program model to address health literacy disparities, said Dr. Herman. The Kansas Head Start agency has launched a pilot to train 5,000 families, Pfizer has decided to sponsor the state of New Mexico and train close to 6,000 families, and HCI just received a federal grant from the Office of Head Start (OHS) to train 10,000 families, she said. In addition, the governor of Washington included in the state's legislature the resources to train 7,000 families. This marks the first time that the program has moved beyond Head Start and into community health clinics, which, noted Dr. Herman, "is definitely the place where these projects should expand."
The methodology can be used to develop training programs on individual health issues, too, said Dr. Herman. For example, the agencies involved in the initial training were recently shown how that same training could be applied to an oral health program.
HCI is also looking at ways the methodology could be adapted to expand its reach and effectiveness. HCI is currently working on developing a program for diabetes and obesity prevention, said Dr. Herman. But, she said, "We realized that we neededto reach more than parents. We also needed to reach staff members, teachers, and children." So HCI adapted its methodology to encourage interaction among these groups and to show participants that they really can make the suggested changes. Handouts illustrate the different food groups and the recommended daily servings. Participants attend a workshop called "How To Stretch Your Food Dollars," and a children's "farmer's market" models good food choices.
While the money saved as a result of the HCI programs is significant, said Dr. Herman, it is not the greatest reward. Even more valuable, she said, is the fact that parents are empowered. HCI's focus is to give parents the tools they need to make good decisions regarding their children's health. In the Head Start training, she said, "We don't tell them not to go to the doctor; we give them the book and the training so that they will know when to go." And in the diabetes prevention program, she said, "We tell the parents that there are threats to their children's health, but that they have the power to affect it."
* * *
Recap of Selected Breakout Sessions
"Promoting Comprehension for Research Subjects: The IRB Consent and Authorization Toolkit"
Michael K. Paasche-Orlow, MD, MPH, MA
As the health community's recognition of limited health literacy increases, so do the attempts to address the disparities it causes. Health literacy advocates have, for a number of years, understood the importance of assessing and improving the readability of health-related materials and forms. Michael K. Paasche-Orlow, M.D. presented a toolkit designed to remove unnecessary complexity from the consent and authorization process for minimal risk research projects and to facilitate a cultural shift that will help to ensure research participants' comprehension.
Why a Toolkit?
Explaining why the toolkit was created is a bit like preaching to the choir, said Dr. Paasche-Orlow. The healthcare community realizes that consent documents are too long and complex and that research participants often do not understand the fundamental concepts. But understanding the problem and addressing it are two different things, he said, and institutions are not necessarily doing a good job addressing the problem.
Dr. Paasche-Orlow looked at the boilerplate language developed by the Institutional Review Boards (IRBs) of 114 medical schools. In about half of the cases, the institution had set grade-level readability standards for consent forms, the most common being an eighth grade readability level. But that is not a particularly logical choice, said Dr. Paasche-Orlow, since the average American's reading level is between eighth and ninth grade-which means that much of the country reads at a lower level. Besides, he said, only five of the sixty-one schools that had a standard actually met it. Typically, readability was off by two grade levels.
In terms of readability, it is not difficult to improve on the forms that are currently being used, said Dr. Paasche-Orlow, and ethical guidelines call for it. What impedes progress, in many cases, is the fact that ethical guidelines consistently bump up against institutional risk concerns, he said. Federal regulators and institutional risk managers each have their own concerns, and making everyone happy may not be possible, he explained, but institutions should at least be held to their own mandates that language should be understandable.
The assumptions about readability and literacy that form the foundation of the toolkit are not new. They are the same ideas detailed in Teaching Patients with Low Literacy Skills (Doak, Doak, and Root, 1996), which is widely regarded as the seminal work on health literacy. However, readability does not necessarily translate to participant understanding, said Dr. Paasche-Orlow.
In fact, he said, a number of writing strategies designed to enhance understanding could negatively affect readability scores. Efforts to improve cohesion, for example, may be at odds with readability. Dr. Paasche-Orlow explained: "You could say, 'Johnny went to the store. Johnny bought a gallon of milk.' But you could improve cohesion by saying, 'Johnny went to the store because he wanted to buy a gallon of milk.'" In this single sentence, the logic of the relationship between the two concepts is made clear and understandable. But by combining the ideas into one sentence, the readability score of the passage almost certainly suffers.
On the other hand, once you understand readability formulas, it is easy to write language that will yield the scores you want, Dr. Paasche pointed out. However, he warned, "there are plenty of short sentences that people don't understand." A form's success cannot be measured in terms of readability alone, he concluded.
Formatting also impacts participants' ability to understand consent forms, he said. Text should be arranged in a way that encourages readers to engage with it and ask questions. The toolkit suggests using a question and answer format to encourage that kind of engagement. Not only does this format simplify concepts for the potential participants, said Dr. Paasche-Orlow, it also helps research assistants realize that they have the ability to facilitate conversation about the information.
Included in the toolkit is a sample template that illustrates a readable, well-formatted document. It is written at the 5th-grade level and represents a significant improvement over the typical consent form. However, it is also designed, said Dr. Paasche-Orlow, to communicate to both the person giving consent and the person seeking consent that the participant's understanding of the information is the goal. It is the beginning of a cultural change that is long overdue in the area of informed consent, he said.
Acquiring Consent as an Education Tool
No matter how much attention is paid to language and formatting, there will never be a document that everyone can understand. If you want to achieve participant understanding-a critical component for true consent-then you need to test for that understanding, said Dr. Paasche-Orlow. Thus, he said, the focus has to shift from whether the document is good enough to whether the process is good enough.
Currently, the consent form is often handed to or read to potential participants as a means of educating them. Not only is this ineffective, according to Dr. Paasche-Orlow, it is too narrow a view of participant education. The healthcare community needs to think of the entire process of acquiring consent as an education tool, he said.
Promoting a research culture that embraces this idea takes practice, simulation, and supervision; simply giving research assistants this toolkit isn't going to solve the problem, said Dr. Paasche-Orlow. They need to practice the teach-back method advocated in the toolkit and gain experience working with potential participants throughout the entire recursive process of listening, assessing participant understanding, asking follow-up questions, and addressing any misconceptions, he said.
This requires an attendant shift in the healthcare community's view of question-asking, said Dr. Paasche-Orlow. The typical attitude is that if potential research participants have questions, it is their responsibility to ask them. But people have been acculturated to not ask questions-whether it is because they do not want to waste the researchers' time or because they do not want to appear unintelligent, he said. For this reason, he emphasized, the process of questioning, listening, and re-presenting necessary information should continue until the participant's comprehension is confirmed or the decision is made to not enroll that person.
Sample Certification Form
The final component of the toolkit is a Sample Certification Form. It was designed to encourage those involved in research recruitment to engage potential participants in order to better assess their comprehension. Basically, the Sample Certification form is a checklist to guide research assistants through the consent process and to facilitate their conversation with participants. Depending on the needs of the particular institution, the form could be used as a training tool for research assistants or as a required tool with all participants, suggested Dr. Paasche-Orlow.
This toolkit goes beyond giving tips for employing plain language and effective formatting, Dr. Paasche-Orlow emphasized. While it does address printed consent forms, it also considers the process by which consent is achieved. The toolkit outlines a process for assessing the comprehension of research participants that is designed to improve communication, to recruit subjects who understand, and to provide opportunities to train and monitor research assistants. This process shifts the responsibility for confirming understanding from the participants to the researchers and, in doing so, empowers participants to truly consent.
* * *
Using Readability Formulas Correctly: Cutting through the SMOG, FOG and Flesch
Both the National Adult Literacy Survey (NALS, 1992) and the National Assessment of Adult Literacy (NAAL, 2003) confirmed that as much as 50 percent of the American adult population reads at or below the 6th grade reading level. Numerous studies conclude that most health information is written at the 10th grade level and above. There continues to be a disconnect between the health information we write and the audience for whom we are writing it. Doing readability testing on printed health information materials remains one way to evaluate and improve written materials.
There are a number of readability formulas available, with Flesch-Kincaid, Fog, Fry and SMOG among the most commonly used. Each readability formula uses a different mathematical algorithm to compute a readability score that is usually converted to a reading ease score or reading grade level. While each of the readability formulas calculate scores based on basic semantic and syntactic elements such as length of words and sentences, and number of syllables and sentences in a sample of text, it is possible to get a range of scores for the same sample when using different readability formulas. It is important, then, to choose a readability formula and use it consistently to evaluate your written health information materials.
Microsoft Word includes readability statistics as part of the spelling and grammar check. With a click of a button, documents can be evaluated using the Flesch Reading Ease and the Flesch-Kincaid Reading Grade. However, readability experts warn that the tool within Microsoft Word has a flaw that causes reading grade scores to be about two grades lower than they actually are. Still, the convenience and ease of a score that is calculated by the computer makes this a commonly used tool. Another readability calculator is available for the SMOG formula. Sample text can be cut and pasted into the online tool and a score is generated. However, readability scores that are calculated by hand are usually more accurate. The presentation included 6 examples of calculating SMOG readability scores by hand.
It is important to note that readability formulas have limitations. They cannot compensate for the impact of design and layout or the complexity of most health concepts. They also do not take into account a reader's familiarity with the subject. A person newly diagnosed with diabetes, for example, will have more of a challenge reading information that includes medical terms and concepts they are encountering for the first time, while as someone who has been managing their diabetes for several years will already be familiar with most of the terms and concepts. Still, readability formulas are useful despite their limitations. They give a benchmark of a document's reading level that can be compared against retested scores following revisions to the document. Readability formulas are also useful training tools and general guides for writers and educators. They validate and support the use of plain language and the principles of universal design.
Universal design is an important concept to consider when developing written health information materials. Universal design is the design of products, environments and communications to be useable by people, to the greatest extent possible, without the need for adaptation or specialized design. One example is automatic doors found on most public buildings. While the Americans with Disabilities Act was written to protect and provide services for a specific part of the population, we all benefit from the convenience and ease of doors that automatically open and close. Pictograms on medication bottles that show a wine glass with a slash through it tells us that this medication should not be taken with alcohol. The goal is to use symbols that can be understood without the need to read the text.
Likewise, following the principles of plain language, including using short, simple words, writing short sentences and paragraphs, and avoiding legal, technical and medical jargon are part of universal design. Making health information easy to read, however, is not dumbing it down. In fact, studies show that even patients who read at the college level prefer medical information written at the 7th grade level or below.
Using readability formulas to evaluate written health information ensures that we are writing materials that the majority of the public can understand. The presentation included tips for improving the accuracy of readability formulas, including adding periods to titles and subtitles and at the end of each statement in a bulleted list.
Medical terms, because they often contain more than 2 syllables, will raise the readability level of a document. The writer must determine which words are important for the patient to understand. For example, a document about high blood pressure may need to include the term "hypertension" because it is a word patients might hear while speaking with their doctor. However, it may not be necessary to include words like "diastolic" and "systolic" or include "mm Hg" as part of a blood pressure reading. When possible, use alternatives to medical words when appropriate, such as "swelling" for "edema" and "chest pain" for "angina." Don't use alternative words that change the meaning of the medical terms, and don't exclude medical terms just to lower the readability score. Each medical word that is used in a document should be defined clearly. It might be helpful to place medical terms and definitions in a shaded box so readers can refer to it easily, or include a glossary for longer documents.
Perhaps the most important thing to remember is that there is no magic number. While we should aim for documents to score at the 4th to 6th grade levels, this is sometimes impossible to accomplish. As already stated, including necessary medical words will increase the readability of the document. These words can be removed from the document for readability testing. It's very important not to write to the formula. In other words, do not remove terms and information that are important for the patient to know. Do not shorten all sentences just to lower the score. This will result in choppy text and may in fact make the document harder to read and understand.
Readability formulas, then, are best used as guides and as training tools for writers and educators. Test for readability as you write. Test the first draft to determine a baseline score, then as you edit and rewrite the document following plain language guidelines and the principles of universal design, retest the document to see if scores are improving. Choose a readability formula that makes sense to you and understand the basic principles of how it calculates a score. Be consistent and use the same method each time. Keep in mind that readability formulas have limitations and remember the tips and suggestions that improve their accuracy. When used correctly, readability formulas can be helpful, informative and persuasive tools that will help you and your organization produce written health information that is both easy to read and understand.
* * *
"Fixing Common Document Design Problems for Easy Reading"
Jeanne McGee, Ph.D
From its beginning, the field of health literacy was interested in the discrepancy between patients' literacy and the demands placed on them by print materials. This has not changed. The evolution of the field and the expansion of the definition of health literacy should not be taken as evidence that print materials no longer present health literacy challenges. In fact, health literacy research has led to a greater understanding of those features that make print materials more or less useful for people with limited literacy. Jeanne McGee, Ph.D, a healthcare communications and research consultant, made the point that the design of print materials can affect their usefulness among this population.
Dr. McGee created the web-based Toolkit for Making Written Material Clear and Effective, which will soon be released. She explained the focus of that toolkit by reviewing the objectives of the kinds of written materials most likely to be encountered in a healthcare setting. For much health-related print material, she said, the goal is to get the reader to do something.
To do that, the materials must:
- attract the readers' attention
- hold the readers' attention
- make readers feel respected and understood
- Help readers understand
- Help move readers to take action
Dr. McGee used samples of conference attendees' own printed materials to illustrate how design can help you reach these goals. Through interactive critiques of the samples, she offered a number of design tips related to arrangement, images, and color.
Use a page grid. "A page grid is like the underlying skeleton of a page," said Dr. McGee. "It helps you to achieve a clean and tidy look." It defines where you are going to line things up and how. It can also allow you to analyze an already printed piece or one that is still in development, she said. By drawing a page grid over a printed document that works well, you can discover--and then, perhaps, replicate your own materials-the underlying structure. Or you can use a page grid to help you identify and remedy problems in a document whose arrangement seems not quite right.
- Work with "reading gravity." "A reader's typical progression through a printed page, at least in English, is to begin at the top left and work his or her way down to the bottom right by moving across the page from left to right," explained Dr. McGee. So, put a main heading or a strong visual in the upper left portion of the page and arrange text and visuals in a way that works with rather than fights the reader's typical left-to-right, top-to-bottom reading pattern, she said.
- Use subheadings. These can help define a clear path for the reader, she offered.
- Avoid placing an attention-getting element (e.g., a picture or large type) in the bottom right of the page, since this draws the reader's eye there. "When readers reach the bottom right of a page, they get a subconscious signal to turn the page-whether or not they have read everything on it," Dr. McGee explained.
- Make it easy to skim. First impressions are crucial; printed material must appear easy to read. "Avoid creating a 'wall of words,'" she said. Instead of presenting dense, uninterrupted text, use elements like bulleted lists, captions for illustrations, and design accents (e.g., a picture of a telephone next to the helpline number) to help readers pinpoint the main information.
- Make sure that images relate directly to the information and reinforce key messages.
- Try to show only behaviors you want to encourage.
- Avoid abstract images (e.g., parts of the body shown in isolation, cross-sections, highly magnified images).
- Choose images that are culturally appropriate.
- Make sure that the facial expressions and body language of any people pictured are appropriate to the situation.
- Avoid "humorous" images. Humor often does not translate well across cultures, said Dr. McGee, and could end up causing misunderstanding.
- Limit the number of images used. Instead of a lot of small images, she suggested, use one or two and make them large enough for good impact.
- Keep images close to the messages they reinforce.
- Place images so that they do not impede the progression of reading.
- Choose colors that are appealing to the intended readers and are free from unwanted connotations.
- Use color sparingly. Using too much color can lead to "color overload" and distract readers, she said.
- Consider how the colors you choose will reproduce when photocopied.
- Avoid using colored backgrounds, shading behind text, or colored text. Black text on a white background provides the most contrast and so is the most readable, said Dr. McGee.
The most important design tip, said Dr. McGee, is to get feedback from actual users whenever possible. Listening to what your audience says about your printed materials can help you identify any design barriers that are causing them to misunderstand the message of the piece or overlook information. "Just because something is on the page, doesn't guarantee that people will notice it," she said, but getting feedback from users on what they understood in the document can reveal problems as well as suggest solutions to those problems. Talking to users allows you to understand better their literacy-related abilities and challenges and to design materials that are better aligned with their skills.