The first panel was separated into two sections, each followed by a brief discussion section. It began with a presentation defining the concept of numeracy by Lynda Ginsburg, an educational researcher at Rutgers University. This was followed by the presentation of the commissioned paper by Ellen Peters, a professor in the Department of Psychology at Ohio State University. The last speaker was Terry Davis, a professor of medicine and pediatrics at Louisiana State University Health Sciences Center in Shreveport. Davis spoke about her personal experience with numeracy and health literacy as a patient.
Lynda Ginsburg, Ph.D.
Center for Mathematics, Science and Computer Education,
Ginsburg said she is a math educator and the goal of her presentation is to provide a broad description of numeracy from a math education perspective.
Historically, in the United States, numeracy has been subsumed under literacy. For example, the National Assessment of Adult Literacy includes numeracy, and the West Virginia Department of Education defines literacy as “the ability to read, write, and speak in English, and compute and solve problems at levels of proficiency” (West Virginia Department of Education,
2013). Addressing numeracy as a separate issue from literacy is important, she said, because the issues are quite different.
Numeracy is different from school math. Ginsburg quoted Robert Orrill as saying that unlike math, “numeracy does not so much lead upward in an ascending pursuit of abstraction as it moves outward toward an ever-rich engagement with life’s diverse contexts and situations” (Orrill, 2001, p. xviii). The difference is that in mathematics, the problems become more and more abstract. One math concept is the building block that leads to another. Numeracy, however, is applying mathematical reasoning and knowledge in increasingly diverse situations for different purposes.
The term “numeracy” was initially conceptualized in England and its use is relatively recent in this country, Ginsburg said. One of the earlier definitions of numeracy includes the concept of “at-homeness” with numbers and an ability to use math skills, which enable an individual to cope with practical mathematical demands of everyday life. A second definition includes having some appreciation and understanding of information that is presented in terms of numbers. Ginsburg believes this is where the medical field intersects with numeracy. Health and medical information is often presented in mathematical terms, graphs, charts, tables, and references to percentage increase and decrease. Often the concept of percentage increase and decrease is difficult for people to understand, even those within the health care field.
A different definition used in Australia explains that numeracy is a critical awareness that builds bridges between math and the real world and all its diversity. No particular level of mathematics is associated with this concept of numeracy. An engineer must be numerate, but so must a primary school child, a parent, a car driver, or a gardener. Numeracy is based on the different contexts in which each individual functions.
One final definition comes from an American, Lynn Steen, and lists five dimensions of numeracy:
- Practical, for use in everyday life;
- Civic, to understand and engage in public policy issues;
- Professional, to provide skills necessary for employment;
- Recreational, to understand games, sports, and lotteries; and
- Cultural, to be a part of the community and understand cultural context (Steen, 1997, p. xxii).
Ginsburg paraphrased Diana Coben (2000), saying that numeracy is an individual’s ability to use his or her judgment about whether to use math in a situation, what math to use, how to use it, and what degree of accuracy is appropriate. An individual who is out shopping does not want to get to the cash register without enough money. This concern requires keeping
track of the cost of the purchases, but the running total does not have to be exact. In fact, in that particular situation, it makes sense to round up because there will be tax or the shopper may have made an error. Making these kinds of decisions about what makes sense in each situation is the key to being numerate.
The components of numeracy, according to Ginsburg, include (1) context or purpose, (2) mathematical content, and (3) cognitive and affective processes (Ginsburg et al., 2006). When considering numeracy, it is important to think about the context or the purpose for the use of numeracy. This can be for further learning or, more importantly, for tasks in the workplace, the family, or the community. The mathematical content that goes into numeracy can include numbers, operations, patterns, functions, algebra, measurement, understanding measurement, and, for some contexts, shape, and also the use of data, statistics, and probability.
Finally, there are the cognitive and affective processes. Ginsburg cited a National Research Council report titled Adding It Up that defined these processes as the skills necessary to use math proficiently (NRC, 2001). First, a conceptual understanding of the mathematical ideas that are integrated and functional is necessary. A second necessity is reasoning, which is the ability to think logically about the relationships across ideas or within or between ideas and situations. Third is strategic competence, or having the ability to formulate problems and use appropriate strategies to solve them. Fourth is procedural fluency, which is the ability to do the calculations needed to solve the problem. Fifth, and perhaps most importantly according to Ginsburg, is a productive disposition, or the willingness to engage and use math skills to persevere in solving a problem as opposed to giving up. This process may face particular challenges in the United States as the focus of adult education interest has long tended to be on adult literacy rather than across literacy, numeracy, and language as is the case in other countries. Therefore, in American culture it is acceptable to say, “I cannot do math” and “Nobody in my family can do math,” as if low numeracy skills are somehow genetic. While it is acceptable for people to say, “I cannot do math,” it is not culturally acceptable for people to say, “I cannot read.”
In the context of the workplace, a number of studies have shown the extent to which math skills are necessary and useful (FitzSimons, 2005; Hoyles et al., 2001; Marr and Hagston, 2007; Masingila, 1994; Smith, 2002; Zevenbergen and Zevenbergen, 2009). Ginsburg noted that it was interesting how often these skills were evident in work that is not generally considered to be mathematical, such as the work of supermarket employees, plumbers, or carpet layers. Across all of these workplace situations, the numeracy is deeply embedded in the context. What is done and how it is done depends on where numeracy happens and the purpose of the activity. People do things differently in different contexts for different purposes.
Often the numeracy is invisible. If people are asked if they use math in their jobs they will often say no, said Ginsburg, yet if their activities are analyzed, math is at least a part of what they do. These skills are often considered to be “common sense” by workers and not thought of as math skills. Math is often considered to exist only in a school context, with rigid rules and right or wrong answers. However, in the workplace, people develop their own solutions to problems, doing what makes sense to them in ways that make sense to them, Ginsburg said.
Sometimes the math is hidden by technology. For example, often bank employees inform customers about interest rates and securities, yet the employees do not know how the math calculations work because they never see them. The technology performs the calculations. In addition, some procedures become devoid of mathematical meaning in practice. Ginsburg related an anecdote in which she visited a construction site in Trenton where workers were renovating an old brownstone. The workers encountered problems with some beams they were installing and decided to measure the diagonals to determine if the outer walls were of equal length and perpendicular. When asked why they would measure the diagonals, the workers could not explain why, only that it was the way to tell if everything was going to line up correctly. The workers did not know the concepts involved, but they knew the procedure. There are many such examples from daily life.
Community-based numeracy involves interpreting information presented in the community, such as in the media. Ginsburg and Gal (1997) conducted a study in which they asked people to read an article in USA Today about a test that claimed to detect cancer in 90 percent of cases. Then they asked people to interpret that number. The results varied, with some people showing a good understanding of the proportional nature of percentages and others who were unable to do so. Importantly, said Ginsburg, it was not always easy to tell which individuals did not understand the number until follow-up questions were asked.
Another example of numeracy within the community comes from the Philadelphia public schools. Due to budget constraints, teachers were recently asked to take a 13 percent decrease in salary. Perhaps teachers would be offered a 13 percent increase once the fiscal crisis was over. Would this be fair to the teachers? This is more complicated than it appears at first, Ginsburg said, and people need to be able to figure out if they will come out even in this scenario.
Family and personal numeracy include things like shopping, cooking, and health-related decisions that are part of everyday activities. For example, Ginsburg said, an individual who is dieting needs to figure out three-quarters of two-thirds of a cup of cottage cheese. This can be calculated using the methods taught in school math class, but often people will figure
out their own ways of solving problems, and those methods are valid if the solutions are correct. Shopping decisions are another example of family or personal numeracy, Ginsburg added. The store Bed Bath & Beyond sends out two types of coupons. Is it better to get “$5 off a purchase of $15” or “20% off of one single item”? Which coupon is better? The answer depends on the total cost and number of items purchased.
Given how vital numeracy is to everyday life, it is important to understand what Americans know about numeracy. The International Adult Literacy Survey included numeracy under the title of quantitative literacy, Ginsburg said. Quantitative literacy was defined by the survey as “the knowledge and skills required to apply arithmetic operations, either alone or sequentially, to numbers embedded in printed materials” (OECD, 2000, p. x). Of course, not all numbers come in printed material.
In the last two versions of the survey (1992 and 2003), the majority of Americans’ quantitative literacy scores were in the lowest two levels, Below Basic and Basic. Forty-three percent of Americans scored in Below Basic or Basic levels for prose literacy, 34 percent for document literacy, and 55 percent for quantitative literacy, indicating that Americans demonstrated greatest weakness in the mathematical assessment, Ginsburg said.
The International Life Skills Survey, which contains a richer idea of numeracy because it includes how people manage situations and solve problems in a real context, involves responding to information about mathematical ideas. The information may not be in text, but rather involve a skill such as reading a gas gauge and making a decision based on that information, Ginsburg said. This requires the activation of a range of enabling knowledge, behaviors, and processes. This assessment was used in an adult education program study. The scores were expected to be relatively low because the respondents were adult education students in a high school equivalency program. Ginsburg noted that even taking that context into account, math skills were weaker than prose literacy skills.
Ginsburg also cited the General Educational Development test passing rates from 2012. Ninety percent passed the test overall, but only 80 percent passed the math portion (GED Testing Service, 2013). This illustrates that people across the board are weakest in math. At the community college level, the pass rate for all developmental math courses is 30 percent (Bailey, 2009). Overall, Ginsburg said, Americans are weak in math.
Ginsburg reiterated that numeracy can require counting, quantifying, computing, solving problems, and having a clearly right or wrong answer. It can also involve making sense of verbal, pictorial, or text-based messages based on quantitative data, without having to manipulate numbers, but just interpreting them. Numeracy can also mean finding and considering multiple pieces of information to determine a course of action, often without clear, correct answers. Ginsburg concluded by saying that these are
the kinds of situations that arise in everyday life where numeracy is about solving a problem or making a decision, and always with a purpose and within a context.
Ellen Peters, Ph.D.
Department of Psychology, Ohio State University
In her presentation, Peters provided an overview of the roundtable commissioned paper that she wrote with two colleagues, Louise Meilleur and Mary Kate Tompkins (see Appendix A). She began by explaining why numeracy is important within the context of the Patient Protection and Affordable Care Act (ACA). Numbers are ubiquitous in health decisions, she said, whether determining the number of pills somebody takes, deciding what time of day to take those pills, or choosing among different treatment options based on risks and benefits. Numbers instruct, inform, and give meaning to information about health plans, medications, and treatments. But, Peters noted, not all people are able to understand and use numbers effectively when making health decisions. Even highly educated people, those with undergraduate and graduate degrees, are not necessarily numerate. Numeracy and education are related, but they are not synonymous with one another.
People who are innumerate tend to understand less numeric information; however, low numeracy skills are not just about comprehension of numeric information. People who are less numerate use numeric and non-numeric information in ways that are different from the more numerate. Innumeracy influences comprehension, but it also influences the use of information, Peters explained. When careful choices are made to present information in an evidence-based manner, these choices can lessen the effects of numeracy skill levels on how people understand and use information.
The first question is, “What does research show regarding people’s numeracy skill levels?” Americans have limited numeracy skills, and disparities exist in those skills. People who are less numerate are more likely to be female, to be older, to be less educated, and to have lower income. According to Peters, some of these disparities are related to whether an individual currently has health insurance. This is important in the context of the ACA because the people who are going to have greater access to the health system because of the ACA are those people who do not already have health insurance, Peters said.
Numeracy can be measured in many ways, both objective and subjective. Peters focused on the results from the National Assessment of Adult Literacy (NAAL). That assessment estimated the proportion of Americans who fell into four quantitative literacy, or numeracy, performance levels.
Within the NAAL, an estimated 22 percent of the American population falls into a Below Basic level of quantitative performance, Peters said. She added that means that 22 percent of the U.S. population can do fairly simple number operations, such as locate numbers in a text, and perform simple quantitative operations such as addition if they are told that it is addition or it can be easily inferred from the situation. She noted that another third have somewhat more advanced basic quantitative literacy skills, and then another third have intermediate skills that allow them to locate less familiar quantitative information and use that information to solve problems. But, Peters said, only 13 percent of the population is considered proficient in numeracy. That means that 87 percent of the U.S. population cannot solve a problem where they are asked to calculate the yearly cost of life insurance using a table that gives the cost per month for each $1,000 of coverage.
As the first task in preparing the commissioned paper, Peters and her colleagues estimated the numeracy skill levels in the uninsured population. To do so, they examined two datasets. First, the NAAL provides numeracy levels by education and by population levels. U.S. Census data provide information on very similar education levels with health insurance status. Using these two datasets, of course, Peters said, the datasets do not align perfectly, but using both gives an idea of the relative numeracy levels between those people who currently have health insurance versus those people who do not but are likely to obtain it through the ACA.
Table 2-1 provides the results of the analysis. Peters estimated that 29 percent of the uninsured would fall in the Below Basic level of numeracy. That means they would be able to locate numbers in tables, for example, and perform simple operations. But they would not have the skills to perform operations at the higher levels. The uninsured would be less likely than the currently insured to be at the proficient level of numeracy.
As a result, numeracy issues are likely going to be more prevalent in the currently uninsured population that will gain access to insurance through the ACA. This means that health care providers are going to be faced with a different population of patients and consumers in comparison with the currently insured population. Whether providers will be prepared is a question because effective communication is different among people who are less numerate compared to people who are more numerate, Peters said.
In addition, numeracy issues may increase when people are in poor health. Being a patient may reduce deliberative capacity and the ability to think about numbers in particular. Stress may also reduce deliberative
|Quantitative Literacy Level||Percentage of Adults in Each Level (National Assessment of Adult Literacy [NAAL] findings)||Estimated Percentage of Uninsured Adults in Each||Estimated Percentage of Insured Adults in Each||Key Abilities Associated with Level (NAAL)|
|Below Basic||22%||29%||18%||Locating numbers and using them to perform simple quantitative operations (primarily addition) when the mathematical information is very concrete and familiar|
|Basic||33%||33%||32%||Locating easily identifiable quantitative information and using it to solve simple, one-step problems when the arithmetic operation is specified or easily inferred|
|Intermediate||33%||29%||35%||Locating less familiar quantitative information and using it to solve problems when the arithmetic operation is not specified or easily inferred|
|Proficient||13%||9%||15%||Locating more abstract quantitative information and using it to solve multistep problems when the arithmetic operations are not easily inferred and the problems are more complex|
|Total U.S. population||101%||100%||100%|
SOURCE: Peters, 2013.
capacity. Patients are often under emotional stress and are overwhelmed by the quantity of information they are receiving about treatment options or regimens. Often patients are under time pressure because the decision must be made in the physician’s office. According to dual-process theories in judgment and decision making, Peters said, this means that patient populations who are sick may not understand and use numbers as well in decision making, and they may rely on emotional sources of information that are easier for them to process. Little research exists on the topic, but it raises the question of whether current numeracy estimates, from the NAAL, for example, overestimate the skills of patient populations when they are sick.
The second task of the commissioned paper was to examine the numeracy skills necessary to perform a variety of tasks in the context of health and health care. To accomplish this task, Peters’ team separated numeracy skills into two groups. The first group is called education-based numeracy skills, a concept discussed by Apter et al. (2008). These skills consist of knowledge about mathematical content and procedures. Within the education-based numeracy skills, Apter and colleagues identified a hierarchy of numeracy skills that are required to make health decisions. The skills range from very basic tasks, such as locating a number in a table or adding up premium costs, to skills that are somewhat more difficult, such as computational skills or working with frequencies and probabilities. Analytical skills are considered more difficult than computational skills in this hierarchy of numeracy skills. Statistical skills, which are often required for understanding the inherent randomness of life and the role of risk in making health decisions, are considered among the hardest of these skills.
Peters referenced an Agency for Healthcare Research and Quality report by Berkman et al. (2011) that concluded that having a theoretical basis to interventions made for more effective interventions to reduce health disparities. She explained that within the field of psychology of judgment and decision making, there is the idea that numeracy may exert its influence on health decisions and, through the making of health decisions, numeracy may ultimately influence health outcomes (Peters, 2012; Reyna et al., 2004). These emergent decision-based numeracy skills are psychological mechanisms that people have to go through in order to understand and use numbers. Some of these may seem very basic, yet there are differences based on numeracy. People who are highly numerate are actually more likely to seek out numeric information rather than avoiding it. It is not merely putting information in front of people and seeing if they understand it. It is about whether they will find the information for themselves.
Even if patients are given the information, there is the question of whether they will look at it because numeric information is usually given in the context of a great deal of other information, Peters said. People who are highly numerate are more likely to focus on numeric information,
whereas people who are less numerate are more likely to look elsewhere. People who are more numerate also are more likely to ignore irrelevant information or less relevant information on a page. People who are more numerate also are more likely to recall numeric information, which can be important for issues such as medication adherence. There is a great deal of numeric information that must be remembered to facilitate health decision making and health behaviors.
People who are highly numerate tend to be more sensitive to numeric information, whereas people who are less numerate tend to be more sensitive to non-numeric and often emotional sources of information, such as what they have heard from friends and neighbors. The ability of the highly numerate to be more sensitive to numeric information may be due to a particular psychological mechanism, Peters explained. People who are highly numerate seem to derive more affective meaning from numeric information; that is, they are better able to interpret numbers within the context of the decision to be made. They are better able to tell not just that the number is 9 percent, but how good or bad this 9 percent is for them within the context of the decision. Research has shown that if an individual does not have a feel for the goodness or badness of a number, he or she is less likely to use it in judgments and decisions. Part of that greater sensitivity to numeric information that the highly numerate show may be due to their ability to derive meaning from numeric information and from comparisons of numbers, Peters said.
Table 2-2 displays the education- and decision-based skills required for some health decisions. The first column contains the quantitative literacy, or numeracy, level that was estimated for the uninsured population. The second column is a NAAL item that reflects that level of numeracy and the third column is a similar task required in health care decision making. The fourth is a breakdown of the skill categories necessary to the task.
The first example task, comparing and calculating the differences among the premiums of different health plans, falls into the Below Basic numeracy skill level. Peters and her colleagues estimated that about 29 percent of the uninsured population would possess Below Basic skill levels, meaning that most of the population and those with higher level skills would be able to complete this task. Peters stressed that not everyone will be able to complete the task, however. She noted that those who fall within the Below Basic category vary in their actual numeric abilities. She had conducted other studies that showed that about 7 to 9 percent of people who range from ages 18 to 64 were not even able to find very basic information in tables and charts.
Peters noted that in 2008 she coauthored a paper (Greene et al., 2008) that detailed the results of a study in which people looked at two different
|Quantitative Literacy (Numeracy) Level||National Assessment of Adult Literacy (NAAL) Item||Example Task: Health Plan Selection||Skill Categories (Education-based, Decision-based)|
|Below Basic||Calculate the price difference||Compare and calculate the difference between||Basic; Analytical Information|
|(29% of||between two||monthly premiums of||Seeking;|
|uninsured||appliances, using||two plans.||Attention|
|population; most||information in a|
|of uninsured can||table that includes|
|do this)||price and other information about the appliances.|
|Intermediate||Determine what time a person can||“The patient forgot to take this medicine||Basic; Analytical Information|
|(29% of||take a prescription||before lunch at 12||Seeking; Attention;|
|uninsured||medication, based||noon. What is the||Memory (if time of|
|population; 62%||on information on||earliest time he||last meal was not|
|of uninsured||the prescription||can take it in the||provided)|
|likely cannot do||drug label that||afternoon?|
|this)||relates timing of medication to eating.||GARFIELD, Robert M. Dr. LUBIN, Michael DOXYCYCLINE 100MG Take one tablet on an empty stomach 1 hour before a meal or 2 to 3 hours after a meal unless otherwise directed by your doctor.”|
|Proficient||Determine the number of units of||Diabetes management: understanding glucose||Basic Computation; Analytical|
|(9% of||flooring required to||meter readings,||Information Seeking;|
|uninsured||cover the floor in||interpreting sliding-||Attention; Memory;|
|population;||a room, when the||scale regimes,||Information|
|91% of||area of the room is||titrating oral||Sensitivity; Affective|
|uninsured likely||not evenly divisible||medications or insulin,||Meaning|
|cannot do this)||by the units in which the flooring is sold.||adjusting insulin for carbohydrate content. (Note: This example is much more complex than any of the NAAL examples used.)|
SOURCE: Peters, 2013.
health insurance plans—one that was a new concept for health insurance and one that was a more traditional health plan. The authors asked the study participants a number of comprehension questions about these two plans, Peters said. They found that most people understood which plan had the lowest monthly premium, but only about one-third could identify which plan was better if the patient needed a great deal of care. The authors estimated that the more difficult task of determining which was the more valuable insurance is at an intermediate level of proficiency.
Understanding medication and treatment instructions is another example of health care information that requires an intermediate level of proficiency, Peters said, citing an item from the NAAL that addresses health care directly and also requires an intermediate level of proficiency. The example is that of a patient who forgot to take medication and must figure out the earliest time he can take the next dose. Participants are given information about the time of the patient’s last meal and the instructions on the medication bottle. Because about 29 percent of the uninsured population falls within the intermediate numeracy level, according to Peters’ estimate, approximately 62 percent of this population would not be able to answer this item correctly. These individuals may lack the ability to determine the correct way to adjust medication if a dose has been skipped.
Peters said her team estimates that about 9 percent of the uninsured will be at the proficient level, meaning that 91 percent of the uninsured population will not be able to do the tasks at that level. Some of the tasks involved in, for example, diabetes management or other chronic disease management require these higher level skills.
The final question addressed in the commissioned paper is how providers can best communicate with individuals with lower numeracy skills. The proportion of the uninsured population who will be able to correctly perform different kinds of tasks will often depend on how that information is presented. Health materials can require greater numeracy skills or fewer numeracy skills to read and understand, depending on how the materials are formatted. Peters listed some of her recommended strategies for communicating with less numerate individuals:
- Provide numeric information as opposed to not providing it. Numbers inform, educate, and give meaning to information. In short, numbers matter.
- Reduce the cognitive effort that is required. Individuals who are less numerate are less comfortable dealing with numbers. Giving careful attention to the ways in which numeric information is presented is critical among this population. A variety of techniques can help those who are less numerate to better understand and use important health information.
- Provide evaluative meaning for numeric information. This can occur through the use of symbols or interpretive labels. This is particularly helpful when the numeric information is unfamiliar.
- Draw attention to important information. People who are less numerate are less likely to attend to numeric information, even when it is provided. There are techniques that can be used to draw attention to important numeric information.
- Set up appropriate systems to assist consumers and patients. One of the most important parts of these appropriate systems is to determine the goal of the communication. Once a goal has been identified, then the provider or health educator can use the evidence base to find the best way to communicate to the less numerate population to meet that goal.
Health decisions and health behaviors involve a great deal of numeric information either explicitly or implicitly, Peters said.
Peters concluded by saying that the average numeracy skills in the population brought into the health insurance and health care systems by the ACA are likely to be lower than that of the current population in those systems. It is also important to note that they likely have more limited knowledge and experience in health settings. As a result, how information is presented may matter as much as what information is presented, particularly to these less numerate populations. Peters also stressed that communication strategies should be evidence based and that various strategies should be tested within this population to determine which are the most effective.
Moderator: Paul Schyve
Rima Rudd, roundtable member, commented that the example used in Peters’ presentation of comparing and contrasting two different health plan options is considered by educators and developers of assessment tests to be a very sophisticated task that is well beyond basic skills. The hierarchy of skill levels begins with the simplest task of finding one piece of information, then moves on to the more difficult task of finding two similar pieces of information. A process like comparison and contrast is considered to be a high-level skill. Rudd said she thinks that most people at basic-level numeracy would have difficulty accomplishing that task with accuracy and consistency.
Peters responded that a great deal depends on context. People’s abilities will depend on whether they are given information in short form without
other irrelevant information around it or if they are trying to locate multiple pieces of information within a complex format, as will occur with the health insurance exchanges. Tasks can differ in difficulty depending on how they are presented.
Patrick McGarry, roundtable member, commented that some patients distrust numbers generally and asked Peters if she took into account the qualitative nature of numeracy. Peters answered that as far as she is aware there is little research on the qualitative aspect of numeracy. She said that there is evidence that people who are less numerate tend to trust numbers less within the health context and may be more likely to avoid numbers and not focus on numbers even when presented with them. This can make a difference in how much people interact with numeric information or avoid it in making judgments or decisions.
McGarry followed up by quoting Mark Twain as saying there are “lies, damn lies, and statistics” and noting that many people believe this. He said it is a challenge that practitioners face in discussing numbers with their patients. Peters responded that in her view it depends a great deal on how the numbers are presented. For example, she noted that evidence shows that people perceive less risk of adverse events from prescription drugs and are more willing to take the drugs if they are provided information on risk numerically as opposed to some other way. This appears to be the case across numeracy levels. Peters said it also depends on the quality of communication. If the information is communicated in a very complex way, then the patient may be less likely to trust it because he or she does not understand it. Context is always an important issue in numeracy. Ginsburg added that this challenge is related to the fact that people often perceive information according to their biases. She said it is important to be aware of this and remember that people may be responding according to emotion rather than the data that are being given to them. Schyve added that he thought this was an example of confirmation bias, where an individual is more likely to believe something that fits his or her current beliefs. He said this is true in narrative literacy, but also in terms of numbers. Peters responded that some people simply do not trust numbers and do not trust information in the health system. She noted that there can be large cultural differences in the way people respond to information. She thinks, however, that some of that distrust can be alleviated if more careful attention is paid to how information is presented.
Ruth Parker, roundtable member, asked the speakers, “If you could do one thing to have an impact on patient protection and affordability and public health within the Affordable Care Act, what would it be? What are the opportunities in the current environment that could help people in the new system?” Peters said she would have default options available for people through the health insurance exchanges that best suit their needs,
but it is difficult to estimate individual health care needs. The amount of information presented to the consumer on the exchanges can be overwhelming, even for the highly literate and numerate. This can create a highly stressful situation that will prompt strong emotional reactions and cause people to think they cannot cope. She would also reduce the number of choices that people have available through the exchange because having too many choices is also overwhelming and reduces people’s ability to make good choices.
Ginsburg agreed that fewer choices and a default option on the exchanges would help the process become more manageable for most people. She would also like knowledgeable people to be available to walk consumers through their options and help them make the best choices for their situation. Peters added that the experience from Medicare Part D suggests that a number of people will choose their plan based primarily on monthly premium cost. Yet that is not always the best option, particularly for people who require a great deal of health care. Parker responded by saying that she believed those answers helped reframe the discussion from what cannot be done by the consumers to what can be done by the exchanges to help the consumers. Peters added that she believed that studying the challenges of a situation leads to better solutions.
Roundtable member Wilma Alvarado-Little asked how the presenters envisioned the role of the health insurance exchange navigators (those people who will help consumers) in alleviating some of the problems caused by low numeracy and how they would empower navigators to address these issues. Peters said it is important for the navigators to have the communication skills necessary to understand what is important to the consumer, whether that is lower monthly costs or avoiding a potential large lump sum cost. The health insurance exchanges can be thought of as information about different health insurance packages and the navigator is trying to uncover the goals of the consumer in order to guide them to the most appropriate product. The navigator should understand some of the evidence base for effective communication, she said. It is tempting to think that all that is necessary is to provide accurate information and people will make their decisions based on that information. Yet just providing information is often not enough, particularly for people who are less numerate. Tools that allow navigators to quickly display options in a way that individuals who are less numerate are more likely to understand would be a valuable resource for navigators to have.
Roundtable member Steven Rush commented that the issues of health literacy and numeracy go beyond the acquisition of health insurance. Once people have been brought into the system, how can information be presented that allows them to make appropriate decisions about their health care? He added that the role of stress in decision making is very important
and that even highly numerate people can be innumerate in crisis situations. Ginsburg agreed that the big decisions in the short term are about choosing and enrolling in a plan, but over the long term interacting and negotiating with insurance companies may be the larger challenge. She noted that it will be important for patients to be able to advocate for themselves and family members and enlist provider support at times to support them in their interactions with insurance companies. This may be an empowerment issue as well as a literacy and numeracy issue.
Rush responded that high numeracy to a certain extent brings empowerment. He wondered what would be the best way to provide numeric-based information to people when they are already in a health plan so that they can communicate better with their physicians and understand their treatment options. Peters said this is a huge topic and there are a number of strategies that can be applied, depending on the specific health situation and the goals of communication within that health situation. First it is important to ask, “What is the most important information for the patient to understand and use?” Once the most important information is determined, then delete the less important information, Peters said. Some techniques involve informing the consumer or patient and helping him or her understand the numbers and their meaning, while other techniques are persuasive. For example, warning graphics on cigarette packages are meant to be informative, but also persuasive. Different strategies are chosen depending on the goal of the communication. Peters added that this brings up the topic of educating health care providers about communication strategies, which is often an overlooked strategy in improving health. The people providing the information must know the best way to communicate in different situations.
Schyve commented that there are ethical questions contained in the suggestions to limit information and choices for consumers, and that different situations would call for different solutions. For example, public policy makers may decide there would be a limited number of insurance plans offered through the exchanges. Regarding the question of choice in treatment, however, it may be unethical to withhold some information even if it makes the options more difficult for the patient to understand. This may mean that it is up to the provider to spend more time helping the person understand his or her options. There could still be a default choice, but all available information must be given to the patient.
Benard Dreyer, roundtable member, asked about the relative importance of written versus verbal communication when communicating numeric information. Peters answered that there has not been much research on verbal communication of numeric information; most of the research has been done on written communication. That raises some questions because most health care providers communicate verbally. Peters thinks numeric
information would be better communicated verbally with written supporting materials, but there is no research to support it. It is an important research question, but it is often overlooked because it is difficult to study.
Dreyer then asked what strategies work best when dealing with information that is so complex that it demands numeric proficiency from the patient. For example, Dreyer said, diabetes care requires a high level of numeracy that is difficult to simplify. Peters said it is likely there are studies concerning numeracy and diabetes, but she is not aware of them. She said it is helpful to give people concrete indicators of some action. For example, if the blood sugar level hits a certain point, then a specific action is required. This gives people something to remember or write down and have available to use as a guide. Helping people understand the goodness or badness of the numbers involved in their care is also important. For example, a patient given a range of numbers may not understand that a certain number in front of him or her is in that range. The provider must help the patient understand what numbers are in the range and whether they are good or bad. Ginsburg added that from a mathematical education perspective, multiple representations of a concept help people understand and learn math. It would be helpful for patients struggling to make sense of new information to be given the information in different ways. Patients can receive verbal instructions along with tables and written information that they can study or graphical information that they can review with the provider.
Cindy Brach, roundtable member, noted that people with low numeracy skills would likely have a great deal of insight into this situation because they must develop coping strategies. For example, Brach said she once spoke with an adult learner who told her that although he was a truck driver, he could not read a map. To cope with this situation he would stop at diners along his route and chat with people to get directions, which he would write down and stick to his windshield. It would be worthwhile to ask patients the best way to help them learn and remember what they need to know.
Robert Logan from the National Library of Medicine (NLM) said that both presenters spoke about research showing that increasing a person’s interest, engagement, involvement, and eventual numeracy capability could result from providing tailored materials for them and other strategic interventions. Both presenters also mentioned that an opportunistic, contextual transformation occurs based on a person’s role. For example, a website called patientslikeme.com illustrates that people with no medical training can develop an impressive knowledge of medicine in the right circumstances. Logan asked if there was any research literature on how to take advantage of that opportunity with a patient and help the patient overcome any literacy or numeracy issues in that context. Peters answered that she was unaware of any research explicitly on the transformation pro-
cess. Some studies look at people’s ability to choose high-quality hospitals given a variety of information. Researchers found that people who are more numerate are more likely to choose high-quality hospitals, and people who are more health literate according to reading levels are also more likely to choose high-quality hospitals. Yet women, who tend to be less numerate, are more likely than men to choose high-quality hospitals. This is likely because women are more likely to have more experience within health settings. Women are more likely than men to be responsible for family health care and as a result have developed additional skills. Peters added that there has been a great deal of research on the patient activation process, which is a related topic. She recommended the work of Judith Hibbard of the University of Oregon. Hibbard’s research focuses on taking patients from one level of activation to another.
Terry Davis, Ph.D.
Professor of Medicine and Pediatrics,
Louisiana State University Health Sciences Center, Shreveport
Does poor health affect numeracy? The answer, Davis said, is probably, but she could find no studies in the literature on the topic. However, many studies indicate that poor health and chronic disease can impact cognition, but none were specifically related to numeracy.
Davis spoke about her personal experience as a patient, noting that since she has been on the faculty of a medical college for 30 years and has done extensive research into health literacy, she should have proficient health literacy skills. At age 60, Davis was very healthy and had never missed a day of work because of illness. She took no prescription or over-the-counter (OTC) medications and was unfamiliar with the benefits and restrictions of her health insurance plan. A visit to the cardiologist changed everything when she learned that she would need open-heart surgery to correct a heart defect. Although Davis found her diagnosis confusing and overwhelming, she was able to conduct some online research and consult with friends to help her make decisions about her care. She did not, however, ask her employer or health insurer about possible restrictions on providers or hospitals.
Davis chose the best place for her procedure based on quality of care considerations without taking into account that the provider was not part of her insurer’s network so, unbeknownst to her, it entailed a 30 percent copay. The admission and preoperative process was very well organized
and the surgeon spent time with Davis and her family answering questions. The discharge process after the surgery was a much more disorienting and rushed experience. The discharge nurse listed Davis’ medications very rapidly and seemed annoyed when asked to write down the indication for each medication. Davis said she felt lost and overwhelmed and remembered hoping that her husband was better able to understand the instructions than she was.
Davis said she found the names of the various prescription medications difficult to remember and pronounce. In addition, there are differences between brand names and generic names for the same medication, and it can be embarrassing for patients to mispronounce medication names in front of health care providers. She felt unsure of the medication instructions and she was far from home and her usual pharmacy. Her husband filled her prescriptions. The pharmacist did not give any oral instructions and Davis’ husband did not ask questions. The first night out of the hospital, Davis was unsure if she had been given all of her medication for that day when at the hospital. She called the hospital where the procedure was performed, but her file had already been deleted from the computer so the nurse was not able to tell her what medication she had been given at discharge. Davis said she felt vulnerable and overwhelmed by the situation.
The instructions on Davis’ medications varied and were written in a convoluted manner. For example, the label of one medication read, “Take 1 tab (10 mg) by mouth once daily except Tuesday and Thursday; 1 & ½ tabs (15 mg) once daily on Tuesday and Thursday,” which she found very confusing and hard to read. Davis’ medication list was extensive and she was often not told why she had been prescribed certain medications or how long she would be taking them. In addition, she was told to take some medications “as needed,” but was not given any instructions on what symptoms would indicate need.
Davis also spoke of her experience with pain medications. She was discharged from the hospital with OTC and prescription pain medications, but not told whether to take them concurrently. In addition, the prescription medications also contained acetaminophen, a common ingredient in many OTC medications that can be dangerous in high doses. Davis was never warned about her acetaminophen dosage or instructed to be aware of how much she was taking. After conducting research with colleagues, Davis said she found that the majority of consumers do not read OTC medication instructions. In addition, people develop their own schematics for taking OTC medication based on what they have always done or what others around them have done, but not based on the instructions provided with the medication.
Davis said additional medications for conditions unrelated to her heart were added to her drug regimen. The dietary instructions related to these
medications were often confusing and difficult to follow. For example, the instructions for an osteoporosis medication to be taken once a week read, “Take with 8 ounces of water at least 30 minutes before first food or beverage of the day. Don’t lie down for 30 minutes.” Davis said she found this very difficult to work into her routine, so she devised a modified regimen of her own. This was particularly difficult to do when traveling, she said, so she did not take this medication when she traveled. Many medications look alike and can be difficult to tell apart. This is especially worrisome when using medication organizers. For some medications it may not matter if a patient accidentally takes a double dose, thinking he or she is taking two different medications, Davis said, but for other medicines that can be dangerous. When the prescription changes from a brand name to a generic, the size, shape, and color of the pill also may change. This can be very confusing for the patient.
Davis noted that she and many others must also face the well-meaning suggestions of friends and family members regarding their illness and treatment plan. Patients may be encouraged to stop taking some medications based on a news story or website, or encouraged to supplement their medications with alternative medicines. She cautioned that clinicians must be aware of these external influences and their impact on a plan of care.
System redesign does not have to be complex, Davis said. It can be as simple as responding to Joint Commission recommendations and implementing teach-back methods before and after a procedure. Providers can simplify medication instructions and solve numerical problems for the patient. Emerging research suggests that a simplified, clearer medication label can affect understanding and adherence.
Other issues concern medicine and numeracy, Davis said. For example, does having alcohol with dinner affect one’s ability to take medication at night? Being distracted or sleep deprived might also affect the ability to follow medication instructions. Feeling stress or being away from home may also have an impact. Finally, the medications themselves may affect numeric skills and cognition.
Davis said she learned several important lessons about following a complex drug regimen. First, hospital discharge is the beginning of a process that requires time and energy to manage. Second, patients must understand the treatment plan and be able to engage in problem solving to make it work. This includes learning to embed medicines into everyday life and manage disruptions to the routine. Patients must plan ahead and develop confidence in communicating with providers and insurance companies. Finally, Davis said, high literacy and assertiveness do not guarantee adequate health literacy.
Insurance was another area that tests even the strongest numeracy skills, Davis said. It is difficult to find answers to questions on which policy
to choose and what various policies cover. Bills from providers and statement of benefit documents from insurance companies can be hard to read and understand. Even employees of insurance companies often have a difficult time explaining the statement of benefits to consumers. It is difficult to know which source of information is trustworthy, Davis added. Insurance companies sell many types of policies, and it is hard to understand the differences or know which one is best for an individual.
Davis cited an article in The New York Times by Gina Kolata (2013) about trying to determine the cost of a medical procedure. Kolata was attempting to find the cost of a vaginal delivery for her daughter, who was uninsured. In the article, Kolata quotes Dr. Uwe Reinhardt from the University of Pennsylvania as saying that hospitals are not required to tell patients the cost of a procedure upfront and often hide that information until they send a bill. Private insurers claim they let patients know what out-of-pocket costs are likely to be, Davis said, but when Kolata checked with one insurer’s website and called the company hotline, she could not find out any information on cost. In the article the question is asked, “How can people make good choices about health care if they cannot find out about cost or quality?”
Davis said if she could effect change in the health care system, she would provide clear and accessible information on cost and quality and have informed and friendly navigators available by phone to personally assist with medicine and insurance questions. She would do away with the phone trees, long holds, and suggestions to call another department that she endured. It took Davis 13 months to settle the insurance for her surgery. Davis also said she would mandate patient-centered hospital discharge instructions; universal and easy-to-read and -navigate instructions on all prescription and OTC bottles; and what she calls an “Apple store” approach to buying and using insurance. Davis noted that when she goes into an Apple store the cost is obvious and the staff are objective and patient when explaining the technology to customers, who may be overwhelmed or do not know what they need.
Davis encouraged those at the workshop to form an action plan to address some of these issues. She noted that health literacy is the interaction between the skills and abilities of the patient and the demands of the system, and that providers need to be prepared to address the demands of the system. This could take the form of standardizing prescription and OTC labels and medicine guides to make them easier to see, navigate, understand, and follow.
Davis concluded by reminding the audience that technology is a tool that does not replace a nice, knowledgeable person. As many in the room know, Davis said, health is personal.
Moderator: Paul Schyve
Benard Dreyer, roundtable member, commented on the issue of patients using the appearance of a medication as a tool to understand what they are taking. In his practice the doctors have been experimenting with different asthma action plans and have found that many families use the color of the inhaler to determine which medication it is. Dreyer said it would be helpful if inhaler labeling or packaging were standardized so that all albuterol is one color and all steroids are another color. He asked if there was a way to use the appearance of medications to overcome low health literacy and numeracy. Davis replied that she thinks people respond to color. She related the story of buying a new printer and being guided by colors to set it up—the blue goes with the blue and the green goes with the green. Technology companies use this method to help people who are not experts to use their products correctly, Davis said.
Robert Logan from the NLM commented that the NLM was completing work on a website that contains high-quality images of pills to help consumers identify medications. The website is operational as of summer 2013 and serves as an important resource for patients.1
Rima Rudd, roundtable member, commented that one of the issues around pills and numeracy is that instinctively people assume that small is small and big is big and big is more powerful than small. When patients move from a brand name drug to a generic drug, often the pill sizes change and there are times when the pill size for a larger dosage is smaller than that of a smaller dosage. This is very difficult for people to manage and leads to medication mistakes. Rudd said that size is a system issue that must be addressed along with color and name. Names of drugs are often chosen because of the sound, Rudd said. For example, drugs that are meant for heart-related conditions have strong sounds in them such as Ks and Ts and other powerful consonants. Drugs that are meant to soothe and relax have soft sounds such as Ss, Cs, and Xs. This means that medications for similar conditions almost all sound the same making it very confusing for people to remember the name of their specific medication.
Kim Parson, roundtable member, commented that the research on the patient-centered label is very interesting and she would like to understand more about it. She noted that in terms of adherence one of the challenges is helping people understand what to take and when. Parson agreed that health care is personal and added that the even the same procedure is dif-
ferent for every patient. Davis responded that presentations later in the day would give more information about the current state of the research. She said that researchers are working toward understanding how to help people with their medications but there is no one solution that will help everyone. No matter how patient centered the final label is or how unique the pill shape, some people will still need personal help.
Susan Pisano, roundtable member, commented that although Davis had many advantages she still felt overwhelmed by the complexity of her care regimen and the system. She asked if there was anybody or anything that was helpful in guiding Davis through the system. Davis answered that there was nothing within the system that was helpful but that she relied on outside friends and contacts to give her advice. Pisano asked about the day before Davis’ procedure, which had gone well, and what it was about that day that worked well and was helpful. Davis answered that the care before the procedure was organized and convenient. Everything was in one place and was organized with the patient in mind.
Margaret Loveland, roundtable member, commented that speaking from a pharmaceutical industry point of view she understands the frustration of the different look and size of generic medication. For example, the day after Singulair, a medication manufactured by Merck, went off patent there were 20 generics approved that came onto the market. Each one looked a little different and it is easy to see how patients get confused. The active ingredient in generic medication is nearly always the same or at least very similar, but the inactive ingredients are different. Davis asked if the pharmaceutical companies are trying to make the names easier. Loveland answered that it is difficult to get new names. There are few one- or two-syllable names and they are all gone, so the names continue to get more complex.
Loveland also said that she believes that for people coming into the health care system via the ACA there will have to be better communication about why they should have health insurance. The costs of health care and reasons for insurance will have to be explained in a way that people can understand.
Apter, A., M. Paasche-Orlow, J. Remillard, I. Bennett, E. Ben-Joseph, R. Batista, J. Hyde, and R. Rudd. 2008. Numeracy and communication with patients: They are counting on us. Journal of General Internal Medicine 23(12):2117-2124.
Bailey, T. 2009. Rethinking developmental education in community college, CCRC Brief. Community College Research Center, Teachers College, Columbia University. http://ccrc.tc.columbia.edu/media/k2/attachments/rethinking-developmental-education-in-community-college-brief.pdf (accessed October 30, 2013).
Berkman, N. D., S. L. Sheridan, K. E. Donahue, D. J. Halpern, A. Viera, K. Crotty, A. Holland, M. Brasure, K. N. Lohr, E. Harden, E. Tant, I. Wallace, and M. Viswanathan. 2011. Health literacy interventions and outcomes: An updated systematic review. Evidence Report/Technology Assessment No. 199. (Prepared by RTI International–University of North Carolina Evidence-based Practice Center under contract No. 290-2007-10056-I.) AHRQ Pub. No. 11-E006. Rockville, MD: Agency for Healthcare Research and Quality.
Coben, D. 2000. Mathematics or common sense? Researching invisible mathematics through adults’ “mathematics life histories.” In Perspectives on adults learning mathematics: Research and practice, edited by D. Coben, J. O’Donoghue, and G. E. FitzSimons. Dordrecht, The Netherlands: Kluwer Academic Publishers. Pp. 53-66.
FitzSimons, G. E. 2005. Numeracy and Australian workplaces: Findings and implications. Australian Senior Mathematics Journal 19(2):14-40.
GED Testing Service. 2013. 2012 annual statistical report on the GED Test, GED Testing Service. http://www.gedtestingservice.com/uploads/files/8d4558324628dfcf1011dc738acca6eb.pdf (accessed October 30, 2013).
Ginsburg, L., and I. Gal. 1997. Uncovering the knowledge adult learners bring to class. In Adults returning to study mathematics, edited by G. E. FitzSimons. Adelaide, Australia: Australian Association of Mathematics Teachers. Pp. 55-61.
Ginsburg, L., M. Manly, and M. J. Schmitt. 2006. Components of numeracy. Occasional paper. Cambridge, MA: Harvard Graduate School of Education, National Center for the Study of Adult Learning and Literacy. http://www.ncsall.net/fileadmin/resources/research/op_numeracy.pdf (accessed October 30, 2013).
Greene, J., E. Peters, C. K. Mertz, and J. H. Hibbard. 2008. Comprehension and choice of a consumer-directed health plan: An experimental study. American Journal of Managed Care 14(6):369-376.
Hoyles, C., R. Noss, and S. Pozzi. 2001. Proportional reasoning in nursing practice. Journal for Research in Mathematics Education 32(1):4-27.
Kolata, G. 2013. What does birth cost?: Hard to tell. New York Times, July 8.
Marr, B., and J. Hagston. 2007. Thinking beyond numbers: Learning numeracy for the future workplace. Adelaide, Australia: National Centre for Vocational Education Research.
Masingila, J. O. 1994. Mathematics practice in carpet laying. Anthropology & Education Quarterly 25(4):430-462.
NRC (National Research Council). 2001. Adding it up: Helping children learn mathematics, edited by J. Kilpatrick, J. Swafford, and B. Findell. Mathematics Learning Study Committee, Center for Education, Division of Behavioral and Social Sciences and Education. Washington, DC: National Academy Press.
OECD (Organisation for Economic Co-operation and Development). 2000. Literacy in the information age: Final report of the International Adult Literacy Survey. http//www.oecd.org/edu/skills-beyond-school/41529765.pdf (accessed May 14, 2014).
Orrill, R. 2001. Preface. Mathematics and democracy: The case for quantitative literacy, edited by L. Steen. The National Council on Education and the Disciplines: The Woodrow Wilson National Fellowship Foundation. http://www.maa.org/sites/default/files/pdf/QL/MathAndDemocracy.pdf (accessed October 29, 2013).
Peters, E. 2012. Beyond comprehension: The role of numeracy in judgments and decisions. Current Directions in Psychological Science 21(1):31-35.
Peters, E. 2013. Numeracy and the Affordable Care Act: Opportunities and challenges. Presentation at the Institute of Medicine Workshop on Health Literacy and Numeracy, Washington, DC, July 18.
Reyna, V. F. 2004. How people make decisions that involve risk: A dual-processes approach. Current Directions in Psychological Science 13:60-66.
Smith, J. P. 2002. Everyday mathematical activity in automobile production work. In Everyday and academic mathematics in the classroom, edited by M. E. Brenner and J. N. Moschkovich. Reston, VA: National Council of Teachers of Mathematics. Pp. 111-130.
Steen, L. A., ed. 1997. Why numbers count: Quantitative literacy for tomorrow’s America. New York: The College Board. P. xxii.
West Virginia Department of Education. 2013. Adult literacy defined. https://wvde.state.wv.us/abe/definingstatus.htm (accessed October 29, 2013).
Zevenbergen, R., and K. Zevenbergen. 2009. The numeracies of boatbuilding: New numeracies shaped by workplace technologies. International Journal of Science and Mathematics Education 7:183-206.