Data for this review were identified by searches of Medline, PsychInfo and the Cumulative Index to Nursing and Allied Health Literature using combinations of the search terms: “health literacy”, “cancer”, “physician–patient relations” and “test of functional health literacy”. A total of 98 abstracts were initially identified. Review articles and abstracts not related to cancer care were excluded, with a total of 47 articles remaining for review. Additionally, the reference lists of retrieved articles were reviewed. Only articles published in English were included. All articles published up until October 2010 were included in the review.
Definition of health literacy
Health literacy is a relatively new concept in clinical practice. The term was originally coined in 1974, though it has received significant attention only in the last decade.5 One of the more widely used definitions is “the capacity to obtain, process and understand health information and services to make sound health decisions”.6 According to the same report by the United States Institute of Medicine, general literacy skills form the basis of a person's health literacy, with health literacy being the active mediator between individual skills, including their cognitive abilities, social skills and emotional states, and the health context, for example, the media, government agencies and health-care providers.6 Hence, health literacy is more than just literacy and numeracy skills in the health context. Figure 1a illustrates the framework of health literacy demonstrated above. “Health literacy” is an umbrella term encompassing different aspects of patients' interaction with the health system from the patient's perspective. This includes the ability to communicate, to be involved in clinical care and to access health care.7 A simple yet comprehensive framework of health literacy from the patient's perspective, as illustrated by Nutbeam, divided health literacy into three main aspects, namely functional health literacy, interactive health literacy and critical health literacy, in order of increased complexity.8 Functional health literacy refers to the ability to understand health information provided. Interactive health literacy refers to the capacity to incorporate social skills and functional health literacy skills to assume good health-seeking behaviour. Finally, critical health literacy is about applying both functional and interactive health literacy skills to critically analyse information provided for making decisions about personal health care. Figure 1b shows Nutbeam's framework of health literacy.
Figure 1. Models of health literacy showing (a) US Institute of Medicine framework of health literacy6 and (b) Nutbeam's framework of health literacy.8
Download figure to PowerPoint
Measurement of health literacy
In recognition of the importance of health literacy in clinical practice, the measurement of health literacy has been an area of burgeoning research. Various types of instruments have been developed to measure health literacy, including proxy assessments of health literacy, the direct testing of individual health literacy abilities and self-report of health literacy abilities.9 While a few studies have used proxy measurements of health literacy, such as education level,10,11 numerous instruments that directly test individual health literacy abilities have been developed to specifically measure health literacy.12–14
Self-report measures of health literacy are less often preferred, as patients with poor health literacy are often embarrassed to admit problems in this area.5,15–17 The use of socioeconomic factors as a proxy measurement of health literacy is convenient but may not be an adequate surrogate measure of health literacy. For example, several studies have shown a significant mismatch between education attainment and reading ability.18 Buchbinder et al. reported that, despite the strong association between health literacy and education level, the use of education level alone as a proxy measure of health literacy would have misclassified more than 10% of their survey respondents as having good or limited health literacy.15
A large international study on adult literacy and life skills (ALLS) coordinated by Statistics Canada and the Organization for Economic Co-operation and Development had a specific domain measuring health literacy in its purpose-built instrument.19 Other direct health literacy testing tools that have been developed include the rapid estimate of adult literacy in medicine (REALM), which measures a person's ability to correctly pronounce common medical words and lay terms14 and the test of functional health literacy in adults (TOFHLA), derived from hospital materials to assess reading comprehension and numeracy skills.12 Given that the TOFHLA takes up to 22 min to administer, compromising its potential routine use in clinical practice, a shortened version of TOFHLA has been developed (S-TOFHLA), which takes up to 12 min to administer.13 The S-TOFHLA has proven to be a reliable and valid test of health literacy, with good criterion validity compared to the TOFHLA and REALM.13
Other direct testing instruments, like the newest vital sign20 and less robust instruments using screening questions to identify patients with poor health literacy have been developed.8,21 These instruments tend to be simple and can be administered quickly, potentially allowing their use in routine clinical practice. However, they lack sensitivity and specificity when compared to the TOFHLA, which has been regarded as a standard for assessment of health literacy.22 Despite the widespread use of TOFHLA as an instrument, there is a lack of consensus on the preferred instrument to measure health literacy levels accurately.9 Various instruments could be used together. For example, the newest vital sign may be administered as a screening tool and the TOFHLA can be used as a second tool to measure health literacy levels more accurately.
Health literacy is considered to include functional health literacy, interactive health literacy and critical health literacy. However, the TOFHLA has been criticized for measuring only patients' functional health literacy, which includes their reading ability, comprehension and numeracy skills, but not the broader context of health literacy.9 Recent research has demonstrated that there are seven key abilities that people need to be able to utilise health information.8 These are: knowing when to seek health information, knowing where to seek health information, verbal communication skills, assertiveness, literacy skills, the capacity to process and retain information and application skills.8 Several authors have thus suggested that measures of health literacy should assess these domains.9 The health literacy management scale is one measure that seeks to assess health literacy more broadly.23
In addition, a number of self-report measures of health literacy have been developed. One example is the set of brief screening questions, which asks respondents to rate their confidence and difficulty in understanding health information, using a five-point Likert scale.24 In addition, Scales for measuring functional, communicative and critical health literacy was developed to assess the three aspects of health literacy as per Nutbeam's framework.25 However, the use of these instruments has been limited.
Prevalence of suboptimal health literacy
Numerous studies have examined the prevalence of health literacy in various clinical settings as well as at a population level.1,10,14 Variability of measurement methods has contributed to differences in the reported prevalence of limited health literacy. For example, the ALLS study in the US and Australia, using purpose-built questionnaires with a dedicated health literacy component, suggest that significant proportions of the population (30–50%) have limited health literacy.19 In the Australian survey, higher education level was associated with better health literacy scores.19 However, more than 30% of those with a bachelor degree were classified as having poor health literacy.19 Respondents with a bachelor degree generally performed better in the other domains of literacy and numeracy, including prose literacy, document literacy and numeracy. About 20% of those with a bachelor degree had poor prose literacy.19 A similar proportion had poor document literacy and almost a quarter had poor numeracy.19 While several studies have also revealed consistent findings using the TOFHLA, both at a population level and specific clinical settings,17,26–28 a few studies using TOFHLA as the main instrument reported only a modest proportion of the population (5–10%) with limited health literacy.5,15,29,30
The TOFHLA was developed and validated through interviews with native English speakers.12,13 In subsequent research the study groups using TOFHLA were varied. Some investigators included only native English speakers as participants31 while others recruited non-native English speakers as participants, provided they could understand English.5,15,29,32
Despite discrepant findings regarding the prevalence of poor health literacy, the assessment of individual health tasks via the TOFHLA have consistently found that many people struggle with simple health tasks, such as reading and interpreting a prescription label or an appointment card. A cross-sectional survey of 2659 patients in two urban hospitals in the USA found that up to 41.6% of patients could not understand directions for taking medication on an empty stomach and 26% of patients could not comprehend information about when the next appointment is scheduled.26
In addition to population surveys, various cross-sectional surveys in specific disciplines of medicine have been conducted, including rheumatology, general practice, pediatrics and geriatrics.1,15,27,32–34 This has brought about a better understanding of the extent of limited health literacy in specific clinical settings, which may differ due to different patient socioeconomic characteristics and the nature of the illness.35
There has been very limited research exploring health literacy in cancer patients.36–38 Only a few studies have involved cancer patients or cancer survivors, generally studying perceptions of the risk of cancer recurrence or satisfaction with the medical decision-making process.36–38 Previous research on health literacy and cancer has focused mainly on the correlation between the health literacy level and participation in cancer screening programs among non-cancer patients.39–44 A recent study measured the prevalence of health literacy in head and neck cancer survivors with a total laryngectomy.45 However, the study was conducted retrospectively and involved only eight participants. Moreover, previous research on health literacy and cancer has tended to use convenience sampling or targeted only very limited groups, such as elderly people or certain ethnic groups, like African-Americans, who tend to have lower educational levels.30,39–44,46
Associations and consequences of limited health literacy
Poor health literacy could be a result of limited abilities and/or high health literacy demands. Jordan et al. demonstrated that an individual's health literacy is dependent on the relationship between individual capacities, the health-care system and society.7 Hence, besides a person's innate abilities, the complex health-care system and other services in society may make a high demand for health literacy on patients. Numerous studies have shown that patients with poor health literacy tend to have difficulties obtaining, understanding and retaining health information.37,47 Suboptimal health literacy, as measured by the TOFHLA, is an independent risk factor for poorer health outcomes, including increased risk of hospitalization.32,48
Patients' socioeconomic factors, especially old age and a low education level, are strongly associated with lower levels of health literacy.5,12,29,30 One study showed that people without formal education were sevenfold more likely to have limited health literacy than those with formal education.29 According to Helitzer et al., people with a high education level may still have difficulties dealing with health-related information and decisions.49
The association between old age and poor health literacy is particularly important in the cancer setting. This is because 60% of new cancer diagnoses and 70% of cancer deaths involve patients over 65 years old.50 Furthermore, the population is ageing in many developed countries. Several studies have explored the association between health literacy levels and gender, though results have been inconsistent.6,29,51
Poor understanding of health information by cancer patients could negatively impact on patients' distress levels. As a result of poor understanding of health information, patients may feel dissatisfied with their care and have reduced overall well-being.4 In a study of women with early stage breast cancer, patients who felt that they had been inadequately informed were twice as likely to be depressed or anxious than those who felt they were well informed.52 According to Gamble, patients who did not receive adequate information about adverse effects of their treatment or who were expecting only minor adverse effects were likely to suffer greater anxiety than the rest.53 Furthermore, a study of patients with prostate cancer found that those with poorer health literacy tended to associate prostate cancer with death, fear and suffering, without much consideration for the stage and nature of disease.54 Hence, well-informed cancer patients seemed more likely to cope better with their disease than those who are inadequately informed.52
Communication issues that hinder understanding and informed decision-making
Poor communication by physicians may place a high health literacy demand on patients.55 Physicians may use medical jargon or long and complex sentences when communicating with patients.56 In fact, medical terms that are regarded as “plain English” by medical professionals may be challenging for patients with suboptimal health literacy.47 A study by Samora et al. on knowledge of medical vocabulary among patients from a public general hospital reported that only 35% of patients understood the word “orally”, 18% had a good understanding of the word “malignant” and 13% understood “terminal”.57 Kilbridge et al. sought understanding of terms commonly used in prostate cancer and found that less than 50% of men with prostate cancer understood common terms used in written materials such as “erection” and “impotent”.56 Moreover, only 25% of patients understood “bowel habits” and a mere 5% of patients understood “incontinence”.56 Doctors may be unaware of the prevalence of low health literacy in the community and may fail to recognize that patients do not understand because they lack the capacity to learn,58 and thus place a high demand for health literacy on patients when communicating with them.
A study on understanding of informed consent in cancer patients seeking treatment found that only 60% of them could recall the purpose and nature of the suggested treatment procedure one day after the information was explained to them.59 When discussions include an initial cancer diagnosis or when bad news is expected, some patients are then unable to absorb further information, as their emotional responses may interfere with understanding of health information.60 Ineffective communication between patients and health professionals could result in patients turning to alternate sources of information. A common practice by patients with low health literacy is to learn about cancer from the previous experiences of family members and friends, who may have advanced stage cancer.55 Furthermore, they struggle to clarify their concerns about cancer with health care professionals.47,55 A US population-based survey reported by Gansler et al. revealed several misconceptions about cancer treatment.61 These misconceptions could also affect patients' acceptance of and adherence to their treatment plan. For example, almost half of the respondents in this study believed that treating cancer with surgery could cause it to spread throughout the body.61 In addition, 27% of respondents thought that the medical industry is withholding a cure for cancer from the public in order to increase profits.61 Studies have demonstrated that optimally prescribed pain control treatments can successfully control pain in more than 80% of cancer patients, however Gansler et al. found that 32% of respondents believed (or were uncertain) that pain medications are not effective at reducing the amount of pain that cancer patients suffer.61 Although levels of health literacy were not assessed directly in this survey, the results indicate that some well-established medical facts about cancer were poorly understood by a sizable proportion of the public, suggesting poor health literacy. The education levels of respondents were recorded, with 13% of respondents having less than the average education level of the general US population.61 Respondents with higher education levels had fewer misconceptions, though some misconceptions were still held by those with a high education level.61 This could affect patients' adherence to treatment plans, which may impact negatively on their physical health. For example, a study on communication with cancer patients revealed that those who felt they had not received adequate information were more likely to pursue alternative therapies, such as the Moerman diet.62 In fact, patients with low health literacy were 1.5 to threefold more likely to experience poor health outcomes than those with adequate health literacy.63
Consequences of limited health literacy on cancer care
The concept of health literacy is especially important in the cancer setting, where making decisions on prevention, screening and treatment are becoming more complex.64 Adequate health literacy is central to effective disease prevention and chronic disease self-management.65 Cancer patients are often presented with multiple, complex and individualized options on cancer treatment,66 hence patients with low health literacy may be disadvantaged.
Patients with suboptimal health literacy could have difficulty interpreting information about their disease when communicating with health professionals. Recent studies have revealed that more than 60% of patients overestimated their chances of a cancer cure by 20% or more than that of their doctors.55,66 Mackillop et al. reported that more than half of patients overestimated their chance of cure.67 One-third of patients with metastatic cancer from different primary cancers actually thought that the cancer was localized and one-third of patients receiving treatments with palliative intent believed that they were undergoing curative treatment.67 In this study, 46% of patients did not complete high school education. These patients were significantly more likely to have misconceptions regarding treatment intent than those who completed high school education (50 vs 25%).67 Similarly, 72% of patients without high school qualification overestimated their chances of cure, compared to 38% of those with a high school qualification.67 Despite the associations, misconceptions were still prevalent among patients with higher education levels. As a result of the misconceptions, patients may draw incorrect conclusions from the data provided to them about the chance of their cancer responding to treatment.47 This may lead to unnecessary interventions or under-treatment.38
Participation in cancer screening programs
Studies have revealed that people with poor health literacy are significantly less likely to participate in disease prevention and health promotion programs.68 According to Dolan et al., men with poorer health literacy are 1.5 times less likely to know about colorectal cancer screening tests and 3.5 times more likely to have never heard about colorectal cancer.69 Additionally, they were four times less likely to use a fecal occult blood test for colorectal cancer screening, even if it was recommended by their physicians.69 Garbers et al. examined participation by women in cervical cancer screening and found that women with limited health literacy were significantly less likely to have ever had a cervical cancer screening test.70 Lindau et al. demonstrated that health literacy was a good predictor of knowledge about cervical cancer screening.71 There is evidence that people with suboptimal health literacy obtain less information on cancer prevention or health promotion.47
By contrast, Guerra et al. found that health literacy, as measured by S-TOFHLA, is not an independent predictor of patients' participation in colorectal screening.42 In fact, their physician's recommendation was the main motivation behind patients' decision to participate.42 On the other hand, some patients with low health literacy might make a considered choice not to participate in cancer screening programs on the basis of belief systems and personal values. However, this is unlikely to constitute a large proportion of those with poor health literacy.
Limited participation from people with poor health literacy in cancer screening programs could be due to difficulty understanding risk communication and understanding the effectiveness of preventive approaches. Of course socioeconomic disadvantage, which may be associated with limited health literacy, may result in reduced uptake of cancer screening for a number of reasons, aside from poor health literacy. Patients with lower health literacy may lack the numeracy skills needed to understand and apply information about cancer recurrence risk or the chance of cancer remission.21,38 Lillie et al. revealed that patients with breast cancer with lower health literacy retained significantly less information than patients with higher health literacy when communicating about cancer risks.37
Communication of risk is also made more difficult by the varying use of simple verbal descriptors, such as low or high risk, and numerical descriptions.38 The use of verbal descriptors may be subjective and open to interpretation, and may fail to provide useful information to patients. Brewer et al. found that verbal descriptors were least easily understood among post-treatment breast cancer survivors.38 However, communicating a quantifiable risk is difficult with patients with suboptimal health literacy, although this group of patients may prefer to receive a quantified risk.38,55 Interpretation of risk of cancer recurrence is challenging for patients with poor health literacy, even using a variety of different formats.38 As a result, patients with limited health literacy may be disadvantaged when required to make decisions about participation in cancer screening programs and treatment plans based on the numerical risk information provided.
Written health education materials are extensively used in clinical practice to inform patients about their disease, treatment options and available supports. Written consent forms are also widely used but many of these materials tend to be written at a high reading level.72 An assessment of cervical cancer prevention materials revealed that information was written at a reading level at or above grade eleven.49 A study by Wilson reported that average American adults have an education level of grade 12, however their average reading level is estimated to be grade eight or nine.73 Studies have consistently found that patients with limited health literacy may struggle with simple health tasks, including interpreting instructions on a standard appointment card or a standard prescription label.12,15 Patients with suboptimal health literacy could struggle to comprehend the health education materials and treatment consent forms provided to them, thus impairing their health decision-making.
Several groups have made efforts to ensure that cancer education materials match the reading level of the general public. For example, the US National Work Group on Cancer and Literacy advocated the use of plain language for both non-written and written materials for all patients.74 However, a review by Guidry et al. demonstrated that despite the increasing use of plain language, there is still need for improvement.75 Factors that could affect patients' understanding other than language, namely the format of written materials and cultural factors, were often neglected when developing health education materials.75 This is likely to hamper the effective participation of patients with poor health literacy in informed decision-making about their treatment.
Strategies to support people with limited health literacy
Several interventions have been directed at supporting targeted groups of patients with limited health literacy. Han et al. explored the use of lay health workers to provide education, counselling and navigation assistance for breast cancer screening to first generation Korean–Americans.76 There was a significant 30% increase in mammography participation rates after the intervention.76 Various communication practices have been recommended for physicians to enhance patients' understanding from a consultation. These include the use of a plain language decision aid to present information about cancer treatment or screening programs in a simple text and numerical format.77,78 Other studies have advocated the use of a “teach back technique” by asking patients to repeat the information they received during a medical consultation in their own words, to ensure they understood the information.55,79 A study aiming to improve colorectal cancer screening rates among the medically underserved included health literacy training to improve physicians' communication, thus reducing the health literacy demand on the patients.80 A national action plan to improve health literacy was recently announced by the US Department of Health and Human Services, which outlines several goals and strategies to enhance the health literacy of the general public.81