Hidden variables: Why patients may misunderstand written screening tools

Health literacy is defined as a patient's ability to obtain, process, and understand basic health information in spoken, written, and numerical forms and use this information to make appropriate health decisions.^1,2 In the United States, nearly 90 million adults have inadequate written health literacy.^1,3 While the average US adult reads below the ninth grade level, 21% of adults read below the sixth grade level.^1,3 Of Medicare enrollees with low health literacy, 42% reported that they could not understand medication directions, 22% did not understand their appointment slips, and 62% could not understand an informed consent form.⁴ Additionally, both county hospital patients and Medicare enrollees with low health literacy are more likely to be hospitalized compared with patients who have adequate health literacy.^5,6


Little research has focused on numerical (quantitative) health literacy, the ability to make sense of numbers and other mathematical concepts. However, recent studies suggest that poor numerical literacy is significantly more prevalent than poor written literacy because the former affects patients with written literacy and high education level.⁷ In a large study, Rothman and colleagues reported that 77% of participants exhibited high school level written literacy. Only 37% of these respondents exhibited high school level numerical literacy (numeracy).⁸ Another study reported that 25% of medical students at a prominent university had imperfect numeracy.^9,10 It appears that unlike written health literacy, numerical literacy operates independently of education level.⁷


NEGATIVE EFFECTS OF LOW HEALTH LITERACY


Access and adherence Both low written and low numerical literacy significantly affect patient outcomes.¹¹ First, low health literacy adversely impacts access to medical care. Patients with low health literacy are 30% to 50% less likely to have received a Pap smear, mammogram, influenza vaccine, or pneumococcal vaccine.^12,13 Low health literacy also impacts compliance with treatment. Patients with low health literacy who have diabetes maintain higher blood glucose and hemoglobin A1C levels.¹⁴ Consequently, low written health literacy has been firmly linked with increased health care costs.^1,11 In addition, in a large study of 653 patients, Gazmararian and colleagues demonstrated a significant inverse relationship between low health literacy and mortality after controlling for other variables.¹⁵ While the literature remains incomplete, low numerical literacy may contribute most to health care costs, morbidity, and mortality given its higher prevalence.


Interactions with the medical system Along with nega­tively affecting patients' medical care and compliance, low written and numerical literacy also affect how patients in­teract with medical systems, specifically through written 
screening tools (WSTs). Most clinicians, including but 
not limited to urologists,¹⁶ endocrinologists,⁴ and derma­tologists,¹⁷ increasingly rely on WSTs to collect valuable clinical information from patients. Greater time constraints and compensation managed by third parties have increased the popularity of these tools to rapidly collect critical information, often independent of the clinician.


Many clinicians presume that patients understand how to use WSTs and can accurately self-administer them. However, while the average American reads below the ninth grade level,^11,18 these tools are typically written at the tenth grade level.^11,18,19 This indicates that physicians have increasingly relied on patients to diagnose their conditions using tools that they sometimes cannot understand. Given the significant role WSTs play in clinical diagnosis and management, low written literacy and numeracy could have safety consequences for patients who misunderstand how to use these tools.


DRIVERS AND IMPLICATIONS OF LOW 
HEALTH LITERACY


Evaluating the AUA-SS A series of investigations has been conducted to assess the impact of low written literacy and numerical literacy on patient understanding using a model written screening tool.^7,20-22 Researchers used the American Urological Association Symptom Score (AUA-SS), the world's predominant WST for benign prostatic hyper­plasia (BPH), a common clinical problem in older men.¹⁶ 
The AUA-SS can be viewed online at www.adultpediatricuro.com/apuauass.pdf. Validated in several studies,^16,23,24 the AUA-SS is a seven-item tool (plus one question on quality of life) used to screen for symptoms, chart clinical progress, and predict prognosis for BPH.^25,26 The AUA and European Association of Urology (EAU) consider the AUA-SS central to their BPH guidelines.^27,28 For example, according to the AUA's BPH guidelines, patient completion of the AUA-SS is required, but diag­nostic testing is optional.²⁷


Of the 1,500 patients interviewed in these studies, 70% misunderstood at least half of the AUA-SS's seven questions.^7,20-22 In addition, 60% of patients claimed to understand the AUA-SS questions fully when they did not.²¹ Most patients who did not understand parts of the AUA-SS responded anyway. Since the AUA-SS is a self-administered testing tool, physicians are often unable to identify those who could not comprehend some or all of the form. Therefore, unbeknownst to their physician and themselves, these patients may be placing themselves at risk for inappropriate management of their conditions. Beyond the AUA-SS, it is plausible that patients may also misunderstand other written screening tools.


Other causes of misunderstanding Several studies aimed to identify other drivers of misunderstanding beyond low written and numerical literacy to discover which characteristics put patients at greater risk for not understanding WSTs.^20-22 To do so, researchers examined a number of variables including age, race, ethnicity, homelessness, employment, income, and education level. Of these variables, understanding was independently impacted by only education level. A number of studies have confirmed the significant impact education has on patient understanding of the AUA-SS. For example, one study of patients at a county hospital assessed the impact of education on likelihood of understanding the majority of the AUA-SS's seven questions. Compared to patients with some college education, patients with a high school education were seven times more likely to misunderstand the majority of the AUA-SS (odds ratio [OR], 7.00; 95% confidence interval [CI], 1.84-26.67). Similarly, patients with 9 to 11 years of education and less than 9 years of education were 23 times (OR, 23.41; 95% CI, 6.03-90.91) and 102 times (OR, 102.16; 95% CI, 23.93-436.10) more likely to misunderstand the AUA-SS.