Public Knowledge and Beliefs About Diarrheal Disease

Introduction

E ach year 60% of the U.S. population experiences an episode of acute diarrheal illness, for which 19.5% seek medical care and 7.8% receive antibiotics (Jones et al., 2007). Most of this treatment is empiric and not consistent with national guidelines (Guerrant et al., 2001; Karras et al., 2003; Carpenter et al., 2008; James et al., 2008). Treatment is also often affected by clinicians' perceptions of patient expectations about receiving antibiotics (Karras et al., 2003).

Better understanding the context in which acute diarrheal illness is managed can have important implications for addressing the substantial burden of these diseases. Understanding reasons for care-seeking behavior, for example, could help target public education toward ensuring appropriate use of medical resources. For respiratory diseases it has been demonstrated that patient expectations about receipt of antibiotics affect provider prescription behaviors, but this has not been studied extensively for diarrheal disease. Understanding public knowledge about common causes of diarrheal illness can help focus preventive education on areas of important misperception.

Despite how common they are, little is known regarding knowledge and beliefs of the public about the causes and treatment of diarrheal illnesses. Studies have been performed in developing countries (Kirkpatrick et al., 1990; Ibrahim et al., 1994; Bachrach et al., 2002; Osumanu, 2008; Usfar et al., 2010), but the few done in the United States have been limited in scope and population (Anidi et al., 2002; Li et al., 2009). To address this gap in the current literature, we performed a telephone survey to assess public knowledge, attitudes, and behaviors regarding diarrheal disease.

Methods

In September 2009, a survey of Tennessee residents was administered by computer-assisted telephone interviewing. The sample of landline phone numbers was generated proportionally to the number of phones with each prefix across the state. Up to three calls were made to each number, or more if callbacks were scheduled. The survey was administered to the first adult in the household willing to be interviewed. After obtaining consent, a trained interviewer administered the questionnaire.

Respondents ≥18 years of age were eligible for the study. All survey respondents were asked questions about themselves. Respondents living in a household with a child ≤12 years of age were also asked about care-seeking behavior related to children.

The survey included questions on demographics and knowledge and beliefs about management and causes of diarrheal disease. Diarrhea was self-defined by the respondent. For purposes of analysis, because of small sample sizes in some groups, educational levels were dichotomized to high school or less (highest year of school completed was grade 12 or below), and education beyond high school (at least 1 year of college or technical school).

Data were analyzed using SAS 9.1. Statistical comparisons were made using chi-square tests for odds ratios (OR) and 95% confidence intervals (CI). The study was approved by the Institutional Review Board of the Tennessee Department of Health.

Results

Of the 5867 households with a landline contacted, 2117 (36.1%) adults over 18 years of age completed the survey, 796 (13.6%) hung up, and 2618 (44.6%) refused to participate. The mean age of respondents was 58 years (standard deviation = 15.5); 71.0% were women, 58.2% had some education beyond high school, 17.4% had a child ≤12 years old at home, and 87.9% were white. Among respondents, 505 (23.9%) reported having an episode of diarrhea within the previous 3 months, for which 17.0% (95% confidence interval (CI): 12.1–18.4) sought medical care.

Bloody stool, dehydration, and persistent diarrhea were the most common reasons for which the adults would seek medical care (Table 1). Respondents were more likely to seek medical care for children than for adults, when children presented with fever (OR = 7.14, 95% CI: 5.26–10.0), vomiting (OR = 2.5, 95% CI: 1.6–3.9), dehydration (OR = 2.33, 95% CI: 1.72–3.23), and diarrhea lasting more than 1 day (OR = 3.13, 95% CI: 2.27–4.35).

Male and female respondents did not differ significantly in their thresholds for seeking medical care. White respondents were more likely than black respondents to seek care for dehydration (10.2% vs. 6.1%, OR = 1.76, 95% CI: 1.02–3.04). Respondents with some education beyond high school were more likely than those with an education of high school or less to seek care for bloody diarrhea (OR = 1.51, 95% CI: 1.18–1.95) or dehydration (OR = 3.31, 95% CI: 2.31–4.76).

The large majority of respondents did not consider stool cultures or antibiotics necessary for most episodes of diarrhea (Table 1). Many respondents said that other antidiarrheal agents (46.7%), drinking lots of liquids (25.5%), and special diets (e.g., yogurt, bananas, rice; 20.8%) were important.

Compared with men, women were more likely to believe that stool cultures should be performed for bloody stools (16.3% vs. 7.9%, OR = 2.28, 95% CI: 1.65–3.16). Blacks were significantly more likely than whites to believe that antibiotics are always necessary for diarrhea (16.1% vs. 5.4%, OR = 3.39, 95% CI: 2.28–5.03) as were men compared with women (9.5% vs. 5.8%, OR = 1.69, 95% CI: 1.20–2.40). Respondents with some education beyond high school were more likely than those with an education of high school or less to think stool cultures are necessary for bloody diarrhea (OR = 1.41, 95% CI: 1.09–1.83). Beliefs about antibiotic use did not differ by education level.

When asked about what they thought was the most common way to get diarrhea, 61.4% said food, 8.3% said contaminated surfaces, and 5.0% said from other ill people. When asked what they thought was the most common germ causing diarrhea, 60.8% of respondents said they did not know, 4.9% said bacteria, and 5.0% said viruses. The most common specific organisms named were Escherichia coli (11.5%), Salmonella (7.8%), and “flu” (2.2%).

Of respondents, 74% (95% CL: 72.4–76.2) thought that it is normal for uncooked ground beef/hamburger sold in grocery stores to have bacteria on it; 77.8% (95% CL: 76.0–79.6) believed the same thing for chicken. Almost half (45.2%, 95% CL: 43.0–47.3) believed it is legal for uncooked meat sold in grocery stores to have bacteria on it. Men were significantly more likely than women to believe that the presence of bacteria on beef (83.4% vs. 74.5%, p < 0.001) and chicken (85.2% vs. 78.9%, p < 0.001) is normal and legal (61.0% vs. 46.7%, p < 0.001). Whites were more likely than blacks to believe that the presence of bacteria on beef (78.2% vs. 69.1%, p = 0.002) and chicken (82% vs. 70.7%, p < 0.001) is normal and legal (53.4% vs. 37.0%, p < 0.001, respectively).

Discussion

The majority of respondents in this survey believed that most diarrheal illness is foodborne. A surprising proportion thought that it is normal for uncooked ground beef and chicken to contain bacteria, and that it is legal to be sold that way. Reports of consumer food safety behaviors suggest that many people do not appreciate the risks of contaminated foods or appropriately do things to mitigate this risk (Shiferaw et al., 2000). Further studies should consider focusing on whether such knowledge influences safe food-handling practices in the home, and whether consumers believe that regulatory agencies such as the U.S. Department of Agriculture should implement stronger controls over bacterial contamination of foods such as raw meats.

Few studies have investigated public knowledge and beliefs about diarrheal disease in the United States (Anidi et al., 2002; Li et al., 2009). In this study of self-reported diarrheal illness, few patients indicated that they routinely expect antibiotic therapy. Despite this, although diarrhea is usually self-limited, it is frequently treated with antibiotics, often empirically (Karras et al., 2003; Carpenter et al., 2008). Physicians have reported that they were more likely to prescribe antibiotics for diarrhea when they believed that patients expected them, but they correctly identified such expectations only a third of the time (Karras et al., 2003). This study suggests that education could likely prevent inappropriate antibiotic use in most instances. It also confirms that a substantial proportion of patients are aware of reasonable supportive-care measures for diarrhea, although more education should be focused on the importance of replenishing fluids.

Stool cultures are performed in only 3%–5% of patients presenting for medical care for acute diarrhea (Carpenter et al., 2008). Because most diarrheal illness is self-limited and does not require antimicrobial therapy this is not unreasonable, and patients appear to accept this. Others have suggested that medical providers are not sufficiently aware of national guidelines regarding stool testing for diarrheal disease (MacDonald et al., 2007; James et al., 2008). Additional studies to better understand the correlates of patient and medical provider knowledge about appropriate testing is warranted.

Although this study involved a large population-based survey, there are some limitations that might affect interpretation of its results. Diarrhea was self-defined, although healthcare-seeking behaviors were determined by patients based on their own definitions. Respondents were asked about beliefs rather than actual practices, which could differ. The response rate in this survey was low, but consistent with other recent large telephone surveys (Jones et al., 2007). This is a widely recognized phenomenon associated with changing telephone habits in the United States, which may bias results because of differential rates of participation among different segments of the population (Jones et al., 2007). Only landline telephones were called during this survey, which has similar drawbacks due to the increasing number of households with cell phones only (Blumberg et al., 2006). Although data were stratified by basic demographics to help understand the results in specific subpopulations, the results may not be generalizable beyond Tennessee.

This study is an important initial look at knowledge and beliefs regarding the causes, management, and prevention of diarrheal illness in the general population and should stimulate additional investigations. These results suggest that patient expectations should not drive inappropriate overuse of antibiotics, and that most consumers understand that it is not uncommon for food to come into the home contaminated. Educational activities should focus on supportive care for diarrheal illness and on the importance of good food-handling practices in preventing foodborne disease.