Abstract: People's questions about health not only reveals their knowledge of medical concerns, but may also reflect on the adequacy of the health information system to provide answers. In this study we collected and analyzed 255 health questions from adults in a discount department store in a mid-sized town in middle America. Results suggest that large portions of the public do not know how to promote and maintain health and are looking for answers that the current health care system does not easily provide. This can be damaging to both health and the economy. This paper provides recommendations for how to improve the health information system based on an analysis of these questions.
The adequacy of a health information system may be judged by the type of questions people ask. For example, if people do not know the answer to simple health questions or if they carry around unanswered questions, this may be evidence that the health information system is defective. In this study we sought to identify and analyze the unanswered health questions of adults and through this came to understand more about their health information needs and how to address them.
The health literature is thick with references that identify the public's level of knowledge on specific health issues. For example, a Medline search on the medical subject heading "knowledge attitudes and practice" reveals over 8,000 entries related to health since 1981. For the most part, this literature is clinical and focuses on questions such as the patient's level of knowledge of particular risks, conditions, and treatments. Other studies assess a community's need for particular types of information, for example, about the causes of hypertension or asthma. Few studies, however, focus on the broader questions: What health questions do the public have? Where do they go for information? Do these questions and practices vary by demographics? Only one study in the recent health literature has sought to identify the type of questions people have - in this case, to ascertain the type of information needed for a local healthy living center (Winn & Bradford, 1991). Studies that seek to determine where people get health information are more common and have been used to assess the adequacy of these sources for meeting the health education needs of particular sub-populations (Arnold & Hom, 1996; Gollop, 1997; Richmond, McCracken, & Broad, 1996).
The current study identifies the health questions of adults and the sources of health information they use in a mid-sized town in south-central Indiana. This information was used to assess the adequacy of the health information system and to make recommendations to improve it.
We set up a table in a discount department store on the fringe of a mid-sized town in Indiana (population 60,000) and offered to research answers to the health questions of adults. A store on the outskirts of the town was chosen to give us access to more of the rural population, which makes up a large portion of the population of southern Indiana.
While those who participated in the study were not chosen randomly, they may be more like the population of users of health information services. To use most health information services, a person must initiate an action to get an answer to their questions. Participants in this study had to approach a table and ask for help with a question. Participants in random surveys may not have taken an action on their own.
Participation was restricted to adults and each adult was allowed to ask one or two questions on any health topic. Health education students in training recorded the questions, researched the answers, wrote a response, and mailed the responses back to participants. In addition to the health questions, socio-demographic information was collected as well as responses to an open-ended question about where the participants go for health information.
We collected 255 questions from 158 people (100 female and 58 male) within the space of 26 hours spread over a week. The sample size was limited by the students' ability to write responses to each question - a process that occupied fifteen students a fair portion of the semester. The age range of participants was from 18 to 85 with a mean age of 38. Roughly half (54.9%) were under forty, about a third (34.6%) were between 40 and 59 and the rest were over 60. Most respondents were White (78.5%), with African-Americans making up 13.9% of the sample, Asians 5.7% and Hispanics 1.9%. Over half the sample (58.7%) had a household income below 25,000; only 14.7% had an income greater than $50,000. The sample approximated the demographics of the county which is 94% White with a median household income of $24,781. The adult population of the county is 53% female with 64.6% of adults being under 40 and. 21.1% between 40 and 60.
Sixty-one percent of the sample asked more than one question. The questions were tremendously diverse, sorting into forty-two different categories - determined by major subject. For example, all questions on health care system were grouped together, and all questions related to diet were grouped together. Categorization was not always easy. In cases where the question could be fit into two categories, the question was placed into what was considered to be the narrower category. For example, a question about diet change to reduce the risk of heart disease was placed in the diet, not the heart disease category, as the diet questions were more similar in nature and less varied in scope than the heart disease questions. Many categories contained only a single question, for example, individuals wanted to know what is: cri du chat disease, celiac disease, the success rate for Islets of Langorhans transplantation, the latest research on Lopressor and Dilantin, and the latest research on Hereditary Hemorrhagic Telangiectasia. Table 1 lists the top ten most frequently asked question categories and the number of questions in each one. The categories with less than eight questions were collapsed into an "other" category.
Table 1 Number and Proportion of Questions in Health Category
Almost three-quarters (70.6%) of the health care system questions were related to insurance, its coverage, and costs. For example, people wanted to know if they were eligible for Medicaid, how to find low-cost insurance, and why insurance coverage keeps getting less. The remainder of the health care system questions reflected diverse and general concerns such as: What is the organizational structure of public health in the United States? What are the priorities of the local health department? What are the types of public health services available locally? Questions about diet were split roughly in thirds among those related to weight loss, vitamins, and questions about what constitutes a healthy diet. Women's health questions dealt primarily with hormone balance (36.8%) and osteoporosis (31.8%), with miscellaneous questions about topics like birth control and endometriosis. Mental health questions were mostly concerned with stress (52.6%) and depression (31.6%). Most (63.2%) of the environmental questions were concerned with water quality in general or with polychlorinated biphenyls in the area's water supply - a long-standing local concern. The other environmental questions were either general (e.g., What are the environmental issues in our area?) or covered a range of diverse topics (e.g., Are pesticides on produce regulated?). The exercise and fitness items were very diverse and included requests for tips about how to incorporate exercise into a busy schedule, the activities that benefit health or weight loss, and techniques to avoid injury. Infectious disease questions ranged from information on STDs to athlete's foot. Almost all (83%) of the cancer questions concerned risk factors for different cancers. A few participants wanted to know about survival rates and cancer screening services. Heart disease questions mostly dealt with risk factors - chiefly cholesterol. A few had general questions, such as the signs of heart disease, or how to maintain heart health, and some asked specific questions about services like angioplasty, and protective factors like red wine. Neurological questions ranged from those about the symptoms of Alzheimer's (the most common question) to Parkinson's disease.
Demographic analysis was conducted to determine if there were significant differences in the type of questions asked by males and females, those with lower and higher income (under or over $25,000), and younger and older persons (under 40 or 40 and over). Respondents were no more likely to ask questions in one category than any other with the following exceptions: Younger people were more likely to ask about food/ vitamins and diet than older persons (c2 = 9.07, p =.003), only women asked questions about women's health, older women were more likely than younger women to ask questions about women's health (c2 = 4.16, p = .041), and lower-income citizens were more likely than higher income citizens to ask environmental health questions (c2 = 7.27, p =.007). This analysis was conducted on the six most frequently occurring question types - those containing more than five percent of the questions. There were too few questions for statistical analysis in the remaining questions.
Sources of Health Information
Table 2 presents the major sources of health information that participants identified. As the question was open-ended the participant could identify multiple sources. For example, if a person identified their physician and television as their major sources of health information, both responses were counted in the analysis.
Table 2 Sources of Health Information
Physicians were, by far, the most common source of health information in this population. Television, newspapers, popular magazines, and family and friends were the next most popular sources. Females were significantly less likely to use television than other sources of health information (c2=3.74, p =.05). People of higher income were significantly more likely than people of lower income (c2 = 20.08, p =.000) to use a physician as a source of health information. Older people were also significantly more likely than younger people to use a doctor as opposed to other sources of health information (c2=4.08, p =.04). No other variation in health source by demographic was significant.
This study from the heartland of America suggests that many Americans carry around unanswered health questions. We had no trouble finding hundreds of health questions in 26 hours at a single location. Further, these questions range across a great variety of topics. Many of these questions, such as those about how to diet and exercise and the risk factors for common diseases should be common knowledge.
One of the reasons this population is having difficulty in getting their health questions answered is that they are relying on sources of health information that have a very limited ability to respond to their questions. Most people in this study identified their physician as a major source of health information. Visits to the doctor, however, are infrequent for most people and market forces make it hard for physicians to spend time with patients answering their questions (Davidoff, 1997). Even if, however, physicians could respond to the health information needs of their clients, this study indicates that many of the poor and the young would still not get their questions answered as they were not as likely to use this source of information.
Besides relying on physicians, many people in this population use television for health information. According to the National Health Council (1997), television is the most common source of public health information. But television programs are rarely interactive and do not respond readily to individual health concerns. Further, they can only be effective if the audience is ready and able to capture the health information that is given. Even if television was a good medium of instruction about health, this study indicates that many women would still not get their health questions answered - as women, apparently, do not use the television as a source of health information as much as men.
Finally, we found that a small proportion of this population relies on newspapers and popular magazines to address their health questions. These too are poor sources of health information because the quality of the information is variable and the probability that a person with a particular health question will find the answer when it is needed is low. This analysis of the sources of health information that people use suggests that many of the public's health questions are apt to remain unanswered for a very long time.
While it will not be easy to address the public's health information needs, there are important reasons to do so. Chief among them is the certainty that many people who have unanswered health questions could be delaying action or taking the wrong action. Our study contains many examples of questions that if not answered will be costly. For example, it is clear that many of our participants were either not insured or were delaying or reducing medical care because of lack of full coverage. One participant asked, "How can I afford insurance on my income of less than 20,000?" Another low-income participant asked, "I have seven costly medications a day I must take, how can I afford this?" Lack of adequate insurance continues to be the most important correlate with unmet health need (Berk, Schur, & Cantor, 1995). Members without insurance delay seeking health care and members with limited coverage may be limited in their response to illness or injury.
Beyond questions about the health care system, it is clear that the people we surveyed have many other questions that if not answered will be costly. The population in the state in which this study occurred is one of the most obese in the nation. Not only is obesity costly to individuals and the nation, it is difficult for the people to solve this problem without adequate information. Yet, through their questions, many people in our sample made it clear that they do not understand the basics of weight-loss through diet and exercise. People asked, for example, "How do you lose weight?" We also observed quite a few participants looking for trendy weight-loss solutions. For example, one participant asked, "Do chromium pills work to lose weight?" Another wanted to know the latest breakthrough in medicines to lose weight. In almost every category, we found questions that, unanswered, will cause health-care costs to rise. Many people did not know how to manage stress, lower cholesterol, or decrease the risk from heart disease and cancer, and many women were confused about the benefits of hormone therapy.
The following recommendations for an effective health information system follow from this study:
1. To meet the public's need, a health information system should be able to respond to a tremendous range of questions in a timely manner. A system that only responds to the most common questions, while helpful, will still leave many people with unanswered questions. Questions about diet, for example, were relatively common, but only made up 12.6% of all the questions. Responding to common questions still leaves many questions unanswered - some of which may be urgent. For example, we received only one question about pain relief. When we were late in responding to this question, we got the following response on our evaluation card, "My mother died in pain since we last talked. I needed immediate action to stop her pain. Why must a dying person suffer when they don't have to with modern medicine? I needed her pain to stop, but as we all know, doctors are afraid, the officials don't care, and people suffer until they die." A good health information system should be able to address a wide array of questions in a timely manner.
2. A health information system should give people with similar concerns a chance to get connected. Health information should be more than personally empowering; it should be socially empowering as well. In some cases this connection should be to the political process. Many of our participants asked questions that require a political response. For example, one participant asked, "Is it fair that old people should sacrifice all they possess to qualify to be in public nursing homes?" Another asked, "Why do we allow HMO's to set quotas?" To be most useful, a health information system should connect people with concerns like these together so that they can work through the political process to achieve change.
3. Because of the time pressures on doctors noted above, it is unrealistic to believe that physicians can provide more health information to clients - at least during the medical interview. Physicians should take note, however, that patients are not generally pleased with the current level of information provided by their physicians and that they want more (Rauber, 1999). Physicians should seek to provide health information in new ways such as direct mail and health education classes. In one promising study, health care providers from four sites gave "information prescriptions" to patients in need to go to local health information centers for more information (Bang, Farrar, Sellors, & Buchanan, 1998).
4. The Internet will not provide the information needs of most of the public for some time. While a National Health Council survey in 1997 found that only 2% used the Internet as a health and medical news source (National Health Council, 1997), this number is increasing. According to the Cyber Dialogue web site, a Manhattan firm that tracks Internet usage, it is estimated that 33.5 million people will seek medical information on the Web in 2000. Clearly many people have an interest in using the Internet for health information, in a nationwide survey of 1,000 adults, 79% expressed interest in on-line health-related information (Editor & Publisher, 1994). In addition, many people are already using the Internet for gathering health information. Medline, the health journal indexing system of the National Library of Medicine, gets about 10 million hits a month (Modlin, 1999). But while the general public does express interest and many people are using resources like Medline, most citizens still do not use the Internet to gather health information. Only 3% of the participants in our study used the Internet as a source of health information.
While the Internet will undoubtedly play an increasing role in the health information system of the future, it is not free from difficulties as a source of health information. In a recent study, Hersh and colleagues (Hersh, Gorman, & Sacherek, 1998) found that 89% of pages retrieved from an Internet search were not applicable to the clinical question that prompted the search. Further, fewer than 1% consisted of evidence-based research, 69% did not indicate an author, and 80% did not give the authors' credentials. Thus, it is not easy to get trust worthy health answers from the Internet and considerable skill is needed to interpret results. Even when Internet searches are restricted to traditional medical sources, the quality of information is variable. McClung, Murray, and Heitlinger (1998) found that only 20% of articles obtained from an Internet-generated search of published articles from traditional medical sources gave correct recommendations for the management of a common condition - diarrhea in children. Eng, Maxfield, Patrick, Deering, Ratzan, and Gustafson (1998) note that many factors limit the capacity of people to successfully use the Internet to answer health questions, including cost, geographic barriers, literacy, culture, disability, and other factors. The Internet should continue to be developed as a source of health information, but development should not be limited to the Internet. In addition, many people will need hands-on assistance to use and interpret findings from the Internet.
5. The library should continue to be developed as a health information resource. A major effort is underway to make libraries a site for health information (Modlin, 1999; Lindberg & Humphreys 1998). One medical library reports receiving about one hundred public requests per month for help answering health questions (Earl, 1998). This effort should continue, but it should be noted that the library is not now a common source of health information for much of the public. In our sample only 8% identified the library as a resource for health information - even though there are several libraries within 10 minutes of our site with very good health resources.
6. Health agencies could sponsor Health Information and Support Telephone Services (HISTS) such as that offered by the Health Extension program in Canada. The telephone is much easier to use than the Internet and is more accessible than the library. The Canadian system provides 420 pre-recorded health information messages available free of charge 24 hours a day (Bang, Farrar, Sellors, & Buchanan, 1998).
7. Health education and information sites should be established in places that the public routinely visits. While many health information needs may be addressed by canned messages over the telephone, others will require personal interaction. We found that many of the questions asked needed interpretation or rephrasing. Indeed, in some cases, the correct question only became apparent through dialogue. To be most accessible, these sites should not be in clinical settings. They should be where the public shops: in malls and in department and grocery stores. We had no difficulty collecting questions from the public in our discount department store site. These health education and information sites could be sponsored by physician groups or health agencies and staffed by health educators. They should be consumer oriented and provide a mechanism to link consumers in mutual aid. The Consumer Health Information/Resource Centers in Canada provide a good model to build upon (Bang et al., 1998).
This study suggests that there may be large portions of the public who do not know how to promote and maintain their health and who are looking for answers that the current health care system does not easily provide. It is damaging both to the health of individuals and to the economy for these questions to remain unanswered. In the current management environment it is unreasonable to expect physicians to do much more in the way of public health education. Further, while the Internet and access to it through both home and library looks promising as a source of health information, substantial barriers exist in making this a reliable, easy to use and understand source of information. More effort is needed to address the public's health questions through the telephone and through increasing personal access to a health educator. Health education and information sites could be established within existing shopping environments and these could also serve to connect people with similar interests and needs. Priority should be placed on establishing these sites in low-income areas.
Arnold, D. E., & Horn, R. P. (1996). The quantity and quality of medical information available to the public. South Dakota Journal of Medicine, 49(2), 69-74.
Bang, D. L., Farrar, S., Sellors, J. W., & Buchanan, D.H. (1998). Consumer health information services: Preliminary findings about who is using them. Journal of Medical Systems, 22(2), 103-115.
Berk, M. L., Schur, C. L., & Cantor, J. C. (1995). Ability to obtain health care: Recent estimates from the Robert Wood Johnson Foundation National Access to Care Survey. Health Affairs, 14(3), 139-146.
Davidoff, F. (1997). Time. Annals of Internal Medicine, 127(6), 483-485.
Earl, M. (1998). Caring for consumers: Empowering the individual. American Libraries, 29(10), 44-46.
Editor and Publisher (1994). Health data, news are top interactive interests. Editor and Publisher, 127(43), 39-41.
Eng, T. R., Maxfield, A., Patrick, K., Deering, M.J., Ratzan, S., & Gustafson, D. (1998). Access to health information and support: A public highway or a private road. Journal of the American Medical Association, 280(15), 1371-1375.
Gollop, C. J. (1997). Health information seeking behavior and older African American women. Bulletin Medical Library Association, 85(2), 141-146.
Hersh, W.R., Gorman, P. N., & Sacherek, L. S. (1998). Applicability and quality of information for answering clinical questions. Journal of the American Medical Association, 280(15), 1307-1308.
Lindberg, D., & Humphreys, B. L. (1998). Medicine and health on the Internet: The good, the bad, and the ugly. Journal of the American Medical Association, 280(15), 13031304.
McClung, H. J., Murray, M. D., & Heitlinger, L. A. (1998). The Internet as a source for current patient information. Pediatrics, 101(6), 1065.
Modlin, M. (1999). Medical questions? Medline has answers. American Libraries, 29(10), 40-43.
National Health Council (1997). Americans Talk About Science and Medical News.
Rauber, C. (1999). Information, please. Modern Healthcare, 29(13), 56-58.
Richmond, D. E., McCracken, H. E., & Broad, J. (1996). Older adults and healthy lifestyle issues: Results of a community study. New Zealand Medical Journal, 109(1019), 122125.
Winn, S., & Bradford, M. (1991). What people want to know about health. Health Visitor, 64(10), 331-333.
Thomas Tai-Seale, Dr.P.H., M.P.H., M.M.S., M.A., Assistant Professor, Department of Applied Health Science, Indiana University, Poplars 615, Indiana University, Bloomington, IN 47405, (PH) 812-855-3817, (FAX) 812-855-7092, email@example.com.
COPYRIGHT 2000 University of Alabama, Department of Health Sciences
COPYRIGHT 2001 Gale Group