Challenges and Failures of Health Fairs and Community Screenings
Health fairs are one of the most recognizable forms of community-based health promotion conducted in the United States. Health fairs are voluntary programs, which typically last a few days, and offer health education and medical screenings at little or no cost. Most fairs measure height, weight, blood pressure, vision, and anemia, while other popular tests include blood chemistry, oral screenings, podiatry exams, hearing tests, and glaucoma screenings. Though these fairs in theory may seem like a good idea, the medical literature has often viewed them with considerable skepticism. “Health fairs are neither regulated nor routinely certified in the United States, and complete data on their numbers and content are not available.” (1) In addition, the laboratory screening tests offered at many health fairs may cause more harm than good. These tests may unnecessarily alarm participants with erroneous abnormal results, or provide a false sense of reassurance if results are shown to be normal. Despite these concerns, health fairs continue to attract large numbers of people.
Inconclusive Evidence Exists Regarding the Benefit of Health Fairs and Community Screenings
Unfortunately, quality studies cannot be found which assess the effectiveness of health fairs. Many studies assess the impact of health fairs based on outputs (how many people had their blood pressure checked, how many received literature about cardiovascular disease, and the %age who stated that they intended to make lifestyle changes based on the information gained at the health fair). However, these studies fail to report the actual outcomes and do not take into account how many people received an exam by a physician and how many actually had improved health outcomes due to the health fair. “The few existing follow-up surveys have tended to report only the satisfaction levels and self-reported behavior of the participants, not health-related outcomes. The dearth of experimentally controlled follow-up data is the major factor inhibiting direct assessment of the benefits, costs, and risks of health fair screening.” (2)
It is also uncertain if the benefits of health fairs outweigh their costs. In a study conducted on health fairs in the Greater Cincinnati area, 706 out of the 940 health fairs offered blood chemistry assessments. Blood chemistry assessments can detect diabetes and hypercholesterolemia, but “the induced costs and hazards of follow-up and reevaluation of insignificant biochemical abnormalities may be substantial.” (3) A study conducted at the Kaiser Permanente Medical Group found that 76.4% of patients with an abnormal blood chemistry screening had at least one follow-up visit, and 18.6% had additional tests performed. Thus, overall the initial test resulted in a follow-up encounter for 41.9% of the screened population and additional tests for 10.2%. New diagnoses (not necessarily of benefit to the diagnosed patient) were made in only .0075% of the people screened and in less than 1.4% of those with at least one abnormal test result. (4) A study on the value of the comprehensive metabolic profile (a panel of 14 blood tests) found that it was inefficient at detecting at risk people and generating new diagnoses. The study, which assessed a rural county health fair in Kansas, used data from a local health center to determine whether comprehensive metabolic profile findings at the screening were associated with new diagnoses in the 4-months following the health fair. The study “reaffirms the limited usefulness of a comprehensive metabolic profile in asymptomatic health fair participants.” (5) After all evaluations were completed, only 1% of the tested subjects received a new diagnosis as a result of the test. Thus, the study corroborated the sentiment that “there is no consensus recommendation from any major professional or government organizations regarding the use of a comprehensive metabolic profile (CMP) or its individual tests (other than glucose levels) for screening purposes in the general population.”(6) Nevertheless, these tests are still routinely offered at health fairs.
It is evident that not all screenings are beneficial, and many are inefficient at detecting at risk individuals. “The mere existence of undetected disease in a population does not, by itself, make screening desirable. For screening to be effective and efficient, the following conditions should be met, at least in part: (1) the disease involved must be sufficiently prevalent and unlikely to be detected without screening. (2) The screening test must have acceptably low cost and error rates. (3) Presymptomatic treatment must be advantageous, available, and acceptable to the screened populations.”(7) Though it is recognized that not all screening services offered to low-risk individuals are beneficial, health fairs often include them anyway. This is because the inclusion of a wide number of services often helps the health fair to receive more support from the community. It is essential, however, for the services included at health fairs to utilize interventions that are proven to be effective and beneficial.(8)
Laboratory Tests and Arbitrary Thresholds
Laboratory tests are offered at many health fairs. However, “widespread use of these and other tests raises complex issues of cost, risk, and benefit. Rates of false alarm of health people and false reassurance of those at risk may be high for some tests, and the benefits of collecting new disease are easily overestimated.” (9) Though laboratory tests may potentially be harmful, a study on health fairs held in the Greater Cincinnati area found that obtaining laboratory testing services was the sole reason that 47% of participants attended. Obtaining lab results was thought to be of much greater importance than the health educational materials also offered at the health fair, perhaps because the tests were identified as providing a sense of control over personal health care. Receiving their own normal test results was perceived as assuring a healthy future for 86% of participants. This study suggests that instead of relying on potentially harmful tests to draw participants to health fairs, health care professionals may be able to enhance participation and satisfaction by “addressing the principle motivational factor identified by individuals participating in health fairs: positive health feedback.” (10)
An additional problem with conducting laboratory tests at health fairs are the arbitrary threshold values used to determine if someone is “normal” or “diseased”. Though the National Health Screening Council for Volunteer Organizations (NHSCVO) training manual includes guidelines for follow-up of abnormal test results, these threshold values are often greater than those that a physician would use to determine if treatment is needed. For example, the manual recommends follow-up for cholesterol levels of 360mg/dL or greater, whereas many physicians would recommend treatment for levels far below this cutoff. (In fact, the American Heart Association categorizes those people with 240mg/dL and above as having high blood cholesterol).(11) Thus, in addition to producing false-positive results, “normal” laboratory tests may also lull people into a false sense of security.
High Associated Costs
Though the direct financial costs of health fairs are typically low since health fairs tend to use volunteer labor, high referral rates can induce substantial follow up costs. Thus, “in choosing to screen for illness, health fair sponsors encounter important issues of risk, cost, and benefit.” (12) In addition, misleading results can provoke panic and have psychological consequences. False alarms and “other misleading results can induce both financial and psychological costs of far greater magnitude than the benefit offered to the small %age who have true-positive results. This problem is most severe in the case of blood chemistry screening, where the simple laws of multiplicity virtually assure a high false alarm rate.”(13) False reassurance can also be dangerous because people will not change their lifestyles or otherwise address the underlying condition. For example, “glaucoma victims with normal intraocular pressures, mild hypertensives and hypercholesterolemics just below the cutoff levels for follow-up, and those with unsound health habits who are reassured by a normal battery of test results could all be harmed by a false sense of security from health fair screening.”(14) Since health fairs can provoke false alarms and a false sense of reassurance, it is important that individuals leave health fairs informed of the real risks of a given disease. It is also important that health fair evaluators do not evaluate the benefit of a screening with the rate of detection of abnormality. “Neither physicians nor health fairs should congratulate themselves on the detection of conditions that do not alter well-being or whose course and outcome cannot be influenced by presymptomatic treatment.” (15)
Unqualified Organizers and Screeners
Student organizations have recently been involved in organizing health fairs. This is potentially harmful because these groups may not act upon reliable information or be properly trained. A study on such organizations found that when assessing the needs of individual communities, most groups searched internet sites for information instead of using statistics supplied by the local county health department. In addition, most groups attempted to address the health needs of the target community by speaking with the previous year’s organization officers. This is problematic because health needs can change from year to year, and the groups did not use reliable sources to assess them. Moreover, “the majority of the organizations (n=10/11) indicated that their active members needed to learn more about the health status and/or health issues of their target population.”(16)
Another problem with health fairs is the variability and unreliability of the training that the screeners and test administrators receive. With regards to screening competency, only a few of the student organizations “felt their trainer was extremely knowledgeable about the health condition that their organizations were being trained to detect. Of nine student organization officers, only two believed that their sessions were effective at training volunteers to discuss health-related behavioral change issues with health fair attendees. Only half of the groups that engaged in training (n=5/10) had methods in place to maintain the skills of their volunteers, and fewer (n=3/10) performed ongoing competency evaluations of their student health fair workers. Officers from several groups mentioned that while training sessions were helpful, much of the experience acquired by student volunteers came from on-site experiences. In addition, nine of ten groups indicated a need for the development of a health fair assessment tool that would enable them to better evaluate the efficacy of their fairs (through collecting and analyzing data on all aspects of their events). A majority of the student organization respondents (n=10/10) stated that they would like to learn about methods for evaluating their health fairs or health outreach activities.” (17) Thus, the ability of undergraduate groups to organize and work at health fairs is questionable. Undergraduate student volunteer groups have the potential to assist in preventive screening needs, but they must develop more effective training tools, language skills to communicate findings, and understanding of the health status and needs of their target population.(18)
Common Health Fair Screenings
Many health fairs conduct hypertension screenings by taking blood pressure measurements, but it is unclear if these screenings actually help to prevent the illness. One of the main problems with screening for hypertension is that in order for it to be effective, elevated blood pressure must be controlled through treatment, which the health fair itself does not provide. Achieving medical compliance with hypertension treatment is especially important because many hypertensives do not receive treatment or monitor the disease. A study conducted in the U.K. found that 51% of known hypertensives were not controlled, and 39% of known hypertensives were neither currently medicated, nor being monitored by their general practitioners.(19) Screening for hypertension is also ineffective because a diagnosis of hypertension can only be made on the basis of a series of measurements taken over a period of time. “Single reading screens have been shown to result in 33% false negatives (missed hypertensives) and up to 66% false positives.”(20)
Another problem with hypertension screening is that it labels people with high blood pressure as hypertensive without providing the necessary education about treatments and interventions that can control the disease. Though many may think that providing volumes of print materials to take home and read later is sufficient education, this is not the answer.(21) In addition, labeling someone with a condition can have a negative impact on an individual’s health. Researchers report that for some individuals “simply being labeled as suffering from high blood pressure (HBP) led to increases in illness behaviors, including illness-related absences from work and increased reports of psychological distress.” (22) Thus, it is possible that screening programs actually can cause decreased productivity and increased psychological symptoms, and may contribute to an overall decrease in the quality of life of participants. “Hypertension is clearly a disorder in which the effects of labeling—such as psychological effects, medication side effects, and other negative impacts on the patient’s quality of life—might outweigh the benefits of early intervention through screening programs.”(23)
Moreover, labeling a person is especially unwarranted in the screening setting because “screenings are just that-screenings….it is clearly inappropriate to use one measurement of blood pressure, or blood cholesterol, to arrive at a ‘diagnosis.’” (24) With this in mind, screening counselors need to be sensitive to this issue and avoid using language that labels a participant. Since hypertension screening is not very effective at detecting at risk individuals, and since the labeling associated with hypertension may actually negatively impact the health of individuals, other interventions should be used. One major cost analysis concludes that an “intervention to improve patient compliance may be a better use of limited resources than maximum efforts to detect hypertension, which should give pause to the current nationwide exuberance for screening programs.”(25)
Like hypertension screening, in order for cholesterol screening to have a positive impact, those individuals identified as having an elevated cholesterol level must receive follow up care and counseling regarding treatment. Unfortunately, there is also a lack of follow up care and counseling regarding high cholesterol. A study which focused on a mall-based cholesterol screening program found that “of all the newly screened participants, only 53.7% of those who required further evaluation, according to the NCEP recommendations, had been followed up after 1 year.” (26) Similarly, in a study conducted on a cholesterol screening program in a Texas high school, only 27% of students who were identified as having elevated cholesterol levels reported following up with their physician within 16 weeks of the screening date. (27) Another study found that only 40% of respondents who had high blood cholesterol measurements at blood cholesterol screenings had seen a physician within 3 months.(28) Since many who participate in public screening programs have been previously tested, fall into low-benefit groups, or fail to comply with recommended follow-up “cholesterol screening programs of the type now commonly offered are unlikely to contribute greatly to the national efforts to further reduce coronary heart disease.”(29)
Cholesterol screening tests are also ineffective and inaccurate because of the variability that exists within test procedures and the differing training levels of the screeners. According to the Office of Inspector General, in the typical cholesterol screening program, staff training can range from medical technologists to those with no medical background. In addition, many screeners do not comply with proper testing techniques to ensure accurate results. For example, milking the finger can have a significant effect on test accuracy; nevertheless, 58% of screenees who gave blood by fingertip had their finger milked. Another study which investigated four public screening programs found that only one of the four programs demonstrated an acceptable level of test error. (30) Thus, in order for cholesterol screenings to be effective, more consistent and accurate testing methods are needed, as well as ways to ensure follow up and treatment compliance
Skin Cancer Screenings
Skin cancer has several features that can make it appropriate for a screening program. Skin cancer is highly prevalent, the skin is an accessible organ to examine and treat, and the screening is brief, noninvasive and inexpensive. However, “although screening for melanoma/skin cancer is theoretically of value, few data are available to evaluate its effectiveness. In large part, this is because few previous studies have systematically followed subjects after the screen to determine their outcome and eventual disease classification.”(31) In addition, the validity of the visual examination used to detect skin cancer is of debatable accuracy. Though dermatologists generally regard visual examination as an accurate means of detecting melanoma/skin cancer, few data exist on the validity of visual examination as a screening tool. Skin cancer screenings may be inaccurate because they are subjective. There is no quantitative criteria used to measure melanoma, and screeners may therefore differ in their ability to recognize a subtle melanoma. There is also significant intraobserver variability: the same dermatologist may detect a clinically subtle melanoma one month, but not the next. Skin cancer screening generally involves a total cutaneous examination (TCE), which is a 2 to 3 minute visual inspection of the patient’s entire body. “TCE is preferable to examining only the sunlight-exposed areas of the body because of its ability to detect more lesions. However, there is often inter-examiner disagreement regarding skin lesions during screening and a tendency to overdiagnose in the screening setting.” (32)
Skin cancer screening has also created a number of ethical and legal issues. “In usual health care, symptoms or scheduled checkups motivate patients to see physicians. By contrast, in screening, providers invite ostensibly healthy persons to attend, believing that the overall health of the community will be improved. Not all persons benefit, however, and some may be adversely affected, particularly those with a false-positive test result who undergo unnecessary biopsy or surgery procedures.” (33) Another problem with melanoma screening is the fact that a complete skin examination is more likely to detect melanoma than a partial examination. Conducting a complete skin examination is much less feasible in a health fair or community screening setting due to privacy concerns.
Similar to other screenings, skin care “screening itself confers no health benefit, those with a positive screen must obtain subsequent care for definitive diagnosis and treatment.” (34) Unfortunately, in a study conducted in Massachusetts, despite a recommendation to do so, 22% of the positive screenees did not seek follow up with a physician to confirm the screening assessment. (35) Moreover, the ability of primary care physicians to detect melanoma is questionable. “The little evidence available suggests that primary care physicians lack the time and the training to recognize melanoma and other common dermatologic disorders.”(36)
Undiagnosed diabetes contributes significantly to morbidity and mortality rates worldwide. In the United States, the prevalence of undiagnosed diabetes is 2.7%. Though diabetes represents an important health problem that imposes a significant burden on the population, “there are no empirical data to demonstrate the benefits of screening, and there are few data on risks.” In addition, “population-based and selective screening programs in community settings (outreach programs, health fairs, or shopping malls) have uniformly demonstrated low yield and poor follow-up.” (37) The American Diabetes Association (ADA) and others also oppose community-based screening, such as in churches or at health fairs.(38)
A study on 3,301 subjects aged 20 years or older who were not pregnant and who did not have a history of diabetes were screened in a community-based program in Michigan. The yield of diagnosed diabetes was very low (the proportion diagnosed with diabetes was 0.5%). Thus, the authors concluded that “screening for diabetes was extremely inefficient at identifying individuals with undiagnosed diabetes. The ADA diabetes screening questionnaire resulted in many false positive tests, and the ADA criteria for positive plasma glucose tests likely missed a substantial portion of individuals with undiagnosed diabetes.”(39)
Though screening for diabetes has many pitfalls, screening individuals with one or more risk factors may prove beneficial. A study found that screening individuals with one or more risk factors resulted in a 100% sensitivity rate for identifying individuals who are at risk for complications associated with hyperglycemia. (40) Nevertheless, screening low-risk populations for diabetes is not recommended b the ADA and others.
Though health screenings at health fairs have the ability to reach a large segment of the public and identify people as being at risk for disease, there is a significant lack of data regarding their efficacy. “The proliferation of these screenings has proceeded largely without the guidance of any systematic evaluation of their quality, accuracy, or ultimate efficacy either for detecting disease or for reducing risk factors for chronic disease. Virtually no data exist concerning disease outcomes.” (41) In addition, for most health fair tests, follow-up is required either to treat the condition or to confirm the diagnosis. Thus, health fairs are not a substitute for seeking comprehensive care with a physician and are potentially harmful because they may cause those who receive false negative results to think that they do not need to seek comprehensive medical care when they actually do.
(1) Berwick, D. “Screening in Health Fairs. A Critical Review of Benefits, Risks, and Costs.” Accessed on 20 October 2010.
(5) Alpert, J., Greiner, A., and Hall, S. “Health Fair Screening: The Clinical Utility of the Comprehensive Metabolic Profile.”Family Medicine. 36.7 (2004): 514-519. Accessed on 22 October 2010.
(7) Berwick, D. “Screening in Health Fairs. A Critical Review of Benefits, Risks, and Costs.” Accessed on 20 October 2010.
(8) Dulin, M., Olive, K., Florence, J, and Sliger, C. “The Financial Value of Services Provided by a Rural Community Health Fair.” Journal of Health Care for the Poor and Underserved. 17.4 (2006): 821-829. Accessed on 21 October 2010.
(9) Berwick, D. “Screening in Health Fairs. A Critical Review of Benefits, Risks, and Costs.” Accessed on 20 October 2010.
(10) Heath, J., Lucic, K., Hollifield, D., and Kues, J. “The Health Beliefs of Health Fair Participants.” Journal of Community Health. 16.4 (1991):197-203. Accessed on 20 October 2010.
(11) Berwick, D. “Screening in Health Fairs. A Critical Review of Benefits, Risks, and Costs.” Accessed on 20 October 2010.
(16) Mays, V., Ly, L., Allen, E., and Young, S. “Engaging Student Health Organizations in Reducing Health Disparities in Underserved Communities through Volunteerism.” Journal of Health Care for the Poor and Underserved. 20.3 (2009): 914-928. Accessed on 21 October 2010.
(18) Mays, V., Ly, L., Allen, E., and Young, S. “Engaging Student Health Organizations in Reducing Health Disparities in Underserved Communities through Volunteerism.” Journal of Health Care for the Poor and Underserved. 20.3 (2009): 914-928. Accessed on 21 October 2010.
(19) Getliffe, K., Crouch, R., Gage, H., Lake, F., and Wilson, S. “Hypertension awareness, detection and treatment in a university community: results of a worksite screening.” Public Health. 114 (2000): 361-366. Accessed on 22 October 2010.
(21) Lefebvre, C., Hursey, K., and Carleton, R. “Labeling of Participants in High Blood Pressure Screening Programs. Implications for Blood Cholesterol Screenings.” Arch Intern Med. 148. (1988): 1993-1997. Accessed on 22 October 2010.
(25) Berwick, D. “Screening in Health Fairs. A Critical Review of Benefits, Risks, and Costs.” Accessed on 20 October 2010.
(26) Lefebvre, C. Banspach, S., Gans, K., Carleton, R., and Lasater, T. “Enhancing adherence to referral advice given at blood cholesterol screenings: impact on participant follow-up and physician behavior.” Health Education Research. 6.4 (1991): 405-413. Accessed on 21 October 2010.
(27) Fischer, P., et. al. “Impact of a Public Cholesterol Screening Program.” Arch Inter Med. 150. (1990): 2567-2572. Accessed on 22 October 2010.
(28) Lefebvre, C. Banspach, S., Gans, K., Carleton, R., and Lasater, T. “Enhancing adherence to referral advice given at blood cholesterol screenings: impact on participant follow-up and physician behavior.” Health Education Research. 6.4 (1991): 405-413. Accessed on 21 October 2010.
(29) Fischer, P., et. al. “Impact of a Public Cholesterol Screening Program.” Arch Inter Med. 150. (1990): 2567-2572. Accessed on 22 October 2010.
(31) Koh, H., et al. “Evaluation of Melanoma/Skin Cancer Screening in Massachusetts.” Cancer. 65 (1990): 375-379. Accessed on 25 October 2010.
(32) Ferrini, R., Perlman, M., and Hill, L. “American College of Preventive Medicine Policy Statement: Screening for Skin Cancer.” Am J Prev Med. 14 (1998): 80-82. Accessed on 25 October 2010.
(33) Koh, H., et al. “Evaluation of Melanoma/Skin Cancer Screening in Massachusetts.” Cancer. 65 (1990): 375-379. Accessed on 25 October 2010.
(36) Koh, H., Lew, R., and Prout, M. “Screening for melanoma/skin cancer: Theoretic and practical considerations.” Journal of the American Academy of Dermatology. 20.2 (1989):159-172. Accessed on 22 October 2010.
(37) Engelgau, M., Narayan, K., and Herman, W. “Screening For Type 2 Diabetes.” Diabetes Care. 23.10 (2010): 1563-1580. Accessed on 25 October 2010.
(38) Porterfield, D., et. al. “Screening for Diabetes in an African-American Community: The Project DIRECT Experience.” Journal of the National Medical Association. 96.10 (2004)” 1325-1331. Accessed on 25 October 2010.
(39) Tabaei, B., et. al. “Community-Based Screening for Diabetes in Michigan.” Diabetes Care. 26.3 (2003):668-670. Accessed on 25 October 2010.
(40) Behan, K. “Screening for Diabetes: Sensitivity and Positive Predictive Value of Risk Factor Total.” Clinical Laboratory Science. 18.4 (2005): 221-225. Accessed on 25 October 2010.
(41) Lefebvre, C., Hursey, K., and Carleton, R. “Labeling of Participants in High Blood Pressure Screening Programs. Implications for Blood Cholesterol Screenings.” Arch Intern Med. 148. (1988): 1993-1997. Accessed on 22 October 2010.