Young Epidemiology Scholars Competition

"I decided to study smallpox, a potentially devastating terrorist weapon in the context of a dynamic and complex social world."

In his first place YES project, Benjamin developed a mathematical model to evaluate the effects of different smallpox vaccination strategies based on data from previous outbreaks of smallpox. He found that mass vaccination before an outbreak typically resulted in significantly fewer infections than vaccinations after the outbreak, but that the two strategies were equal in their ability to eliminate the virus from the population within five months. A National Merit Scholar, Benjamin was a member of the Akiba Hebrew Academy debate team and a volunteer tutor for children at St. Barnabas Homeless Shelter. His long-term interest in computer science led to his involvement with a University of Pennsylvania research group using computer models to study social science and geopolitical issues. He also did research at MCP-Hahnemann University on the development of computer games for use in surgical education curricula. He will pursue undergraduate study in political science and philosophy at Yale University.

VIR-POX: An Epidemiologic Study of Smallpox Preparedness and Response Policy

Abstract

The possibility of a bioterrorist smallpox attack poses a serious, immediate threat to national security and public health. Empirical data from previous outbreaks were used in concert with computational techniques to create a "hybrid" simulation model, called VIR-POX. Stochastic data were then collected with VIR-POX and used to study the viability of vaccination measures as defenses against the spread of smallpox. More than 19,000 virtual futures were extrapolated, uncovering complex interaction effects between different response strategies and scenarios posing different kinds of demographic challenges. Ring vaccination rivaled mass vaccination in effectiveness when a very substantial proportion of new cases could be detected and isolated almost immediately after infection, but ring vaccination was also significantly more susceptible than mass vaccination to complication when confronted with an unfortunate stream of accidents or a particularly vulnerable community. Pre-attack mass vaccination typically resulted in significantly fewer infections than post-attack mass vaccination, but the two strategies were nevertheless equivalent in their capacities to eliminate the virus from the population within five months. The relationship between the portion of the population vaccinated and the scale of the subsequent outbreak was not linear, instead yielding clear "sweet spots" at which substantial additional security could be obtained through little additional investment. These conclusions also point to the importance of better understanding residual herd immunity and the feasibility of early detection measures.

"Seeing the hallways filled with students, many of whom are friends, who looked remarkably tan in the middle of winter made me think … I want to tell students that they can't forget the long-term ramifications of their actions for something as superficial as a good-looking tan."

Concern for the health of his friends and classmates motivated this avid student athlete and outdoorsman in the design of his YES project. A National Honor Society student, Robert complements his studies with active participation in organizations such as the Interact Club, Future Problem Solvers and his synagogue youth group. His passion for outdoor activities is evident in hobbies that include soccer, rock climbing, road cycling, running and varsity athletic competition on the track and field team. An environmental activist in his school, Robert is president of the Students Against Violations of the Environment (SAVE). In his YES project, he surveyed a large Midwestern high school, known for its academic excellence, to quantify indoor tanning behaviors and sunscreen use. Reasons for tanning as well as other influences were examined. This is important in light of the increasing popularity of tanning salons and the known association between exposure to ultraviolet radiation and the increased risk of skin cancer. Planning to major in biology or environmental studies, Robert will attend Middlebury College.

"This project is of interest to other communities surrounding landfills that have similar concerns for their residents."

Fascinated by her introduction to epidemiology during summer study in Costa Rica, Katherine's project resulted from her interest in creating a personally meaningful project that addressed a community health issue. She conducted a study of residents living near a landfill in Pottstown, Pennsylvania, and determined that there was no evidence of a cancer cluster due to the landfill. A varsity tennis player, Katherine is active in the Girls Athletic Association and the Special Olympics club. Her academic achievement is recognized by membership in the National Honor Society. In her free time, Katherine relaxes by reading and shopping. After graduation, she plans to pursue undergraduate study in biology and anthropology at Washington University in St. Louis.

Evaluation of Possible Elevated Cancer Rates in the Pottstown, PA Area

Abstract

The purpose of this study was to evaluate the possibility that the Pottstown Landfill is causing elevated cancer rates in residents of the area around the landfill. Concern has arisen from the Alliance for a Clean Environment, a local environmental group, calling for the closing of the landfill due to effects of chemicals on the residents. This concern has been echoed by the residents themselves, leading to increased suspicion of cancer rates around the landfill. The data for this study was obtained from the Pennsylvania Department of Health. Twenty-three primary cancer sites were chosen, and for each of these sites the number of diagnosed cases from 1990 to 2000 was found for residents living in the five townships bordering or encompassing the landfill. Incidence rates were calculated, and then those rates were compared to the incidence rates of the three counties encompassing the townships, and the state of Pennsylvania itself. The risk rates of leukemia and cervical cancer for the affected population were significantly elevated. However, there is no evidence that this is due to an environmental effect, such as the landfill. The landfill has not resulted in a cancer cluster of the twenty-three types of cancer studied in the Pottstown area.

"Body piercing and the associated infections and complications have become an epidemic among American teenagers. Many of my friends have recently been pierced and subsequently developed some nasty infections and complications."

Victoria's project reflected her concern about the current teenage trends in body piercing and the high incidence of infection experienced by many of her friends. Her survey of high school students linked an increased risk of infection to an increased number of body piercings. She also found a lack of enforcement of health policy regulations regarding piercing establishments. Victoria's interest in science is evident in her work with animals as a veterinary technician and as a volunteer at a wildlife rehabilitation center. Her volunteer activities include working as a peer tutor and a teaching assistant. She currently participates in leadership roles as president and charter member of the American Sign Language club and secretary of the Model United Nations club. Her wide ranging personal interests include running, weight training, writing, poetry and baking. She relaxes by reading and observing nature. Victoria will pursue undergraduate work in biology at Cornell University.

Body Piercing and Risk of Infection

Abstract

This investigative research project defines the relationship between body piercing and the spread of infectious disease. Viral and bacterial infection resulting from exposure through body piercing has become an epidemic primarily affecting high school and college students. The descriptive aspect of the study supported the hypothesized argument that if a subject has a greater number of body piercings, then their risk for infection will increase. The data that supported this hypothesis was obtained through a survey of an anonymous high school population. The survey testified to the correlation between an increased number of piercings and an increased risk of complications, primarily viral and bacterial infection. Results of the survey did not confirm the secondary hypothesis that proposed that an individual pierced in an unlicensed facility has a greater risk factor for contracting infection. However, interviews with anonymous body piercing establishments rendered this data unreliable. The significance of the project lies in the definitive conclusion that body piercing is a prominent and alarmingly common factor pertaining to susceptibility of infection. Due to repeated exposure, the risk of infection increases with an increased number of body piercings. This data could be used to promote the creation and enforcement of sanitation standards and legislation targeting body piercing practitioners. Documentation of this study and the accompanying statistical analysis could additionally be used to support the adaptation of educational programs to include discussion of the dangers of body piercing and ways through which this risk may be reduced.

"The ultimate opportunity that exists to improve the quality of life for individuals alive today, as well as prevention of the spread of infections in the future, is inspiring."

Inspired by the opportunity to improve the quality of life and to prevent the future spread of infection, Bevin developed her project to build upon previous research by her mentor. She conducted a study that compared hand hygiene practices in two neonatal intensive care units and suggested ways to decrease exposure to infection among infants in those units. Bevin's experience in science competitions includes participation in the Long Island Science Congress, INTEL Science Talent Search, Junior Sciences and Humanities Symposium and the Long Island Science and Engineering Fair. She writes for her school newspaper and is active in a variety of organizations including 4H, Model United Nations and Habitat for Humanity. She enjoys cooking, organizing dinner parties, fine arts and crafting. A vocalist as well as an instrumental musician, Bevin plays guitar and piano. She plans to continue her studies at the University of Vermont.

A Comparison of Hand Hygiene Practices in Two Neonatal Intensive Care Units

Abstract

Compliance with hand washing protocols in hospitals, specifically in intensive care facilities, is recognized as one primary means for controlling the spread of healthcare-associated infections (HAI). However, noting that many studies found compliance among health care workers to be less than 50% overall, and that one third of all HAI are estimated to be preventable, there is an alarming demand to improve hand hygiene practices among health care workers. The study was conducted in two New York City neonatal intensive care units (NICU) for a period of one month. Observers used a standardized tallying sheet to record all touches made in the room by level of contact with neonates, person making the contact, and the hygienic condition of the person's hands. At hospital A an alcohol based emollient hand rub was used for hand sanitation, while at hospital B traditional hand washing with a 2% chlorhexidine gluconate (CHG) detergent was used. The data was statistically analyzed using SPSS software (Chicago, Il) for differences in practices between the two hospitals, differences in practice between three levels of baby contact, and differences between groups (nurses, attending physicians, medical residents, other health care workers, and visitors.) Statistically significant differences in hand hygiene practices were found between all groups of people, between the three levels of contact, and between the two hospitals. The significant differences indicate a need for increased attention towards hand hygiene in hospitals and provide a basis for future implementation of more effective hand hygiene protocols for hospital staff. Improved practices are ultimately expected to result in a decreased rate of HAI.

"I wondered how a disease would actually spread in a confined space like a cruise ship, and began to theorize with my mentor about how the disease would be passed in waves to the rest of the people on the ship. This discussion led me to questions pertaining to the spread of disease in other closed-system populations and the processes of epidemiology."

Anna enjoys the challenge of developing viable solutions using the research process of hypothesis testing and data collection. She developed a computer simulation to model the spread of infectious disease and showed that the rate at which an epidemic will spread and "burn out" increases as the greater number of social connections that exist in a community increase. Encouraged by a teacher to enter the YES competition, Anna is no stranger to research competitions. She reached the regional finals in the 2003 Siemens Westinghouse competition, was a semifinalist in the 2004 Intel Science Talent Search, was a Finalist in the 2004 Intel ISEF and was named overall winner at the 2004 Science Horizons and Connecticut State Fairs. Her research is currently being used in support of legislation to contain the spread of invasive aquatic species in Maine lakes. A member of the National and Science Honor Societies, Anna is a peer tutor. She is the editor-in-chief of the high school literary magazine and enjoys composing music for the violin. Anna will pursue studies in biomedical engineering and writing at Stanford University.

Disease Propagation Through Connective Paths: Application of the Small World Theory to Epidemic Modeling

Abstract

The researcher developed an original methodology designed to provide a deeper insight into the spread of disease. Initially inspired by Milgram's Small World Theory, the research combined his idea of population connectivity with computer simulation-based epidemiology, two previously unrelated fields of study. Two programs were written in Visual C++: one to create 300 sample populations with different connectivity metrics, and another to run 7,500 simulated propagation trials. The average number of individuals removed by the epidemic was found to be an exponential function of the connections per capita in the given population. The average time for the epidemic to burn out was shown to rise and then fall, as the number of connections per capita increased. Analytically determining the epidemic behavior in a large connected population is virtually impossible. The computer simulations enable the investigation of the variability of the epidemic outcome, as well as its average behavior. Epidemic distributions were discovered which describe this variability and reveal how it changes as the connectivity of the population changes. The results exhibited three distinct regions of epidemic behavior as the connectivity of the population increased. The epidemic distribution predicts how far and how fast a disease will spread in a population, and how those results are likely to vary. Studies indicated that the presence of even one highly connected individual can dramatically alter the spread of a disease. Discovering this epidemic behavior is a very significant result, and provides the researcher with simple, but intriguing new strategies to combat epidemics.

"I am committed to trying to fully understand the brain from medical, psychological and philosophical directions in an attempt to develop better insight that would lead to better understanding and treatment of head injuries."

Zarabeth's YES project is the latest in a series of projects on the treatment and causes of brain injuries that extends back to fifth grade. She reviewed the cases of elderly subjects diagnosed with dementia who later underwent a more comprehensive evaluation. Zarabeth found that the rates of the types of dementia that can potentially be cured or prevented - as opposed to Alzheimer's, which currently can be neither cured nor prevented - are higher than reported by less sophisticated diagnostic techniques. She said this could affect the allocation of prevention and health care resources, particularly given the increasing elderly population in the United States. In 2002 and 2003, she presented at national conferences on topics related to Alzheimer's and vascular dementia. She has participated in the Florida State Science Fair, the INTEL International Science Fair, the INTEL Talent Search and the Florida Science and Engineering Symposium. At school, she is president of the Science and Robotics clubs, captain of the swimming team and a member of student government. Zarabeth finds time to volunteer at the Ocean Hyperbaric and Nova Southeastern University. Her hobbies include massage, gymnastics and drawing. Zarabeth will enroll as an undergraduate at Emory University to pursue the study of neuroscience and psychology.

The Epidemiology of Different Forms of Dementia in an Urban Elderly Population

Abstract

The purpose of the present study was to investigate the incidence of specific dementia subtypes in a population which received detailed diagnostic work-ups. The study also investigated the demographic associations of each subtype in an attempt to identify factors which predicted the incidence of the subtype. Data from the files of a group of 705 elderly subjects diagnosed with dementia and referred for more in-depth diagnostic evaluations were included in the study. Subjects were divided into four subtypes of dementia. The groups were compared on demographic, cognitive, and lifestyle variables. The study identified significantly different frequencies of each subtype than did previous studies, apparently due to the use of more sophisticated diagnostic procedures. The group with pseudo-dementias (no dementia) was much larger than previously estimated. The results suggested that previous studies with less extensive diagnostic techniques may have overestimated the incidence of Alzheimer's disease (AD) and Vascular dementia by 25%. The results also indicated strong relationships between type of dementia and age, education, history of AD, history of head injury, cardiac history, and marital status. The apparent cognitive decline in each group was the same, as were estimates of premorbid intelligence. Increasing age was associated with higher rates of AD and Vascular Dementia, but not other forms of dementia. Education appeared to offer "protection" from AD but not the other categories of dementia. Widowed subjects were over-represented in the dementia sample, especially in the AD group. The results have extensive implications for resource allocations in the elderly population.

"This project brings attention to the prevalence of mothers who continue to smoke during pregnancy and the stressful life events that make it much more difficult for them to quit."

After a summer of volunteering in the obstetrical department of a local hospital, Alanna decided that a medical career is in her future. Enrolled in the Science and Technology Program at her high school, she decided to combine her career interest with the substance abuse specialty of her mentor in the design of her YES project. Alanna analyzed data from a study of pregnant women in North Carolina and found that women who were going through a greater number of stressful life events were less likely to stop smoking during pregnancy, despite the known risks to their offspring. Alanna is a member of the French and National Honor Societies, the Future Business Leaders of America, the Medical Careers Club and the Girl Scouts of America. She is an avid reader, enjoys dancing and is quite active in services and activities at the Fort Washington Baptist Church. As preparation for her anticipated medical career, Alanna will pursue undergraduate work in biology and psychology at Stanford University.

Smoking and Stress in Our Future

Abstract

Research has determined the harmfulness of smoking during pregnancy. Despite these widespread findings, a significant percentage of women continue to smoke throughout their pregnancy. In and of itself, pregnancy can be a challenging time for women without having to consider the presence of other negative factors. This study investigated the relationship between stressful life events and continuing to smoke during pregnancy. A secondary analysis of the North Carolina Pregnancy Risk Assessment Monitoring System (PRAMS) dataset was conducted. PRAMS is a CDC-sponsored initiative that asks pregnant women about a variety of topics. This study examined the pre-pregnancy smoking sample and created a quitter variable based upon whether or not these women were still smoking during the last trimester. The study results showed a significant association between stress and smoking, that is, the women who were going through certain stressful events were more likely to continue smoking. In addition, the total number of stressful events experienced was related to a higher risk of continuing to smoke, controlling for the effects of demographic variables. Non-quitters had an average of about one more stressful life event than quitters. This study could lead to designing programs to help women cope with stress better so that they avoid smoking during pregnancy and preserve the lives of themselves and their children.

"My research was inspired by a previous retrospective study of treatment techniques for prostate cancer, as well as my readings of other studies, particularly Hank's 'Patterns of Care Studies'."

While volunteering at a local hospital, Eugene became interested in cancer research. Eugene feels that conclusions from his YES project will be useful to patients, clinicians and investigators who use the American Joint Committee on Cancer staging system. He evaluated recent changes made in the system to determine the odds of surviving breast cancer without recurrence and found that the new system gives a more accurate picture of the prognosis. A seasoned competitor, Eugene has participated in the Washington State and National Junior Science and Humanities Symposia, the INTEL Science Talent Search, and as a member of the state, regional and USA Biology Olympiad teams. Enrolled in the International Baccalaureate Program, Eugene is a member of the National Honor Society and the Portland Youth Philharmonic. He continues volunteer work at the St. John Medical Center. Eugene will attend Case Western Reserve University.

Evaluation of Breast Cancer Survival Based on the Fifth and Sixth Edition AJCC Staging Systems

Abstract

Between 1974 and 2002, 608 patients staged as II and III based on the American Joint Committee on Cancer (AJCC) fifth edition staging system for breast cancer were identified in the cancer registry at Saint John Medical Center in Longview, Washington. Neo-adjuvant chemotherapy or radiation therapy patients were excluded from this study. All patients were restaged according to the new sixth edition staging system. This study compares disease-free survivals (DFS) between the old fifth edition AJCC stage classification and the new sixth edition. The mean and median disease free follow-ups were 72.2 and 52.7 months respectively, ranging from 1.2 to 335 months. Using the fifth edition, Stage II accounts for 522 patients (85.9%). Ninety-five patients (18.2%) were upstaged in the sixth edition from Stage II to Stage III. DFS between the Stage II and up-staged groups were significantly different (p<0.00001). The five and ten year DFS for Stage II, fifth edition were 81% and 70% and for Stage II, sixth edition were 84% and 75% respectively but were not statistically significant (p=0.1656). The five and ten year DFS for Stage III for both editions were similar. Overall, DFS rates were found not to be significantly different when compared stage to stage between the fifth and sixth editions. However, a small group of patients upstaged from the Stage II of the fifth edition to the Stage III of the sixth edition showed significant differences in DFS, reflecting a more accurate survival prognosis with the new sixth edition staging system.

"Bioterrorism is a very real threat that can result in tragic consequences. The possibility of a smallpox outbreak within the US, as well as the dire necessity that viral outbreak behavior be predicted beforehand, were the motivating factors for my research."

News coverage of potential bioterrorist threats stimulated the development of Stephanie's YES project. She developed a computer model of a smallpox epidemic that showed the disease spreads faster as the population density and the mobility of the people in a community increase. A certified EMT, she volunteers in the Finderne First Aid and Rescue Squad and serves as president of her high school's emergency rescue club. An accomplished musician, Stephanie plays violin and piano and is a member of the All-State High School Orchestra. After graduation in 2005, she plans to pursue undergraduate study in biology and applied mathematics at an Ivy League university.

Smallpox Epidemic Modeling: Application of Population Mobility/Spatial Characteristics in Predicting Disease Spread in Metropolitan Cities

Abstract

A computer smallpox epidemic model was developed to predict virus transmission intensity and patterns for assessing risk. Utilizing visual computer modeling and simulation to incorporate biological and demographic/spatial realism in the study of smallpox outbreaks, a hypothesis on whether increased mobility would increase number of viral infections was tested. Results obtained from the smallpox simulation model were then employed to construct an epidemic distribution curve that predicted the scale of casualties for various population densities. Specific demographic parameters consisting of population density and mobility were incorporated in the simulation model to test for correlation with outbreak intensity. All results from the simulation runs were analyzed using the Two-Tailed Statistical Hypothesis Testing method. Final analysis displayed a strong positive correlation between the percentage of population infected and population density. Furthermore, population mobility patterns were found to have significant impact on outbreak intensity for heavily populated urban areas.

"This model helps determine who should be vaccinated to maximize influenza prevention. Moreover, the model evaluates other interventions such as school closing to determine their effectiveness in minimizing disease."

Evan did his first immune system project in fifth grade. His YES project was influenced by his father's interest in vaccines and preventative medicine and was developed after a summer research experience with Dr. Michael Haber. Evan used a computer simulation to show that people who have a higher number of contacts with people outside their own age group are more likely to spread influenza. He indicated this might suggest that such people should be targeted for vaccination. Active in science competitions, he has participated for the past three years in the regional and state Science Olympiad programs. His other competitions include the Junior Engineering Technical Society TEAMS competition, the State Science Bowl Competition and the Problem Solving Bowl Competition. Evan participates in the Westminster Service Club and Project Isaiah. He enjoys the theater along with outdoor activities like rock climbing, hiking, and backpacking. He will study molecular biophysics and biochemistry at Yale University.

Sensitivity Analysis of an Influenza Epidemic Simulation: Analysis of Exposure and Intervention

Abstract

This sensitivity analysis examines possible interventions for controlling the spread of an influenza epidemic within a computer simulation developed by Dr. Michael Haber and research assistant Rajan Patel. The simulation models a census-sized population separated into age groups with different characteristics. Influenza is introduced into the community and spreads via contacts within households, schools, workplaces, and the general community. This analysis systematically varies different initial parameters dealing with withdrawal from the community, initial vaccination rates, school closings, and vaccine efficacy. The results are presented to show the relative strengths of different intervention strategies such as quarantine or increased vaccination. This information serves as the basis for recommendations for the simulation itself as well as conclusions that extend beyond the simulation with implications for the understanding of herd immunity. In particular, this research demonstrates the importance of contact diversity when considering the direct and indirect benefits and overall effectiveness of vaccination. Similarly, other interventions such as school closings and quarantine change certain populations' contact rates both in number and in variety, thereby changing the dynamics of disease spread. These changes can be interpreted to understand how and why an intervention strategy affects attack rates for a population with given contact patterns.

"I think that the field of epidemiology is going to be greatly needed in the future. With the outbreaks of diseases worldwide and the looming threat of bioterrorism, sound public health methods and research will save many lives."

Bailey's first attraction to public health occurred at the age of 10 when he became fascinated by the "sleek" U.S. Public Health Service uniform worn by a Centers for Disease Control Epidemic Intelligence Service Officer. In his YES project, he studied the mortality rate during the heat wave of 1995 in Chicago. He demonstrated that there was a decrease in the expected mortality rate in the following year because the most vulnerable people had succumbed to the effects of the heat wave. He described how this phenomenon could help target public health efforts to prevent premature deaths. Bailey has done community service work in the Illinois Department of Public Health lab and was mentored there by Dr. Greg Huhn. At school, Bailey is a member of the IMSA Student Council, Mu Alpha Theta and the Varsity tennis team. His hobbies include mountain biking and composing music for the piano and harmonica. With anticipated high school graduation in 2005, Bailey is interested in university-level study in either law or medicine.

In and Out of the Hot Zone: A Study of Heat Wave Mortality Displacement

Abstract

Despite the improvement of living conditions in the U.S., heat waves still kill in large numbers. Because many of the victims were either elderly, belonged to low socioeconomic status, and had pronounced pre-existing medical conditions, it was hypothesized that these people would have died shortly afterwards even without a heat wave. Their accelerated deaths during the heat wave therefore would produce a period of decreased mortality following the heat wave known as mortality displacement. This study investigated whether or not there was mortality displacement associated with the 1995 Chicago heat wave. Daily mortality numbers for Chicago during 1995 were compared with a baseline mortality constructed through averaging same day mortality in 1994 and 1996. Net mortality changes were compared to estimate displacement at various time intervals. Mortality displacement was not found 3 months after the heat wave. After 5 months, however, there was 10.7 percent displacement (95 percent confidence interval, 95% CI: 6.7, 14.9). The magnitude of displacement increased to 50 percent 11 months after the heat wave (95% CI: 34.8, 75.4). The difference between the actual and baseline mortality also indicated that people over age 65 or those with cardiovascular diseases were indeed more likely to die in the presence of the heat wave. Of all excess deaths observed during the 1995 Chicago heat wave, about 10 percent would have occurred within 5 months even without the heat wave. And in less than a year, about 50 percent of the heat victims would have expired anyway.