My name is Kathleen Holt and I am a staff scientist in the informatics department of the University of Rochester Medical Center's Clinical and Translational Science Institute. I also have an appointment in the University Center for Community Health and Prevention. My PhD is in social psychology and psychometrics and my day-to-day work involves making data into actionable information. Today, I will give an overview of population and public health data sources. Both sources of getting the data themselves, such as in spreadsheet form for inclusion analysis, as well as some tools for exploring those data geographically and with maps. Why are public health data important? Why do we need to measure health? We measure so we may improve public health to know what programs, research, and efforts are effective for disease control and all health promotion. How do we assess these efforts? How do we know they make a difference or not? We rely on data collection. This graphic from the Centers for Disease Control and Prevention describes factors and processes that contribute to public health. These factors are from a variety of data sources, are not merely a function of the medical and clinical care that people receive, but socio-economic and physical environment factors that surround them. My presentation will concentrate on the what and where of public health. Initially, I will cover the what kinds of data are available and how can you access these data. Later, I will discuss the where of these data. But first, a caution and some background. This quote from a paper published in the Journal of the American Medical Association reflects a very real aspect of our work in public health. The central issue in accessing public health initiatives is not the scarcity or lack of data, rather it is the usability and applicability of those data to help understand public health issues. How long has this interest in data been evident? Is this just a recent phenomenon with all the discussion about big data? This Google trend search shows the relative frequency of the search terms population health and health data over the past 15 years. There are several interesting things to note about this graph. First, is the similar periodicity, the peaks and troughs of the two lines showing the frequency of word searches. The lowest points, the troughs correspond to July and December of each year. This may suggest that these are searches done by academics and not by the general public. Typical slow times in the academic year are the summer and winter breaks and if academics and researchers are doing these searches, this is the pattern that would result. A second interesting factor in this graph is the increase since 2010 or so in the frequency of both search terms, which may suggest a recent increase focus on these ideas. Finally, it is interesting to consider why there are somewhat higher frequencies for the searches about health data rather than population. This is an area of great interest and great activity. Survey methodology for these population-based surveys has traditionally included landline telephone, random digit dialing per participant recruitment. In recent years, this technique has evolved to include cellphone numbers and in-house interviews based on random selection by a geographic region and household composition. A little about each of these sets of data. The National Health and Nutrition Examination Survey, or NHANES, began in the 1960s and includes interview data containing demographics, socioeconomic data, and health-related questions, as well as including an actual medical examination consisting of physiological evaluation and laboratory tests. Began in 1984 the Behavioral Risk Factor Surveillance System, or BRFSS, sometimes referred to as BRFSS, is a national survey of health conditions, health risk behaviors, and chronic health conditions. The National Survey of Family Growth began in the 1970s, contains information on issues and family life such as marriage, divorce, contraceptive use, and general reproductive health. The national immunization surveys, begun in 1994, are a series of phone surveys to monitor vaccination coverage in children and young adults. The National Health Care Surveys is a survey of healthcare providers and healthcare entities. These potential interviewees register with the survey to provide information and are randomly selected for participation. The State and Local Integrated Telephone Survey supplements national data collection with focused surveys on topics relevant to local or regional concern. These survey tools are all under the auspices of the Centers for Disease Control, the CDC. Literally, thousands of datasets exist, many geographically specific. Additional data, those specific to states are available through portals created and maintained by state government or state departments of health. Here are just a few examples from New York, California, Illinois, and North Carolina. For more granular data, organizations in many regions such as Rochester and Monroe County, make local data available. The New York State prevention agenda, part of a larger New York State site mentioned in the previous slide, contains county level health data in six areas with downloadable data and benchmark progress toward goals. ACT Rochester, an initiative of the Rochester Area Community Foundation presents data and report cards on a wide variety of social determinants of health, such as housing, education, economic security, and community vitality. This is for the City of Rochester as well as surrounding counties and Monroe County. The Monroe County health department has released several reports on adult and child health in areas such as chronic disease, environmental health, and immunization. In addition to the survey datasets mentioned earlier, there are many sets of data on social determinants, including environment, health insurance, crime statistics, education, and poverty. Two large meta repositories of national data in the United States are HealthData.gov and the broader dataset, Data.gov. Both are available at the US government's websites.
