Assigment 6 modules reflexion 2-3 pages.Apa seven . All instructions attached.
Module 3: Population Health Surveillance
Welcome to Module three, where we will discuss population health surveillance.
Slide 2: Acknowledgements:
This educational module is made possible through a cooperative agreement between the Centers for Disease Control and Prevention and the Association for Prevention, Teaching and Research. My name is
Kristen Schneider from Rosalind Franklin University of Medicine and Science. And I want to acknowledge the other individuals who generously contributed to the creation of this training.
Slide 3: Population Health Modules:
Before we discuss this module's objectives, I want to highlight how population health surveillance is connected to the other modules covered in this training. Surveillance is often connected to population health assessment. As discussed in Module 2, surveillance data can be used to justify the need for a population assessment, and it can be used as part of the population health assessment. Population health surveillance data can also be used to inform the need for interventions and policies, which will be discussed in models 4 and 5. Upon implementation of interventions and policies, surveillance data can be used to then track the impact of those interventions and policies on population health.
Slide 4: Overall Learning Objective:
The overall goal of the series of modules is to provide foundational knowledge in population health science and best practices in population health, including the effective design, implementation, and evaluation of population health activities for public health professionals, students and public health or health professions, allied health, and health care providers.
Slide 5: Module 3 Objectives:
This model has five objectives. First, to enable you to describe the appropriate use of population health surveillance methods. Second, to identify ways to use epidemiological and health care data and observations to make inferences about the population. Third, to define the principal methods of analyzing population health surveillance data. Fourth, to identify ways to improve population health surveillance via data modernization, and five, to discuss strategies for effectively communicating population health surveillance data.
Slide 6: Objective 1:
After first defining population health surveillance, I will discuss appropriate ways to use population health surveillance.
Module 3 Transcript | 10
Slide 7: Public Health Surveillance:
As mentioned during the population health assessment module, there is a lot of overlap between public and population health. To understand population health surveillance, it is useful to understand what is meant by public health surveillance. Public health surveillance is the ongoing systematic collection, analysis and interpretation of health-related data essential to planning, implementation and evaluation of public health practice.
Slide 8: Population Health Surveillance:
Similar to public health surveillance, population health surveillance involves the ongoing systematic collection, analysis and interpretation of health-related data. Population health surveillance focuses on the surveillance of a specific population, often those with poor health outcomes, since populations with poor health outcomes are the ones that will benefit most from specific strategies, interventions and policies tailored to their needs. Most notably, the way in which population health surveillance is unique is because it typically involves the health care delivery system, as one of several partners. The most common way health care delivery systems are involved in surveillance is via use of electronic health record or EHR data for surveillance.
Slide 9: Population Health Surveillance Components:
Population health surveillance consists of several components, all designed to improve population health. Surveillance is used to track and monitor disease and injury. The surveillance of specific conditions and diseases are prioritized based on population health needs, often based on input from the World Health Organization and the CDC. Surveillance is used to detect and investigate notable disease events, symptoms, and outbreaks so that the health care system, public health workers and others can respond appropriately to address the problem and prevent further disease. Surveillance is also used to collect, collate, and analyze data that can summarize the population's health. Two standard population health status measures include the infant mortality rate, which estimates the rate of death in the first year of the infant's life, and life expectancy, which summarizes the impact of death and each age of life in a particular year. Other population health measures capture disability, given the rise in chronic health conditions. Lastly, surveillance data is used to generate reports that could facilitate or trigger a population health assessment and can be used to educate key stakeholders such as health care systems, policymakers and the community about health problems and progress. A bidirectional sharing of information typically occurs between the local or community level and higher levels of the system, such as at the state or federal level.
While surveillance data is helpful for improving population health, one common limitation of surveillance data is that there often is no context for the findings to understand what may explain changes in population health.
Slide 10: Types of Population Health Surveillance:
There are three main types of population health surveillance: indicator-based surveillance refers to the traditional reporting of disease diagnoses or indicators by a health care delivery system. As an example, consider how the electronic health record can be used to track
diabetes prevalence within and across health care systems. The second type, event-based surveillance refers to nontraditional reporting of health events that could impact the population health. Reports of a health event like a food borne illness and social media are an example of how this type of nontraditional reporting can occur. Syndrome surveillance serves to combine EHR data and nontraditional surveillance data to identify disease outbreaks earlier, this type of surveillance focuses on reports of patients’ symptoms before a diagnosis is made. For example, patients’ flu symptoms documented in the electronic health record could be combined with absenteeism data from local employers to identify a flu outbreak in a local community.
Slide 11: Objective 2:
Now that you have a basic understanding of what population health surveillance entails, I'll discuss the different ways in which epidemiological and health care data and observations are used to make population level inferences.
Slide 12: Best Practices for Population Health Surveillance Data:
I want to start by highlighting best practices for population health surveillance data. Population health surveillance data should be timely in that it allows for real time assessment and if necessary, intervention to prevent health threats like disease outbreaks and environmental risks. Consistency is key to maintaining the integrity of data collected over long periods of time. The integrity of the data can be compromised by changes in measurement or changes in data collection procedures. You also want consistency across systems, across health care systems, counties, states since different standards for reporting might mean that observed differences in population health observed between systems could be due to the way that the data are defined or reported instead of actual true differences between systems. Surveillance data collected should be broad and then it encompasses social determinants of health, not just disease states. Data should be representative of the population of interest, all too often, data might not be fully available for a population of interest. Any surveillance system should include appropriate safeguards in place to ensure that personally identifiable information is not disclosed inadvertently. With an increasing variety of data collected for surveillance, combining identified data sets may be used to deduce personally identifiable information. Thoughtfulness is critical to creating systems that can protect patient privacy. Lastly, it is ideal to connect multiple sources and types of data to create a detailed understanding of what factors could influence or contribute to the population's health. I’ll now discuss each of these different types of data and how they can be useful for population health surveillance.
Slide 13: Electronic Health Records (EHR):
Electronic health record or data is a key aspect of population health surveillance, since EHRs provide a direct connection to clinical indicators of disease. The use of EHR data benefits public health agencies by enhancing surveillance, data, quality, and variety. When using EHR data, public health agencies do not need to rely on self-reports of disease states, which may be less accurate. EHR data can provide greater consistency in how conditions are defined and how outcomes are measured. Through the HER, public health agencies gain access to data that may
not be readily available elsewhere, which also increases the variety of data they can access. There are also benefits to health care providers to provide public health agencies with access to EHR data. Major shifts by Medicare and Medicaid services, as well as by private insurers, promote the adoption of a value-based health care delivery model, where clinicians receive higher reimbursement for improved health outcomes of a defined patient population, rather than for number of individual visits and services. With value-based care, clinicians and public health goals are more aligned and focused on disease prevention in the overall population. The extent to which population health surveillance efforts can reduce disease in a provider's population could have significant implications for their reimbursement.
Slide 14: National Notifiable Disease Surveillance System (NNDS) Data Flow
Systematic collaboration between public health agencies and health care systems is critical to the use of our data for population health surveillance. The CDC has created the National Notifiable Diseases Surveillance System to monitor trends, study risk factors and evaluate prevention and control efforts. The system contains two components: case reporting and case notification. With case reporting, hospitals, health care providers and labs send data about positive lab results or disease diagnoses to appropriate public health departments in accordance with disease reporting laws. Ideally, oversight of surveillance at hospitals by health care providers and laboratories will be led by an infection control professional, or ICP, who is certified in infection control and epidemiology to ensure data integrity and timely reporting.
Case notification refers to the process by which health departments shared identified data about diseases to the CDC disease surveillance specialists in public health departments are responsible for data management, sharing the data with other agencies and communication about disease surveillance. CDC scientists use the data provided to monitor trends, study risk factors, and evaluate prevention and control efforts. The CDC will also determine whether the disease notification prevents a risk to others outside of the United States. If the CDC deems that the public health event is an international concern, it is shared then with the World Health Organization. Use of EHR data can help to streamline the data flow between health care, public health departments and the CDC. The CDC has launched a modernization initiative to enhance the surveillance systems ability to provide more comprehensive, timely and higher quality data to inform public health decisions. You'll learn more about that in a CDC video that you'll watch after you conclude this module.
Slide 15: Government and Social Service Data:
Government and social service data can nicely complement data from electronic health records to understand how social determinants of health connect to health metrics like infant mortality. Population density, poverty rates, air and water quality metrics and violent crime rates are just some examples of government and social service data that can be useful for population health surveillance.
Slide 16: Geographic Information Systems (GIS):
A third type of population health data includes geographic information systems or GIS data. GIS refers to a computer-based system used to collect, edit, integrate, visualize, and analyze
spatially reference data. GIS data can provide information on how health conditions are geographically distributed. One of the earliest examples of how diseases were tracked by place comes from Jon Snow in London. During the mid-19th century in Broadstreet, London, Snow investigated a cholera outbreak by creating a spot map of cases and deaths. He noticed that most of the deaths clustered around one water pump indicated here by the black arrow, the source of the contaminated water supply. The map image on this slide is a modern day representation of Snow's map using GIS data created by Dr. Robin Wilson. In population health, GIS is used to improve our understanding of relationships between health and place, which is important since we often observe disparities in health care access and health outcomes based on place as discussed during the social determinants of health module. GIS can integrate non spatial data like age, sex and the use of social services to better understand how health varies by geographical location depending on other factors.
Slide 17: 4 Main Uses of Geographic Information System (GIS) Data:
There are four main uses of GIS in population health surveillance. One, disease surveillance where GIS helps understand where diseases spread and how the spread can be stopped. Two, in risk analysis. GIS helps identify whether environmental and or geographical factors impact a population's disease risk, which can help justify implementations of interventions and policies. Three, health care access and planning. GIS can help us examine a population's ability to access healthcare. And four, community health profiling. GIS can help us understand the health of a population by identifying geographically related strengths and weaknesses. The information can inform strategies to address the community's weaknesses and use strengths to improve the population's health.
Slide 18: Example: Coronary Heart Disease (CHD) Death Rate by County, Maine
Here's an example of how GIS data can be used for population health surveillance. In this example, the coronary heart disease death rate has been identified in each county. In Maine, GIS data helped to create this chloroplasts map to display visually how the death rate differs depending on the county. In the map, the darker shades indicate a higher death rate. Counties such as Somerset County that include the diagonal lines have a significantly higher death rate than the main average, while those with a dotted pattern such as York have a significantly lower death rate. With this map, we have a visual representation to easily identify areas with high burden and need, areas that could benefit from intervention or policy implementation.
This map could also include the distribution of intervention sites, local policies, health care resources or other related information which can be particularly useful for tracking the ability of population health efforts to reduce disease incidence over time. These chloroplasts maps are created using a software program called ArcGIS and can be used to facilitate dissemination and communication of surveillance information.
Slide 29: Other Digital Data Sources:
Lastly, other digital data sources are increasingly being integrated into population health surveillance to provide unique and potentially more timely surveillance information, particularly when that data can be connected to location data. Other digital data sources
include things like social media sites like Twitter and Facebook. Here's an example of how Twitter could be used to identify flu outbreaks. You'll notice that the tweet with a rectangle around it highlights a potential concern about use of social media that it may be an unreliable source.
Internet search engines like Google and Bing are another digital data source. For example, Internet searches for things like vomiting and diarrhea could be used to help identify food poisoning outbreaks.
Mobile phone and mobile app data are a third type of digital data source. Mobile phone data can be used to track movement via the phone's GPS, while mobile apps can be used for tracking specific health behaviors like physical activity. Mobile phone data could be connected to electronic health records to examine health behaviors like medication adherence or smoking cessation in a health systems population.
Slide 30: Example: Use of Twitter to Track and Measure H1N1 Disease Activity:
As an example, this study used Twitter data to track H1N1 disease activity. To do so, they identified relevant search terms like flu, swine, and H1N1. Tweets for timestamped and geocoded to identify the time the tweet was made and the location from which the tweet was made. In the figure, you'll see an example of a tweet and the location from which it came. All the red dots on this figure refer to instances when the H1N1 tweets occurred. Then they use complex modeling to estimate H1N1 disease activity in real time based on the identified tweets and compared their data to CDC data of reported H1N1 cases. In their analysis, they were able to estimate cases one to two weeks earlier than the CDC's method for tracking cases.
Slide 31: Pros and Cons of Other Types of Digital Data:
As you just heard about, there certainly are some notable pros of using other types of digital data. The timeliness of the data suggests that disease outbreaks could be identified more quickly. These sources of digital data can provide unique data on things like attitudes and health behaviors in real time, as opposed to collecting this data via an annual survey, which is typically how this type of data is collected. Since digital data is often coded, it can be integrated with this data. Another pro of using other sources of digital data is the citizen science potential for improving population health. Citizen science enables residents to actively contribute to all aspects of population health. From conceptualization of a problem and data collection to knowledge, translation, and evaluation. If community members are aware of how their digital data may be used to improve the population's health, they may be more likely to use it to communicate relevant health information. In addition to collecting relevant, unique, and timely data, these digital data sources can also be used as tools for population health dissemination, communication, and intervention.
Despite these pros, there are some potential cons that are worth considering. These sources may not at all be representative of the population of interest, particularly populations that are typically neglected or understudied. As pointed out with the Twitter H1N1 example, the
validity and reliability of these data sources may also be concerning. Another con is the need for big data expertise to effectively use this data. Public health workers need expertise in areas like the analysis of big data, machine learning, and natural language processing to fully take advantage of these types of digital data. There's not currently any systematic integration of other types of digital data with public health agencies, which is another con, and there may be concerns about privacy and ethics, since when individuals use these platforms, they may not expect that their data is going to be used for population health surveillance. Lastly, with these data sources, you are relying on private businesses that may decide to prevent sharing their data. These businesses could also have underlying algorithms that may affect your data and make it less useful. As the population health field advances their ability to integrate multiple sources of data for surveillance, it will be worth considering these pros and cons to make decisions about whether and how other data sources can be best be utilized.
Slide 32: Reflection Question #1:
Let's pause and take a moment to consider what types of surveillance data may be relevant to your work. If you need more time to consider this question, go ahead and pause the video at this time. If not, let's move on to some possible examples.
Slide 33: Reflection Question #1- ANSWERS:
So here's an example of what surveillance data could be relevant to your work. Consider someone who's focused on preventing communicable diseases in pediatric populations. Some possible surveillance data sources could include electronic health record data on immunizations, assessments of growth and development, emergency department visits, government and social service data on household income, population density, environmental contaminants, and again, these data would ideally be connected to place via GIS. GIS data connected to school districts, health systems, would also be useful for identifying at-risk locations and health care access. One novel source of data could be Internet search data on the rotavirus vaccine. This is certainly not an exhaustive list, but just a sampling of how a variety of data can be used to understand population health.
Slide 34: Objective 3:
Once you've collected population health data, the next step is analyzing it, so I'm going to shift to introducing some of the principal considerations and methods for analyzing population health surveillance data. Please note that this is not an exhaustive list.
Slide 35: Common Types of Epidemiological Data:
Before we discuss analyzing population health data, I want to share a few basic epidemiological terms that you may have heard or read about. You may see epidemiological data presented as counts, percentages or rates. While all of these can be important and useful, they are different. Counts are raw numbers of cases that are presented. Counts may be useful for understanding what is happening within a specific population, for example, counts can be used to examine a local measles outbreak. Perhaps there are two students one week at an elementary school that are reported to have measles. By the following week, that number of students at that school
with measles increased to 18. Counts are not necessarily as helpful in trying to make comparisons about different communities as it does not account for other factors, such as population size, that could affect a different difference in the number of cases. For example, comparing a raw count of measles cases in Chicago to a raw count of cases in Kansas City may not be as useful a comparison with other types of data we'll talk about shortly. Percentages may be more helpful when making comparisons. As you can directly compare a percentage to another percentage, as long as you are looking at the same data point. For example, from a state level survey, you could compare the percentages of adults in each county that report 30 minutes of exercise per day. A third type of epidemiological data used regularly is rates. Rates are the number of cases per a specific population size. Usually, you will see this reflected as per one thousand individuals or per one hundred thousand individuals. Using rates is helpful because it helps correct for differences in population size. So using rates, you could compare the rates of measles cases in Chicago per 1000 residents to the rates of measles cases in Kansas City per 1000 residents.
Slide 36: Data Analysis:
There are three data analysis topics I will discuss in the section. Two main ways to organize population health surveillance include descriptive epidemiology and analytic epidemiology. I will also introduce you to big data analysis, which is increasingly being used in population health to take advantage of multiple and varied sources of data to understand how to improve the population's health. Keep in mind that with all data analysis, the quality of the data is critical to examining research questions. Without high quality data, your results may be misleading, no matter how complex the analysis.
Slide 37: Descriptive Epidemiology:
Descriptive epidemiology provides the foundation for population health surveillance. With descriptive epidemiology, public health workers can evaluate and compare trends for specific health issues, which can provide a basis for planning, provision of resources, and regular evaluation of that health issue. Descriptive epidemiology can also identify potential problems that could be examined and analytic epidemiology studies. Two other terms you may often hear are incidence and prevalence. Incidence refers to new counts or rates. For example, the incidents count or the number of new cases of COVID-19 are reported daily at the state level. Prevalence, on the other hand, refers to the current number of cases for COVID-19. Prevalence would refer to the active number of COVID-19 cases within the state on any given day.
Prevalence includes existing and new cases. Prevalence provides us with an idea of how prevalent a health issue is within a community at any point in time. Consider how surveillance information is often analyzed by time, place, and person. In the next couple of slides, they'll go over what we mean by each of these. A key benefit of descriptive epidemiology is that it is a straightforward approach such that the public health workers likely won't need extensive statistical training and expertise or statistical packages to conduct analyzes. One main limitation is that because it is a descriptive technique, what causes change in trends cannot be identified.
Slide 38: Characteristics of Persons:
Population health is particularly interested in focusing on characteristics of persons that might be vulnerable to worse health outcomes, since improving the health of the most vulnerable may have the largest impact on the overall population's health. Examples of characteristics include things like age, sex, marital status, race and ethnicity, activity and migration, religion, and socioeconomic status. Characteristics of persons can be used to define a population of interest. For example, you could track infant mortality rates in a population of interest. Or within a given population of interest, characteristics of persons can be used to identify which individuals may be experiencing poor health. For example, within a specific health care system. You could evaluate whether lower income patients are at greater risk for high blood pressure as illustrated in the table on the right.
Slide 39: Characteristics of Place:
Place is a determinant of health and well-being, which is one of the reasons why GIS data can be quite useful in population health surveillance. Pace can be characterized in a variety of ways. International comparisons can be made by country or continent. Within a country, comparisons can be made by region, state and or county. Smaller comparisons could include comparing cities, neighborhoods, or census tract. In the United States, urban and rural differences in disease rates are often of interest. Standard metropolitan statistical areas were established by the US Bureau of the Census to make regional and urban, rural comparisons and disease rates. Data analysis by places is a typical method used to examine where cases were reported or ideally where the illness occurred. For example, identifying what states are experiencing a measles outbreak as illustrated in the map on the right. Data analysis by place allows resources to be directed to where the exposure occurred.
Slide 40: Characteristics of Time:
The third consideration for descriptive epidemiology is time. The occurrence of disease changes over time. Some of these changes occur regularly, while others are unpredictable. Displaying the patterns of disease occurrence by time is critical for monitoring disease occurrence in a population and for assessing whether the public health interventions made a difference. Analysis by time can be by the minute, by days, months, years, or other specific ranges of time. Disease occurrence can be graphed over the course of a year or more to examine whether there is a seasonal pattern. Some diseases such as influenza and West Nile infection, have characteristic seasonal distributions. The figure on the right illustrates the percentage of health care visits for influenza-like illness over the course of a year. Each line represents a different year. Notice how each line consistently shows how visits are higher in the first few months of the year during the winter and lower between weeks 20-38, which corresponds to spring and summer. Seasonal patterns may suggest hypotheses about how the infection is transmitted, what behavioral factors, increased risk, and other possible contributors to the disease or condition. Point epidemics are another way to characterize time. Point epidemics refer to the response of a group of people circumscribed in place and time to a common source of infection, contamination, or other etiological factor to which they were exposed almost simultaneously. Examples include food borne illnesses and infectious diseases.
Lastly, secular or long-term trends are helpful for understanding population health. Graphing the annual cases or rate of a disease over a period of years shows long term or secular trends in the occurrence of the disease. Health officials can use these graphs to assess the prevailing direction of disease occurrence, increasing, decreasing or essentially flat, which can help them evaluate programs or make policy decisions to infer what caused an increase or decrease in the occurrence of a disease and to use past trends as a predictor of future incidence of disease. An example could be graphing obesity prevalence over time in the United States.
Slide 41: Analytic Epidemiology:
The second type of epidemiology used in population health is analytic epidemiology, which is used to test specific hypotheses. Analytic epidemiology goes beyond describing the incidence and prevalence of a disease to understand causal relationships between exposures and disease outcomes. There are two main types of studies, observational and experimental.
Slide 42: Observational Studies:
With observational studies, no manipulation is involved to investigate the relationship between a cause and effect. Cohort studies are one type of observational study. With a cohort study, you follow individuals over time to understand disease risk and protective factors. An example of a cohort study is the Black Women's Health Study, which enrolled 59,000 black women in 1995 and asked them to complete questionnaires every two years to understand the causes of illness such as breast cancer, hypertension, and diabetes in black women. The questionnaires assess risk and protective factors like experiences of racism, physical activity, sleep, and depressive symptoms. Case control studies are another type of observational study. With case control studies, cases are individuals who have a disease of interest while controls do not have the disease. Analyses are conducted to examine whether disease risk factors are associated with having the disease or not. An example of a case control study is a study of asthma exacerbation in a pediatric Medicaid population from Detroit, Michigan. Using data extracted from Medicaid claims data for a three-year period, cases were identified as all children who made at least one asthma claim. Cases were then matched controls who are randomly selected from the rest of the Medicaid population. They then examined whether home proximity to major roads was associated with the asthma exacerbation.
Slide 43: Experimental Studies:
Unlike observational studies, experimental studies involve some degree of manipulation. Most commonly these studies are used to examine the impact of an intervention or policy on health. An example is a multilevel intervention to increase human papilloma virus vaccine or HPV uptake among adolescent girls. As indicated on the figure on the right, the intervention aimed to increase vaccine uptake at three different levels. First, the clinic level where HPV vaccine education materials were visible. Second, the provider level in which providers received a presentation on evidence-based information and strategies for vaccine discussions with parents to encourage provider conversations with parents. And lastly, the parent level such that parents received a brochure and DVD about HPV in the vaccine, in addition to telephone session to reinforce HPV education and address any barriers. In this example, the manipulation
involved to receive the multilevel intervention. 12 counties in Ohio were randomly assigned to either receive the intervention or the control condition.
Slide 44: Use of Big Data for Population Health Surveillance:
Use of big data is increasingly common in population health surveillance. Big data involves the use of large datasets, often incorporating data from multiple sources such as GIS, social media, and electronic health records to understand the population's health. Because of the size and scope of the data, machine learning may be necessary for analysis. Machine learning refers to the techniques that fit models algorithmically by adapting to patterns in the data. One pro for using big data for population health surveillance is that you may improve the accuracy of disease prediction, which could improve the population's health. Big data may also enable faster identification of at-risk populations, which could then be targeted for intervention and provide a faster evaluation of progress to understand whether given intervention or policy is having the desired effect. One main concern is that big data analysis requires advanced training and statistics, computational power and use of statistical software. Another con is that the data you may be using are often not designed to answer your specific question, so the outcome measurement could be less than ideal.
Slide 45: Example: Use of Wearables to Monitor Influenza-like Illness:
Here's an example of how big data could be used in population health surveillance. Radin and colleagues evaluated whether the use of wearable technologies, like the Fitbit watch pictured here, could be used to monitor the seasonal trends observed in influenza-like illness, since influenza can elevate resting heart rate and alter daily activities. They examined whether wearable devices that collect resting heart rate and sleep data could identify influenza-like illness rates in users from five states, which provided over 13 million measurements. Results indicated that including data from wearable devices significantly improved their ability to predict influenza like illness. This study suggests that wearable data could improve objective real time estimates of influenza-like illness, which could hasten responses to suspected outbreaks and potentially prevent further spread.
Slide 46: Objective 4:
While there are a lot of great ways to conduct population health, surveillance, improvements can certainly be made. Now we'll discuss ways to improve population health surveillance via data modernization.
Slide 47: Variability in Population Health Surveillance:
One challenge in population health surveillance is the tremendous variability that exists historically. Public health departments rely on health care providers to submit case reports relevant to disease surveillance. Yet only 20-30 % of infectious disease cases may actually be reported due to barriers in the flow of information between health care systems and public health departments. The variability observed across state and local health jurisdictions in how and when cases are reported is one significant barrier. Antiquated processes for reporting cases is another barrier. In some places, providers may be required to fax or mail a form to the
health department, which places undue burden on the provider. In other circumstances, the reporting procedures in place may not contain enough information to be useful for understanding the disease outbreak or adequately track the population's health.
Slide 48: A Proposed Solution: Digital Bridge:
Digital Bridge offers a potential solution to address surveillance barriers. In 2009, the American Recovery and Reinvestment Act incentivized health care systems to transition to electronic health records, which improved their ability to manage patients and their health care information. The move to electronic health records offered the possibility of improving surveillance and eliminating onerous processes for sharing relevant information with public health agencies. Unfortunately, most our systems have limited functionality to exchange data with public health agencies. Digital Bridge is designed to address that problem. Digital Bridge represents a partnership between public and private organizations to automate the exchange of electronic health information between health care systems and public health departments. Public health agencies, health care systems and electronic health record companies are collaborating to identify how use of electronic health records can facilitate reporting of cases.
Slide 49: Digital Bridge: How it Works:
Here is how Digital Bridge works. A key component of the Digital Bridge project is the decision support intermediary represented here by the circle in the middle to facilitate reporting. Cases identified in the health care providers electronic health care record are automatically reported to the decision support intermediary and evaluated against public health reporting criteria that was developed by the Council of State and Territorial Epidemiologists. This ensures that the criteria used to identify cases in the health care record and whether they should be reported to the public health agency are nationally consistent, which improves the reliability of surveillance measurement. When the decision support intermediary receives a case report from the health care provider, the intermediary also uses national standards to discern whether the case should be reported to the relevant public health agencies. If a report is justified, the decision support intermediary then automatically generates a report that is sent to the designated public health agencies. Use of the digital bread streamlines the reporting of disease outbreaks while significantly decreasing the work required of health care and public health workers to identify cases and outbreaks. In doing so, errors decrease, which improves the quality and consistency of the data. Importantly, even with the intermediary reporting times to public health agencies decrease, which allows public health agencies to intervene more quickly and lessen the disease spread. Digital bridge pilot projects demonstrated the feasibility and benefits of this approach. Efforts are continuing to expand this digital bridge approach.
Slide 50: Population Health Record (PopHR), Community Health Record (CHR):
Data modernization also assists with the creation of population and community health records. A population health record was defined by Friedman and Parrish as a repository of statistics, measures, and indicators regarding the state of and influences on the health of a defined population. The record exists in a computer processable form, stored, and transmitted securely and accessible by multiple authorized users. The repository facilitates population health by
allowing authorized users to examine multiple factors impacting health within a defined population in more complex ways. In doing so, population health records are designed to support surveillance and other aspects of population health practice, such as conducting assessments, identifying population health disparities, and designing, implementing, and evaluating population health interventions and policies. A similar term has been used a community health record which has been defined by King and colleagues as a framework to guide health care, public health and community collaboration and information exchange, and as a tool for integrating and transforming multi-sector data into information that can aid decision makers.
Slide 51: The Community Health Record Framework
Here's a figure illustrating the community health record framework. The framework represents a multitiered, multi-sector model to create a community health record that improves population health at the foundation. A collaboration exists between the community health care and public health sectors represented by the three triangles outside the pyramid. This collaboration supports the exchange of information between the three sectors connected to a common agenda. Data modernization facilitates data exchange between the three sectors.
Data exchange in a community health record enables examining health information in specific populations, such as by a level of the census block neighborhood or zip code. The ultimate goal of sharing data and information is to inform, target and evaluate evidence, informed interventions that are collectively implemented by community health care and public health sectors, ultimately to improve population health. Note that the creation of a community health record is an iterative, flexible and participatory process, meaning that if the interventions do not have the desired effect or the health targets change, the three sectors can revisit the data and revise their approach.
Slide 52: Reflection Question #2:
Take a minute to consider the ways in which your workplace could improve population health surveillance. If you need more time, go ahead and stop the video now.
Slide 53: Reflection Question #2 – ANSWERS:
Here are just some possible ways your workplace might improve population health surveillance. You could improve data sharing capacity via steps like adherence to standard protocols, collaboration with an intermediary who would transmit the data between the health care systems and public health, rather than relying on direct submission with improvements in data sharing, more training, and data management data security. Data analysis would certainly be necessary, which is another way to improve population health surveillance. You can also improve surveillance by greater variety in the data collected. You can expand upon the types of data collected to include more novel sources of data like social media data or web searches.
Slide 54: Objective 5:
The final objective of this model highlights strategies for effectively communicating population health surveillance data.
Slide 55: Communicating about Population Health Surveillance:
In the population health assessment module, you learned about the importance of communicating the results of an assessment. The same is true of communicating about population health surveillance. An important function of surveillance is not just the gathering of data, but to effectively present the data to provide a basic understanding that can be used to make health care decisions. Ideally, the representation of surveillance data will be interactive so that users can examine it in a way that is most useful to them. For example, a state public health department might provide statewide data on diabetes incidence. A county public health department in that state might want the capacity to compare their county's diabetes incidence to nearby or similar counties. They also might want to know how diabetes incidence varies within their county based on certain demographic variables like age or gender identity. Any data representation should provide some flexibility to enable users to interact with the data in a meaningful way. Surveillance data should also be communicated in real time so that it is meaningful and facilitates action. Multiple modalities for communicating surveillance information are beneficial. Different platforms and ways of communicating information should be modified based on the target audience so that the information is accessible. Address accessibility broadly by considering issues such as visual impairments, health literacy, learning styles, and whether the tech platform can be accessed by other devices. Population health surveillance communication should include information on data quality. The relevance of the information to different populations should be clear to facilitate drawing attention to the information relevance also includes highlighting the representativeness of the data. As I discussed earlier, surveillance data are not always consistently collected, so any problems or issues with data integrity should be made clear. Communication about the variety of data available is certainly useful at a minimum population health surveillance data should include a connection to social determinants of health. Greater variety of data types provides a more complete picture of the population's health.
Slide 56: Data Visualization (data viz):
Data visualization refers to creating a visual representation of data, which provides an effective way to communicate about population health surveillance data. There are three core aspects of data visualization. The first states that data visualization includes qualitative and quantitative data. A second aspect is that the created image represents the raw data. The final aspect, which is true of all population health surveillance, highlights how any data visualization should be readable by viewers and supports exploration, examination, and communication of the data.
Slide 57: Best Practices for Data Visualization:
Effective data visualization should tell a story about your data in a clear and easily understood manner, data representation should be simplified to focus on the key findings. Consider using meaningful colors such that one color represents the same outcome or data type. Selected color should also be color blind friendly. The labels, you should make it easy to understand the data, avoid use of jargon and technical language which could confuse your audience. Data
should also be ordered in a way that facilitates understanding order data consistently and intuitively by categorizing alphabetically, sequentially or by value.
Slide 58: Example: RiskScape:
Here's an example of using data visualization to communicate population health surveillance data. RiskScape was developed by the Therapeutics Research and Infectious Disease Epidemiology Group at Harvard Medical School and Harvard Pilgrim Health Care Institute and the Massachusetts Department of Public Health. RiskScape is an open source, interactive, Web based, user friendly data aggregation and visualization platform for public health surveillance. Using EHR data. On the dashboard depicted on the right notes of several different health conditions are examined in real time. This tool allows the user to examine health conditions by demographic or comorbidity categories.
RiskScape also can create heat maps of conditions by zip code to visually identify places with higher disease prevalence.
Slide 59: Strategies for Rigorous Population Health Surveillance:
In closing, consider the following strategies to facilitate rigorous population health surveillance. Involvement of community members is a common population health strategy. Population health surveillance depends on community members trust in their health care system. If they are not accessing the health care system, then the ability of the system to monitor and track diseases is compromised. This is particularly important for underserved and minority populations who may have significant barriers to accessing care. Organizations need to make a long term financial and political commitment to surveillance to ensure that it will improve population health. Surveillance efforts require a strategic plan to enable thoughtful identification of population health priorities for surveillance. Any plan should build in adaptation to the changing needs of the population. High functioning and interconnected systems are necessary to take advantage of technological improvements that can improve population health surveillance. Routine evaluation and innovation is also key surveillance data should be regularly evaluated to enhance population health surveillance. Regular evaluation can identify new variables that should be included in models to identify health disparities and improve algorithms. One good example of innovation refers to how intersectionality should be considered in population health. Intersectionality refers to the way in which structural factors, social and historical processes and systems of power and oppression can interact to influence health disparities. Finally, workforce training and data management of complex systems and analysis of complex data is often necessary to improve population health surveillance.
Slide 60: General Surveillance Resources:
I'll leave you with some resources that you can check out for more information on population health surveillance. All of these are available for download from the training website. The first slide contains general resources for conducting population health surveillance.
Slide 61: Data Analysis Tools:
To bolster your understanding of analyzing surveillance data, consider checking out these informatics and data analysts, says resources.
Slide 62: Data Repositories:
Here are several data repository resources that can provide information at the federal, state and even city level.
Slide 63: GIS Data Resources:
And here are a few geographic information systems, resources.
Slide 64: Communication Resources:
And resources to assist with communicating about surveillance data.
Slide 65-end: References:
Finally, I want to acknowledge the sources that were used to create this presentation. And I want to thank you for your attention.