Tutorial Module 4: How Are Public Health Data Used?

While it is critically important to know what data exist (see Module 3, “What are Public Health Data” for more information on data sources), it is also important to know how best to use these data to answer important questions about public health. This module will examine how public health data are used, as well as the challenges faced when using public health data.

Data Users

The use of public health data varies depending upon the type of data user. Decision makers (e.g., senior level governmental officials in health and human services agencies) use public health data to aid in making decisions about public policy, cross-program initiatives, and funding priorities related to public health. Funders use public health data to determine need, health priorities, and the benefits of investing in programs; to “fine-tune” the management of funded programs; and to assess the results following implementation. Practitioners, analysts, and advocates use data to research, plan, operate, and/or evaluate programs. These are just a few of those who use public health data in their work.

Data Flow

When using public health data or accessing a public health data system, it is also critical to understand the flow of data, i.e., how data are collected, stored and communicated throughout the public health system. By documenting the process of data flow, users can better interpret raw data as well as identify areas for improvement in data systems.

In a surveillance system, for example, the flow of data begins when the patient/person seeks treatment, either due to illness or as a well check-up. If an occurrence of a health-related event is confirmed (note: some surveillance systems may record “near misses” that meet a threshold criteria – making early warnings possible), the information is recorded by a health professional (e.g., physician, hospital, or clinic staff) or directly through a computerized information system (e.g., lab results). This may be done in paper or electronic format, depending on the particular system being used. Regardless of the method of data collection, confidentiality must be assured.

The data are then transferred to the appropriate public health entity for reporting purposes. In some cases, this may only mean the local health department. For certain cases, these data are transferred through the local department to the state or even federal public health agencies. Once the appropriate public health entity has analyzed the data, it will take any necessary action, e.g., interviewing the health professional who confirmed the event, or alerting a broader audience about the event.

Exhibit 1 shows how data flow works in a generic surveillance system.

Exhibit 1: Generic Surveillance System Simplified Flow Chart Source: Centers for Disease Control and Prevention, Morbidity and Mortality Weekly Report (July 27, 2001). Updated Guidelines for Evaluating Public Health Surveillance Systems: Recommendations from the Guidelines Working Group [On-line], Available: http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5013a1.htm.

The Centers for Disease Control and Prevention (CDC) has made recent advancements in the area of surveillance systems by applying data standards. It developed Disease Surveillance System (NEDSS), an “initiative that promotes the use of data and information system standards to advance the development of efficient, integrated, and interoperable surveillance systems at federal, state and local levels.” “The vision of NEDSS is to have integrated surveillance systems that can transfer appropriate public health, laboratory, and clinical data efficiently and securely over the Internet.”(1) (See Module 5 for more information about data standards for public health and Module 7 for more information about NEDSS.)

The flow of health care service utilization or encounter data has also been influenced by uniform national standards. Encounter data flow either electronically or through paper submission from providers to insurers and governmental agencies to trigger insurance claims as well as to trigger the reporting of most public health data. The Health Insurance Portability and Accountability Act (HIPAA) of 1996 mandates that the United States adopt uniform national standards for electronic transactions related to health care encounters, among other administrative data. While public health is not mandated to adopt these standards, except as payers or providers of health services, the Public Health Data Standards Consortium member organizations agreed that it made good sense for public health to adopt the same standards as named in the HIPAA legislation, where appropriate. The Consortium developed the Health Care Services Data Reporting Guide to facilitate public health's ability to communicate with care delivery systems, especially to address data needs that rely on encounter data, e.g., specifications for data elements such as race, ethnicity, geographic location, and a limited set of clinical data elements.

Data Definitions for Health Indicators

Often, examining raw data figures will not provide meaningful information. Data definitions for health indicators turn raw data into information. For example, the number of infants who died in a particular state does not provide insights into the magnitude of the problem of infant deaths in that area. However, when used with other data, such as the total number of infants born in a given year for that state, the health statistic called infant mortality rate facilitates an important comparison of events (number of deaths less than one year old per 1,000 live births in a population). See Exhibit 2 for commonly used data definitions or standards for health indicators.

Exhibit 2: Data Standards for Health Indicators

The indicators above transform raw data about individuals to aggregate data about specific populations.

There are a number of various rates that are used often in public health. Exhibit 3 shows a number of these common rates.

Exhibit 3: Commonly Used Rates in Public Health

Additional rates used to measure the health status of a population are the prevalence and incidence of disease. Prevalence is “the proportion of a population that has disease at a specific time.”(2) Incidence is the number of new cases of disease during a prescribed period of time. For example, understanding the cumulative rate of all of the people with HIV/AIDS helps decision-makers determine demand for treatment and services; understanding the rate of new cases informs targeted outreach and prevention education.

Using Public Health Data to Measure Healthy People 2010 Leading Health Indicators

“Healthy People 2010 is a set of health objectives for the nation to achieve over the first decade of the new century. It can be used by many different people, states, communities, professional organizations, and others to help them develop programs to improve health.”(3) Healthy People 2010 has two primary goals: to help individuals of all ages increase life expectancy and improve their quality of life; and to eliminate health disparities among different segments of the population.(4)

A number of organizations are working toward these health objectives. Among them are the Secretary’s Council (an advisory group to the Secretary of Health and Human Services regarding Healthy People), the federal agencies responsible for the different objectives of Healthy People, and the Healthy People Consortium (a group of 400 national membership organizations, all State and Territorial health departments, and key national associations of State health officials working to advance health).(5)

Healthy People 2010 uses 10 indicators to measure the health of the Nation over the next 10 years. Each of the 10 Leading Health Indicators has one or more objectives from Healthy People 2010 associated with it. The Leading Health Indicators are:

1. Physical Activity;

2. Overweight and Obesity;

3. Tobacco Use;

4. Substance Abuse;

5. Responsible Sexual Behavior;

6. Mental Health;

7. Injury and Violence;

8. Environmental Quality;

9. Immunization; and

10. Access to Health Care.(6)

The following is an example of how data are used to determine the status of a key Healthy People 2010 indicator.

Immunization

One of the objectives for the leading health indicator of immunizations is to “increase the proportion of young children who receive all vaccines that have been recommended for universal administration for at least five years”(7) To determine how well a particular state is meeting this goal there are several pieces of data necessary. First, there needs to be an accurate count of children. Although the census provides much more accurate counts of the population, the fact that it occurs only once every ten years prevents it from being accurate from year to year. In this case, one of the best sources of data is the Current Population Survey (CPS). The U.S. Census Bureau conducts a survey of people throughout the country (the survey includes over 10,000 individuals nationwide, with oversampling in specific areas). Once the survey is complete, the numbers are given to statisticians at the Bureau to develop state and county estimates of population, including age. The survey also provides demographic information, including race and income.

Next, the current number of immunizations must be determined. Many states maintain immunization registries. In these cases, physicians and public health clinics submit records to state departments of health when children receive immunizations. This information, including the type of immunizations received, is aggregated by the state departments of health and made available. The data from these two sources can be used in many ways. First, comparing the estimates of children from the CPS with the immunization registries can provide an approximation of the percentage of children who are receiving immunizations. More specifically, the proportion of children receiving specific immunizations can be found. These pieces of information can be cross-referenced with socio-economic data found in the CPS as well, helping to determine whether certain characteristics like race are statistically significant. Longitudinal studies can also examine some of the statistics listed by examining data from past years. This might be particularly useful in comparing this information with the start dates of programs launched to improve immunization rates.

Constraints and Challenges in Data Use

Although the use of public health data can provide critical information for decision making purposes, there are often numerous constraints to obtain and use these data. Below are a number of challenges faced by various data users.

Silos of Data

Typically, public health data sources are independent of one another, making it difficult to link information. These silos of data are the result of categorical funding, which provides resources for individual data source collection, but fails to appreciate integration of multiple data sources. As a result, many data users are unable to effectively use data sources together.

Inconsistent Data Collection

There are several types of problems that arise when collecting data. For example, public health data can originate from multiple providers, including physician offices, clinics, laboratories, and hospitals. Each provider may use slightly different practices (voluntarily or by mandate) for documenting service provision or definitions for characterizing a patient. Data coming into a central source may not be comparable, e.g., the codes used for the race and ethnicity of a patient are not the same.

Data may also be incomplete, necessitating follow-up and further time delays. Providers of care see their main responsibility as patient care, not data collection. As a result, they may neglect to collect certain public health data. In addition, much of the data collection in a clinical setting is done initially by hand. Later, the information is transferred into an electronic format. Two problems result by using this system: (1) data are not readily available after it is initially recorded; and (2) errors can occur when transferring the information.

Finally, it is often the case that data are lagged in their release (e.g., national survey data or census data) making it difficult to rely on them for real-time analyses. In addition, data are not collected yearly, which limits longitudinal analysis.

Regulatory Constraints to Data Use

Using health data is extremely important for officials to make policy decisions about the public’s health. However, there are some constraints in place to protect individuals. The Health Insurance Portability and Accountability Act of 1996 (HIPAA) calls for privacy standards to protect identifiable data when used by providers, insurers, and other “covered entities.” In most cases, public health agencies are exempt (i.e., not a covered entity unless a provider or insurer) from HIPAA privacy standards but may be covered by other state and federal regulations. The HIPAA Privacy Rule recognizes the legitimate need for public health authorities and others responsible for ensuring public health and safety to have access to protected health information to carry out their public health mission. Accordingly, the Rule permits covered entities to disclose protected health information without authorization for specified public health purposes. However, when reporting of health data is not mandated by state law or regulation, public health agencies may encounter greater reluctance from providers to share data.

The security of data is also critical, for both privacy purposes and to protect the integrity of the data. Consequently, electronic transfers of data, particularly patient records, must be secure. Encryption and passwords and other policies required under HIPAA will help to maintain the necessary level of security.

Case for Data Standards

Many of the challenges and constraints detailed above reflect the same inherent problem: a lack of standards. In the remaining modules, public health data standards will be discussed and information will be provided showing how the use of such data standards might alleviate some of these problems.

Internet References

The following are links to other sources of information regarding how public health data is used.

• CDC-PulseNet

• Health Services Research Projects in Progress (HSRProj) - A joint effort of Academy Health and the Cecil G. Sheps Center for Health Services Research, University of North Carolina at Chapel Hill)

• Healthy People 2010 Information Access Project

• Massachusetts Health Data Consortium

• National Association for Public Health Statistics and Information Systems

Endnotes

(1) Centers for Disease Control and Prevention (February 6, 2003) National Electronic Disease Surveillance System [On-line], Available: http://www.cdc.gov/nedss/
(2) Rothman, Kenneth J. and Sander Greenland (1998). Modern Epidemiology. pp. 42.
(3) Department of Health and Human Services (February 6, 2003). Healthy People 2010: What is Healthy People [On-line], Available: http://www.healthypeople.gov/About/whatis.htm
(4) Department of Health and Human Services (February 6, 2003). Healthy People 2010: What Are Its Goals [On-line], Available: http://www.healthypeople.gov/About/goals.htm
(5) Department of Health and Human Services (February 6, 2003). Healthy People 2010: Implementation [On-line], Available: http://www.healthypeople.gov/Implementation/default.htm#partners
(6) Department of Health and Human Services (February 6, 2003). Healthy People 2010: Leading Health Indicators [On-line], Available: http://www.healthypeople.gov/LHI/lhiwhat.htm
(7) Department of Health and Human Services (November 2000). Healthy People 2010: Understanding and Improving Health. Volume I. pp. 42.