Health Forecasting at UCLA

October 2013

In This Edition:



Download a PDF Version of the Newsletter

Health Forecasting Quarterly Newsletter •  October 2013

Established over a decade ago at the UCLA Fielding School of Public Health, Health Forecasting aims to provide evidence-based, forward-looking health forecasts to support health advocates, researchers, elected and government officials in their efforts to anticipate, prepare, and plan for healthier communities.


The Center for Health Advancement receives a grant from Kaiser Permanente

The Center has received a grant from the Kaiser Permanente to apply cutting-edge statistical methods to generate robust statistics for small, local areas, such as zip codes. In much of California health statistics for are available only at the county-level. Even in Los Angeles County, where the large population, allows estimates to be made at the sub-county level, health risk and prevalence data are still available only at aggregations of about one million residents...a far cry from the neighborhood-level data that public health officials, community planners and community organizations would like to have.


 A wide variety of surveys track the prevalence and distribution of health risks and conditions in the population. Surveys, such as the  Behavioral Risk Factor Surveillance System (BRFSS), the National Health and Nutrition Examination Survey (NHANES) and the California Health Interview Survey (CHIS) provide invaluable information for identifying public health priorities and planning services. These surveys, however are costly and time-consuming. Constraints on their feasible sample size limit survey estimates for small areas (e.g. neighborhoods). The resulting data gap has been recognized by the State's Strategic Growth Council as a major priority for state planning efforts.[1]  


The Kaiser grant will help Center researchers to develop, apply, and evaluate statistical modeling methods for combining survey data from different points in time and varying geographic levels to project health outcomes at the community-level. Various small area estimation techniques have been around for over thirty years, but these new powerful techniques have the ability to draw data from multiple sources and time periods to generate more precise estimates for local area. Improvements in data storage and management as well as technological advancements in data standardization and sharing have allowed small area estimates to be reliably generated in the health arena.

[1] California Strategic Growth Council. 2009 Data Needs Survey


Health Forecasting Hosts Seminar for Inland Empire Nonprofit Hospitals

As part of the Center's Health Forecasting work with the UniHealth Foundation, in August Dr. Lu Shi and Research Analyst Peggy Vadillo Orenstein led the first of a series of seminars on using health forecasting in community benefit planning and reporting. Starting with San Bernardino in the first year of a three-year project that will expand to other areas of Southern California, the UCLA/UniHealth Health Forecasting project is partnering with non-profit hospitals, public health agencies and universities, to develop and apply health forecasting tools to community benefit planning.  


Attendees at the August seminar, which was hosted by Loma Linda University, included representatives from nonprofit hospitals throughout the Inland Empire and officials from the San Bernardino County Department of Public Health. The UCLA team outlined how health forecasting can be used for community health needs assessments (CHNA) and community benefit implementation plans, and led a hands-on demonstration on using the UCLA Health Forecasting Tool for getting long-term projections conditions, such as obesity and coronary heart disease for their local service area. They also solicited ideas from attendees on how to best build out the forecasting model for meeting the needs and emerging health issues addressed by hospitals serving San Bernardino County.


The UCLA/UniHealth Health Forecasting project will continue working with San Bernardino partners and in early 2014 will initiate a similar effort with partners in Orange County.  


The results of the multicounty simulation modeling will become available on the web-based interface called the Health Forecasting Tool. To explore the 'Los Angeles Community Benefits Study', go onto our website then click 'Get Started' on the navigation bar or click here.


If you need more help navigating the tool, download the Health Forecasting Toolkit.


Air Pollution & Community Health in South Kern County

Poor air quality has been linked to numerous adverse health outcomes including asthma, susceptibility to respiratory infections, like colds and flu, lung cancer, heart disease, stroke and premature death. Air pollution-related illness impacts school attendance, academic and work performance, and overall quality of life. In this final community profile of the three-part-series, we present our findings of the impact of air quality on health in South Kern County, one of the communities identified in The California Endowment's 10 year-initiative, Building Healthier Communities.


Community Health and Air Quality Profile-Part 3: South Kern County

Despite significant reductions in air pollution levels over the past few decades, Californians in many areas of the state

are still exposed to hazardous levels of air pollution. One area where air pollution remains high is South Kern County at the southern end of the San Joaquin Valley where ground-level ozone, a respiratory irritant linked to asthma, bronchitis, heart attack and premature death, persists at levels above state and federal environmental standards.[1]


Although South Kern County has experienced a 44% reduction in ground-level ozone between 2003-2009, the number of days that exceeded federally accepted standard in 2009 was 71 days or nearly 1/5 of an entire year. Poverty, which is widespread in South Kern, compounds the health effects of air pollution and is itself associated with a wide range of health risks.


Policies to control air emissions that contribute to these high levels of ground-level ozone can significantly benefit the health and well-being of residents of South Kern. Using computer-based models that incorporate air pollution data, population trends and known relationships between air pollution and different health outcomes, we estimated decreased death rates, decreased hospitalizations due to asthma and chronic lung disease, and fewer school absences.


Scenario 1: Ozone concentrations remain the same

Over the past decade ground-level ozone levels have been decreasing in South Kern County, but some portion of decline since 2008 is likely due to the economic recession. For this scenario we assumed that ozone levels will stabilize as gradual economic recovery offsets air pollution improvements resulting from past pollution control efforts. Thus, forecasts for 2010 through 2030 assume annual average maximum 8-hour ozone concentrations of 103.4 ppb, the same as the three-year average from 2007 through 2009.


Scenario 2: Ozone reductions meet federal standards by the year 2025

If the annual average ozone levels dropped from 103.4 ppb to the federally accepted standard of 75 by 2025, then the number of days classified as "hazardous" (8-hour ozone levels exceeding 375 ppb) would drop from 73 days to only 3 by 2025. This represents a decrease in the cumulative number of days with hazardous levels of ozone dropping from 1,570 days to 449 days over the period from 2010 through 2030.


The key findings of our simulation, shown in the box above, reveal a number of direct and indirect health benefits. The projected health benefits of reduced ozone translate into reduced medical care expenditures, improved quality of life, more financial support for local schools, and higher economic productivity for residents of South Kern County. For schools, increased attendance would increase allocations of state education funds that are calculated based on student attendance. Reductions in ground-level ozone could also produce economic benefits through increased crop yields.


[1] U.S. Environmental Protection Agency. (2012). Ground-level ozone health effects.


Latest Edition of Health Forecasting Toolkit Now Available

The toolkit is a helpful resource for anyone interested in generating health forecasts via the project's web-based interface called the Health Forecasting Tool. The innovative simulation model web interface allows a user to tailor health assessments in their community by zip code and overlay evidence-based interventions to determine the impact on health within their custom geographic area.  


The step-by-step guide includes information for users to:


  • become familiar with health forecasting and 

    understand the various applications to real-world scenarios;

  • login; 

  • navigate the tool;

  • learn about the three types of reports and their different functions; and

  • view a sample community health needs assessment case analysis and report.

Although the Toolkit was designed for hospital staff for use in developing their community health needs assessments and community benefit plans, anyone can use the Toolkit to assess health status, identify high-risk groups, and compare and target interventions across population subgroups. Whether you are a county health official, hospital community benefit staff member, or a nonprofit focused on reducing heart disease in your area, forecasting can inform effective strategies to improve health by revealing how health status and shifts in demographics may change in five, ten, or twenty years.    


The new edition includes a more expansive methodological section and elaborates upon the microsimulation modeling process. The figure shown above is a simplified illustration of the components involved in generating health forecasts. This section also enumerates input parameters for disease relative risk functions for health conditions and behaviors. There is an expanded list of sources we use to build the model and a data dictionary for topics available in the Health Forecasting Tool.  


The Health Forecasting Toolkit can be viewed on our website:


Publications and Presentations

Article Published:  

Cigarette Smoking and Abdominal Obesity: a Meta-analysis of observational studies. (2013). Journal of Substance Abuse


Lu Shi, Ruopeng An, and Jeroen van Meijgaard



Dr. Brian Cole, Program Manager of the Center's Health Impact Assessment (HIA) Project, participated in the 2nd Annual National Health Impact Assessment (HIA) Conference held September 24-26 in Washington, DC. The growing popularity of this sold-out conference, attended by over 350 participants from non-profits, government agencies, universities and businesses working in multiple sectors, reflects the surge of interest in HIA from grassroots agencies to state legislatures. The UCLA HIA Project with Drs. Cole and Fielding, along with Dr. Gerald Kominski and Hal Morgenstern (now at Michigan State University), was a pioneer of HIA in the U.S. Since the UCLA HIA Project researchers first began their work in 2001 the field has become a major component of what practitioners call "The New Public Health" that works across sectors and disciplinary boundaries to advance population health.


During this year's conference Dr. Cole presented early findings from two of the Center's current HIA projects focusing on urban water conservation1 and alternatives to California's state gas tax.2 He also moderated a panel on 'Analytic Tools & Data Sources' and facilitated a roundtable discussion focusing on Quality Assurance in HIA. A detailed conference agenda can be found at the HIA Conference website. More information about 'The New Public Health' is available on the Robert Wood Johnson website, HIA Conference Blog.   


1Funded by a grant from the  Pew Charitable Trusts' Health Impact Project 
2Funded by a grant from the Robert Wood Johnson Foundation.
*Health Impact Assessment is a partner project of Health Forecasting UCLA. Both projects are housed in the Center for Health Advancement at the Fielding School of Public Health.




Dr. Lu Shi & Donglan Zhang received an Honorable Mention for the Best Paper Award:  'An Agent-based Model to Evaluate Positive Externality of Tobacco Control Interventions: Reducing Passive Smoking at an Accelerating Rate' at The Chinese Economists Society  at (CES) 2013 Annual Conference in Chengdu, China.


Our Projects


The California Endowment

Expanding the capabilities of the UCLA Health Forecasting Tool by incorporating education and income, two critical social determinants of health, andi focusing on interventions relevant to underserved individuals and communities in California.




Kaiser Foundation

Efficiently combining multiple sources of survey data to improve local area estimates of health to improve local area estimates of health and forecasting outcomes for Los Angeles County.




National Institutes of Health

Forecasting and improving Latino health by examining the role of acculturation and physical activity

to account for health disparities among the Latino






Robert Wood Johnson Foundation

Incorporating additional risk factors (i.e. smoking) and disease outcomes (i.e. lung cancer) into the forecasting model and applying the model to other states, beginning with Arkansas and Wisconsin.




UniHealth Foundation

Supporting local not-for-profit hospitals in assessing current and future characteristics of the populations they serve and identifying long-term planning needs of local communities.  Providing information on future health and health disparities among subpopulations in the absence of additional effective health promotion and disease prevention efforts in Los Angeles, Orange, and San Bernardino Counties.




About Us

Health Forecasting is a collaborative effort with the California Department of Public Health and the Los Angeles County Department of Public Health.


Health Forecasting & Health Impact Assessment (HIA) are housed at the UCLA Fielding School of Public Health Center for Health Advancement.




  Evidence-based model to support advocacy of public health, research, and programs