May 11, 2012 (CIDRAP News) – The Institute of Medicine (IOM) yesterday unveiled a decision-support model with a test version of software to help policymakers and other stakeholders determine what vaccines to develop at both national and global levels.
The IOM's work arose from a request from the Department of Health and Human Services (HHS) National Vaccine Program Office (NVPO). The model is the first implementation step for the first goal of the NVPO's 2010 National Vaccine Plan, which is to "develop new and improved vaccines." The NVPO released the plan in February 2011.
Prioritizing vaccine development is a challenge, because technology, disease threats, economic issues, and social factors are constantly changing, and currently there are no widely accepted methods or models to help guide decisions, the IOM said in its 163-page report, which is available on its Web site. This is the first time the NVPO has addressed vaccine prioritization since 2000.
The decision-support model and software, called Strategic Multi-Attribute Ranking Tool for Vaccines, or SMART Vaccines, isn't designed to make decisions, the IOM committee that developed the tools wrote. "It is intended to be used exclusively as a decision-support tool and only that," they wrote, adding that a primary use of the model will be to facilitate discussions about attributes and values among diverse users.
The 16-member committee was led by Lonnie King, DVM, dean of the College of Veterinary Medicine at Ohio State University. According to the report, the group met five times in 2011, hosted an international stakeholder session, and held public workshops during its first two meetings. Consultants helped develop the model and software, and an 11-member expert group helped evaluate them.
The model includes up to 29 attributes, which relate to health, economic, demographic, public, scientific, business, program, and policy considerations, factoring in intangible values such as the eradication or elimination of disease. Because of the model's complexity, the IOM committee developed software to make the tool easier to use.
Users must input substantial amounts of data, such as age- and sex-specific population information, vaccination and healthcare use patterns, disease burden, and costs. The model is designed to generate information on matters such as cost-effectiveness, premature deaths averted, and gains in worker productivity.
The report describes how the committee tested the models using three diseases: influenza, tuberculosis, and group B streptococcus. Hypothetical vaccines used for testing included a universal flu vaccine and vaccines against the other two diseases. For countries, they used data from the United States, plus a developing country, South Africa. The committee said it selected South Africa because it is vastly different from the United States and had data and disease-burden and vaccine estimates that were needed to test the model.
The report includes a detailed description of the modeling rationale and technique, as well as an account of the testing process.
The pilot version of the software isn't available to the public yet, though the report has several screen shots of the program.
The project has now moved to its second phase, which involves demonstrating SMART Vaccines to a wide range of stakeholders and potential users to solicit their feedback. Then the group will make changes in the model based on the input received and will test them and the software again using three more vaccine candidates.
Eventually, SMART Vaccines 1.0 will be available for public use in an open-source environment. "We hope to inspire a community of users who will improve, enhance, and potentially manage the capabilities of this product," the IOM committee wrote.
See also:
May 10 IOM "Ranking Vaccines" report
NVPO 2010 National Vaccine Plan background
