You sit down to research a new topic. Maybe it is for a grant proposal or to get up to speed on a program or practice area. Your search results include published research, but you can’t access full-text articles without going through a library or directly purchasing the content from the journal publisher’s website. You settle for reviewing the abstract and keep searching. But what have you missed by not being able to easily access the research and data you need? Meet open science, the principles of planning, conducting and disseminating research that increase transparency of data and methods and foster productivity by offering both research results and the tools used for more efficient knowledge exchange.
Open science includes multiple principles: for publications (open access), data sets (open data) and software/application code (open source). Other examples include open-source models for sharing intellectual property and open governance practices. While the application and interpretation of the principles of open science can vary across platforms and countries (for example, whether single articles or entire journals are open access) the goals are the same: to bring research results to the surface and to improve transparency by making information available to the most people as efficiently as possible.
The following list, adapted from the Texas State University Library, offers pros and cons for those considering publishing in open access journals:
Pros:
- Anyone can read your work, so you can increase your reach and influence. Search engines can find the full text article (not just the abstract), too.
- The copyright is yours, so you can share your work and cross publish on different platforms as appropriate and relevant. For example, you could publish part or all of your article on your organization’s website or in a newsletter for a professional association.
Cons:
- Traditional subscription-based journals can be cost prohibitive outside a university or corporate setting.
- Open access doesn’t always mean free. APCs can be expensive and a barrier for the author.
- Despite the rising number of open access journals, some tenure committees at academic institutions prefer traditional journals.
- You have to be on the lookout for so-called “predatory” journals that “use the open access model to prey on authors. These journals do not provide editorial services, peer review or indexing services” and may use email spam techniques to attract new authors. Always verify a publisher before submitting your research.
Source: Texas State University Library. guides.library.txstate.edu/c.php?g=430544&p=2937705
Open access journals
Open access journals offer full access to articles with no fees to the end-user. They are primarily published online and anyone — from the public to a clinician, program manager or policymaker — can easily access the published work. As of December 2021, there were 17,214 open access journals from 130 countries listed in the Directory of Open Access Journals. All open access journals listed in the DOAJ must follow the “Principles of transparency and best practice in scholarly publishing,” a set of standards and best practices created by the DOAJ, Committee on Publication Ethics, Open Access Scholarly Publishers Association and World Association of Medical Editors. And just like traditional publications, many open access journals follow rigorous peer-reviewed processes for article solicitation and selection.
There are supports in place for journal publishers and funding bodies to guide the transition from traditional to open access publishing. In 2018, cOAlition S, an international group of research organizations, published its Plan S principles to help move the needle on open access publishing. Organizations that agree to implement the principles commit to transition to open access within one year of the agreement.
While readers have no fees to pay, open access journals may still have article processing charges, or APCs. These fees are paid by the author or a funding body for the article to be published as open access. Waivers can be obtained for APCs in some situations. It’s worth taking time to research the fee waiver process, because these fees can amount to thousands of dollars, depending on the journal. Even if the entire publication isn’t fully open access (known as Gold Open Access), an author can pay for an open access license in a hybrid journal (a mix of open access and subscription) for the article they have written. Again, the APC can be cost-prohibitive, but an author, author’s organization or a funding body may be willing to pay the APC so the published work can be more widely shared.
There are different types of licenses for content in terms of requirements for attribution and any restrictions such as limitations on commercial use, so it is important to check any publisher or funding entity requirements related to licensing. Many open access licenses fall under Creative Commons; learn more about these licenses here. In addition, if the research received outside funding, the funder may require any resulting publications be open access. The Bill & Melinda Gates Foundation, for example, has had an open access policy in place since 2015 for its funding recipients and awardees. Under this policy, “all peer-reviewed published research funded, in whole or in part, by the foundation, including any underlying data sets” must have unrestricted access. Finally, federally funded research, such as by the National Institutes of Health, is made publicly available through PubMed Central as part of NIH’s Public Access Policy.
Open data
Open data involves the placement of data sets in publicly accessible online locations for download or the unlimited provision of datasets via electronic file delivery by direct request. Open data is not new, particularly when it comes to the government. Since 2009, there have been government policies for open data. According to the World Bank’s Open Data Toolkit, more than 250 governments and 50 countries have open data initiatives. The Open, Public, Electronic and Necessary Government Data Act of 2018 expanded the open data policy for the U.S. federal government and requires agencies to “publish information as open data by default, as well as develop and maintain comprehensive data inventories.” There also are global and national open data repositories, such as World Bank Open Data, UNData (United Nations), GODAN (Global Open Data for Agriculture and Nutrition) and Data.gov, the U.S. Government Open Data Portal.
The more data that’s made available, the more opportunities for secondary analysis; consolidation of smaller data sets into big data that can be mined for patterns and learnings; and future research projects that take what has been done to the next level. This can mean a bigger patient pool, a larger demonstration project, a revamped intervention approach and more.
Any data collected as part of surveys, electronic health records or other forms or processes can be made “open” as well, depending on the policies in place and the terms of use agreements signed by data contributors. This data can then help further knowledge and learning and be a powerful tool for designing interoperable systems; modeling use cases (such as descriptions or flowcharts of how a user might interact with a technology tool or perform a certain task); creating maps for exploring social determinants of health or tracking health or disease data; applying artificial intelligence or “smart” applications; and identifying common pathways for access to services and provision of care.
While there are many benefits of open data, there are ethical considerations to ensure health data is handled appropriately. For example, when researchers and practitioners have access to and can compare data sets that include both social and health information, they can more readily identify and address health equity challenges in designing and delivering programs.
But wait. Are open access journals as reputable as traditional journals?
It is a misconception that open access journals do not follow rigorous peer-reviewed processes. If a journal — whether traditional or open access — is peer-reviewed, the process will follow a framework for ensuring high-quality research is published. Before submitting any manuscript, research the journal, ask colleagues who are familiar with it or request information from your local library or academic institution.
Data privacy
Many open access journals now require, at the manuscript submission stage, that authors attest that data from the research will be made publicly accessible. One example is the Dataverse Project by Harvard University, which offers a free open-source platform for data to be published and shared. It also links the data to the original research and offers users the option to download data in formats that can be inputted into statistical analysis software.
Data published as part of an open data platform need to be de-identified. This means all personally identifiable information, or PII, is removed. The National Institutes of Health defines PII as “information that can be used to distinguish or trace an individual’s identity, either alone (direct) or when combined with other personal or identifying information that is linked or linkable to a specific individual (indirect).” Examples include names, emails and home addresses. Social security numbers, driver’s license numbers, biometrics and medical or financial records are considered “sensitive PII” and require more stringent handling. The International Association of Privacy Professionals offers information and guidance on de-identifying data for open publishing.
In the health care setting, PII is referred to as protected health information, or PHI, and is governed by the Health Insurance Portability and Accountability Act, also known as the HIPAA Privacy Rule. There are times when components of this information can be shared, such as for disease tracking and surveillance, but criteria for HIPAA-permitted uses and disclosures must be adhered to. The U.S. Department of Health and Human Services provides guidance for effectively de identifying PHI when sharing data, including using “Expert Determination” to certify data and assess risk using statistical analysis or the “Safe Harbor” method, which globally removes key identifiers such as dates and social security numbers.
Another concern can be the re-identification of an individual when data sets are linked together. A study published in 2020 in Environmental Health Perspectives reviewed data and methods from 12 environmental health studies and identified data types that were vulnerable to linkages. These included data related to family members, genetics, medical care, housing and occupations. Given the large amounts of data gathered through formal research, as well as through web browsing, social networking, mobile apps, retail purchases and more, it is essential that data privacy and security standards be followed and upheld through the open data process.
Putting it all together
Access to data is essential to the Academy’s work in meeting its Research Priorities, a collection of emergent research needs that, if addressed, would have the greatest impact on knowledge advancement and empowerment of nutrition and dietetics practitioners. Work toward these priorities could be accelerated with increasing opportunities for data sharing. On eatrightPRO, members can find collections of secondary data sets as well as information on the Academy’s Nutrition Research Network.
For nutrition and dietetics professionals, especially those in research, you can put an open science lens on your work. From participating in knowledge exchange forums to pursuing publication in open access journals and submitting your data to open data portals, you can generate and advance evidence and best practices for the value of nutrition and the work of RDNs and NDTRs.
The more research and data are transparent and available, the more they can be used in collective efforts to improve programs and services, strengthen quality of care and increase positive patient outcomes. One example of this is Electronic Clinical Quality Measures, which use data from electronic health records and health information technology systems to measure the quality of care.
As a program manager or funding entity, you can put into your requirements that any research and data be made open and accessible. Not only would you have access, but so would the nutrition community and the world. And when barriers to science are broken down, we all can be better communicators of nutrition messaging that is evidence-based and showcases data utilizing graphic design and data visualization best practices.
The result: bringing the data to life, making it “here and now” and, through robust dashboards, charts and apps, putting the control of it in the hands of patients and clients, partners and fellow health professionals.
Additional Resources to Learn More
- Center for Open Science: https://www.cos.io/
- Data.Gov: https://www.data.gov/
- Federal Data Strategy: https://strategy.data.gov/
- Gates Open Research: https://gatesopenresearch.org/
- Harvard University Dataverse Project: https://dataverse.harvard.edu/
- HealthData.gov: https://healthdata.gov/
- Mind the Gap Report: https://mindthegap.pubpub.org/
- Office of Data Science Strategy (NIH): https://datascience.nih.gov/
- Open Data Barometer: https://opendatabarometer.org/
- Open Data Institute: https://theodi.org/
- Open Knowledge Foundation: https://okfn.org/opendata/
- Open Science Project: https://openscience.org/
- PLOS Open Data: https://plos.org/open-science/open-data/
References
A De-Identification Protocol for Open Data. International Association of Privacy Professionals website. https://iapp.org/news/a/a-de-identification-protocol-for-open-data/. Accessed October 4, 2021.
Boronow K, Perovich L, Sweeney L, et al. Privacy Risks of Sharing Data from Environmental Health Studies. Environ Health Perspect. 2020;128(1):17008.
Cantor M, Chandras R, Pulgarin C. FACETS: using open data to measure community social determinants of health. J Am Med Inform Assoc. 2018;25(4):419-422.
Cocoros N, Kirby C, Zambarano B, et al. RiskScape: A Data Visualization and Aggregation Platform for Public Health Surveillance Using Routine Electronic Health Record Data. Am J Public Health. 2021;111(2):269-276.
Demski H, Garde S, Hildebrand C. Open data models for smart health interconnected applications: the example of open EHR. BMC Med Inform Decis Mak. 2016;137(16).
Directory of Open Access Journals website. https://doaj.org/. Accessed October 3, 2021.
Electronic Clinical Quality Measures (eCQMs). Academy of Nutrition and Dietetics website. https://www.eatrightpro.org/practice/quality-management/quality-improvement/malnutrition-quality-improvement-initiative. Accessed October 13, 2021.
Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule. Department of Health and Human Services website. https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de-identification/index.html#determinations. Accessed October 13, 2021.
How HIPAA Supports Data Sharing. Office of the National Coordinator for Health Information Technology website. https://www.healthit.gov/topic/interoperability/how-hipaa-supports-data-sharing. Reviewed September 6, 2018. Accessed October 13, 2021.
Levin N, Leonelli S, Weckowska D, Castle D, Dupré J. How Do Scientists Define Openness? Exploring the Relationship Between Open Science Policies and Research Practice. Bull. Sci. Technol. Soc. 2016;36(2):128-141.
Open Access Policy FAQ. Gates Foundation website. https://www.gatesfoundation.org/about/policies-and-resources/open-access-policy-faq. Accessed October 22, 2021.
Open Data in 60 Seconds. World Bank Open Government Data Toolkit website. http://opendatatoolkit.worldbank.org/en/open-data-in-60-seconds.html. Accessed September 22, 2021.
Open Data: Agencies Need Guidance to Establish Comprehensive Data Inventories; Information on Their Progress is Limited. U.S. Government Accountability Office website. https://www.gao.gov/products/gao-21-29. Accessed October 4, 2021.
Payne P, Lele O, Johnson B, Holve E. Enabling Open Science for Health Research: Collaborative Informatics Environment for Learning on Health Outcomes (CIELO). J Med Internet Res. 2017;19(7):e276. Published July 31, 2017.
Peregrin T. Social Determinants of Health: Enhancing Health Equity. J Acad Nutr Diet. 2021;121(6):1175-1178.
Public Access Policy. National Institutes of Health website. https://publicaccess.nih.gov/. Accessed October 13, 2021.
The Dataverse Project website. https://dataverse.org/. Accessed October 13, 2021.
Toolkit for Patient-Focused Therapy Development. United States Department of Health and Human Services website. https://toolkit.ncats.nih.gov/glossary/personally-identifiable-information/. Accessed December 10, 2021.
Transparency and Best Practice. Directory of Open Access Journals website. https://doaj.org/apply/transparency/. Accessed September 22, 2021.
