Young outcomes researchers often ask me if “the kind of work we do is even fundable.” In other words, it seems intuitive to anyone that laboratory science would need funding: you need reagents, lab equipment, technicians, mice, cells, etc. It might seem that “just crunching data” would not follow this model and, as such, would not lend itself to traditional funding mechanisms. However, an overview of the typical expenses found in outcomes research (and reassurance that, yes, the NIH considers these to be “viable expenses”) reminds us that the laboratory funding model is not so different from outcomes research.

Even if the research does not involve patient interactions, “crunching data” is expensive. Even the data themselves are expensive, and funding agencies do not expect that you have already paid for data. Some datasets (such as claims or pharmacy data) can cost hundreds of thousands of dollars. Additionally, dataset linkages are also time-consuming and fundable. Demonstrating feasibility and potential effects in either a small subset of the main data, or in data from a different source with similar structure, can suffice for preliminary data; there is certainly no requirement to have conducted the entire study before applying for the grant. No matter the source, data arrive dirty, and require cleaning and extensive exploration to ensure high-quality data prior to the primary analysis; again, pilot data can be derived from a subset or a different source, and in either case the main data still require this work, which requires the time and effort of a research assistant or an analyst (or both).

The analysis itself is often complex, requires computers and statistical software (you might have the software, but the licenses might require updating) and, most importantly, personnel. For an analysis to be reliable, redundancy is likely required. To ensure the best methods are used, analysts are required but not sufficient; faculty (typically from departments of biostatistics and/or epidemiology) with extensive, published experience in methods relevant to the science are critical, and must contribute enough effort (at least 1.2 calendar-months per year) to demonstrate full engagement with the research team. Since a clinical understanding is required to inform exploration of the data and the analytical approach, substantial effort from the PI and/or clinical experts is also required.

If patients are involved, expenses add up quite quickly. These can include patient incentives to participate in the research (gift cards, other tokens, meals) as well as expense reimbursement (travel, parking). Research assistants are often required to collect the data directly from the patients, abstract data from medical records, and enter data into whatever computing system has been established for capturing the data. The data collection system is also an expense of the research; often, the pilot study uses a more rudimentary data collection system, and this is expanded once funding is secured. As with data analysis, data collection does not just involve those collecting the data, but requires supervision, redundancy, and faculty collaborators with extensive experience in conducting human subjects research contributing enough effort to demonstrate engagement in the process of data collection design, subject recruitment and retention, and protection of human subjects.

The holy grail of medical research support is the NIH. Readers to whom this statement is a surprise are encouraged to rethink their interest in research. Not only does the NIH have the largest budget for medical research (just under $30 billion in 2013), it has the process of grant review widely considered to be the most robust; as such, achieving NIH funding, serving on an NIH study section, or even having the letters “NIH” on your license plate are considered prestigious in the academic community and highly valuable to promotions committees. NIH publishes a Weekly NIH Funding Opportunities and Notices newsletter that is well worth receiving through their free e-mail subscription. The general type of grants offered by the NIH (Who Gets the Money) and the NIH grant review process (The Road to Riches) are discussed in more detail below.

However, it is also not breaking news that the NIH budget has recently stagnated (which effectively means a decrease in funding), and that matters are likely to become far worse. As such, it is critical to be aware of (and seek) alternative sources of research funding. The following list illustrates a broad range of non-NIH funding opportunities but is undoubtedly incomplete, and the reader is encouraged to explore funding opportunities independently.

Three major government-based funding sources specific to medical research are the Agency for Healthcare Research and Quality (AHRQ), the Patient-Centered Outcomes Research Institute (PCORI), and the Health Resources and Services Administration (HRSA). AHRQ functions basically like a “mini-NIH” in terms of funding opportunities, grant review, and grant structure; the other agencies use their own mechanisms. The Centers for Disease Control and Prevention (CDC) also offer occasional disease-specific grant or contract opportunities, and the National Science Foundation (NSF) is a staple for science and engineering funding that might align well with engineering/medicine collaborations. Recently, the Center for Medicare and Medicaid Services (CMS) has introduced some interesting outcomes research and innovation funding opportunities as well. Finally, the Department of Defense funds a Congressionally Directed Medical Research Program for areas of medicine directly relevant to service members, their families, and other military beneficiaries.