Estimating Clinical Research Project Duration from Idea to Publication

The scientific response to the COVID-19 pandemic has elicited commentaries on the quickening of biomedical research,^1–3 contrasting with literature on prolonged time to publication for clinical research projects.^4–6 We investigated research project duration for three clinical departments (emergency medicine, family medicine, and pediatrics) with centralized research leadership in a community-based US medical school. Following institutional review board (IRB) approval, we identified original research reports published or accepted in academic year 2019–2020, originating at our institution, and involving faculty from these departments. Of 39 eligible publications, we reconstructed study timelines (table 1) for 17 publications based on departmental records, and 10 publications based on a survey of corresponding authors.

OPEN IN VIEWER

Table 1

Data points on project stages, by project type

Project stage	Number of projects with available data
	Prospective research	Retrospective research	Basic science research	Other project type
Conception  First written outline or synopsis of project rationale and aims	5	18	1	2
Funding  First funding received in direct support of the project	2	5	1	0
Regulatory submission  First submission to IRB/IACUC	5	8	1	0
Regulatory approval  Receipt of all approvals needed to conduct the project	4	9	1	0
Start of data collection	5	18	1	1
End of data collection	4	18	1	1
Manuscript draft  First draft of manuscript, including introduction, methods, and results	5	18	2	2
Manuscript submission  First submission of manuscript to peer-reviewed journal	5	18	2	2
Manuscript acceptance*  Date of acceptance decision from journal	5	18	2	2
Publication date  Date of earliest publication online or in print	4	17	2	2

*

Data set included two prospective studies, nine retrospective studies and one basic science study accepted on or after April 1, 2020.

IACUC, Institutional Animal Care and Use Committee; IRB, institutional review board.

The median overall project duration was 18 months (IQR 10–26). Median durations of specific phases were 2 months for project development (IQR 1–4), 6 months for execution (IQR 1–18), 2 months for writing (IQR 1–4), and 4 months for publication (IQR 2–5). Durations are compared by project type and stage in table 2. On multivariable Cox regression analysis, time to publication was prolonged for prospective versus retrospective projects (HR of publication 0.14, 95% CI 0.02 to 0.83, p=0.030) and funded versus unfunded projects (HR 0.13, 95% CI 0.02 to 0.80, p=0.027). Twelve articles were accepted or published during the COVID-19 pandemic (April–June 2020). Median project duration was longer compared with pre-COVID publications in our sample (20, IQR 19–25 months, vs 13, IQR 9–40 months), but this difference did not reach statistical significance (rank-sum p=0.271).

OPEN IN VIEWER

Table 2

Project duration (in months) by project characteristics (N=27)

Characteristic	Median (IQR) project duration (months)
	Total	Development	Execution	Writing	Submission
Project type
Retrospective	17 (11–19)*	2 (<1–4)	2 (1–9)	2 (1–4)	3 (2–5)
Prospective	42 (24–42)*	5 (2–8)	20 (7–28)	9 (2–15)	4 (4–7)
Bench science or other	7 (5–24)*	6 (<1–12)	12 (1–23)	1 (1–2)	3 (<1–4)
Regulatory approval required
No	11 (5–19) †	<1 (<1–4)	1 (<1–3) †	2 (1–3)	3 (2–5)
Yes	20 (17–38) †	3 (2–5)	7 (2–26) †	2 (1–6)	4 (3–5)
External funding received
No	16 (9–20) ‡	2 (<1–4) ‡	2 (1–9) ‡	2 (1–4)	3 (2–4) ‡
Yes	38 (29–58) ‡	9 (5–12) ‡	25 (15–27) ‡	1 (1–12)	7 (6–14) ‡
Trainee participation
No	16 (5, 24)	<1 (<1–4)	2 (1, 7)	3 (1, 8)	2 (1, 4)
Yes	19 (11, 32)	2 (1, 4)	7 (2, 22)	2 (1, 4)	4 (3, 5)
Prior presentation
No	18 (10, 19)	3 (1, 4)	2 (1, 8) §	2 (1, 4)	3 (2, 5)
Yes	21 (11, 40)	2 (<1, 5)	7 (3, 26) §	2 (1, 4)	4 (3, 6)

*

Statistically significant difference across project types (p<0.05) on Kruskal-Wallis test.

†

Statistically significant difference according to need for regulatory approval (p<0.05) on rank-sum test.

‡

Statistically significant difference according to receipt of external funding (p<0.05) on rank-sum test.

§

Statistically significant difference according to prior presentation (p<0.05) on rank-sum test.

Qualitatively, centralized research support programs have been described as ‘expediting’ clinical research ⁷ by addressing limitations of time, funding, and expertise among investigators.^8–10 Short-term evaluations of such programs (<2 years) have focused on activities which can be completed in a few months, such as IRB protocol or grant submission ⁸ ; or manuscript submission, but not necessarily publication. ¹⁰ Indeed, within this time frame, many projects receiving central support may not reach publication. We propose that tracking project duration from conception to publication, in addition to discrete steps such as conference presentation or grant submission, can help identify opportunities for improving the research process for investigators.

Our data also have important implications for faculty career development and mentorship. In light of the typical research project duration, pursuing projects concurrently rather than sequentially is important for building a strong file for promotion and tenure. While faculty with extramural funding ultimately tend to achieve higher academic productivity, ^{11 12} an important finding was that grant-funded research took significantly longer from conception to publication than unfunded research. Delays on grant-funded projects may be addressed by optimizing institutional grant-related processes and developing a diverse portfolio of funded and unfunded research. Lastly, we present an early comparison of project duration for publications accepted before and during the early months of the COVID-19 pandemic. Considering the long life cycle of a typical project, the impact of the pandemic on project duration and publication is likely to continue accumulating in the months ahead and may pose serious career challenges for active scientists.