Guidance to Departments for Effort Estimation of Collaborative Biostatisticians | Duke Department of Biostatistics and Bioinformatics

Departments frequently ask the BERD Core for guidance in estimating biostatistics effort on grants, and sometimes when we are not asked, insufficient effort is budgeted. This document is in response to these requests and needs. To accurately estimate effort needed to perform duties for a grant, biostatisticians take into consideration many different aspects of the project. Many of these considerations are outlined on the UC Davis website. These guidelines were referenced to create the table below that explains the overall effort needed for projects of various sizes and complexities¹.

While the UC Davis document and the table below provide substantial information to help understand a potential range of effort that could be required for a biostatistician to collaborate on a project, the actual budget depends greatly on:

the particular skillset of the collaborative biostatistician,
the biostatistician’s salary,
the effort availability of the biostatistician during the span of the grant/project work.

Therefore, these guidelines should not be used as a replacement for appropriate effort estimation by a biostatistician. We recommend that these documents are used in conjunction with direct interaction with your biostatistics collaborator to estimate project effort.

In an ideal setting, we advise that a collaborative biostatistician should have 4-6 months’ notice to collaborate on the development of a grant. With less time to learn about the project needs, the effort estimation may be less accurate. Therefore, we recommend addressing effort estimation at the start of formulating your proposal.

Annualized Biostatistics Effort Allocation Guidelines

Project Type	Large/complex projects	Regular projects	Simple projects	Limited scope projects
Total Effort	50-100% + per year	30-65% per year	20-35% per year	< 10% annually per year
Lead Biostatistician Effort³	20% + per year	10-15% per year	5-10% per year	≤5% per year
Analysis Biostatistician Effort⁴	30-100% + per year	20-50% per year	10-25% per year	5-10% per year
Activity Scope	Projects that require a high level of involvement in development and implementation of the project and communication of study results. Involvement may take many forms, including any of the following: Development and/or implementation of complex study designs. Assembly of datasets from large, complex or poorly documented sources (e.g. administrative or survey databases). Development and/or implementation of interim data analyses during data collection phase of prospective studies. Development of and/or use and interpretation of novel or complex statistical methods. Developing algorithms to identify units of analysis and define analysis variables Active participation in publications, with opportunity for first authored papers.	Projects with standard study designs and routine analyses. Involvement includes: Collaboration through all phases of the study, including the biostatisticians attending regular study meetings. Involvement of biostatisticians in study design, implementation, and data collection. Well-documented primary datasets provided for statistical analysis Analyses carried out using off-the-shelf procedures available in statistical software packages. Active participation in publications, with opportunity for first authored papers.	This profile is appropriate for simple projects requiring minimal effort from the lead biostatistician and straightforward statistical analyses are performed by an analysis biostatistician. Involvement is limited to: Occasional meetings with PI about study issues, such as choice of statistical methods to use. This level of effort is typically too low for the lead biostatistician to carry out analyses or support weekly meeting attendance. The lead biostatistician manages and supervises the analysis biostatistician on all study activities from data processing to analysis using standard statistical methodology. This level of effort for the lead biostatistician is generally not compatible with smooth workflows and readily available collaborative support, unless an experienced and capable analysis biostatistician is adequately supported on the project. Assistance planning and conducting early Phase I trials (e.g. adaptive trials) with active monitoring. Feedback from the biostatisticians on 1-2 manuscripts.	Occasionally, a limited amount of funding may be justified to support research needs that may include the following: Assistance planning routine, small early-Phase trials with little monitoring, or modest proof-of-concept animal studies Small studies with limited scope requiring minimal statistical analysis. Feedback or advice on a project in which oversight/monitoring is not necessary.
Example	A research group with multiple R01s, an NIH P or U, or over 500K R01s, such as large, prospective cohort studies or trials that also include substantial retrospective or record data collection.	Most R01s with proposed study design and statistical methods that statisticians would consider standard or routine.	Smaller scope R-level grants, such as R21s or R34s.	Small foundation grants with small scope of work and limited data.

¹Adopted from University of California Davis School of Medicine Guidelines for Estimating Biostatistician Effort and Resource Guidelines on Grants. Drafted and revised by Nichole Carlson (Univ of Colorado, Denver), Jody Ciolino (Northwestern), Gina-Maria Pomann (Duke), Sandra Taylor (Univ of California, Davis), and Leah Welty (Northwestern) with feedback from members of the Division of Biostatistics, Northwestern University Feinberg School of Medicine as well as members of the Biostatistics, Epidemiology, and Research Design Special Interest Group within the Association for Clinical and Translational Science.

²Effort allocation guidelines assume that other necessary other resources such as research study coordinators and project coordinators are also included in the budget. If a project has a strong research study coordinator or team of coordinators, if may be possible for biostatistics effort to be on the lower end of the range provided (e.g., if project coordinator sets up and maintains the study database with limited input from the biostatisticians). Notably, biostatistics effort does not include data entry.

³The most senior biostatistician on the study.

⁴ The biostatistician who executes the relevant components of the study, e.g. data management, SAP and report writing, under the directions of the lead biostatistician.