Biostatistics

The Institute for HealthCare Delivery Science is dedicated to improving the effectiveness, safety, and efficiency of patient care throughout the Mount Sinai Health System. While medical knowledge is increasing at an exponential rate, numerous studies have documented unexplained variations in practice and patient outcomes, with unacceptable rates of medical errors and inefficiencies in health care delivery nationwide.

Given the diversity of scientific expertise and research within the Tisch Cancer Center, the Biostatistics Shared Resource Facility (BSRF) provides support to a substantial number of projects, encompassing a broad research spectrum. With this in mind, our principal objective at BSRF is to provide biostatisticians with expertise in diverse types of cancer as well as a comprehensive coverage of statistical methodologies. Specifically, the four aims of BSRF are to:

  • Establish a scientific and administrative structure that supports investigators from a broad background and creates a collegial environment
  • Provide high-quality consultation in research design and biostatistical analysis
  • Train laboratory and clinical investigators in the quantitative aspects of research
  • Support methodological research for development of innovative research design and implementation of newly published analytic methods

The BSRF is available to any cancer investigator in the Icahn School of Medicine at Mount Sinai, whether you are a basic scientist, clinician, epidemiologist, psychologist, behavioral scientist, or other type of researcher. As an investigator you are able to access the facility via online requests for new projects through the facility website or by contacting the members directly for work on ongoing investigations.

The shared resource director will respond to online requests for statistical supports using the priority formulation of:

  • Planning of Investigator-Initiated clinical studies
  • Preparation of NCI/NIH grant submission
  • Analysis of NCI/NIH funded projects (presentation to scientific meetings and manuscript submission are given higher prioritization)
  • Preparation of grant submission to non-NIH agencies
  • Analysis for projects funded by non-NIH agencies (presentation to scientific meetings and manuscript submission are given higher prioritization
  • All others (unfunded exploratory or training related work)

To request services please complete our service request form. Your request will be assigned to a TCI-BSRF statistician within two business days. Once assigned, a statistician will respond to your inquiry within three business days to arrange for the first meeting. Timeline for the work to be done will be discussed in this meeting. For interventional investigator initiated protocols, movement to PRMC requires approval from the ‘Biostatistics Design Workshop’ (BDW), a committee of all statisticians in the TCI-BSRF that meets weekly on Thursdays from 2-3 pm. For grant applications, time for development of the statistical sections for a grant also ranges between 2-4 weeks depending on the readiness of the proposal in terms of having all information needed for writing the statistical sections.

Core members include:

Affiliated Members

Affiliate members are called upon for their special skills in biostatistics, bioinformatics, data science, genetics, economics, qualitative research, decision science, and epidemiology if TCI projects are in need of these skills.

Institute for Healthcare Delivery Science in the Department of Population Health Science and Policy

Institute for Translational Epidemiology in the Department of Population Health Science and Policy

Department of Population Health Science and Policy

Center for Biostatistics in the Department of Population Health Science and Policy

Department of Environmental Medicine and Public Health

Department of Population Health Science and Policy and Department of Genetics and Genomic Sciences

Department of Diagnostic, Molecular and Interventional Radiology

Our Cancer Biostatistics Shared Resource Facility (C-BSRF) serves as a cost-effective resource for members of the Cancer Center (CC) by offering a diversified balance of financial support plans to accommodate different types of projects and funding levels.

Services Provided for Collaboration (Free of Charge) Supported by CCSG

Financial Model for Services Provided Free of Charge
Grant and protocol development (and review) services are provided free of charge to CC investigators, however the expected arrangement is that when an investigator utilizes these core resources to develop the statistical sections of their grant application, they will provide the statistician(s) that they collaborated with adequate FTE support. The level of support to budget for a biostatistician should be clearly aligned with the scope of the work, and should account for any methodological development, data management and statistical programming that may also be needed for the project. We request that investigators budget at least 10% FTE for statistical support on each grant application

Services Provided for Collaboration Supported by Grants and Contracts

  • Statistical analysis of data NOT directly related to I-IIT including NI_IIT and retrospective studies
  • Manuscript writing and review
  • Preparation for research conferences (e.g. data analysis and pre-conference critiques and presentations)
  • Assistance with identification of research gaps in order to initiate research
  • Assistance with journal clubs and paper review from a statistical methodology perspective
  • Teaching short courses in experimental design and analysis methodology catering to a particular disease system

Financial Models for Other Services Provided

While, the NCI-CCSG subsidizes work on grant and protocol development (and review), financial support for all other statistical services can be arranged through either a fee-for-service or long-term collaboration contract where collaborating academic units pay for statistical support either on an hourly short term basis (fee-for-service) or by funding a percentage of a statistician’s salary over a long-term period (long-term collaboration).

Fee-for-Service Contract

C-BSRF statisticians can be contracted to provide statistical services for CC researchers on a fee-for-service basis. The researcher must be a faculty or staff member currently affiliated with the TCI and must have a funding source with a full accounting unit. The process begins with an initial meeting to discuss the research project, and the scope of the work. If after the meeting, both the researcher and C-BSRF members (statistician and Director) mutually agree that the fee for service model is appropriate, a contract is prepared with a good faith effort made by the C-BSRF statistician to complete the work within a certain number of hours over an allotted time period. Investigators are charged at a subsidized hourly rate of $125; requiring a minimum of eight hours of work. This arrangement is appropriate for investigators with a very limited short-term need for biostatistical collaboration for projects that are not supported by the NCI-CCSG.

Long-term Collaboration Contract

Basic science laboratories or clinical/population health units can use their departmental operational funds to support a percentage(s) of a specific statistician’s (or a team of statistician’s, i.e. Junior/Senior or statisticians with special skill such as proteomic data analysis etc.) salary to collaborate on their ongoing research with matching dollars provided by the NCI-CCSG. Investigators interested in initiating a long-term collaboration contract can email the BSRF managing director to determine an appropriate support level and discuss the terms and process of initiating a collaborative agreement contract. A fund account number and title will be required for internal billing and transfer. This type of collaborative arrangement provides a stable source of funding for biostatisticians fostering development of productive long-term collaborations while motivating related methodological research.

The table below lists the % FTE matching support provided by NCI-CCSG for different levels of incremental % FTE departmental operational support. For example, 1 day/week of a statistician’s effort is allotted when a department provides 15% FTE with 5% matched by NCI-CCSG funds. 2 days/week of a statistician’s effort is allotted when a department provides 30% FTE with 10% matched by NCI-CCSG funds.

Long-term collaboration with matching financial support model

Days per Week Dedicated to TCI Project

% FTE

%FTE Departmental Operational Funds

%FTE NCI-CCSG Matching Support

1

20%

15%

5%

1.25

25%

20%

1.5

30%

25%

2

40%

30%

10%

2.25

45%

35%

2.50

50%

40%

3

60%

45%

15%

3.25

65%

50%

3.50

70%

55%

4

80%

60%

20%

4.25

85%

65%

4.50

90%

70%

5

100%

75%

25%

Related Information and Guidelines

The mechanism used to provide support does not alter the collaborative atmosphere that the TCI-BSRF has fostered over the years. The contribution of each person needs to be evaluated as a manuscript is prepared. Consideration for authorship should be based on the accepted criteria for most medical journals. These criteria generally cite both study design and statistical analysis as intellectual input sufficient for authorship. It is impossible to define every situation in advance; however, it should be clear that reimbursement for time does not preclude or replace authorship. Please refer to some helpful guidelines below

All cancer related publications using the statistical services of the TCI-BSRF should acknowledge the NCI-CCSG:

"Research reported in this publication was supported in part by the National Cancer Institute Cancer Center Support Grant P30CA196521 awarded to the Tisch Cancer Institute of the Icahn School of Medicine at Mount Sinai and used the Biostatistics Shared Resource Facility. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.”

To request any of the above services please submit a service request form

The Biostatistics Design Workshop (BDW) is comprised of statisticians in the TCI-BSRF and provides a forum for statistical collaboration on the development, design, and analysis of interventional (I-IIT) protocols. All investigators are encouraged to contact the biostatistician early in the development stage of their protocol, so they can get advice on study design, data management, and appropriate statistical methods for analyzing endpoints. The time for development of the statistical sections for a protocol ranges between 2‐4 weeks depending on the readiness of the protocol and the engagement of the PI. Typically it takes 2 weeks to gather the information needed for writing the statistical sections, and then 1‐2 weeks for editing. The statistician assigned to your protocol will present it to the BDW once they have completed the statistical sections of the protocol. You do not need to submit or present to the BDW. Please be advised that an automated notification from the service request form will be sent to the PI following the initial BDW review; this is for tracking purposes only. Statisticians will communicate progress to you and provide a final version of the protocol once it is approved.

Following are itemized steps of the BDW review process:

  1. All Interventional Investigator initiated research protocols must be reviewed by the Biostatistics Design Workshop (BDW) prior to submission to the DFG with exception for protocols that underwent scientific and biostatistical review by one of the following organizations.
  2. Collaborators should submit requests for statistical support for 'PROTOCOL DEVELOPMENT' through service request form
  3. DFG biostatistician is assigned and develops statistical sections of protocol
        1. Only Interventional (I-IIT) require BDW (DFG and PRMC) approval and need to contain the Essential Elements of a Protocol (industry provided protocol template), or be in proper PRMC format: Protocol Template
        2. The IRB format (HRP-503a) is NOT an acceptable format for I-IITs
  4. DFG Biostatistician submits protocol for BDW review when it is deemed ready by biostatistician and PI
  5. Once a protocol is approved by BDW, it can be presented at DFG
        1. On BDW approval, all track changes are accepted
        2. A version number and date of BDW approval is added to the protocol
        3. The BDW approved protocol is emailed to the PI, DFG Chairman and PRMC administrators
  6. After DFG approval, the protocol can be submitted for PRMC review (contact Venesha White Venesha.white@mssm.edu for information on this submission process)

Protocol Related Information and Guidelines

The primary information we need to begin the process of developing the data management, sample size justification, and analysis plan for the proposed grant is a draft of the grant itself. The Specific Aims, Research Design and Methods sections are critical, but having the Background, Significance and Preliminary Studies sections are also important because they can provide valuable information for developing both the sample size justification and overall analysis plan. Biostatistics Shared Resource Facility (BSRF) statisticians provide support with grant writing to investigators, but it is highly recommended that the biostatistician be included in the proposal with percent FTE support.

Related Information and Guidelines

Prior to grant submission, the following information needs to be provided to the biostatistician involved at least 4 weeks before the deadline.

  • Funding agency
  • Submission date
  • Title of proposal
  • Proposed start and end date
  • Principal Investigator(s)
  • Our statistician’s role
  • % FTE over the proposed years
  • Annual direct costs
  • An updated version of the full application
  • Your department’s (financial) administrator’s name

Guidelines

Regulatory Services

For information on regulatory services including:

  • Support to obtain IACUC and IRB approvals
  • External regulatory support with IND/IDE submissions to the FDA
  • NIH grant applications
  • Guidance on protocol management and monitoring services
  • Visit the Office of Research Services

Clinical data management is the process of collection, cleaning, and management of data in compliance with regulatory standards. It is an important part of any research study. Every interventional (I-IIT) should have a clinical data management plan and support of a research study assistant. Statisticians are available to assist in the development of the data management plan by offering guidance on: selecting the right electronic data capture system, the best data collection methods, data entry practices that ensure consistency with paper CRFs, data validation methods and proper transfer of data for statistical analysis. However, biostatisticians are not taking responsibility for data management.

Related Information and Guidelines

1. Electronic Data Capture Systems (EDC)

  • eRAP - a fee for service web-based application that allows you, as an investigator, to have a custom built database.
  • RedCAP – a no-fee self-service web-based application that allows investigators to design their own database.

2. Case Report Form (CRF) Design

  • Here you will find CRF templates consistent with FDA standards. These forms serve only as templates and should be edited to meet study data collection needs as described in the protocol.

3. Data Entry

  • Single
  • Double – Double data entry ensures better consistency with paper CRFs.

4. Data Validation

  • Embed Range and Logic Checks in the EDC system to ensure data validity.
  • Please create queries and run routinely to ensure ongoing quality control.

5. Discrepancy Management

  • Discrepancies flagged during the data validation and/or double data entry processes need to be reviewed and resolved with documentary proof.

6. Medical Coding

  • Coding for Adverse Events should be done using the Medical Dictionary for Regulatory Activities (MedDRA). To request access to MedDRA email refdesk@mssm.edu.
  • Coding for Medications should be done using the WHO Drug Dictionary Enhanced (WHO-DDE)
  • Coding for Adverse Reactions should be done using WHO-ART

For clarification of any of the above data management activities please consult the NIH guidelines.

All requests received through eRAP will be assigned to a TCI-BSRF statistician within two business days. Once assigned, a statistician will espond to your inquiry within three business days to arrange for the first meeting. Timeline for the work to be done will be discussed in this meeting. For protocols, movement to PRMC requires approval from the ‘Biostatistics Design Workshop’, a committee of all statisticians in the TCI-BSRF that meets weekly on Thursdays from 2-3 pm. For grant applications, time for development of the statistical sections ranges between 2-4 weeks depending on the readiness of the proposal in terms of having all information needed for writing the statistical sections.

Resources For You

Department of Population Health Science and Policy
Biostatistics and Data Science
Mount Sinai Tisch Cancer Center