Communicating Research

Communicating Research

Conducting a Medical Research Study

• Assemble a study team.
• Produce a study protocol and related documents.
• Review study documents by human or animal research ethics committee.
• Research data management.
• Presentation and publication of data.

Study Team

• Principal Investigator
• Chief investigators
  • Usually includes a biostatistician.
• Associate investigators
  • Recruiting study participants.
  • Essential skill for part of study.
• Research staff (usually grant-funded)
  • Research scientists and technicians
  • Study nurses
  • Others

Guidelines for Conducting Human Medical Research in Australia

• Australian Code for the Responsible Conduct of Research (NH&MRC)
• National Statement on Ethical Conduct in Human Research (ARC)

Production of Study Protocol

• Aims and hypotheses
• Background and rationale
• Participant inclusion and exclusion criteria
• Clinical and laboratory methods
• Statistical analysis plan
  • Sample size calculation, power
• Expected outcomes, with timelines
• Patient information and consent form

Research Data Management

• Plan ahead of study commencement
• Ownership of intellectual property
• Documentation and metadata
• Storage and backup of data
• Sharing and reuse
• Retention and disposal of data
• http://guides.is.uwa.edu.au/RDMtoolkit

Data Presentation

• Discuss data with your supervisor(s) before presentation.
• Presentation at departmental and local meetings to obtain advice and assistance.
• Presentation at scientific/medical conferences.
• Publication in scientific/medical journals.

Presentation at Meetings and Conferences

• Abstract submission to select for:
  • Poster presentation
  • Oral/poster presentation
  • Oral presentation
  • Online presentation
  • Rejection

Poster Presentations

• Poster presentations are good for networking.
• Adhere to guidelines for that conference.
• Keep text to a minimum and use bullet points where possible.
• Minimum font sizes: heading 120; subheadings 48; text 36.
• Maximize the use of figures or simple tables.
• Must be readable from 1 meter away.
• Making the most of your conference poster (uwa.edu.au)

Oral Presentations

• MS PowerPoint (or equivalent) is now standard.
• Plan for about 1 slide per minute.
• Use 3-7 bullet points per slide.
• Avoid writing and especially reading out long and complete sentences on slides because it is really boring to the audience.
• Slide appearance (font, colors) should be consistent throughout presentation.

Production of PowerPoint Slides

• Font
  • Use sans serif fonts (Arial, Calibri, or similar) not serif fonts (Times New Roman, Courier).
• Type size should be 18 points or larger
  • I prefer 28, 32, or 36.
• AVOID USING CAPITAL LETTERS ONLY BECAUSE IT IS MUCH HARDER TO READ.
• Use as much of the slide area as possible.

Colour

• Dark letters against a light background are best for smaller rooms, especially when the lights are on for teaching.
• http://www.fw.msu.edu/orgs/gso/documents/GSOWorkshopDocsSp2006/PresentationTipsinPowerPoint.ppt#302,5,Powerpoint basics: 1. What font to use

Colour

• Light letters on a dark background also work in a large room, but most people use white backgrounds.
• http://www.fw.msu.edu/orgs/gso/documents/GSOWorkshopDocsSp2006/PresentationTipsinPowerPoint.ppt#302,5,Powerpoint basics: 1. What font to use

Figures and Tables in PowerPoint Slides

• Figures
  • ‘1 figure ≈ 1000 words’
  • Must be readable, understandable, and uncluttered.
  • Keep simple and use color for clarification (NB. red/green color blindness is more common than you think!).
  • Explain axes and variables.
• Tables
  • Only use for small amounts of data.
• Try to avoid figures and tables from publications, especially those with large amounts of data.
• Include reference to source of data if not yours.
• http://www.cs.aau.dk/~luca/SLIDES/howtotalk-ru.pdf

Slide Simplicity

• Keep the slide simple where possible: simple text + visual enhances retention rate.
• Visual: 65% retention.
• Text: 35% retention.

Data Presentation Graphics

• Use data presentation graphics appropriate for the data, e.g., pitfalls of bar graphs.
• Weissgerber TL et al. PLoS Biol. 2015; 13:e1002128

Principles of an Effective Presentation

• State what you are going to say.
• Say it.
• Summarize what you have said and make conclusions.
• Don’t try to build suspense and then unveil a surprise ending.
• http://www.safetyoffice.uwaterloo.ca/hspm/tools/images/scaffold_stair.png

Basic Structure of a Presentation

• Title slide
  • Including authors and institutions
• Introduction/background
• Methods
  • Present only essential information
  • A figure is often more informative than text
• Results
  • Slide heading should indicate what slide shows
• Summary/Conclusions
• Acknowledgements

Example Study

• Ramachandran Vignesh, Andrew Lim, Suniti Solomon, Kailapuri G Murugavel, Kenneth H Mayer, Nagalingeswaran Kumarasamy, Martyn A French
• TB-IRIS after commencing ART is associated with expansion of pre-existent CD4+ and CD8+ effector T cell responses against Mycobacterium tuberculosis antigens.

TB-IRIS Patients

• TB-IRIS patients have higher proportions of IFNγ+ PPD reactive T cells before and after ART
  • CD4+ T cells
  • CD8+ T cells

Final Slides and Discussion

• Summary slide
  • List main findings as dot points
• Conclusions slide
  • List important conclusions and future studies
• Acknowledgement slide
  • Collaborators and their institutions
  • Funding sources
• Leave 2-3 minutes for discussion and have 1 or 2 slides for questions likely to be asked

Presenting Research Findings by Publication: Producing a Manuscript

• Inform and involve all authors at all stages of manuscript production.
• Select a journal for publication in conjunction with other authors
  • Impact factor vs citations
  • Open access publication (NB. publication fees)
• Read the ‘Instructions for Authors’ of that journal before starting to write the manuscript.

Structure of a Manuscript

• Title page
• Abstract
• Introduction
• Materials and Methods
• Results
• Discussion
• Materials and Methods (or online)
• References

Abstract

• Structured or unstructured
• Provide a brief, accurate summary of
  • Study rationale
  • Important methodology
  • Important findings
  • Conclusions
• Remember, the Abstract is often the only part of the paper that is read!

Structured Abstract Example

• Trend S, Leffler J, Cooper MN, Byrne SN, Kermode AG, French MA, Hart PH.
• Narrowband UVB phototherapy reduces TNF production by B-cell subsets stimulated via TLR7 from individuals with early multiple sclerosis.
• Clin Transl Immunology. 2020; 9:e1197
  • Objectives: At the end of a 60-day course of narrowband UVB phototherapy, administered to individuals with early multiple sclerosis, there were changes in the relative proportions of circulating B-cell subsets. This study investigated phototherapy-associated changes to cytokine responses of B cells when exposed to a TLR7 ligand.
  • Methods: PBMCs from participants of the PhoCIS (Phototherapy for Clinically Isolated Syndrome) trial taken before (day 1) and after phototherapy for 8 weeks (day 60) were incubated with, or without, the TLR7 ligand, R848, for 18 h. Production of TNF and IL-10 in seven B-cell subsets was examined, with cytokine responses in each individual at day 60, adjusted for responses at day 1. Paired PBMCs were from participants administered phototherapy (n = 7) or controls (n = 6).
  • Results: At day 60, significantly fewer B cells, particularly marginal zone-like B cells (CD27+ /IgD+), from participants administered phototherapy produced TNF in response to TLR7 stimulation. When responses by seven B-cell subsets were analyzed together using multivariate methods, a phototherapy-specific signature was observed. An increased responsiveness from day 1 to day 60 in IgM-only memory B cells (CD27+/IgD-/IgM+) after TLR7 stimulation also predicted slower progression from CIS to MS. Phototherapy was without significant effect on B-cell IL-10 production.
  • Conclusions: Reduced TNF responses after TLR7 stimulation in marginal zone-like B cells from participants administered phototherapy suggested treatment-associated priming effects that were detected upon subsequent polyclonal B-cell activation. Changes in responsiveness to TLR7 stimulation also suggested that IgM-only memory B cells may be important in conversion from CIS to MS

Introduction

• Describe the clinical significance and epidemiology of the disease under study.
• Outline the rationale for conducting the study
  • Importance of understanding more about the disease or pathogenic mechanisms
  • Gaps or controversies in published information
• Provide sufficient background information for the reader to understand the research approach, methodology, and findings.
• Propose a hypothesis to be tested.

Methods (Participants)

• Participant inclusion and exclusion criteria
• Statement that participants gave informed consent (i.e., signed consent form)
• Statement that the study was approved by a human (or animal) research ethics committee
  • Name of committee and approval number(s)

Methods (Clinical and Laboratory)

• Describe methods used with sufficient detail for another investigator to reproduce your investigations, e.g.
  • Clinical investigations undertaken
  • Laboratory reagents used, including information about supplier (name and location of company or individual)
  • Experimental conditions, e.g., culture times and temperatures
  • Machines, techniques, and data analysis methods, e.g., flow cytometry

Example Figure

• B cell gating strategy for analysis of uncultured peripheral blood mononuclear cells (PBMC) for flow cytometry data.
• Figures are taken from a representative sample.
• ASC: antibody secreting cells; MBC: memory B cells; MZ-like: marginal zone-like B cells; DN: double negative B cells; Sw MBC: class switched MBC.
• FcγRIIb Expression Is Decreased on Naive and Marginal Zone-Like B Cells From Females With Multiple Sclerosis Trend SF et al. Frontiers in Immunology 2021; 1:614492

Methods (Statistics)

• Outline statistical methods used and why they were used, if appropriate
  • Consider use of parametric vs. non-parametric analysis methods
  • Different datasets might require different methods
• Provide evidence the study is powered to show a statistically significant difference between comparator groups, if appropriate.

Results

• Present study findings as text and in numbered figures or tables referred to in the text.
• Figures are accompanied by a legend that should contain sufficient information to understand the data without referring to text.
• Tables have a heading that outlines what the table is presenting.
• Most journals allow the presentation of data in supplementary figures and tables that can be accessed online.

Results (Participants)

• Describe cases and controls (table or figure) with particular reference to:
  • Number of cases and controls (does the study have statistical power?)
  • Matching of cases and controls for factors that might also affect findings:
    • Demographic characteristics such as sex, age, racial group
    • Characteristics of disease under investigation, e.g., baseline CD4+ T cell count in HIV patients

Example Demographic Characteristics Table

• Demographic Characteristics of Patients with Tuberculosis-Associated Immune Reconstitution Inflammatory Syndrome (TB-IRIS) or ART-Associated Tuberculosis (ART-TB) and Controls
• Oliver B G et al. J Infect Dis. 2010; 202:1728-1737

Results (Data)

• Organize results into sections with appropriate subheadings.
• At the start of each section, briefly indicate why an experiment or clinical study was done.
• Describe findings with reference to data presented in tables or figures.
• At the end of each section, briefly summarize findings to provide a take-home message.

Example Results Section

• Expression of CD32b was next assessed on total B cells from people with MS or CIS and compared with controls. There was a clear trend towards decreased expression of CD32b in patients with CIS or MS compared with controls on total B cells (Figure 3C), though this difference was not statistically significant (p=0.07).
• Assessing individual B cell subsets, expression of CD32b was significantly lower on naïve B cells in patients with CIS or MS compared with controls (Figure 4B). No other comparisons of CD32b expression in B cell subsets were statistically significant (Figure 4 and not shown).
• Given the higher prevalence of MS in females, the cohort was separated and examined by sex to assess whether there was any difference between males and females in CD32b expression. Significantly lower expression of CD32b was detected on total B cells, naïve B cells, and IgMhi MZ-like B cells from females with CIS or MS compared with control females (Figures 5A–C).
• A trend toward slower CD32b expression on IgMlo MZ-like B cells from female patients was observed, but the difference was not statistically significant (p=0.06, Figure 5D).
• There were no differences in CD32b expression levels on B cells from males with CIS or MS and controls.
• Together, these results demonstrated that IgM+ B cells from females but not males with CIS or MS expressed lower levels of CD32b compared with female controls.
• FcγRIIb Expression Is Decreased on Naive and Marginal Zone-Like B Cells From Females With Multiple Sclerosis Trend SF et al. Frontiers in Immunology 2021; 1:614492

Example Figure

• CD32b expression on B cell subsets from males and females with CIS or MS compared with controls.
• CD32b expression is shown using Tukey’s boxplots on (A) total B cells; (B) naïve B cells; (C) IgMhi MZ B cells; and (D) IgMlo MZ B cells from healthy controls (n = 7 males, 9 females; grey circles), and patients with CIS or MS (n = 9 males and 13 females; orange circles).
• A dot is shown for each individual.
• Mann-Whitney tests were used to compare groups; *p-value < 0.05.
• FcγRIIb Expression Is Decreased on Naive and Marginal Zone-Like B Cells From Females With Multiple Sclerosis Trend SF et al. Frontiers in Immunology 2021; 1:614492

Example Continued

• Since significantly lower CD32b expression on total, naive, and IgMhi MZ-like B cells and non-significantly lower expression on IgMlo MZ-like B cells from females with CIS or MS was detected, CD32b expression on these and other B cell subsets was examined for associations with serum factors that may be related to B cell dysfunction.
• Variables investigated included serum levels of BAFF, which are increased in people with MS (42), and anti-EBV antibodies, since EBV has been implicated in the pathogenesis of MS (14).
• In addition, total non-specific IgM levels were compared with CD32b expression levels on cells, since MZ-like B cells are the main producers of serum IgM (43).
• Serum levels of other Ig, including IgA, total IgG, and IgG subclasses, were also included for comparison. No serum was available for controls.
• Serum BAFF levels in patients with CIS or MS negatively correlated with the expression of CD32b on naive B cells (p = 0.043; Figure 6A) and IgMlo MZ-like B cells (p = 0.02; Figure 6B), but not with other B cell subsets.
• Females with CIS or MS had significantly higher levels of serum IgM compared with males, and although females had higher median BAFF levels in serum than males, this was not statistically significant (Supplementary Figure 2A).
• While no relationship was detected between serum IgM levels and CD32b expression on other B cell subsets, serum IgM levels correlated negatively with CD32b expression on naive B cells (Figure 6C).
• Positive correlation between serum IgM and BAFF levels was also detected (Figure 6D).
• Serum levels of total IgG, IgG2, IgG3, and IgG4, as well as IgA, did not correlate with any B cell subset CD32b expression or serum BAFF levels (not shown).
• FcγRIIb Expression Is Decreased on Naive and Marginal Zone-Like B Cells From Females With Multiple Sclerosis Trend SF et al. Frontiers in Immunology 2021; 1:614492

Serum Factors and B Cell Subset

• Relationships between serum factors and B cell subset CD32b expression. (A) Correlation between naive B cell CD32b expression (MFI) or (B) IgMlo MZ B cell CD32b expression with serum levels of BAFF (A, B; n = 13).
• (C) Correlation between naive B cell CD32b expression and serum IgM levels (n = 19).
• (D) Correlation between serum IgM and BAFF levels (n = 13).
• Values shown were calculated for males and females combined using Spearman’s rho; results from males are shown as blue circles, and results from females are shown as red circles; p values<0.05 were considered statistically significant.
• (E) Levels of naive B cell CD32b expression, (F) IgMhi MZ-like B cell CD32b expression, and (G) IgMlo MZ B cell CD32b expression, according to anti-EBV VCA IgM antibody detection status (Yes, n = 8; No, n = 14; Controls, n = 16).
• For control samples, no serum was available for EBV serology testing. Data in (E–G) are presented using Tukey’s boxplots with a dot shown for each individual.
• Potential differences in CD32b expression between groups were tested using Mann-Whitney tests; *p<0.05.
• FcγRIIb Expression Is Decreased on Naive and Marginal Zone-Like B Cells From Females With Multiple Sclerosis Trend SF et al. Frontiers in Immunology 2021; 1:614492

Anti-EBV

• The serum of all individuals with CIS or MS contained anti-EBV EBNA-1 IgG antibodies at a titer of 1:32 or greater, indicating chronic EBV infection, as expected (44).
• In addition, the serum of 8/22 (36%) people with CIS or MS also contained anti-EBV VCA IgM antibodies, which is considerably higher than the 3–6% rate of seropositivity demonstrated in other clinical settings (45).
• Furthermore, anti-EBV VCA IgM antibodies were detectable in the sera of six females with CIS, one male with CIS, and one female with MS.
• The increased detection of anti-EBV VCA IgM in females compared with males in the CIS or MS patient group was borderline significant (Supplementary Figure 2B).
• Analysis of people with CIS or MS separated into groups of those with and without detectable anti-EBV VCA IgM antibodies demonstrated that CD32b expression was significantly lower on naive, IgMhi, and IgMlo MZ-like B cells in EBV VCA IgM seropositive compared with seronegative individuals (Figures 6E–G).
• Median levels of serum BAFF and total IgM levels were non-significantly higher in those with detectable anti-EBV VCA IgM (Supplementary Figures 2C, D), but other serum total Ig levels and IgG subclass proportions (%IgG) were similar between anti-EBV VCA IgM seropositive and seronegative individuals.
• Anti-EBV VCA IgM and total IgM were not correlated with one another (not shown).
• Taken together, these data from patients with CIS or MS suggest that lower CD32b expression on naive and MZ-like B cell subsets may be associated with higher serum levels of BAFF and IgM, and EBV VCA IgM antibody seropositivity, and this phenotype is predominantly observed in females.

Discussion

• First paragraph summarizes major findings.
• Subsequent paragraphs discuss individual findings in detail with reference to previously published data (confirmatory and contradictory).
• Penultimate paragraph often outlines study limitations and strengths.
• Final paragraph summarizes most significant findings with take-home messages and suggestions for future studies.

References

• List references at the end of the manuscript, using the citation method specified in journal Instructions for Authors.
• Reference management software, such as EndNote or Reference Manager, are helpful.
• Double-check that references are appropriate and correctly listed.
• Only cite authors’ publications where appropriate.

Acknowledgements in Publications

• Affiliation to institutions in authorship list
• Output measurements for university schools and research institutes
• Excellence in Research Australia (ERA) data
• University rankings
• Acknowledgements at end of manuscript
• Study contributors who were not authors
• Funding bodies
• Author contributions and conflicts of interest