Making visible the invisible: sex and gender dimensions in scientific research and reporting

Shirin Heidari, PhDChair – EASE Gender Policy Committee

Executive Editor / Sr Manager - International AIDS Society

25,281

25,281

25,281

Yang et al Genome Research 2006

Source: Vera Regitz-Zagrosek; Sex and gender differences in health EMBO reports 2012

Sex and Gender differences in other clinical entities

Source: Vera Regitz-Zagrosek; Sex and gender differences in health EMBO reports 2012

Are we surprised?

“Women appear to be more susceptible to this risk because they eliminate zolpidem from their bodies more slowly than men. … FDA has informed the manufacturers that the recommended dose of zolpidem for women should be lowered from 10 mg to 5 mg for immediate release products (Ambien, Edluar, and ‐Zolpimist) and from 12.5 mg to 6.25 mg for extended release products (Ambien ‐CR).”

http://www.fda.gov/downloads/Drugs/DrugSafety/UCM335007.pdf

Why were these differences observed after approval and marketing?

1. Gender bias in clinical studies

Kruskal Wallis, p=0.05

020

4060

8010

0P

erce

nt fe

mal

e en

rolle

d

Private (commercial)Private (non-commercial)

PublicMixed

No data

n=111 n=18 n=66 n=76 n=64Funding source

Kruskal Wallis, p=0.92

020

4060

8010

0P

erce

nt fe

mal

e en

rolle

d

Private Public Mixed No datan=9 n=19 n=51 n=6

Funding source

HIV treatment studies (Antiretrovirals)Median women: 19.2%

HIV Cure StudiesMedian women: 9.9%

ARV trials funded, partially or wholly, by NIH, have significantly lower median proportion of female participants than non-NIH funded trials (15.3% vs 22.3%

p=0.001).

Why were these differences observed after approval and marketing?

2. Sex bias in pre-clinical and basic science

3. Sex/Gender bias in reporting“…75% of studies in three highly cited immunology journals did not specify whether the animals used were male or female.”

Ideas

Research design

Research resultsPublication

Knowledge Education

Products Market

Peer-reviewedJournal

Research

& Clinical

Governanc

e

Research Funding

Body

“Research must systematically

incorporate attention to sex and gender in

design, analysis, and interpretation of

findings” WHO 2009

Regulatory agencies

ICMJE policy on CT registration13 Sept 2005

Source: Zarin et al NJEM 2005

Clinical Trial Registration required by FDA in 1997

EXAMPLE: Clinical trials registration

2005

http://journals.lww.com/jaids/pages/currenttoc.aspx

EASE Gender Policy CommitteeEstablished by EASE Council in June 2012

Chairs: Shirin Heidari & Thomas BaborMembers: Rachel Carol, Paul Cummins, Mirjam Curno, Paola De Castro, Srecko Gajovic, Joy Johnson, Ravi Murugesan, Ana Marusic, Paul Osborne, Petter Oscarson, Ines Steffens, Kerstin Stenius, Chris Sterken, Sera TortSupporting members: Carina Sorensen, Meredith Sones

secretary@ease.org.uk

EASE Gender Policy Committee

• Mission: – Advocate for better reporting of gender and sex differences

and/or similarities in scientific research – Promote gender mainstreaming (and better science)

through inclusion of sex/gender considerations in policies and standards for scientific publishing

– Promote gender mainstreaming and gender balance in editorial boards and editorial offices.

• Activities:– International Gender Survey (next presentation)– Development of Common Standards

International Gender SurveyLaunched in spring 2013

• Purpose: to map existing editorial gender policies and opinions towards the adoption of such policies.

388 Unique journals - 114 Unique publishing housesNumber of respondents and response rates by target group

Target group Nr invited Nr responded Response rate (%)

EASE 429 167 40%

ISAJE 32 27 84%

100 journals 334 58 17%

Open - 464 -

TOTAL - 716 -

Existing editorial gender policiesand opinions towards them

1. Does the journal have:

2. Do you think journals should have:– instructions for authors, in which authors are required

or encouraged to disaggregate data by sex and provide gender analysis when applicable?

– a gender policy concerning the composition of the editorial staff and boards?

– a gender policy that strives for gender balance in the pool of peer reviewers?

Overview of existing gender policies

Yes No Do not know Not applicable

7%

53%

21% 18%

7%

56%

32%

5%5%

68%

22%

5%

Instructions for AuthorsComposition of editorial staff/boardsPool of peer reviewers

All sample groups

25%

43%

32%Yes

No

Do not know

A majority (75%) are unwilling or unsure to introduce sex and gender considerations as requirements in Instructions for Authors.

Do you think requirement of data disaggregated by sex should be included in instructions for authors as a matter of routine across all journals/publishers?

All sample groups

Opi

nion

Women are more in favour of gender policies in instructions for authors than men

Yes No Unsure0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

31%

59%

33%

69%

41%

67%

Woman (n=121)Man (n=101)

EASE/ISAJE/OPEN

Gender of respondent significantly correlated with readiness to adopt gender policy in IfA

Why sex disaggregation should NOT be included in ‘Instructions for Authors’

• “It's not applicable to all journals, only ones that publish research about people”

• “This policy will - paradoxically and unwillingly - create inequity for all other classes of 'different' humans”

• “I cannot see any reason whatsoever for doing it”

• “Not applicable to animals”

Do you think journals should have a gender policy concerning the composition of the editorial staff and boards?

40%

45%

18%

YesNoUnsure

All sample groups

Opi

nion

Women are more in favour of gender policies for composition of editorial board

than men

Significant correlation between gender of respondent and readiness to adopt gender policy with regard to composition of editorial board and staff

OPEN groupYes No Unsure0

20

40

60

80

100

120

140

160

180

200

31%59%

33%

69% 41%67%

Woman (n=121)

Man (n=101)

Reasons why gender policies for composition of editorial boards should NOT be included

• “Members of the editorial staff and editorial boards should be appointed on the merit of their qualification, academic standing and availability, not on the basis of their gender.”

• “Because as a woman I want to know I was given a position because I was the best person for the job not to fill a quota.”

• “The very idea of gender requirements strikes me as paradoxically sexist, i.e., the idea that gender must be considered when making up editorial staff and boards puts undue emphasis on the person's gender over their qualifications.”

• “I would need to see evidence of the necessity of such policies.”

Gender distribution among our respondents

Editor-in

-Chief

Editoria

l Board

Publisher

Technical assi

stant

Other0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

69%

48%

26%38% 35%

31%

52%

74%63% 65%

TransgenderWomanMan

All sample groups

So what next?

• Provide evidence and rationale for the implications of sex and gender bias in research reporting

• Provide guidance and recommendations for:– policies promoting reporting of sex and gender

information in scientific publishing– policies and procedures to promote gender balance in

editorial offices, editorial boards and pool of rev• Disseminate the Common Standard internationally

and to monitor its implementation

SHIFTING MINDSChanging the “default assumption”:There are sex/gender differences until the contrary is proven.

“The absence of evidence is not the evidence of absence”Carl Sagan

Thanks to members of the EASE Gender Policy Committee for their dedication and support, and in particular thanks to: Carina SorensenMeredith SonesJoy JohnsonThomas BaborPaul OsborneChris Sterken

CONTACT US:Shirin.Heidari@iasociety.orgsecretary@ease.org.uk