The number of women entering and advancing in academic psychiatry remains significantly less than the number of men, ^{Reference Killaspy, Johnson, Livingston, Hassiotis and Robertson1} despite the increasing number of women enrolling in medicine. ^{Reference Jagsi, Guancial, Worobey, Henault, Chang and Starr2} A number of factors have been discussed in the literature to explain the underrepresentation of women in academic psychiatry; ^{Reference Howard3} these include having few visible female role models and mentors, and the subtle cultural factors that seem to exclude women from the corridors of power. ^{Reference Greenfield, Peters, Lane, Rees and Samuels4} Other factors might include gender discrimination and the tendency for women's participation to decrease progressively as the selectivity and importance of the academic activity increases - the so-called ‘glass ceiling effect’. ^{Reference McSweeney, Donahoe and Swindell5}

An academic career in psychiatry is based on a triangle of clinical practice, research and publication, and teaching. ^{Reference Reiser, Sledge, Fenton and Leaf6} Publication in peer-reviewed journals is an important indication of academic activity and measure of productivity; it is also highly emphasised in the academic promotion process ^{Reference Jagsi, Guancial, Worobey, Henault, Chang and Starr2} and to some extent in non-academic career advancement - hence the phrase ‘publish or perish’. Traditionally, when a paper has multiple authors, the first author position is assigned to the person who contributed most significantly to the research/paper, ^{Reference Dickersin, Fredman, Flegal, Scott and Crawley7,Reference Porter, Christian and Polling8} indicating primary responsibility, leading role and active participation.

Studies in internal medicine and psychiatry have previously suggested that female faculty members are less likely than male colleagues to publish academic papers. ^{Reference Barnett, Carr and Boisnier9,Reference Leibenluft, Dial, Haviland and Pincus10} However, some more recent studies have reported an increase in the number of female authors. ^{Reference Alonso-Arroyo, Gonzalez-Alcaide, Bolanos Pizarro, Castello Cogollos, Valderrama-Zurian and Aleixandre-Benavent11-Reference Bhattacharyya and Shapiro14} Alonso-Arroyo et al reported an increase from 29.92% in 1999 to 38.86% in 2006 of female authors in Actas Espanolas de Psiquiatria, with 34.02% of female first authors. ^{Reference Alonso-Arroyo, Gonzalez-Alcaide, Bolanos Pizarro, Castello Cogollos, Valderrama-Zurian and Aleixandre-Benavent11} Kurichi et al reported an increase from 11.1% in 1985 to 25.6% in 2003 of female first authors in Archives of Surgery. ^{Reference Bhattacharyya and Shapiro14} Jagsi et al found that the proportion of first authors who were females increased from 5.95% in 1970 to 29.3% in 2004 in academic medical literature. ^{Reference Jagsi, Guancial, Worobey, Henault, Chang and Starr2} Alonso-Arroyo et al ^{Reference Alonso-Arroyo, Gonzalez-Alcaide, Bolanos Pizarro, Castello Cogollos, Valderrama-Zurian and Aleixandre-Benavent11} and Opgenhaffen et al ^{Reference Opgenhaffen, Vansteelandt, Germeau and Pieters12} studied female authorship in psychiatric literature but did not explore authorship of editorials.

In this study we aimed to determine the gender of all first authors of original articles and editorials published in the British Journal of Psychiatry and the Psychiatric Bulletin and editorials in Advances in Psychiatric Treatment from 1997 through 2008, and the geographical location of female first authors. All of these journals are peer reviewed and published by the Royal College of Psychiatrists. The British Journal of Psychiatry is one of the world's leading international journals, with a high impact factor. The Psychiatric Bulletin is also an international journal; it covers not only clinical aspects of psychiatry but also developments in psychiatric practice and service provision. Advances in Psychiatric Treatment is a bimonthly journal aimed primarily at promoting continuing professional development in psychiatry. The majority of editorials in all three journals are commissioned, but some are unsolicited. We hypothesised that over this 12-year period the proportion of female authorship would have increased.

Method

We examined all original articles and editorials published from January 1997 to December 2008 in the British Journal of Psychiatry and the Psychiatric Bulletin. Editorials in Advances in Psychiatric Treatment for the same period were also included in the study. British Journal of Psychiatry supplements were excluded as they cover selected areas and the editorial approach may differ significantly from the norm for the journal. International Psychiatry, also published by the Royal College of Psychiatrists, was excluded as it does not cover all of the study period. We determined the gender of the first author of each original article and editorial. Further information was collected regarding the discipline and geographical region, at the time of the publication, of female first authors of original articles.

The gender of the first authors was determined using a wide range of techniques, including personal knowledge, telephone contact, email contact, searching the General Medical Council website and using the Google internet search engine. In cases where the gender of the first author could not be determined, the authors were included in the total number of authors but excluded from the analysis.

We calculated the total number and percentage of articles by female first authors for each year and for each journal. The percentages were calculated based on the total number of authors whose gender was known. All data were collected and entered into a computer spreadsheet (Microsoft Office Excel 2003). We used descriptive statistics and simple linear regression analysis to test for significance with the aid of NCSS (2004) statistical software for Windows.

Results

Female authorship of original articles

A total of 1480 original articles from the British Journal of Psychiatry and 844 original articles from the Psychiatric Bulletin were included in this study. We were unable to determine the gender of the first author in 17 articles (1.15%) in the British Journal of Psychiatry and in 1 article (0.12%) in the Psychiatric Bulletin. Females were first authors of 33.6% of articles in the British Journal of Psychiatry and 37.6% of articles in the Psychiatric Bulletin published during the study period.

The percentage of articles with female first authors in the British Journal of Psychiatry showed a statistically significant increasing trend (P = 0.0128) over time (Fig. 1). The lowest female authorship contribution to the British Journal of Psychiatry was in 2002, with 22.5%, while the highest was in 2005, with 42.1%.

Fig 1 Female first authors in the British Journal of Psychiatry, 1997–2008.

In the Psychiatric Bulletin, the lowest female author contribution was in 2005, with 25.5%; contributions in other years were consistently over 30%, with the highest contribution of 46.8% achieved in 2004. There was no statistically significant increase (P = 0.85) in female first author contributions to the Psychiatric Bulletin during the study period (Fig. 2).

Fig 2 Female first authors in the Psychiatric Bulletin, 1997–2008.

Female psychiatrists were responsible for almost half (225/492; 45.7%) of female first authors' contributions to the British Journal of Psychiatry and almost three-quarters (232/317; 73.2%) of female first authors' contributions to the Psychiatric Bulletin. The other authors were female researchers with an educational background in psychology, epidemiology, public health, pharmacy, nursing, sociology, health economics, statistics or health administration, and a small number of unknown professions. Some of these researchers hold positions as senior research fellows, readers and professors. Analysis of female first authors of original articles with regard to their geographical region is presented in Table 1.

Table 1 Female first authors of original articles by geographical region

	British Journal of Psychiatry (article total: N = 492)		Psychiatric Bulletin (article total: N = 317)
Region	n	%	n	%
UK	302	61.4	304	95.9
Europe	101	20.5	10	3.2
North America	49	10.0	1	0.3
South America	3	0.6	0	0
Africa	3	0.6	1	0.3
Asia	11	2.2	0	0
Australia	23	4.7	1	0.3

Discussion

In this study, we focused specifically on the trend of first female authorship in the journals of the Royal College of Psychiatrists. The significant increase in women's contributions to the British Journal of Psychiatry mirrored what has been reported in a study by Alonso-Arroyo et al. ^{Reference Alonso-Arroyo, Gonzalez-Alcaide, Bolanos Pizarro, Castello Cogollos, Valderrama-Zurian and Aleixandre-Benavent11} Similar increases have also been reported in journals of epidemiology, ^{Reference Dickersin, Fredman, Flegal, Scott and Crawley7} intellectual disability (mental retardation) ^{Reference Porter, Christian and Polling8} and surgery. ^{Reference Bhattacharyya and Shapiro14} This could reflect an increase in the number of original articles being submitted by females or a higher acceptance rate of articles by females for publication, or a combination of both factors. However, women appeared underrepresented as first authors in both the British Journal of Psychiatry (33.6%) and the Psychiatric Bulletin (37.6%), accounting for about a third of all original articles published over the 12-year study period.

We did not investigate the underlying causes of these gender discrepancies in authorship; however, several studies have explored and discussed reasons for the low proportion of female authorship in scientific journals, including a lower rate of acquiring research grants, ^{Reference Dickersin, Fredman, Flegal, Scott and Crawley7} family commitments (in particular child-rearing), ^{Reference Kurichi, Kelz and Sonnad15,Reference Wilson and Eagles16} and personal career choices, especially with women devoting more of their time to teaching and clinical activities than to research. ^{Reference Carr, Ash, Friedman, Scaramucci, Barnett and Szalacha17} Although a number of these factors have not been investigated well, a Swedish study published in Nature in 1997 provided strong evidence of gender discrimination, showing that women in Sweden needed to be two and a half times more productive in terms of publications than their male counterparts to get the same rating for scientific competence. ^{Reference Buckley, Sanders, Shih and Hampton18} This is particularly interesting, as this was at the same time that the United Nations officially ranked Sweden as the country that performed best worldwide in terms of gender equality. This apparent discrimination may in turn affect opportunities for research grants, publication and academic progression. Yedidia & Bickel suggested that gender inequities often occur early in the academic training process and affect career trajectory and, ultimately, the gender distribution and skill mix of the health workforce. ^{Reference Wenneras and Wold19}