Tool for considering sex differences in guideline development

1. Introduction

Scientific evidence is an important basis for clinical decision-making in current medical practice. Sex characteristics can sometimes lead to different outcomes in certain populations, but this aspect is not always specifically considered in scientific research. Translating the results of scientific research into clinical practice requires specific knowledge about the influence of sex on the course and effectiveness of interventions and requires the skills for translating these results into a guideline. When a clinical guideline is developed, it is important to systematically involve someone who has the specific knowledge and skills to contribute to researching, evaluating and summarising scientific evidence. Research shows that guideline developers often lack the appropriate skills.

2. Objective of this tool

This tool uses examples to explain why it is relevant for medical practice to consider sex differences and describes how this can systematically be taken into account during the various guideline development stages. If there is a lack of evidence of the influence of sex on the course and effectiveness of interventions, guideline developers should use practical knowledge and experience when developing the sex-specific recommendations. Although this tool does not consider other differences between patient groups (e.g. ethnic or socio-economic differences), this method can also be used to take these topics into account.

3. Definitions

To understand the relevance of sex differences, it is necessary to be aware that sex differences are referred to in different ways in the literature. The most common terms in the English scientific literature are ‘sex differences’ and ‘gender differences’. The term ‘sex differences’ refers to the biological and physiological differences between men and women, while ‘gender differences’ refers to the differences based on socially constructed roles, behaviours, activities and characteristics that a society considers suitable for men and women. Both sex and gender differences can have effects on health care outcomes. When this tool uses the term ‘sex differences’, it is referring to both the biological and physiological differences as well as the socially constructed differences.

4. Why consider sex differences?

The scientific literature describes many examples of when different approaches are needed for the prevention, screening, diagnosis and treatment of male and female patients. For example, coronary heart disease in women can present at a later age and with different syndromes and symptoms. A number of these examples are described in Box 1 below.

Box 1. Examples of evidence that point to sex differences in prevention, screening, diagnostics and treatment.

The use of aspirin for the primary prevention of stroke and myocardial infarction: a meta-analysis that was broken down by sex showed that aspirin decreased the risk of stroke in women by 19% when compared with a placebo (RR, 0.81; 95% CI, 0.69–0.96, p = 0.01). The risk of myocardial infarction did not decrease (RR, 0.99; 95% CI, 0.83–1.19, p = 0.95). In men, aspirin reduced the risk of myocardial infarction by 32% (RR, 0.68; 95% CI, 0.54–0.86, p = 0.001) but did not reduce the risk of stroke (RR, 1.13; 95% CI, 0.96–1.33, p = 0.15).

The AUDIT screening instrument for identifying alcohol problems: scores can range between 0 and 40, and the score of 8 is used as the standard cut-off score indicating an alcohol problem. The instrument is less sensitive (66 versus 91 p <0.05) and more specific (97 versus 80 p <0.05) for women than for men, suggesting different cut-off points (8 for men and 5 or 6 for women).

Women with chronic obstructive pulmonary disease (COPD) are more likely to have severe dyspnea (one of the symptoms of COPD) than men (OR 1.30, 95% BI 1.10–1.54), but are less likely than men to undergo a spirometry to determine the diagnosis of COPD (OR 0.84, 95% CI 0.72–0.98). In this study’s population, the women were younger than the men (mean age 61.2 (SD 10.5) versus 64.4 (SD 11.0)), and the women had smoked for significantly fewer years than the men (36.9 (SD 29.6) versus 46.6 (SD 35.1) p <0.05).

Men respond better to imipramine (a tricyclic antidepressant) than sertraline (SSRI) (62% versus 45%, P = 0.04), while women display the opposite results (46% versus 57%, P = 0.02).

In addition to the above-mentioned sex differences, there are also differences based on reproductivity. Pregnancy, breast feeding or fertility may be important factors in the clinical strategy considered.

5. Systematic consideration of sex differences in guideline development

The differences described in the scientific literature give reason to systematically consider the differences between male and female patients. This tool offers the following instruments for each step in the guideline development process:

  1. Composition of the development group
  2. Analysis of clinical care gaps
  3. Literature research
  4. Analysis of the literature
  5. Formulation of conclusions and recommendations
  6. Identification of possible ‘knowledge gaps’

5.1. Composition of the development group

Research has shown that incorporating sex-specific expertise in the guideline development group can have a positive impact on the consideration given to sex and the eventual introduction of scientific evidence in the area of sex differences. Considering that sex expertise will not always be available, the chair, secretary or other development group members can also take this responsibility. It is essential to have the support of the entire group in considering sex differences throughout the guideline development process.

5.2. Analysis of clinical care gaps

During the analysis of clinical care gaps, the development group will need to consider whether sex differences need to be taken into account for each individual subtopic. A checklist has been created for the development group members to test their own expertise in determining whether sex differences play a role. If they expect sex differences to play a role, they can formulate specific review questions or refine an existing one.

5.3. Literature research

The guideline development group can search for literature on sex differences using sex-specific search terms. They can combine these search terms with the ones used to answer review questions. In this way, the search terms can function as a sex-specific search filter. The sex-specific terms are then used for the topics in which the analysis of clinical care gaps revealed that sex differences may play a role. The search terms depend on the database to be searched. Search terms are formulated below for Medline.

When searching in Medline via PubMed, a development group can use the following search terms:

"sex ratio [MeSH]," "sex distribution [MeSH]," "sex factors [MeSH]," "sex characteristics [MeSH]," 
"sex differentiation [MeSH]," "gender differences [tw]," "sex differences [tw]," "gender [tw]."

OvidSP can also be used to search in Medline (Moerman et al., 2007). This method is also used by information specialists. The following filter can be used for OvidSP:

#1 (gender$ or sex$).af.
#2 (boys or girls).tw.
#3 (women or men).ti.
#4 (male$1 or female$1).ti.
#5 (women or men).ab./freq=4
#6 (male$1 or female$1).ab./freq=4
#7 (women adj8 men).ab.
#8 (female$1 adj8 male$1).ab.
#9 or/1–8

The choice of the search terms depends on the National Library of Medicine and on trends in the literature, and should be regularly revised by information specialists.

Since sex differences can also be ‘hidden’ from search terms, for example, if they are described in subgroup analyses, it is important that the different health effects for men and women be taken into account when analysing the remaining literature.

5.4. Analysis of the literature

A guideline development group can use the instrument below to determine whether the selected literature contains information relating to sex differences:

Box 2. Sex differences (Keuken et al., 2007)

Do the title and the abstract provide information about sex, men/women, both sexes or a comparison between the sexes?

If so, then that article can be analysed for information about sex differences using the following questions:

  • Does the research question apply to both men and women?
  • What is the composition of men and women in the research group?
  • Have both sexes been sufficiently represented in the research group (in your opinion)?
  • Is a comparison made between men and women?

If so:

  • Has the subgroup analysis been performed correctly?
  • Are sex differences clinically relevant (in your opinion)?

If not:

  • Do the conclusions apply to both sexes (in your opinion)?

If an article contains a subgroup analysis between men and women, the development group will first verify whether the subgroup analysis was performed correctly. Next, they will assess whether there are significant differences between the two sexes. A publication by Altman & Bland (2003) may be helpful in assessing the accuracy of the subgroup analysis and how this analysis can be corrected if needed.

5.5. Formulation of conclusions and recommendations

The development group needs to clearly describe the population when presenting the conclusions from the literature. They should try to avoid generalising the research results. Lack of evidence for one of the two sexes will be discussed in the conclusions if the development group deems it relevant. If there is insufficient evidence for one of the two sexes, the development group may decide to extrapolate the recommendations to the other sex or limit the recommendation to one of the sexes.

5.6. Identification of possible ‘knowledge gaps’

During the development of the guideline, the development group may suspect a sex difference but have been unable to find any direct evidence, or only found evidence for one of the sexes. This provides a good opportunity to identify knowledge gaps in the area of sex differences.

6. Literature

  • Keuken DG, Haafkens JA, Moerman CJ, Klazinga NS, ter RG: Attention to sex-related factors in the development of clinical practice guidelines. J Womens Health (Larchmt ) 2007, 16: 82-92.
  • World Health Organization; What do we mean by 'sex' and 'gender'? . 2011. 10-2-2011.
  • Charney P: Corony artery disease in women. New York: American College of Physicians; 1999.
  • Keuken D, Bindels P, Klazinga N, Haafkens J: A systematic approach for uptake of evidence on sex-specific issues in guidelines - a pilot study. J Eval Clin Pract 2010.
  • Keuken DG, Haafkens JA, Mohrs J, Klazinga NS, Bindels PJ: Evaluating the effectiveness of an educational and feedback intervention aimed at improving consideration of sex differences in guideline development. Qual Saf Health Care 2010, 19: e18.
  • Hellema MJ, Haafkens JA, ter RG, Moerman CJ. Aandacht voor sekse in richtlijnontwikkeling; Training in het kader van het project Diversiteitsconsultatie voor richtlijnontwikkelaars. 2005. Huisartsgeneeskunde, Academisch Medisch Centrum-Universiteit van Amsterdam.
  • Moerman C, Deurenberg R, Haafkens J: Locating sex-specific evidence on clinical questions in MEDLINE: a search filter for use on OvidSPTM. BMC Medical Research Methodology 2009, 9: 25.
  • Keuken DG, Haafkens JA, Hellema MJ, Burgers JS, Moerman CJ: Incorporating a gender perspective into the development of clinical guidelines: a training course for guideline developers. Implement Sci 2007, 2: 35.
  • Altman DG, Bland JM: Interaction revisited: the difference between two estimates. BMJ 2003, 326: 219.
  • Ridker PM, Cook NR, Lee IM, Gordon D, Gaziano JM, Manson JE et al.: A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med 2005, 352: 1293-1304.
  • Reinert DF, Allen JP: The Alcohol Use Disorders Identification Test (AUDIT): a review of recent research. Alcohol Clin Exp Res 2002, 26: 272-279.
  • Cherpitel CJ: Analysis of cut points for screening instruments for alcohol problems in the emergency room. J Stud Alcohol 1995, 56: 695-700.
  • Watson L, Vestbo J, Postma DS, Decramer M, Rennard S, Kiri VA et al.: Gender differences in the management and experience of Chronic Obstructive Pulmonary Disease. Respiratory Medicine 2004, 98: 1207-1213.
  • Kornstein SG, Schatzberg AF, Thase ME, Yonkers KA, McCullough JP, Keitner GI et al.: Gender differences in treatment response to sertraline versus imipramine in chronic depression. Am J Psychiatry 2000, 157: 1445-1452.