Queer in the birth cohorts

Until recently, the vast majority of British birth cohort studies did not capture any aspect of sexuality or gender identity, making the study of queer individuals within the cohorts almost impossible (Tabor et al., Reference Tabor, Kneale and Patalay2023). While studies have asked questions on identities such as national and class identity and have covered sensitive topics such as familial abuse, illegal substances, and violence, it was not until 2012 that a UK national birth cohort fielded a sexuality question. This is despite huge social and attitudinal changes in the last fifty years and clear evidence of the social, economic, and health and well-being disparities experienced by LGBTQ+ people (Bécares, Reference Bécares2021; MacCarthy et al., Reference MacCarthy, Saunders and Elliott2021; Saunders et al., Reference Saunders, MacCarthy, Meads, Massou, Mant, Saunders and Elliott2021).

A further issue is the relatively small numbers of LGBTQ+ people within each study. This is partially a function of the ways in which birth cohorts are recruited as it is not possible to oversample LGBTQ+ people from birth as with other marginalised groups. The recent 2021 England and Wales census, the first national data on the size of the LGBTQ+ population, found that at least 3.2% of population, approximately 1.5 million people, identified as gay, lesbian, bisexual, or ‘other sexual orientation’ (ONS, 2023). While this is a large number of people, when applied proportionally to the sample recruited for birth cohort studies, LGBTQ+ numbers remain small – often in the low hundreds or less. As a result, opportunities for robust quantitative analysis are limited and result in the neglect of LGBTQ+ data even when it is captured.

Large quantitative exercises can produce and legitimise certain kinds of lives; through such tools, populations are made ‘knowable’ (Browne, Reference Browne, Browne and Nash2016). Heteronormative lifecourse events such as marriage and reproduction are at the centre of birth cohort logic but are arguably detached from ‘queer time’ (Halberstam, Reference Halberstam2005). ‘Queer temporalities’ create logics of life course timing and stages which are messy and/or stretched (Halberstam, Reference Halberstam2005; King and Hall, Reference King and Hall2023). For example, parental rejection and families of choice, coming out, and late-life marriage of partners due to legislative change.

Pen portraits

To address the methodological and theoretical challenges outlined above, the authors propose person-centred mixed-method pen portraits (Tinkler et al., Reference Tinkler, Cruz and Fenton2021).

Elliot (2005, p. 1) has argued that the ‘temporal or chronological dimension’ of longitudinal studies of the life course lends such studies ‘a certain narrative quality’, even when the data are purely quantitative. An early example of using quantitative longitudinal data to construct narrative life histories or pen portraits was Singer et al. (Reference Singer, Ryff, Carr and Magee1998). Arguing that typical quantitative approaches to longitudinal data obscure the complexity of individuals’ lives, Singer and colleagues used more than 250 variables (on topics including education, family, health, and employment) collected over three decades to construct written ‘life histories’ detailing the mental health trajectories of women in the Wisconsin Longitudinal Study. Sharland et al. (Reference Sharland, Holland, Henderson, Zhang, Cheung and Scourfield2017) used 300+ variables from the British Household Panel Survey to construct written narratives illustrating the pathways of families using social work. These narratives were then analysed to investigate questions about the impacts of social work on families. These two studies are examples of applying ‘narrative practices’ (Sharland et al., Reference Sharland, Holland, Henderson, Zhang, Cheung and Scourfield2017, p. 668) to purely quantitative datasets. The British birth cohorts include a wide range of quantitative data that could be used for narratively orientated biosocial research. The cohorts capture genetics, blood biomarkers such as cortisol, environmental and occupational exposures, and measures of physical and mental health and functionality, amongst a wide array of other topics.

The birth cohort studies also include small but potentially valuable amounts of qualitative data in the form of open-text survey responses and semi-structured interview material that can be used to explore cohort members’ subjective experiences and perspectives (Carpentieri et al., Reference Carpentieri, Elliott, Brett and Deary2016; Carpentieri and Elliott, Reference Carpentieri, Elliott, Y., D., Cooper, Hardy and Richards2013; Elliott, Reference Elliott2013, Reference Elliott2008). The presence of qualitative data affords opportunities for the construction of mixed methods pen portraits that are richer than those based solely on quantitative data. For example, Carpentieri et al., (Reference Carpentieri, Carter and Jeppesen2024) used a combination of closed and open text survey responses to examine trajectories of education, employment, and social mobility in post-war Britain, while Carter (Reference Carter2023) shed valuable light on gendered experiences of education and employment.

Such studies illustrate the potential of narrative pen portraits to illuminate diversity, non-linearity, and complexity instead of smoothing differences (Sharland et al., Reference Sharland, Holland, Henderson, Zhang, Cheung and Scourfield2017). By supporting a more holistic, person-centred approach to complex questions and the mapping of important transitions and timings (Singer et al., Reference Singer, Ryff, Carr and Magee1998), person-centred pen portraits can reveal the unique experiences, perspectives, and trajectories underpinning broader trends. Pen portraits also afford the potential to excavate marginalised lives (such as those of LGBTQ+ study members) from the broader birth cohort samples (Tinkler et al., Reference Tinkler, Cruz and Fenton2021). Furthermore, pen portraits encourage the reuse of data, reducing the burden on participants and highlighting the value of their time and effort (Guyan, Reference Guyan2022).

However, there are a number of limitations that must be considered. First, mixed methods pen portraits may present a limited picture of participant voice, while purely quantitative pen portraits lack participant voice altogether (Carpentieri et al., Reference Carpentieri, Elliott, Brett and Deary2016; Elliott, Reference Elliott2008; Sharland et al., Reference Sharland, Holland, Henderson, Zhang, Cheung and Scourfield2017). Individual participant voice is further lost in the construction of generalised cases which combine the experiences of multiple individuals (Elliott, Reference Elliott2008; Sharland et al., Reference Sharland, Holland, Henderson, Zhang, Cheung and Scourfield2017). Pen portraits require strict adherence to principles of anonymity and confidentiality, in order to protect cohort members’ identities, especially for a stigmatised minority group such as LGBTQ+ people. Finally, while attempts can be made to find queer lives in the cohorts, the datasets and surveys were not developed to make such work easy. Nonetheless, below is an example of how queer lives could be excavated and recomposed from a British birth cohort.

1970 British cohort study – a working example

The 1970 British Cohort Study (BCS70) is an ongoing longitudinal study of 17,000 individuals born in England, Scotland, and Wales in 1970 (Elliott and Shepherd, Reference Elliott and Shepherd2006), collecting data on a wide range of topics including family, health, education, socioeconomic conditions, and social networks (Elliott and Shepherd, Reference Elliott and Shepherd2006). The most recent survey wave captured data from 8,581 respondents aged 46 years (Brown and Peters, Reference Brown and Peters2019). At the time the BCS70 was being established, the legal and attitudinal landscape for LGBTQ+ people was changing. In 1967, homosexual acts between men over 21 had been partially decriminalised in England and Wales, 1969 saw the Stonewall riots in the USA, and the first Pride was to be held in London in 1972 (Stonewall, 2016). As a result, this cohort came of age during the 1980s and 1990s movement for LGBTQ+ rights and equality, as well as the HIV/AIDs epidemic. However, it was not until 2012, when the participants were aged 42 years, that BCS70 cohort members were asked if they identified as gay, lesbian, or bisexual.

Of 9,764 BCS70 respondents asked, 286 (2.9%) identified as gay, lesbian, bisexual, or other. This group is made up of a roughly equal proportions of men and women across a range of employment and income brackets. Around 30% of the group also identified as an ethnic minority. While potentially a rich source of information about queer adults and their lives over the last fifty years, such numbers are difficult to use statistically and using traditional quantitative methods force a lumping together which obscures differences. However, pen portraits would allow for a detailed exploration of LGBTQ+ respondents in BCS70.

Recent UK evidence using large cross-sectional population studies has identified that lesbian, gay, and bisexual women may be at a higher risk for heart attacks and stroke; however, this was not consistent across the studies (Jacob et al., Reference Jacob, Lopez-Sanchez, Kostev, Schnitzler, Haro, Koyanagi, McDermott, Il Shin and Smith2021; Saunders et al., Reference Saunders, MacCarthy, Meads, Massou, Mant, Saunders and Elliott2021). This is part of a wider pattern of evidence which suggests that whereas LGBTQ+ people may be more likely to be overweight, smoke, and consume more alcohol, and this does not necessarily translate into a higher risk of chronic conditions such as cardiovascular disease and diabetes (Booker et al., Reference Booker, Rieger and Unger2017; Eliason, Reference Eliason2014). Some have argued that this mismatch is due to “resilience,” for example, through community resources and networks (Foley et al., Reference Foley, Stanton, Shepard, Morris, O’Cleirigh, Potter and Batchelder2022; Follins et al., Reference Follins, Garrett-Walker and Lewis2014; Fredriksen-Goldsen et al., Reference Fredriksen-Goldsen, Kim, Bryan, Shiu and Emlet2017). The construction of pen portraits would allow for the examination of pathways into and out of good health while highlighting BCS70 participants’ historical and social context.

A pen portrait project could compare the pathways and experiences of LGBTQ+ study members with and without chronic health outcomes. While the method is exploratory and therefore all variables and data could be considered relevant to the analysis, there are a number of particularly relevant topics captured by the cohort study. For example, a battery of biological markers such as anthropometry and blood biomarkers were collected by a nurse and can indicate underlying functionality and physiological stress. This information can be placed temporally and conceptually alongside participant’s social and environmental context, for example, their communities, socioeconomic conditions, partners and family, and neighbourhood. These variables would provide the basis to consider how biology, environment, and the social interact and combine over the life course to produce ill-health within stigmatised bodies. Likewise, they allow the researcher to consider a more contextual and even destabilised understanding of “health,” “lifecourse,” and “resiliency” where an individual’s agency and coping are foregrounded. Open-text survey responses over the life course could also contribute to such pen portraits. Additional analysis could combine these data-rich individual narratives into aggregated accounts of biosocial pathways into better and poorer health for comparison and interpretation.