During the emerging HIV/AIDS epidemic in the 1980s, little was known about the sexual behaviour in the general population.
Data were urgently needed by infectious disease modellers and the public health community, and the idea of a nationally representative sex survey was born. A feasibility study trialled an interviewer-administered interview, with a self-completion for the most sensitive questions, and found that, despite much scepticism, such a survey was acceptable to participants, and produced high-quality data. However, getting the survey off the ground was not straightforward. Although government funding through the Economic and Social Research Council had been promised, there continued to be political concern that such a survey would “invade the privacy of the 20,000 people due to be questioned” leading to 10 Downing Street cancelling the funding of the survey. This was met by an outcry from the scientific community, but fortunately the Wellcome Trust quickly stepped in to fund the survey. Natsal-1 (fieldwork 1990-1991) became a valuable resource for understanding not just the transmission of HIV, but also sexual behaviour more broadly.
By the late 1990’s, HIV models were still using data from that first survey, which had become outdated. Clinic data showed that Sexually Transmitted Infections (STIs) were rapidly increasing, and there was particular concern about the prevalence of chlamydia, for which a new screening programme was being planned. In parallel, ongoing concerns about the UK’s high teenage pregnancy rates were being addressed by the Teenage Pregnancy Strategy, bringing reproductive health further up the policy agenda. To provide key data needed for these initiatives, Natsal-2 (fieldwork 1999-2001) included a broader range of topics relating to sexual and reproductive health, and biological samples (urine) to estimate the prevalence of chlamydia and HPV in the general population. The methods adapted to technological advances – moving from pen-and-paper to a computerised questionnaire. The sample also boosted the number of participants from four ethnic minority groups. This time the survey was publicly funded, by the Medical Research Council.
Ten years on from Natsal-2, updated data were yet again needed. By now, the way the scientific and public health community thought about sexual health had broadened beyond just STIs and reproductive health (‘bugs and babies’), to include concepts like pleasure, and sex free from coercion, discrimination and violence. Natsal-3 (fieldwork 2010-2012) collected data on a broad range of topics to reflect this, and extended the age range up to 74 years (previously 59 years in Natsal-1 and 44 years in Natsal-2) to examine sexuality and health across the life course. Urine samples were this time tested for a wider range of STIs including HIV, and morning saliva samples were collected to measure testosterone.
The next survey
The decade since Natsal-3 has seen huge societal shifts, particularly in the role of digital technology in all aspects of people’s lives, as well as major changes to health service funding and commissioning. Then in 2020 the Covid-19 pandemic transformed private lives and public services, with the duration and extent of these impacts currently unknown. Natsal-4 (fieldwork 2022-2023) will provide much-needed data to understand how these changes relate to sex and relationships. The revised Natsal-4 questionnaire seeks to be more inclusive of diverse gender identities, and includes a wide range of new topics including digital technology, sexual wellbeing, sexual harassment and violence. Natsal-4 will continue to collect urine samples and, for the first time, will invite eligible participants to give self-collected vaginal swabs. The survey will also request consent to linkage of the study data to a range of administrative datasets. Natsal-4 is complemented by the Natsal-COVID study, a self-selected (non-probability) web panel survey conducted at two points during the pandemic (July 2020 and March-April 2021).
Methodological work
Methodological rigour is critical to a survey like Natsal, which faces the same challenges as other social surveys, along with additional challenges and scrutiny that come from the subject matter. Development for the first study included detailed qualitative work exploring the appropriate language for general population studies to use when asking about sexual behaviour, as well as the feasibility study examining acceptability, representativeness, and data quality. Qualitative work with Natsal-2 respondents has explored factors that facilitate reliable reporting of sensitive behaviours, and cognitive testing, psychometric validation, and public engagement work has underpinned the questionnaire development for Natsal-3 and -4.
With Natsal-2’s move to computerised interviewing, an experiment comparing paper vs computerised (CASI) self-completion demonstrated higher data quality with CASI, but no difference in reporting of sensitive behaviours. Further experimentation on Natsal-3 around response rates found that doubling the incentive amount made little difference to response rates. Statistical work has examined reporting biases and impact of weighting on estimates.
Much work has been undertaken to assess alternatives to the probability sample face-to-face design, including comparisons with self-selected general population web panel surveys, with sexual health questions on a general population survey, and with convenience sample surveys of sub-populations of interest. More recently, as an adaptation to enable Natsal-4 fieldwork to be undertaken if COVID-19 restrictions are in place, alternative remote interviewing modes have been developed, which include an embedded remote self-completion and remote self-collected biological samples.
Impacts
Since its early impacts on the HIV and Teenage Pregnancy Strategies, Natsal has achieved a well-documented legacy of impact on policy and service planning. The data have been used to support sexual health strategies, the National Chlamydia Screening Programme, the Human Papillomavirus (HPV) vaccination programme, legislation on the age of consent, and statutory Sex and Relationships Education. Beyond policy, Natsal aims to encourage evidence-based journalism, teaching, and public discussions about sex. Some examples of this work are on our website, and we’re always open to supporting others to use the data in new ways, so please get in touch with ideas.
How to access the data
All Natsal data, including biosampling data, are available from the UK Data Service. The archived data have been used in more than 1500 projects by users in around 40 countries. The Natsal team are happy to support these by giving advice or taking part in collaborative projects.
Acknowledgements: Our thanks go to the thousands of participants who have given their time to the study, as well as the hundreds of interviewers whose hard work has made the study a success. Thanks also to the many researchers and operations staff who have contributed to the study over the years, and to the study funders.
Author biography: Soazig Clifton is Academic Director for Natsal at NatCen Social Research and University College London (UCL). She is a survey researcher and epidemiologist, and specialises in the design, delivery, analysis, and dissemination of sexual health research.