Trends and projections in sexually transmitted infections in people aged 45 years and older in England: analysis of national surveillance data

Aims: We describe the epidemiology of sexually transmitted infections (STIs) and HIV in people aged ⩾45 years in England and provide future projections about the burden of STIs in this age group. Methods: Analysis of national surveillance data in England from 2014 to 2019 for chlamydia, gonorrhoea, herpes, syphilis, anogenital warts and HIV was carried out. Time trends were assessed by the Poisson regression and reported using incidence rate ratios (IRRs). Two scenarios were modelled to predict the number of new STI diagnoses and associated costs in 2040. Results: In 2019, there were 37,692 new STI diagnoses in people ⩾45 years in England. Between 2014 and 2019, there was a significant increase in the rate of new STI diagnoses in men (IRR = 1.05, p = .05) and those aged 45–64 years (IRR = 1.04, p = .05). Absolute numbers of new STI diagnoses in men who have sex with men increased by 76% between 2014 and 2019 (IRR = 1.15, p < .001). In adults aged ⩾50 years, the number of episodes of care for HIV increased over time (age = 50–64 years, IRR = 1.10; age = 65+ years, IRR = 1.13; p <.001). The modelled scenarios predicted an increase in STI diagnoses and costs in older people by 2040. Conclusion: STI rates in England are increasing in people aged ⩾45 years. The population is ageing and older people will contribute an increasing burden to STI costs if this trend continues. The reasons for this trend are not fully understood and further longitudinal epidemiological research is needed. Sexual health promotion campaigns and healthcare interventions targeted at older people should be prioritised.


Peer review
for Health and Care Excellence (NICE) recently introducing the recommendation that healthcare professionals should ask older people about their sexual history and identify their risk of STIs. 4 In stark contrast to this, the WHO's 2021 report on global progress on STIs did not even report prevalence rates of STIs for people aged over 49 years. 50][11] However, it remains unclear what factors are driving the increasing STI rates in older people, with physiological, behavioural and social theories being proposed. 11ike many high-income countries, England has an ageing population, with those aged over 45 years forecast to grow by 4.4 million people between 2018 and 2038. 12There is a growing body of data on sexual activity in older people, including those who remain sexually active into their 70s and 80s, with regular sexual expression associated with various positive health outcomes. 9,13Therefore, it is increasingly important to understand the evolving epidemiology of STIs in older adults so that behavioural factors, service delivery, and clinical and public health interventions can be ready to respond.This article has the following two key aims: 1. To describe the epidemiology of STIs and HIV in people aged ⩾45 years in England between 2014 and 2019.2. To estimate the future burden of STIs in this age group in England.

Methods
This is a cross-sectional analysis of national surveillance STI data in England from 2014 to 2019.

Data sources
The National Institute for Health Protection (NIHP) -formerly Public  18 HIV was excluded from modelling as this is an established discipline with recommended methodologies which were outside the scope of this article. 19

Data analysis
The unit of analysis is a new episode of infection so individuals may be duplicated in the data set if they have been diagnosed with more than one STI.The rates of new infections per 100,000 population were calculated using the ONS population estimates. 17

Economic modelling
Current National Health Service (NHS) recommendations for the treatment of STIs (excluding HIV) were used to estimate the associated costs (from an NHS perspective) using a bottom-up approach. 20As HIV is a lifelong condition, rather than modelling costs for new cases, the number of episodes of secondary care was used as a measure of resource use.NHS unit costs were derived from the most recent versions of the NHS reference costs database, 21 Personal Social Services Research Unit (PSSRU) unit costs of health and social care, 22 and the NHS drug tariff. 23Unit costs included in the analysis are reported in Supplemental Appendix 1.The unit cost (per case/ episode) was multiplied by the predicted number of cases/episodes for each model scenario.

Ethics and patient and public involvement
No ethical approval was required for this secondary analysis of PHE data.There was no direct patient or public involvement in this analysis.

Peer review results
In

Trends over time
Between 2014 and 2019, there were 198,144 new STI diagnoses in people aged 45 years and over (Table 2).Over this time period, there was a significant increase in the rate of new gonorrhoea diagnoses (IRR = 1.16, p = .01)and chlamydia diagnoses (IRR = 1.10, p = .03).The majority of new STI diagnoses were in men (n = 143,465, 73.7%).People in the 45-64 years age group accounted for 92.6% of STI diagnoses in those aged over 45 years (n = 184,120).The most common route of transmission in 2014 was in heterosexual men, who accounted for 40.4% of new STI diagnoses (n = 12,481).Over time, this has changed, and the dominant route of transmission in 2019 was among MSM (n = 13,843, 36.7%), which increased by 76% between 2014 and 2019 (IRR = 1.15, p < .001).

Trends over age and time
Younger people continue to bear the highest burden of STIs; between 2014 and 2019, people aged under 45 years accounted for 92.5% of new STI diagnoses (n = 2,456,033).However, in the youngest age groups, there is a significant downwards trend in the rate of new STI diagnoses while in the older age groups, including those aged 45-64 years, the rates of STI are increasing (Table 2).

HIV
In 2019, there were 850 new HIV diagnoses in people aged 50 years and over in England (Table 1).This decreased from 1051 new diagnoses in 2014, although the trend failed to reach significance (IRR = 0.94, p = .61).Between 2014 and 2019, the over 50s population increased by 1.7 million people, from 19.4 to 21.0 million.In this period, there were 5336 new HIV diagnoses in older people, 72% (n = 3834) of which were in men (vs women).The ratio of male-to-female diagnoses remained stable at 2.5:1.There were no significant trends over time for new HIV diagnoses in adults aged ⩾50 years.
In 2019, there were 38,260 episodes of HIV care for older people in England, which represents 42.4% of all HIV care episodes (Table 3).This is an increase of 13.1 percentage points in the proportion of adults aged ⩾50 years receiving HIV care from 2014.Episodes of HIV care increased significantly in both the 50-64 years (IRR = 1.10, p < .001)and the over 65 years age groups (IRR = 1.13, p < .001).Less than 2000 older adults received a late HIV diagnosis between 2014 and 2019.There was a significant decreasing trend for late diagnosis in adults aged 50-64 years (IRR = 0.95, p = .001)and those aged over 65 years (IRR = 0.92, p = .01).

HIV episodes of care -trends and costs over age and time
There has been a downwards trend over time in HIV episodes of care (EoC) for those in younger age groups, while in adults aged 50 years and over, the number of EoC has increased over time (age = 50+ years, IRR = 1.11, p < .001),which partly reflects the changing demographic of people living with HIV over time.In terms of NHS costs for HIV care, there was a net increase of £3.4m between 2014 and 2019.This was driven by the over 50 years age group whose costs increased by £4.3m, whereas in younger age groups, costs decreased by £0.88m in this period (Table 3).

Activity and cost projections
The total number of diagnoses for chlamydia, gonorrhoea, herpes, syphilis and warts in 2019 among older people in England was 28,660.The cost of treating these STIs in 2019 was estimated to be £2.97m.Both scenarios projected an increase in STI diagnoses in people aged over 45 years by 2040 (Table 4).By 2040, annual NHS treatment costs for STIs in people aged 45 years and over are estimated to be between £3.1m (assuming STI rates remain the same but reflecting demographic population change) and £40.5m (assuming the changes in STI rates observed between 2014 and 2019 continue).

dIscussIon
Between 2014 and 2019, there was a significant increase in the rate of new STI diagnoses in people aged 45-64 years, with new gonorrhoea and chlamydia diagnoses roughly doubling over the 6-year period.Specifically, 2019 saw 37,692 new STI diagnoses in England in people aged 45 years and over, which was 8% of the total STI diagnoses.MSM and the 45-64 years age group have seen the highest increases in new STI diagnoses.Episodes of HIV care have also significantly increased in people aged 50 years and over.The modelled scenarios predicted an increase in STI diagnoses and costs in older people by 2040.

Strengths and limitations
This study used a large administrative data set which included all STI and HIV diagnoses in England in both specialist and non-specialist settings, thereby minimising bias.As far as we are aware, this is the first national analysis of the epidemiology of STIs in people aged over 45 years in England.We describe significant increases in STI rates in older people in England, consistent with global data from both high-and low-/middle-    ethnicity in this analysis.There was also a change in the way STI data were collected in 2015 which meant that 2014 data could not be used when analysing trends for all new STIs.Finally, we did not use transmission dynamic modelling which is more accurate than epidemiological models in predicting future trends in new STI diagnoses. 25his was due to a lack of data on additional factors which may be associated with changing STI rates in older people, for example, behavioural data and social networks. 26

Context
Only one comparable study was identified, from the West Midlands STI Surveillance Project. 3 The West Midlands is a region in England with a population of 5.9 million people.The rates of chlamydia, genital herpes, gonorrhoea and syphilis in people aged 45 years and over attending GUM clinical were analysed over an 8-year period from 1996 to 2003.The analysis included 4445 STI episodes in older people and found an overall doubling in STI rates over the study period.
Our analysis is broader in scope, in terms of geography, diagnoses and settings.Data cover the whole of England, with over 155,000 new STI diagnoses in older people, HIV is included, and activity data come from both specialist and non-specialist settings.Older people are less likely to attend a sexual health clinic after having condomless first sex with a new partner and/or ⩾2 partners and no condom use, 10 so this analysis may be more representative including older people presenting, for example, to primary care.The West Midlands analysis used rate ratios, comparing their first and last data points on the number of episodes to assess temporal trends.We used incident rate ratios from the Poisson regression to assess trend, which uses data from all available years and takes into account the population at risk in each year.Our method found more modest changes in STI rates, but we believe it is more robust, and would allow comparison with future analyses looking at changes over different periods of time.The West Midland analysis did collect disaggregated data allowing a more in-depth data analysis describing activity in four narrower age bands, whereas we were limited to two broader groups.However, despite having ethnicity data, Peer review they reported they were unable to use this in their analysis due to small numbers.
The results of this study are consistent with the emerging global literature of STIs in older people.Surveillance data from the US, Canada and Australia all show increases in diagnosis rates of STIs in older people and older people living with HIV/ AIDS. 24,27,28Older people are often excluded from epidemiological analysis of STI trends so there is a dearth of comparable studies. 5,29

STI risk factors for older adults
A range of physiological, sociobehavioural and structural factors have been proposed as explanations for the increasing STI incidence among older adults. 30Ageing is associated with reductions in immune response which could increase susceptibility to infection. 31Older women may be at an increased risk of acquired STIs due to reduced vaginal lubrication and thinning of the vaginal mucosa. 27Decreasing testosterone levels in men can lead to erectile problems which make condom use more difficult, contributing to increased STI risk. 27The widespread use of Sildenafil (Viagra) has allowed men to engage in penetrative sex later in life. 32onsistent with research in younger men, older MSM who report Viagra use are more likely to engage in unprotected sex. 33ocial changes in recent decades, such as higher divorce rates, changing social attitudes and increasing foreign travel, have led to people taking new partners in later life, intergenerational relationships and engagement with sex workers. 11Older people are less likely to have had sex education at school, and are typically excluded from sexual health promotion programmes more often targeted at young people. 34These factors may lead to people being less able to negotiate safe sex in later life. 35lder people are less likely to seek help for STIs, have decreased condom use, lower rates of STI testing and delayed presentation for treatment. 10,36n a US sample of people aged 40-80 years, over 75% of those who experienced a sexual health problem did not seek help from a health professional. 37Healthcare professionals are less likely to initiate sexual health conversations with older people; the mutual reluctance from older people and healthcare professionals to raise sexual health concerns in a clinical setting present significant barriers to treatment. 38

Implications
The estimated annual treatment costs for cases of STIs in people aged over 45 years in 2019 were around £3m.The modelled scenarios for 2040 suggest that costs could increase modestly due to demographic changes (if STI rates remain as they are in 2019) or increase by more than 10-fold (if STI rates in this age group continue to change at the same rate as between 2014 and 2019).These estimates may be considered conservative and worstcase scenarios, respectively, but provide reasonable bounds of future healthcare spending in this area upon which commissioners can draw when planning sexual health services to meet the changing demands of the future as our population ages.
Policy-makers should continue to promote older people being asked about their sexual history in primary care settings 4 and also consider the sexual health of this age group in future public health campaigns.A key area for further research is longitudinal studies of sexual behaviour and evaluation of health delivery and prevention interventions targeted at older people.Also, there are very limited data on the relationship between ethnicity and STIs in older adults.We have made assumptions about trends in STI rates using population-based projections, future analyses could use transmission dynamic modelling using behavioural parameters specific to older people.conclusIon STI rates in England are increasing in people aged over 45 years.Although the numbers are small relative to younger people, this population demographic is increasing and people aged 45 years and over will contribute an increasing burden to STI costs if this trend continues.A combination of changing societal norms, structural barriers in accessing health care and knowledge about STIs and increased biological susceptibility are all likely contributors to these increasing STI rates.Further epidemiological research is needed to assess trends and causal associations.This article highlights a clear need for sexual health promotion campaigns and healthcare interventions targeted at people aged over 45 years.

Table 1
new stI diagnoses for people aged 45 years and over; new hIV diagnoses, episodes of care and late diagnoses for people aged 50 years and over in england by diagnosis, gender and age group.
STI: sexually transmitted infection; HIV: human immunodeficiency virus; IRR: incidence rate ratio; CI: confidence interval; MSM: men who have sex with men; WSW: women who have sex with women; M: male; F: female.a Includes other STIs so total different.b Not included in regression model.

Table 2
trends over age, gender and time for new stI diagnoses.Trends and projections in sexually transmitted infections in people aged 45 years and older in England

Table 3
trends and costs over age, gender and time for hIV episodes of care (eoc).
HIV: human immunodeficiency virus; EoC: episode of care; IRR: incidence rate ratio; CI: confidence interval.a Based on 2019 NHS reference cost -outpatient HIV (stable patients) £278 per episode.b Including where gender is 'unknown' or 'other'.

Table 4
cases of stIs and estimated treatment costs for 2019, with projections for 2040.
STI: sexually transmitted infection; CI: confidence interval; Model 1: demographic change model; Model 2: continuing trend model.Data for 2019 include only cases where both age and gender were known.a 30% of cases resolve without treatment.