The number of individuals living with diagnosed HIV in the UK has increased three-fold over the past 10 years, rising from 17 911 in 1998 to 56 443 in 2007 . A contributing factor to this increase is the number of older adults (aged 50 years and over) living with diagnosed HIV rising sharply in the new millennium. To date, it is not clear as to what extent the increase is due to an ageing cohort of HIV-infected individuals experiencing better survival rates since the introduction of highly active antiretroviral therapy (HAART) or driven by newly acquired infections among adults in their 50 s or above.
HIV infection among older adults is associated with faster disease progression leading to higher rates of morbidity and mortality than those infected at a younger age [2–4]. Adults aged over 40 years have lower CD4 cell counts following seroconversion and steeper declines in CD4 progression . Comorbidities are also likely to be more significant among older adults , along with a greater risk of misdiagnosis of HIV [6–8].
We aim to describe epidemiological trends in adults aged 50 years and over living with diagnosed HIV in England, Wales and Northern Ireland (E,W&NI) and compare these with adults aged 15–49 years, exploring late diagnosis and its impact on mortality for these groups. We also estimate age at infection and quantify the proportion of older adults who acquire their infection at age 50 years or older.
Data from three national surveillance systems held at the Health Protection Agency (HPA) Centre for Infections (CFI) were analysed: the HIV and AIDS New Diagnoses and Deaths database (to be referred to as new diagnoses), the CD4 Surveillance Scheme and the Survey of Prevalent HIV Infections Diagnosed (SOPHID).
Full descriptions of these surveillances are available on the HPA Web site [9–11]. Briefly, the new diagnoses database collects demographic and epidemiological information on adults (aged 15 years and older) newly diagnosed with HIV within E,W&NI.
SOPHID collects residential and epidemiological information on all individuals accessing HIV-related treatment and care services annually. The data provide a census of persons living with diagnosed HIV infection in the UK and inform the commissioning of HIV treatment and prevention services.
The national CD4 surveillance scheme monitors immunosuppression among HIV-infected individuals by collecting CD4 T lymphocyte (CD4 cell) counts from all laboratories. The data from the CD4 surveillance is matched to new diagnoses and SOPHID data. The combined data allow for the calculation of late diagnosis.
Analyses were conducted on all adults aged 15 years and over who were newly diagnosed or were accessing HIV-related care between 2000 and 2007 in E,W&NI. Persons aged 50 years or over were defined as older adults.
CD4 cell count and AIDS-defining illness at diagnosis
Adults reported to have a CD4 cell count less than 200 within 91 days of diagnosis were defined as diagnosed ‘late’. The proportion of adults with a clinical AIDS condition (irrespective of CD4 cell count) within 91 days of HIV diagnosis is also reported.
Death rate among older adults
Rate of death was calculated on a yearly basis, using data obtained from the SOPHID survey as the denominator. The number of deaths identified for a year was divided by the HIV prevalent population for that year and presented as a rate per 1000 population.
Estimated age at infection
Age at infection was estimated applying a Markov Chain model of HIV progression based on CD4 cell count at diagnosis . The model used a cohort of seroconverters among men tested six monthly for HIV in San Francisco. Using this algorithm, an estimate of 8 years was allocated to persons with a CD4 cell count less than 200 cells/μl reported within 91 days of HIV diagnosis, 6 years for those with a CD4 cell count between 200 and 349 cells/μl, 4 years for those with a CD4 count between 350 and 449 cells/μl and 2 years for those with a CD4 cell count of 450 cells/μl or above.
Among adults first diagnosed with HIV in E,W&NI between 2000 and 2007 (49 795), probable route of infection and ethnicity was reported for 95% of individuals (47 539 and 47 343, respectively) and CD4 cell count within 91 days of diagnosis and probable country of infection for 76% (37 990 and 38 031, respectively). Among individuals diagnosed when aged 50 years and over, the respective proportions were 93% (3712), 94% (3765), 78% (3118) and 74% (2965).
Among HIV-infected adults residing in E,W&NI and reporting to SOPHID in 2007 (52 769), completion of route of infection, ethnicity and whether in receipt of antiretroviral therapy exceeded or was equal to 98%. Similar high variable completion was observed among older adults.
Data preparation and statistical analysis
Data preparation and calculation of an individual's age and date of infection were completed in Microsoft Access and statistical analysis was performed using STATA version 10 (Stata Corp., College Station, Texas, USA). Proportions are calculated as among those persons for whom the relevant information is available.
Older adults diagnosed with HIV
Adults aged 50 years and over accounted for 8.0% (4001/49 795) of all new HIV diagnoses between 2000 and 2007 in E,W&NI, increasing from 8.3% (299/3610) in 2000 to 9.7% (710/7291) in 2007. The absolute number of older adults diagnosed more than doubled over the period (from 299 in 2000 to 710 in 2007) (Table 1). The median age at diagnosis for all years combined was 55 years. Seventy-three percent (2925) of older adults were diagnosed between the age of 50 and 59 years, with the rest (1076) diagnosed when aged 60 years and over. Males accounted for three-quarters (74%, 2971) of older adults, compared with 58% (26 627/45 794) among those diagnosed prior to 50 years of age (P < 0.001).
No difference was found between median age and age distribution at diagnosis in older adults by sex and probable route of infection. A higher proportion of older adults were diagnosed outside of London as compared to younger adults (57%: 2298/4001 vs. 52%: 23 998/45 794; P < 0.001). The proportion of older MSM diagnosed outside London was also significantly higher than that among younger MSM (60%: 878/1468 vs. 45%: 6793/15 108; P < 0.001). There was no significant difference in the proportion of older and younger adults diagnosed outside of London among heterosexual men (59%: 722/1224 vs. 57%: 5517/9615; P = 0.28) or heterosexual women (52%: 481/924 vs. 57%: 10 117/17 818; P = 0.13).
By exposure category, MSM accounted for 40% (1468/3712) of diagnoses among older adults, with heterosexual men accounting for 33% (1224) and heterosexual women for 25% (924). The respective figures among younger adults diagnosed with HIV differ significantly (all P < 0.001). Among younger adults, 34% (15 108/43 827) were MSM, 22% (9615) were heterosexual men and 41% (17 818) were heterosexual women.
The majority of older MSM were of white ethnicity (94%: 1342/1423) and probably acquired their infection in the UK (84%: 735/873). As compared to younger men infected heterosexually, older men infected heterosexually were more likely to be of white ethnicity (48%: 582/1209 vs. 15%: 1431/947; P < 0.001) and to have reported their probable region of infection as Asia (18%: 204/1109 vs. 6%: 530/8918; P < 0.001). The difference in the proportion of heterosexual men reporting Asia as probable region of infection was particularly marked among those of white ethnicity (86%; 171/200 vs. 61%: 319/521; P < 0.001). Although the majority of older heterosexual women were of black ethnicity (68%: 624/912), a significantly higher proportion were of white ethnicity as compared to younger women (21%: 191 vs. 9.3%: 1644; P < 0.001).
Injecting drug users represented 0.9% (33) of older adults compared with 2.4% (1039) of younger adults (P < 0.001). Among older adults infected through injecting drug use, 87% (27/31) were of white ethnicity, 64% (14/22) reported their probable country of infection as the UK and 23% (5) as western Europe. For younger adults, these proportions were 84% (813/967; P = 0.07), 44% (283/638; P = 0.27) and 24% (151; P = 0.65), respectively.
Older adults accessing HIV-related care
During 2007, 16% (8255/52 769) of HIV-infected adults accessing care and residing in E,W&NI were aged 50 years and older, a 3.5-fold increase from 2000 when 11% (2332/20 696) were older adults. Overall, the median age of all older adults accessing HIV-related care in 2007 was 55 years (range: 50–90; 55 years for men and 54 years for women). Of older adults accessing care in 2007, 73% (6010) were between the age of 50 and 59.
The number of older adults accessing HIV care reflects the distribution of those newly diagnosed with HIV; 55% (4469/8129) were MSM, the majority (93%; 4137/4444) of whom were of white ethnicity, and 42% (3420) were infected through heterosexual contact. Compared with younger heterosexuals, a significantly higher proportion of older heterosexuals were men (58%: 1969/3420 vs. 34%: 7991/23 705; P < 0.001) and of white ethnicity (42%: 1416/3403 vs. 16%: 3703/23 594; P < 0.001). Forty-five percent (1541/3403) of older heterosexuals were black African compared with 72% among younger heterosexuals (17 080/23 594) (P < 0.001).
Older adults accessing HIV care in 2007 who were diagnosed at age 50 years and over were more likely to be receiving HAART compared with younger adults (84%: 6884/8164 vs. 69%: 30 209/43 917; P < 0.001). This was also evident for exposure groups, with 86% (3809/4435) of older MSM on treatment compared with 66% (12 139/18 293) of younger MSM and 82% (2789/3382) of older heterosexuals compared with 71% (16 618/23 313) of younger heterosexuals. Median CD4 cell count at date last seen for care between older and younger adults was similar (437 vs. 435 cells/μl). There were no notable differences by exposure groups or ethnicity.
First AIDS diagnosis
An AIDS diagnosis was reported for 20% (782) of older adults newly diagnosed between 2000 and 2007, 91% of whom (742) were diagnosed within 91 days of diagnosis. The corresponding figures for younger adults were 10 and 90%, respectively. Pneumonia carinii pneumonia (PCP) was the most common AIDS-related illness reported among older adults (46%: 132), followed by tuberculosis (17%: 132).
The proportion of adults diagnosed late with a CD4 cell count less than 200 cells/μl was significantly higher among older adults than among those diagnosed prior to 50 years (48%: 1490/3118 vs. 33%: 11 394/34 827; P < 0.001) (Table 1). These differences were also important across exposure groups.
As compared with younger MSM, older MSM were almost twice as likely to present late (40%: 489/1212 vs. 21%: 2487/12 083; P < 0.001). Although the proportion of older heterosexual men presenting late was higher than that among MSM, it was similar to that among younger heterosexual men (53%: 513/976 vs. 45%: 3333/7343; P = 0.012). A high proportion of heterosexual women also presented late, with the figure being higher among older women than among younger women (51%: 379/745 vs. 36%: 4991/13 846; P < 0.001).
A total of 538 (13%) older adults diagnosed between 2000 and 2007 died, of whom 76% (409) died within a year of diagnosis. Older adults diagnosed late with a CD4 cell count less than 200 cells/μl were on average 14 times more likely to die within a year of diagnosis than those diagnosed promptly (14.4 vs. 1.0%; P < 0.001) (Fig. 1). Older adults diagnosed late had 2.4 times the risk of dying within a year of diagnosis than younger adults diagnosed late (P < 0.001). Importantly, there has been a decline in short-term mortality among younger adults diagnosed with a CD4 cell count less than 200 cells/μl, decreasing from 7.2% (68/942) in 2000 to 4.3% (58/1352) in 2007 (P < 0.001), but no decline among older adults. Although late presenters aged 50 years and over represented 4.4% of all new diagnoses in 2007, they contributed to an estimated 30% of the 110 AIDS-related deaths in all individuals diagnosed in 2007.
A total of 919 adults aged 50 years and over were reported to have died in E,W&NI between 2000 and 2007, corresponding to an average rate of 25 deaths per 1000 population. This rate decreased from 35 in 2000 to 17 per 1000 in 2007. In contrast, among 2783 individuals reported to have died when under the age of 50 years, the average rate of death was 12 per 1000 population.
Almost half (47%) of all deaths reported among all age groups were due to AIDS. AIDS-related pneumonia and PCP were the leading cause of mortality in adults (22%: 200/890) who died at age 50 years and over, followed by non-AIDS-related malignancies (12%: 108), cardiovascular/cerebrovascular disease (10%: 89) and non-AIDS-related pneumonia (9.3%: 83). Similarly, among younger adults, AIDS-related pneumonia and PCP were also found to be the leading cause of death (19%: 503/2706), followed by non-AIDS-related pneumonia (7.8%: 212) and suicide and accident (7.2%: 194).
Age at infection among older adults
Based on CD4 cell count at diagnosis, an estimated 48% (1486/3118) of newly diagnosed older adults acquired their infection at age 50 years and over [33% (1035) were estimated to be aged 45–49 years and 19% (597) between 40 and 44]. Among the 8255 older adults accessing care, approximately a quarter (1486) acquired their infection at age 50 years and over. The proportion infected at age 50 years and over remained fairly stable over the period (Fig. 2) and within prevention groups (MSM 49%: 593; heterosexual men 50%: 488; and heterosexual women 44%: 327). The estimated median age at infection was 54 with a median age at diagnosis of 60 years.
Three-quarters (76%: 1134/1486) of individuals infected when 50 years or over were men, 54% (593/1101) of whom probably acquired their infection through sex with other men. The overwhelming majority of MSM infected as an older adult were of white ethnicity (95%: 552/582) and had acquired their infection in the UK (83%: 313/376).
Heterosexual contact was reported for 57% (815/1438) of adults who acquired their infection as an older adult, 60% of whom were men. Among older men infected heterosexually, 56% (270/485) were of white ethnicity. Eighty percent (195/243) of heterosexual men of white ethnicity probably acquired their infection abroad, with 40% (96) in Asia. Among heterosexual men of black African ethnicity, 99% (140/142) acquired their infection abroad. The majority of heterosexual women who acquired their infection as an older adult were of black African ethnicity 60% (196/325), and 29% (93) were of white ethnicity. A high proportion of heterosexual women (71%: 220/308) probably acquired their infection in Africa. No difference was observed in the median age in individuals infected at age 50 years and over by sex and probable route of infection.
The number of adults aged 50 years and over living with diagnosed HIV in E,W&NI more than tripled between 2000 and 2007, so that by 2007 older adults comprised one in six of the 53 739 adults seen for HIV care. New diagnoses among older adults more than doubled over this period, representing 10% of all diagnoses in 2007. Similar high proportions of older adults have been reported in Europe, with older adults representing 13% of newly reported cases of HIV in western Europe, 9% in central Europe and 4% in eastern Europe in 2007 . Our analyses suggest that almost half of newly diagnosed older adults were diagnosed late (CD4 cell count <200 cells/μl) and that over half were infected when aged over 50 years.
Compared with younger adults, newly diagnosed older adults were significantly more likely to be men, infected through sex between men and of white ethnicity. Older heterosexual adults were more likely to be infected within the UK, with evidence of travel abroad among white heterosexual men.
National surveillance data indicate that the majority of individuals (≥95%) diagnosed in the UK are seen for care within a year . Not surprisingly, the prevalent pool of diagnosed older adults were more likely to be on antiretroviral therapy compared with those aged 15–49 years reflecting an ageing cohort of persons living with HIV in the UK. There was no significant difference in median CD4 cell count (430 cell/μl) between older and younger individuals accessing care, indicating no discrepancy in the way these groups access care.
Late diagnosis, a major contributor to mortality in HIV-infected individuals, was found to be overall higher among older adults than among younger adults, particularly among heterosexual women and MSM. This is consistent with previous studies conducted in Europe in which a significantly higher proportion of older adults were diagnosed at a more advanced stage of HIV infection than younger adults [6,15,16]. It is likely that a number of factors contribute to this elevated rate, including an individual's perception of risk, missed opportunities in healthcare settings [6,17–19] and limited sexual health information targeting older adults [19,20].
Previous studies have shown that MSM diagnosed late are 10 times more likely to die than those diagnosed early and heterosexuals diagnosed late are over nine times more likely to die [21,22]. Age was identified as a factor significantly associated with both the increased odds of late diagnosis and the odds of short-term mortality .
In our study we found that, compared with younger adults, older adults diagnosed late had 2.4 times the risk of dying within a year of diagnosis. Among older adults, those diagnosed late were 14 times more likely to die within a year of their diagnosis compared with those diagnosed early. Among younger adults, there was an eight-fold difference. These results highlight the extent to which older adults are affected by the late diagnosis of HIV.
The death rate of older adults living with diagnosed HIV was similar to that among older adults in the general population  (25 per 1000 population vs. 26 per 1000 population, respectively). AIDS-related causes accounted for approximately half of the deaths among older adults.
The number of new HIV diagnoses among older adults doubled between 2000 and 2007. This rise will be in part due to increases in HIV testing as well as newly acquired infections. For the first time, we estimate the proportion of older adults who acquired their infection when aged 50 years and over. Half of the older adults newly diagnosed (48%) were estimated to have acquired their infection at age 50 years and over, equating to a quarter of the 8255 older adults accessing care in 2007. We further estimate that an additional quarter of older adults accessing care were probably infected when in their 40s. The remaining older adults represent an ageing cohort of individuals infected at a younger age.
Community-based studies conducted in the UK  and USA  found that over half of older adults (over the age of 50) were sexually active (65 and 54%, respectively), and one in 10 of those within the UK-based study did not use any form of contraception (and therefore STI barrier) with their current partner .
High-risk sexual behaviours have been documented among older adults [20,25]. A community-based study in the UK found that 7% of adults over the age of 50 years engaged in risky sexual behaviour (defined as having more than one partner in the last 5 years and not consistently using condoms) . It has been suggested that older adults attending sexual health clinics are neglected in terms of the information made available to them , and that older adults as a whole are reluctant to discuss sexual problems with their physicians .
There are a number of limitations to our study. First, CD4 cell counts were only available for three quarters of older adults. Second, our method of estimating an individual's age at infection is based on a CD4 cell count model and may not reflect individual cases. Seroconversion can lead to lower CD4 cell counts, particularly in older adults [5,26]; however, our surveillance data  show that only 7% of older adults with a reported seroconversion have a CD4 cell count less than 200 cells/μl at diagnosis. It is, therefore, unlikely that seroconversion would significantly influence estimates of the proportion of older adults infected as older adults. It is also unlikely that many individuals would have been misclassified as a result of treatment, given that the median time from first HIV diagnosis to first CD4 cell count was 7 days. No individual would ordinarily receive treatment prior to a CD4 cell count. Third, it is recognized that older individuals will experience a greater decline in CD4 cell count following infection . However, in sensitivity analyses applying an age-specific model of CD4 cell count progression , we found little difference in the proportion estimated to be infected when aged 50 years and over (48%: 1486 vs. 51%: 1583). Fourth, it was not possible to validate our findings using date last negative, as the completion rate for this variable among individuals newly diagnosed between 2000 and 2007 was less than 5% . Finally, further reports of death are expected for individuals diagnosed in recent years.
This study provides evidence of high numbers of new HIV diagnoses among older adults and evidence of HIV transmission in this group. Furthermore older adults have elevated rates of late diagnosis and short-term mortality. Among MSM, the elevated rate of late diagnosis was particularly marked.
These findings highlight the need for increased targeted prevention efforts and HIV testing strategies among older adults to ensure earlier testing and treatment and reduce transmission of HIV. Adults aged 50 years and over account for a significant number of persons living with HIV in developed countries and it is important that global and national surveillance outputs include older age groups.
The continuing collaboration of those who contribute to HIV surveillance in the UK is gratefully acknowledged. Without their generosity, time and effort the current level of understanding of the epidemic could not have been obtained.
Author contribution: All authors work within the HIV and STI Department at the Centre for Infections. R.S. is a scientist and led in the design and writing of the paper and was responsible for the final manuscript. V.D. is the lead consultant and provided comments and advice on drafts of the paper. A.B. is a lead scientist and helped conduct analyses and provided comments and advice on drafts of the paper. B.D.R. is a lead scientist and provided comments and advice on drafts of the paper.
2. Nogueras M, Navarro G, Anton E, Sala M, Cervantes M, Amengual M, et al
. Epidemiological and clinical features, response to HAART, and survival in HIV-infected patients diagnosed at the age of 50 or more. BMC Infect Dis 2006; 6:159.
3. Stoff DM, Khalsa JH, Manjan A, Portegies P. Introduction: HIV/AIDS and ageing. AIDS 2004; 18:S1–S2.
4. Babiker AG, Peto T, Porter K, Walker AS, Darbyshire JH. Age as a determinant of survival in HIV infection. J Clin Epidemiol 2001; 54:S16–S21.
5. CASCADE collaboration. Differences in CD4 cell counts at seroconversion and decline among 5739 HIV-1-infected individuals with well estimated dates of seroconversion
. J Acquir Immune Defic Syndr
6. Grabar S, Weiss L, Costagliola D. HIV infection in older patients in the HAART era. J Antimicrob Chemother 2006; 57:4–7.
7. Casau NC. Perspective on HIV infection and ageing: emerging research on the horizon. Clin Infect Dis 2005; 41:855–863.
12. Satten GA, Longini IM. Markov chains with measurement error: estimating the ‘true’ course of a marker of the progression of human immunodeficiency virus disease. Appl Stat 1996; 45:275–309.
13. Lazarus JV, Nielsen KK. HIV and people over 50 years old in Europe
. HIV Med
14. Rice B, Sinka K, Patel B, Evans B, Delpech V. Trends in treatment access in the United Kingdom: identifying cohorts from the national cross sectional census of prevalent diagnosed HIV infections
. In: International Conference on AIDS
; 11–16 July 2004. Bangkok: International AIDS Society (IAS), 2004. Vol. 15; abstract MoPeC3575.
15. Orchi N, Balzano R, Scognamiglio P, Navarra A, De Carli G, Elia P, et al
. Ageing with HIV: newly diagnosed older adults in Italy. AIDS Care 2008; 20:419–425.
16. Longo B, Camoni L, Boros S, Suligoi B. Increasing proportion of AIDS diagnoses among older adults in Italy. AIDS Patient Care STDs 2008; 22:365–371.
17. Leigh BC, Temple MT, Trocki KF. The sexual behaviour of US adults: results from a national survey. Am J Public Health 1993; 83:1400–1408.
18. Ress B. HIV disease and ageing, the hidden epidemic. Critical Care Nurse 2003; 23:38–42.
19. Pratt G, Gascoyne K, Cunningham K, Tunbridge A. Human immunodeficiency virus (HIV) in older people. Age Ageing 2010; 39:289–294.
20. Gott CM. Sexual activity and risk-taking in later life. Health Soc Care Commun 2001; 9:72–78.
21. Chadborn TK, Baster K, Delpech VC, Sabin CA, Sinka K, Rice BD, et al
. No time to wait: how many HIV-infected homosexual men are diagnosed late and consequently die? (England and Wales, 1993–2002). AIDS 2005; 19:513–520.
22. Chadborn TK, Delpech VC, Sabin CA, Sinka K, Evan BG. The late diagnosis and consequent short-term mortality of HIV-infected heterosexuals (England and Wales, 2000–2004). AIDS 2006; 20:1–9.
25. Lindau ST, Schumm LP, Laumann EO, Levinson W, O'muircheartaigh CA, Waite LJ. A study of sexuality and health among older adults in the United States. N Engl J Med 2007; 357:762–774.
26. Lindback S, Brostrom C, Karlsson A, et al
. Does symptomatic primary HIV-1 infection accelerate progression to CDC stage IV disease, CD4 count below 200 × 106
/l, AIDS, and death from AIDS? BMJ 1994; 309:1535–1537.