Prevalence of HIV infection in pregnant women in London and elsewhere in England

BMJ 1994; 309 doi: (Published 06 August 1994) Cite this as: BMJ 1994;309:376
  1. A Nicoll,
  2. C McGarrigle,
  3. J Heptonstall,
  4. J Parry,
  5. A Mahoney,
  6. S Nicholas,
  7. E Hutchinson,
  8. O N Gill a collaborative group
  1. Communicable Disease Surveillance Centre, Public Health Laboratory Service, London NW9 5EQ
  2. Hepatitis and Retrovirus Laboratory, Central Public Health Laboratory Statistics Unit, Public Health Laboratory
  1. Service Correspomdence to: Dr Nicoll.
  • Accepted 18 May 1994

In 1990 a programme based on the unlinked anonymous test method began in England and Wales to monitor the prevalence of HIV infection. It included a survey of blood specimens collected from pregnant women for rubella screening in London and other parts of England.1 Forty antenatal centres took part: 15 in London, 19 in three other cities (Manchester, Leeds and Bradford), and six in non-metropolitan areas near to these four cities. The survey began in January 1990 but ended outside London in 1992 when a strategic decision was taken to use dried blood spots from newborn infants as the preferred method for monitoring maternal HIV-1 infection in areas with lower prevalence.2 The survey continues in London, and we present data to the end of June 1993.

Methods and results

Specimens were grouped by antenatal centre, calendar quarter in which serum was collected, and age (four age groups) and were irreversibly unlinked from the source women. Specimens were tested with a commercial HIV-1 and HIV-2 enzyme immunoassay (Wellcozyme HIV1+2); repeatedly reactive specimens were tested by other assays, including western blotting, at the national reference laboratory.1 We used logistic regression analyses to model the variation in the number of tests that yielded positive results for HIV infection, allowing for differences among study centres and the relation between the prevalence of HIV-1 infection and time, age group, and centre.

HIV-1 infection in women attending antenatal centres in London, other metropolitan areas, and non-metropolitan areas from January 1990 to June 1993. Values are numbers of women unless stated otherwise

View this table:

In all, 405077 specimens were tested between January 1990 and June 1993 (175 957 from London, 162 386 from the other cities, and 66 734 from the non-metropolitan areas). The volume of serum in 2616 (0.64%) specimens was insufficient for testing, and objections to testing led to the exclusion of 457 (0.11%) specimens. The prevalence of HIV-1 infection in the London centres was 0.23% (table). It varied significantly with age, ranging from 0.15% in women aged <20 and 0.14% in those aged >29 to 0.29% in those aged 20-24 and in those aged 25-29. The prevalence of HIV-1 infection was significantly higher in London than elsewhere, but substantial variation was observed among the London centres (from 0.04% to 0.51%). The prevalence of HIV-1 infection was 0.011% in the cities outside London and 0.007% in non-metropolitan centres. The highest prevalence in a centre outside London was 0.03%, which was lower than the lowest prevalence in London; this centre and an adjoining centre accounted for nine of the 25 infections outside London. The prevalence of HIV-1 infection in specimens from women attending London centres was 0.18% in 1990 compared with 0.21% in 1991 and 0.26% in 1992 and in the first half of 1993 (table). The rise was significant over time when allowance was made for the effects of centre and age (P=0.006), and it occurred in each age group.


The upward trend in the prevalence of HIV-1 infection in pregnant women in London is probably not an artefact as it is consistent across the age groups and most of the study centres. The higher prevalence and the rise in prevalence in the London centres could be explained either by a net immigration of infected women from sub-Saharan Africa1,2 or by an increase in transmission of the HIV-1 infection within Britain.3 Both mechanisms are probably operating. In other countries the prevalence among pregnant women has levelled off or even declined.4,5 We do not therefore expect the rise to continue indefinitely, but the situation will require careful monitoring.

The incidence of AIDS in some European countries is three to four times that in England and Wales. The prevalence of maternal HIV-1 infection in London, however, is comparable with that in the worst affected cities, such as Paris (0.25% in 1992-3) and Rome (0.12% in 1992).4,5 As the prevalences in Paris and Rome are not rising the situation in London seems to be worse.

This work was supported by the Medical Research Council with funds provided by the Department of Health. We thank Mr Meg Wiseman for administrating the survey, Ms Kate Soldan for its initial implementation, and Dr A Swan and Mr J Mortimer for comments on earlier drafts of the manuscript.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract