aDivision of Retrovirology, and bEpidemiological Data Centre, Robert Koch-Institut, Berlin, Germany; cClinic for General Paediatrics, Charité, Humboldt Universität, Berlin, Germany; and dNiedersächsisches Landesgesundheitsamt, Hanover, Germany.
Sponsorship: This work was supported by a grant from the German Ministry of Health.
Received: 27 July 2000;
revised: 8 September 2000; accepted: 15 September 2000.
Chronic diseases are challenging to the public health system in terms of planning and providing appropriate medical and social care and implementing specific preventative strategies. Therefore it is crucial to estimate the prevalence of the disease in question in different cohorts, using a broad spectrum of epidemiological methods available for obtaining this information [1–3].
We investigated the spread of HIV infection in the general population in Germany, using anonymous unlinked testing (AUT) for HIV of all new-borns in Berlin and Lower Saxony over a 7 year period (1993–1999). As maternal IgG antibodies cross the placenta , the antibody pattern in the new-born reflects the infection status of the mother. Therefore child-bearing women were chosen as a subset of the age-matched heterosexually active population .
This study was performed to estimate the occurrence of HIV in those parts of the German population who do not belong to known risk groups such as homosexuals or intravenous drug users. The study regions represent one of the centres of the epidemic, Berlin, and a more rural area with a lower prevalence of HIV infection, Lower Saxony. Running the tests anonymously, the bias that is known to be associated with stigmatized diseases such as AIDS and other sexually transmitted diseases was avoided, and informed consent of the donors was not necessary. The same study protocol for collecting and testing the samples for HIV antibodies was introduced in both laboratories, Berlin and Hanover . Briefly, dried blood spots on filter paper (Guthrie cards) were eluted with diluent used in an enzyme-linked immunosorbent assay (ICE HIV-1.O.2; Murex Diagnostica, Burgwedel, Germany) and screened for HIV-specific IgG antibodies, either as single samples or as pools of six. Reactive samples were re-tested and, in the case of pooled samples, broken down to the reactive sample. Reactive sera were analysed by Western blotting (Diagnostic Biotechnology, Singapore). In the case of multiple labour only one of the new-borns was tested; in the case of insufficient material no test was performed.
According to official statistics, in Berlin the number of births averages 30 000 per year compared with Lower Saxony with 80 000 births. The age of child-bearing women ranges from 15 to 45 years of age. This subgroup is representative of approximately 20% of the total population and 40% of women. The results for Berlin (1993–1998) and Lower Saxony (1993–1999) are summarized in Table 1. In Berlin the mean rate of HIV-positive tests was 0.59 per 1000 samples investigated, compared with only 0.14 per 1000 samples in Lower Saxony. However, in the larger cities in Lower Saxony, such as Oldenburg or Hanover, the rates of HIV antibody-positive mothers (0.5 or 0.4 per 1000 samples, respectively) are comparable to the rates in Berlin. On the other hand, in rural areas of Lower Saxony the rates were less than 0.1 (data not shown in detail). The rates between and within the regions varied; however, there was no significant variation over time.
The data obtained by AUT suggested that in this subpopulation no increase in the infection rate was observed. To evaluate whether this reflects the trend in the general population, we compared these data with those obtained by the mandatory laboratory reporting system (Table 1). In the years 1993–1999, rates of seven to 11 confirmed HIV-positive tests per 100 000 inhabitants for Berlin and of one to two for Lower Saxony, respectively, were reported [8,9]. In women as a subgroup of the population, the rate of positive HIV diagnoses was approximately three times lower than in the total population (data not shown). In contrast, in women in the age group between 15 and 45 years of age, who represent the sexually active part of the population, the rate of positive HIV tests was twice as high as in the general female population. Although the diverse structure of the cohorts of child-bearing women and women reported by the mandatory reporting system as well as the different denominator in both surveillance systems prohibit a direct comparison of the data, a stable rate of HIV infections could be observed in both cohorts, but on different levels.
In view of the consolidated state of the HIV epidemic in Germany , we are aware that there might be a bias concerning the number of HIV-positive test results in new-borns. As counselling exists for pregnant women, with the choice of abortion in the case of a positive HIV test, or because women who are aware of their HIV status might take precautions against becoming pregnant, one could underestimate the prevalence of HIV in the group of sexually active women. On the other hand, we know from additional surveillance data that during the observation period an increase in the transmission rate of HIV infections in the heterosexual population was observed, and that the number of HIV-infected migrants originating from pattern II countries had increased . In contrast, a decrease in the HIV infection rate in the risk group of intravenous drug users was observed . Therefore one should be aware that the steady level of the infection rate observed in both AUT and the mandatory reporting system might be the result of a changing distribution of HIV-infected individuals in different subpopulations.
The screening of new-borns over 7 years revealed a stable rate of HIV infections in child-bearing women, leading to the conclusion that no measurable changes in the prevalence of HIV in this cohort have taken place.
Taking the data from the AUT and the mandatory testing system together, it can be concluded that the HIV epidemic is still mainly restricted to the known risk groups. The normal population, i.e. people who are not known to have or not conscious of being at a certain risk of HIV infection, is not affected by an increasing rate of HIV infections.
1. Gill ON, Adler MW, Day NE. Monitoring the prevalence of HIV.
BMJ 1989, 299: 1295 –1298.
2. McCarty DJ, Tull ES, Moy CS, Kwoh CK, LaPorte RE. Ascertainment corrected rates: applications of capture–recapture methods.
Int J Epidemiol 1993, 22: 559 –565.
3. Koch MA, Schwartländer B, Kirschner W. Blutspender als Sentinelpopulation für sexuell übertragbare Krankheiten – Ergebnisse einer soziodemographischen Untersuchung und Befragung zum Sexualverhalten.
Gesundheitswesen 1993, 55: 504 –513.
4. Saji F, Koyama M, Matsuzaki N. Current topic: human placental Fc receptors.
Placenta 1994, 15: 453 –466.
5. Goubau P, Wantzin P, Gessain A. et al
. Seroepidemiology of the human T-cell leukaemia/lymphoma viruses in Europe – the HTLV European Research Network.
J AIDS Hum Retrovirol 1996, 13: 68 –77.
6. Siedler A, Hamouda O, Schwartländer B. Anonymous unlinked testing as a sentinel approach: experience in Germany.
J Epidemiol Commun Health 1998, 52 (Suppl. 1) : 25S –27S.
7. Statistisches Bundesamt Wiesbaden (ed.). Statistisches Jahrbuch für die Bundesrepublik Deutschland.
Stuttgart: Metzler/Poeschel; 1999. p. 159 159.
8. Verordnung über die Berichtspflicht für positive HIV-Bestätigungsteste vom 18 Dezember 1987 (Laborberichtsverordnung). Bundesgesetzblatt 1987, 2126-1-8-2 I: 2819.
9. Marcus U, Hamouda O. Epidemiologie der HIV-Neuinfektion in den verschiedenen Risikogruppen.
Bundesgesundhbl – Gesundheitsforsch – Gesundheitsschutz 2000, 43 (Suppl. 1) : S3 –S8.
10. Marcus U. 20 Jahre HIV-/AIDS-Epidemie in Deutschland – Entwicklungen, Trends und Erklärungsversuche.
In:Glück gehabt? Zwei Jahrzehnte AIDS in Deutschland.
Marcus U (editor). Berlin/Wien: Blackwell; 2000. pp. 1 –62.
11. Pant A. Die HIV-Epidemie unter i.v.-Drogenbenutzern: Verlauf, Primärprävention und drogenpolitische Reaktion.
In:Glück gehabt? Zwei Jahrzehnte AIDS in Deutschland.
Marcus U (editor). Berlin/Wien: Blackwell; 2000. pp. 184 –210.