Epidemiological surveillance in Belgium

Measles was very common in Belgium before 1985, affecting 95% of children before they reached the age of 15 years. In 1984 the incidence of measles, calculated using data from a general practitioner surveillance system, was 823/100 000 inhabitants (2). With the introduction of the trivalent measles, mumps, and rubella (MMR) vaccine for the immunisation of children between 12 and 24 months of age coverage of measles vaccination rose from 52% in 1983 to 90% in 1990, and the incidence fell to 80/100 000 in 1990. The incidence of measles in all age groups in Flanders - the northern part of Belgium - rose slightly from 76/100 000 (CI 64-85) in 1991 to 87/100 000 (CI 75-101) in 1995 (2). The age specific incidence of measles in children aged 15 to 19 years in Flanders fell from 113/100 000 in 1983 to 66/100 000 in 1986 but rose to 125/100 000 in 1992. No mass campaign has been conducted to vaccinate school age children as was undertaken in the United Kingdom (3,4) but in 1995 the government recommended a second dose of the trivalent vaccine at the age of 11 years.

The outbreak

Increasing numbers of cases of measles were reported around Herentals (a city of 30 000 inhabitants, in the north of the province of Antwerp) by general practitioners and a university laboratory in March 1996. We conducted an investigation to describe the outbreak in detail and to evaluate the field efficacy of measles vaccine.

Methods

There is no statutory duty to notify cases of measles in Flanders and although the general practitioner survey system can estimate regional incidence it cannot supply valid local rates. We therefore looked actively for cases to measure the impact of the outbreak. Questionnaires were sent to 214 doctors (general practitioners and paediatricians), six local laboratories, and the area's seven hospitals (survey A). In addition, 4092 children who attended secondary school in Herentals were interviewed (survey B). Their responses were compared with those from 978 children from Heist-op-den-Berg, a small town chosen as a control population in which to estimate the background incidence of measles (survey C). As no central records of child immunisation are kept in Belgium, the data obtained through our questionnaires were validated by comparing them with information from general practitioners.

A case of measles was defined as an illness characterised by a generalised rash lasting three or more days, a fever of over 38.3ºC, and cough or coryza or conjunctivitis (5).

Laboratory criteria for the diagnosis were the detection of a significant (fourfold) rise in measles antibody titre or the identification of specific measles IgM antibody. Cases that met the clinical definition were described as probable; confirmed cases were either laboratory confirmed or clinical cases epidemiologically linked to a confirmed or a probable case (5). The study was conducted during six months of 1996 and the area was geographically limited to Herentals, Heist, and their surrounding municipalities.

Results

In survey A, 26 (12%) of the general practitioners we interviewed reported 122 confirmed cases of measles, (95) 78% of whom were more than 10 years of age. The highest attack rates occurred in February and March (figure 1). Laboratories reported eight serologically confirmed cases. Survey B of schoolchildren in Herentals had a response rate of 88% (3621 forms), and suggested that 301 (8.2%) of the children had suffered from measles during 1996. Survey C of the reference group from Heist-op-den-Berg had a response rate of 77% (759 of 978 children), seven (1%) of whom had suffered from measles during 1996. Integration of data from surveys A and B, after removing duplicates by date of birth, provided 345 cases for 307 for whom complete data were available. Thirty-eight per cent of cases had been vaccinated. Children who had not been vaccinated were six times more likely to develop measles than those who had been vaccinated once at the age of 15 months. The estimated vaccine efficacy was 80.9%. (95% CI: 76.2-84.6)

Recommendations

A local programme of MMR vaccination and revaccination of the schoolchildren and their siblings with MMR was recommended during the outbreak. MMR vaccine was delivered free of charge.

Discussion

Clinical data must be interpreted carefully, but it is clear that an outbreak of measles occurred in Herentals early in 1996. This outbreak may be explained by the relatively low vaccine coverage among children in the age groups affected. They were too old to have been included in vaccination against measles in the second year of life, which began in 1985. In addition this cohort stood less chance of acquiring natural infection than earlier cohorts because less measles was circulating in the population. The relatively advanced age of these cases - most over 14 years - is the consequence of the same fact. The outbreak described here can be seen as an adverse consequence of a campaign to vaccinate children once in the second year of life without providing catch up vaccination for older children. The morbidity and mortality of measles and the high risk of further outbreaks in the near future are arguments for holding a mass campaign to immunise all children of school age. It will be held in autumn 1996 to avoid the next wave of cases: cases of measles occur mainly in the late winter and early spring.

1. American Academy of Pediatrics. 1994 red book: report of the Committee on Infectious Diseases. Elk Grove Village: American Academy of Pediatrics, 1994.

2. Van der Veken J, Van Casteren V. Surveillance van mazelen en bof door de Belgische huisartsenpeilpraktijken 1982-1993. Brussel: IHE, 1994.

3. Ramsay M, Gay N, Miller E, Rush M, White J, Morgan-Capner P, Brown D. The epidemiology of measles in England and Wales: rationale for 1994 national vaccination campaign. Commun Dis Rep CDR Rev 1994; 12:141-6.

4. Miller E. The new measles campaign. BMJ 1994; 309:1102-3.

5. CDC. Case definitions for public health surveillance. MMWR Recommendations and Reports 1990; 39(13) :23.