Data on the incidence of infectious diseases included in the Netherlands Immunisation Programme (RVP) are available at several institutes. A regular, for instance yearly analysis and reporting of these data (epidemiological surveillance), is an imported tool to evaluate the RVP. In this report information on he sources of data on the incidence of the diseases are described. The epidemiological surveillance focuses on the incidence of the disease and/or its complication(s). Mortality data provide insight into the case fatality rate, and data on the vaccination status provide insight into the vaccine efficacy. A time variable makes it possible to study trends in disease incidence. In addition to the compulsory registration of mortality at the Central Bureau of Statistics (CBS) and morbidity at the Medical Inspectorate of Health (IGZ), infectious diseases and their complications are registered at several (professional) agencies. The coverage of the registration are frequent nation-wide, for example the National Medical Registration (LMR) of the Health Care Information (SIG), the Medical-Microbiological Laboratories (MML), the Netherlands Paediatric Surveillance Unit (NSCK) and the SSPE-Registration. Sometimes registration covers a region like the European Registration of Congenital Anomalies (EUROCAT) and the Continuous Morbidity Registration Nijmegen (CMRN). Sometimes it concerns a random sampling survey like the registration of the Netherlands Institute of Primary Health Care (NIVEL). Problems can arise when comparing the data from the different institutes. It is not permissible, by legalisation, to collect data which are reducible to the name of the patient. The possibility of comparing at the level of sex, age (or year of birth), diagnosis and possibly city(region) and/or time of onset will be investigated. The registration of vaccination status is not good. Comparing the data with the Provincial Immunisation Administration (PEA) on an individual level will be difficult, but the possibility of using the PEA to obtain information on records with missing values on vaccination status must be studied.We suggest to collect data from the year 1976 onwards. The agencies can provide data on mumps, measles, rubella, diphtheria, whooping cough, tetanus, poliomyelitis, infectious caused by Haemophilus influenzae type b, strepto-, pneumo- and meningococcus (registered as main diagnosis or alternative diagnosis) stating - if available - sex, age, city(region) and/or onset of the disease and vaccination status. The epidemiological pattern of the diseases will be described on the base of the available data. The additional value of the different sources will be evaluated. On the basis of the evaluation results data will be collected and reported yearly.<br>

The objective was to obtain insight into the current incidence of pertussis. Design: Comparison study from different surveillance sources. Method: For the years 1989-1994 the incidence of pertussis was estimated from the number of pertussis notifications, from laboratory data from serodiagnosis, isolations of Bordetella and from the national registration of clinical diagnosis of hospital admissions. Results: The number of notifications and positive serological results yielded similar average annual incidence for the period 1989-1994 (2.3 and 2.2/100,000, respectively). According to notifications peak incidences occurred in 1989 (3.5/100,000) and 1994 (3.4/100,000). For positive serodiagnosis the incidences in 1989 and 1994 were 2.4/100,000 and 3.2/100,0000, respectively. In 1993 the incidence of notifications and positive serodiagnosis (2.4 and 3.2 per 100,000) was also higher than in the other years. The incidence based on hospitalizations, notifications and positive serology was highest among children younger than one year. In 1994 and 1993 the estimations for the vaccine efficacy were 84% (95%-confidence interval 80-87%) and 92% (95%-confidence interval 90-94%), respectively. Conclusion: For 1989-1994, the pattern indicated that pertussis is endemic with four-yearly peaks. The incidence of hospitalizations emphasizes the seriousness of the illness in infants. Due to the vaccine coverage in the Netherlands, the incidence among unvaccinated children is much lower then among unvaccinated children in neighbouring areas with lower vaccine coverage. It is important that general practitioners are aware of the occurrence of pertussis in vaccinated and unvaccinated children and adults. The probable decrease in vaccine efficacy (92% in 1993; 84% in 1994) needs special attention and must be monitored. This stresses the importance of continued surveillance.<br>

Polio-free certification in the framework of the global polio eradication programme, will require the implementation of a more active, comprehensive surveillance system. The Global and European Commissions for the Certification of the Eradication of Poliomyelitis have established the principles, criteria and process of polio-free certification, particularly with regard to surveillance. These include the use of performance indicators. Polio surveillance in the Netherlands should be based on the following elements. 1) Clinical surveillance consisting of mandatory notification of suspected patients and of reporting of patients with acute flaccid paralysis born in 1957 and thereafter. An expert committee has to be established for final classification of cases. 2) Virological surveillance comprising (a) diagnostic investigation of suspected polio patients and AFP patients, (b) analysis of a selection of enterovirus strains isolated in Dutch virus diagnostic laboratories, and (c) environmental surveillance. 3) Serological surveillance to determine the level of protection to poliomyelitis in the general population and specific risk groups. 4) Collection of information on vaccination coverage. The proposed surveillance system will meet the three objectives to detect possible (wild) poliovirus circulation, to determine the origin of circulating (wild) poliovirus, and to document the absence of wild poliovirus circulation in the Netherlands.<br>

The surveillance of Acute Flaccid Paralysis (AFP) is included in the Dutch Paediatric Surveillance System (NSCK) since October 1992. Paediatricians in hospitals report a list of rare diseases on a monthly basis as part of an active surveillance scheme. After the initial report, additional information on clinical presentation, diagnostic results and vaccination history is collected through a questionnaire. In 1995 11 cases conforming to the case-definition were reported, and in 1996 15, resulting in an AFP-rate of 0.39 per 100,000 in 1995 and 0.53 per 100,000 in 1996. No cases of AFP caused by polio infection were observed. In about 50% of the reported AFP cases Guillain-Barre Syndrome was diagnosed. To date, the AFP surveillance in the Netherlands does not meet the WHO criteria for adequate surveillance for certification as polio-free according to the polio eradication initiative. An observed AFP-rate of at least 1 per 100,000 is accepted as proof of sufficient sensitivity of the surveillance system. Along with timeliness of the reports, we are concerned of the low proportion of cases of whom faecal samples are adequately virologically investigated (58% one faecal sample, of which 51% within 14 days after onset of disease; 11% two samples). Recommendations for optimalisation: introducing initial reporting by telephone (immediate reporting and advice on adequate specimen collection); improved information of the paediatricians through NSCK newsletters, presentations and publications extending of the surveillance to neurologists.<br>

To gain insight into the incidence and severity of pertussis in the Netherlands in 1999 and 2000, surveillance data based on notifications, laboratory data, hospitalisations and deaths were analysed for these two years and compared to the 1989-1998 period. Results of the paediatric surveillance are also presented here. According to various sources the incidence of pertussis increased in 1999 compared to previous years and decreased again in 2000. The peak incidence according to notifications and positive serology was observed among 4- to 5-year-old children. In 1999 the incidence according to hospital admissions (3.2 per 100,000) was comparable to the incidence during the epidemic of 1996 (3.3 per 100,000) and decreased in 2000 (1.6 per 100,000). The paediatric surveillance showed that most hospitalised children were under one year of age and that complications (apnoea, cyanosis and administration of oxygen) were more frequently reported in the younger age groups. Vaccine efficacy, estimated by the screening method, was higher in 2000 compared to 1997-1999, particularly among 1- and 2-year olds. In conclusion, the incidence of pertussis in 1999 according to notifications increased to reach a higher level than in 1996. In 2000 the incidence decreased again. However, the number of hospital admissions were comparable to the figures for 1996 and 1999, and lower in 2000. Both unvaccinated and vaccinated persons can develop classical pertussis symptoms. Surveillance of pertussis based on various surveillance sources should be continued to monitor the incidence of pertussis and to study the effect of changes in vaccination strategies. Active paediatric surveillance and surveillance of hospital admissions are useful for verifying trends in routine surveillance and describing the severity of pertussis.<br>

The Netherlands Immunization Programme has greatly reduced the incidence of target childhood diseases and their complications. However, continuous vigilance is required. In this report we propose an integrated approach for surveillance of the Netherlands Immunization Programme. Data collection for all target diseases should be combined as much as possible. Such surveillance system has four key elements: . epidemiological surveillance . immunosurveillance studies . microbiological surveillance . surveillance of adverse events following vaccinations. The proposed primary surveillance system is considered suitable to collect the relevant data at the level of the general population. Specific studies in subpopulations can be further initiated on signals from the primary system.<br>

In March 1999 results from an exploratory case-control study indicated that the most probable origin of a large epidemic of legionnaire's disease had to be directly located at or in the surroundings of the Westfriese Flora (WF) in Bovenkarspel (Province of Noord-Holland, Netherlands). The WF is a yearly exhibition of flowers, and agricultural and consumer products, which was held in 1999 from 19 to 28 February. After inspecting the premises and making an inventory of all exhibits, a list of potential sources was drawn up. The water supply system and all the still available equipment using water were sampled. A risk assessment was done for each potential source by interviewing exhibitors. In total, sixteen products were found potentially hazardous: two whirlpools, two bubblemats (perforated rubber mats meant for the bottom of a water-filled bathtub through which air can be blown at high speed), eleven fountains and a sprinkler installation. Legionella pneumophila was isolated from two whirlpools and a sprinkler installation. With the use of molecular-biological techniques (PFGE, ERIC/REP-PCR and AFLP), three different genotypes were identified. These were B-1, B-2 and B-3 (B for Bovenkarspel). To enable comparison of these isolates with patient isolates, all microbiology laboratories in the Netherlands collaborated in collecting patient isolates related to the epidemic. In total 29 isolates were collected, 28 of which were genetically identical to B-1; the other one was genetically identical to B-2. Through the risk assessment based on the reported use and estimating the degree of contamination from the microbiological investigation of the three culture-positive products, the most probable source contributing the most to the spread of L. pneumophila was found to be the whirlpool located in hall 3 of the WF premises. It is not possible to rule out the contribution made to spreading by the whirlpool in hall 4, whereas the sprinkler's contribution to spreading is not probable. The contribution to spreading by the fountains and bubblemats is also unlikely. Although Legionella spp was not isolated from the water supply system, the organism was likely introduced into the three products from this system. These results were confirmed by the outcome of a case-control study and a cohort study, both designed to pinpoint the exact location of exposure to Legionella spp on the WF premises. They showed the whirlpool in hall 3 to be the most probable source of infection (reported separately).<br>

A sudden and high increase in registrations, positive serodiagnostics, Bordetella isolations and hospital admissions were observed in the Netherlands in 1996. This seems to reflect a true increase in the incidence of pertussis. Most cases occurred among 1-9-year-old vaccinated children. In 1994-1995, and especially 1996 a higher proportion of vaccinated cases was seen among the cases registered. Despite the methodological constraints of vaccine-efficacy estimations using the screening method, a lower vaccine-efficacy for 1996 is probable. The proportion of registrations confirmed with one-point serology in the 1993-96 period which does not meet the formal criteria for registration increased from 25% to 48%. In addition to this 'overreporting', the underreporting of patients with positive two-point serology decreased from 74% in 1993 to 51% in 1996. The number of pertussis patients in 1996 was higher than in 1987. Increased awareness, changes in diagnostics and a lower vaccine coverage could not explain the epidemic. There are indications that changes occurred in the bacterium. This could have resulted in a mismatch between the vaccine-induced immunity and the circulating Bordetella strains. The constant ratio of hospital admissions and notifications and hospital admissions and positive two-point serology among infants less than one year old, indicate no change in the virulence of the circulating strains. In January 1997, active (monthly) surveillance by pediatricians of cases among hospitalized children was started to obtain insight into the severity of pertussis among infants. Protection of these infants is the main reason for pertussis vaccination. The methodological limitations of the current estimations based on the screening method and retrospective data, calls for optimisation of vaccine-efficacy estimations. Therefore a prospective study is considered to assess efficacy of the whole cell vaccine, differentiated by severity of disease, is under consideration.<br>

As a result of improved socio-economic state and related hygiene, and the introduction of the National Vaccination Programme (RVP), the incidence of the target diseases of the RVP is low nowadays. Insight in the occurrence of the diseases remains necessary in order to be able to signal possible secondary effects at an early stage. The vaccination coverage is very high in the Netherlands, but considerable geographic differences exist within the country. Especially in municipalities where groups that reject vaccination on religious ground are clustered, the vaccination coverage is low. Herd immunity can be broken in socially and geographically clustered non-vaccinated groups. On the basis of the occurrence results of the target diseases of the RVP, we give recommendations for the future surveillance of the target diseases of the RVP and invasive meningococcal and pneumococcal infections. Besides continuation of the surveillance, additional research in case of outbreaks and epidemics is recommended, in order to obtain more insight into the circulation of the pathogens.<br>

Following the recommendation of the National Health Council to include immunisation against Hib in the National Immunisation Programme to prevent mortality, severe morbidity and permanent disability due to invasive Hib infection, the vaccine is offered to all children born after 1 April 1993. To assess the effectiveness of this immunisation, surveillance has to be carried out. In this annual report on the Hi(b) surveillance in 1995 done by the Dutch Paediatric Surveillance Centre (NSCK) progressive effects of Hib vaccination became evident. Forty-one invasive Hi infections were registered: 28 meningitis, 8 epiglottitis, 2 arthritis and 3 sepsis. This means a substantial reduction compared with 1994 (129 cases) and the estimated numbers before the start of vaccination. Three cases of true vaccine-failures occurred and no possible or apparent vaccine-failures. Since 1 January 1995 the case-definition was changed to invasive Hi disease irrespective of serotype. As a result it is possible to study invasive Hi disease caused by non-serotype b strains as well. Cellulitis was included into the case-definition since 1 January 1995. No cases of cellulitis as the only sign were reported in 1995. The coverage of the paediatric surveillance scheme was estimated through comparison of the records of meningitis cases with the laboratory records of the Netherlands Reference Laboratory for Bacterial Meningitis (RBM).<br>