Assessing the potential impact of vaccination programs on invasive pneu mococcal disease : Data from population-based surveillance

I nfections due to Streptococcus pneumoniae remain a major cause of morbidity, mortality and health care expenditure around the world. The recent emergence of antibiotic resistance in S pneumoniae and the development of conjugate pneumococca l vaccines have highlighted the need to better understand the ep idemiology of this disease so that preventive measures can be appropriately targeted. Pneumococcal infections are not reportable in Canada. While a number of publications have now documented the in crease in antibiotic resistance in S pneumoniae, there are no published reports describing the epidemiology of invasive infection. To better understand the epidemiology of infection in Canada, the Toronto Invasive Bacte rial Diseases Network (TIBDN) has been performing population-based surveillance for invasive pneumococcal disease in Metropolitan Toronto and the regional municipality of Peel in Ontario since 1995. TIBDN is a collaborative network of all hospitals and the three largest private clinical microbiology laboratories serving the population of Metropolitan Toronto and Peel Region (population 3.4 million). Prospective surveillance for all cases of invasive bacterial disease in residents of this geographic area has been conducted since January 1, 1995. Invasive pneumococcal disease is defined as any infection associated with S pneumoniae isolated from a sterile site. Sterile site iso-

The Toronto Invasive Bacterial Diseases Network* I nfections due to Streptococcus pneumoniae remain a major cause of morbidity, mortality and health care expenditure around the world.The recent emergence of antibiotic resistance in S pneumoniae and the development of conjugate pneumococca l vaccines have highlighted the need to better understand the ep idemiology of this disease so that preventive measures can be appropriately targeted.
Pneumococcal infections are not reportable in Canada.While a number of publications have now documented the increase in antibiotic resistance in S pneumoniae, there are no published reports describing the epidemiology of invasive infection.To better understand the epidemiology of infection in Canada, the Toronto Invasive Bacte rial Diseases Network (TIBDN) has been performing population-based surveillance for invasive pneumococcal disease in Metropolitan Toronto and the regional municipality of Peel in Ontario since 1995 .
TIBDN is a collaborative network of all hospitals and the three largest private clinical microbiology laboratories serving the population of Metropolitan Toronto and Peel Region (population 3.4 million).Prospective s urveillance for all cases of invasive bacterial disease in residents of this geographic area has been conducted since January 1, 1995 .Invasive pneumococcal disease is defined as any infection associated with S pneumoniae isolated from a sterile site.Sterile site iso-   Streptococcus in Edmonton, Alberta.
The overall incidence of disease was 14.4 per 100,000 population in 1995, 16.1 per 100,000 in 1996 and 11.8 per 100,000 in 1997.Rates of disease were markedly higher in the elderly, and most of the decrease in incidence from 1995 to 1997 was in this age group (Figure 1).
For the 4 70 cases of invasive pneumococcal disease that were identified in 1995, complete clinical information was available for 97% (455 of 470) of cases.The most common clinical diagnoses were pneumonia (300 infections, 66%), bacteremia without focus (75 infections, 17%), meningitis (14 infections, 3%) and otitis media (13 iqfections, 3%).The overall case fatality rate was 18.6%.Case fatality varied with the age of the patient; only 2.2% of children died compared with 28% of those age 65 years and older.Thirty-two episodes of disease (7%) were nosocomial, and 44 (10%) occurred in residents of other institutions (3 7 in nursing homes, three in retirement homes, and two each in jails and psychiatric facilities) .There were no detected outbreaks in any institutions.
Underlying medical conditions that might predispose patients to invasive pneumococcal disease were present in 18 Can J In fect Dis Vol 10 Suppl A January/February 1999 0 y.Twenty-five of 247 (10%) eligible patients for whom adequate vaccination hi stories could be obtained had received pneumococcal vaccine before their episode of illness: 13 were yo ung, chronically ill adults (six with human immunodeficiency virus infection), eight were over 65 years of age and had at least one chronic underlying illness, and four were over 65 years of age with no underlying illness.All four older persons without underlying illness and s ix of eight with chronic illnesses were nursing home residents.An additional 11 of 269 (4%) vacci ne-eligible and surviving patients were vaccinated after their episode of illness.

CONCLUSIONS
Rates of pneumococcal disease in this area are similar to those in published reports from the United States and northern Europe (2-4), and to that found by the Canadian Sentinel Health Unit Su1veillance System.In 1995, the majority of cases a nd virtually all deaths occurred in vaccine-eligible patients as defined by the National Advisory Committee on Immunization (5,6).Vaccine use up until 1995 was suboptimal.The decrease in rates of disease among the elderly between 1995 and 1997 may reflect increased use of vaccine associated with the Ontario pneumococcal vaccination program.However, 25% of all invasive disease occurs in persons not eligible for vaccination, and another 30% occurred in persons with

USE O Y • DO NOT COPY
immunocompromising medical illnesses for which vaccine efficacy would be expected to be low (6).Thus, even a pneumococcal vaccination program that is effective and cost effective (7) may only reduce overall rates of invasive pneumococcal disease by 15% to 25%.If surveillance is to be used to evaluate the success of pneumococcal vaccination programs, it will be necessary both to measure changes in the incidence of disease in specific at risk subgroups and to know vaccination rates in these subgroups.

Figure 1 )
Figure 1) Annual incidenceqfinvasive pnewnococcal disease in Metropolitan Toronto and Peel Region.Open bars represent rates in children (younger than I 6years qf age), hatched barsyoungeradults (16 to 64 years qf age) and closed bars older adults (65 years qf age or olde1) ams and invasive pneumococcal disease

Assessing the potential impact of vaccination programs on invasive pneu mococcal disease: Data from population-based surveillance
Naus, Ontario Ministry ef Health, Toronto; D Noria, Scarborough Grace Hospital, Scarborough; P O'Brien, Etobicoke General Hospital; B Oliver, Humber River Regional Hospital,-K Ostrowska, The Trillum Health Centre, Toronto; fl Platt, Centenary Health Centre, Scarborough, Ontario; A Schuchat, CentersJor Disease Control and Prevention, Atlanta, Georgia; P Shokry, Markham Stol!ffville Hospital, Markham; AE Simor, Sunnybrook Health Science Centre, Toronto; F Smail!, Chedoke McMaster Hospital, Hamilton; D Sturman, Riverdale Hospital; E Szentgyorgi, North York Branson Hospital; M Tischler, Wellesley-Central Hospital; P Van Nostrand, The Toronto Rehabilitation Institute; M Veamcombe, Women's College Hospital; S Walmslry, The Toronto Hospital, Toronto, Ontario !ates are reported by telephone to the central study office at the Mount Sinai Hospital, Toronto , Ontario.Clinical data are collected from the patient's physicians and/or the hospital's infection control practitioner.For 1995, case isolates were shipped to the central laboratory, where they were confirmed as S pneumoniae, and antimicrobial susceptibility testing was performed to National Committee for Clinical Laboratory Standards (NCCLS) standards by microbroth dilution ( 1 ).Capsular typing of isolates was performed at the National Centre for

TABLE 1
Risk of invasive pneumococcal disease in adults associated with selected underlying conditions, metropolitan Toronto/Peel Region, 1995