810-S HOSPITAL-BASED SURVEILLANCE FOR INVASIVE PNEUMOCOCCAL DISEASE IN KHANH HOA PROVINCE, VIETNAM: EPIDEMIOLOGIC CHARACTERISTICS AND JUSTIFICATION FOR A PNEUMOCOCCAL VACCINE. D D Anh, T T Minh, L H Tho, M P E Slack, L Jodar, N Batmunkh, P E Kilgore, and *C A Talarico (International Vaccine Institute, Seoul, Korea, 151–600) Worldwide, invasive pneumococcal disease (IPD) (pneumonia, meningitis, and sepsis) is a major cause of mortality and long-term neurologic sequelae in children. In Asia, a few studies have suggested that Streptococcus pneu- moniae is a leading cause of meningitis and lower respiratory tract infec- tions in children, but no studies have systematically described the burden of IPD. In order to document the epidemiologic characteristics of children with IPD and estimate the annual IPD incidence proportion, a hospital- based surveillance system was established at Khanh Hoa General Hospital (KHGH) in Khanh Hoa Province, Vietnam. Participating children were 5 years and were admitted to KHGH with clinical symptoms of pneumonia, meningitis, and/or sepsis. Chest x-ray and serum were collected from all participants for microbiologic testing, and cerebrospinal fluid was collected from suspected meningitis patients. Parent interviews were conducted to collect demographic and health history information. The mean age of all patients was 13 months, and 57% were male (n ¼ 722). Eighty-two percent of patients were diagnosed with pneumonia, 2% had meningitis, and 1% had sepsis. Forty-six percent of patients reported taking antibiotics during the two weeks prior to the hospital admission. On average, IPD patients reported living with six other people in a dwelling with 2–3 rooms. Only 30% of patients were attending daycare or school at the time of illness onset. A pneumococcal vaccine may be justified in this high transmission and high antibiotic consumption environment. 811-S FACTORS INFLUENCING FLU VACCINE AVOIDANCE AMONG ADULTS OVER 50 IN HOUSTON, TEXAS, 2004–05. *K J Chronister, B J Selwyn, C Amos Jr, M A Livoti-DeBellis, L K Bartholomew (University of Texas School of Public Health, Houston, TX 77030) Influenza causes more than 114,000 hospitalizations and 34,000 deaths annually predominately among persons over 65, yet in 2003 vaccination rates were 65.4% for this age group. Disparities in vaccination coverage between non-Hispanic blacks, whites, and Hispanics continue unabated. To describe factors associated with vaccine avoidance among adults at least 50 years old, two rapid assessment studies of shoppers at community grocery stores were conducted regarding prevalent opinions on getting flu shots. The first study used semi-structured qualitative interviews of about 60 people per ethnic group. Interviews were analyzed for themes related to vaccine avoidance like experiences with flu vaccines, perceived risk, and doctor recommendation. These findings were integrated into the instrument for the project’s second phase. We then used a venue-based sampling method to gather data from about 100 people in each of three ethnicities. This prevalence study included factors associated with underutilization of flu vaccine as well as items relating to the themes previously identified. Vaccine avoidance in the 2004 flu season varied across ethnic groups (26.8% whites, 37.2% Hispanics, 40.6% in blacks). Previous experiences with flu shots and the perception that the shot causes the flu is associated with a higher prevalence of vaccine avoidance depending on race. The perception of risk with respect to intention to get a flu shot had mixed associations but was very strong among blacks. Indicators of negative atti- tudes about flu shots are also associated with vaccine avoidance in all groups. Targeted message development is needed to improved flu vaccine coverage to address these factors. 812 PNEUMOCOCCAL CONJUGATE VACCINE EFFECT ON STREPTOCOCCUS PNEUMONIAE SEROTYPE CARRIAGE AND MULTIPLE DRUG RESISTANCE IN THE UNITED STATES. *R Mera (GlaxoSmithKline, Research Triangle Park, NC 27709) The heptavalent pneumococcal conjugate vaccine has been available since February 2000 and has reached coverage of 73% in 2004. Changes over time in the serogroup proportion and multiple resistance (MR) to antibiotics is evaluated using a surveillance study. 704 isolates were sampled from a large organism database, 301 from the years before the introduction of the vaccine (1998–1999) and 298 isolates for 2003–2004. The strains originated from the nine US census regions. Age, sex, geographic origin, isolate source, serotype and resistance information were available for the analysis. Among all isolates the prevalence of vaccine serotypes (VS) and non-vaccine sero- types (NVS) was 66% and 34% in the pre-vaccine era, with a complete inversion in the post-vaccine era. The proportion of VS in adults declined from 43% to 28% after the introduction of the vaccine. NVS were 1.8 times (95% CI 1.1–3.1) more likely to acquire MR over time. Due to serotype switch MR was just slightly higher in the post vaccine era, 47% vs. 46% than in the pre-vaccine era. MR among invasive isolates did not change, but increased among NVS otitis media isolates by 17%. NVS are replacing VS serotypes as the source of invasive disease (71% in the post era compared to 43% before the vaccine), but at a lower prevalence. Although MR has been increasing the fastest among non-invasive isolates (middle ear), isolates from all sources display a similar pattern. PCV7 serotype replacement in children < 5 by 2004 was over 101%. A herd immunity effect in adults has contributed to the switch process. NVS have acquired multiple resistance at a rate that is proportional to the replacement process. Overall multiple re- sistance will likely continue to increase despite serotype replacement. 813 IMPLICATIONS OF UNIVERSAL HEPATITIS A VACCINATION IN CANADA: PREDICTIONS OF A DYNAMIC MODEL. A S R S Rao, *C T Bauch, B. Pham, M Krahn, V Gilca, B Duval, M H Chen, A Tricco (University of Guelph, Canada) Vaccination against Hepatitis Avirus (HAV) in Canada is currently targeted toward high risk groups. However universal vaccination has been adopted in other regions with a similar disease burden. Here we develop an age- structured compartmental model of HAV transmission to predict the impact of universal vaccination in Canada. The model incorporates the peculiari- ties of HAV transmission in Canada (and many other low-incidence coun- tries), such as a cohort effect, under-reporting, subclinical infection, and the large role played by travel-related incidence. The model shows that the transmissibility of HAV has declined by a factor of 2.8 since the early twentieth century in Canada, producing the cohort effect observed in the age-stratified seroprevalence profile for Canadian-born individuals. Three possible universal vaccination strategies (age 4, age 9, age 15) are consid- ered, all of which are shown to significantly reduce morbidity and mortality. However, gains are substantially larger for vaccination at 4 years of age. Universal vaccination at age 4 (resp. 9, 15) would reduce reported incidence by 64% (resp. 45%, 39%) and mortality attributable to HAV by 52% (resp. 30%, 27%), relative to no vaccination, over fifty years. 2006 Congress of Epidemiology Abstracts S203 Am J Epidemiol 2006;163(Suppl):S1–S258 * ¼ Presenter; S ¼ The work was completed while the presenter was a student Downloaded from https://academic.oup.com/aje/article-abstract/163/suppl_11/S203/114574 by guest on 08 June 2020