Stratification of data to refine epidemiologic investigations

20 8 0
Stratification of data to refine  epidemiologic investigations

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

Thông tin tài liệu

Lecture 6: Stratification of data to refine epidemiologic investigations National Institute of Infectious Disease January 17, 2017 Endemic salmonellosis in São Paulo, Brazil in the  early 1980s Epidemiological question: What is the source of drug‐resistant  salmonellosis among children recently hospitalized in Sao Paulo— hospital or community? Salmonella serotype distribution, 8-outpatient clinic study, Sao Paulo, Brazil, 1981-82  S typhimurium 22  S agona 10  S infantis  S heidelberg  S panama  S havana  S eimsbuettel  Mixed (agona, infantis)  Total 40 History of recent hospitalization by Salmonella serotype, 8outpatient clinic study, Sao Paulo, Brazil 1981-82 S typhimurium Previously hospitalized Not hospitalized Total % hospitalized Other serotypes total 20 27 11 22 16 38 91 44 71 P=0.0029, Fisher’s exact test, 2-tailed Hospitalized children with S typhimurium infection, 8-outpatient clinic study, Sao Paulo, Brazil, 1981-82 Hospitalized children by S typhimurium plasmid profile pattern, 8-outpatient clinic study, Sao Paulo, Brazil, 1981-82 Profile A Profile B Profile C Profile D All others Endemic salmonellosis in São Paulo, Brazil conclusions  What appeared to be unrelated, endemic cases of drug-resistant salmonellosis in one city was comprised of several small outbreaks, unrecognized as outbreaks  About 45% of these cases of drug-resistant severe salmonellosis could be traced to specific hospitals in the city  Plasmid profile analysis facilitated a more refined study design— case-control—to identify specific risk factors E coli O157:H7 surveillance in Minnesota (Bender J et al, NEJM, 1997)  344 E coli O157:H7 cases reported to Minnesota Dept of Health between 1994-5  317 E coli O157:H7 isolates genotyped by PFGE  143 distinct PFGE patterns found  10 outbreaks discovered  These responsible for 56 (18%) of 317 isolates  outbreaks identified solely by PFGE study  Sudden increases may not necessarily be due to outbreaks All isolates By PFGE types E Coli O157:H7 Surveillance in Minnesota, 1994 From Bender et al, NEJM 1997 All isolates By PFGE types E Coli O157:H7 Surveillance in Minnesota, 1995 From Bender et al, NEJM 1997 Surveillance for E coli 0157:H7 infections, Minnesota: From Bender et al, NEJM 1997 From Bender et al NEJM 2001 Stratification of data to aid epidemiologic investigations:  Stratification: analytical tool used to determine the significance or strength of association of an attribute with disease or disease outcome, when that attribute belongs to a larger group sharing common characteristics Data stratification basic unit of epidemiologic analysis  Without laboratory support:  demographic (age, sex, race, country of origin, etc)  clinical (symptoms, signs)  socioeconomic (income, education, place of residence, occupation)  behavioral (smoking, ETOH, injection drug use, sexual practice)  belief systems  With laboratory support:  clinical (blood tests, physiologic tests, imaging, etc)  microbiologic (smears, cultures, nucleic acid detection, etc)  immunologic (serologic, skin tests, cell function assays, etc) Effect of data stratification on intervention options: example-Salmonellosis Type of stratification Intervention option -Diarrhea vs no diarrhea -Fluid, electrolyte support, handwashing -Salmonella in stool -Antibiotics? -Antibiotics -Salmonella bacteremia -Drug resistant salmonella -Case-control study to identify risk factor for drug-resistance -Plasmid profiles -Case-control study to identify source more precisely Stratification of data creates opportunities for  more specific and focused intervention options ... Bender et al NEJM 2001 Stratification of data to aid epidemiologic investigations:  Stratification: analytical tool used to determine the significance or strength of association of an attribute with... belongs to a larger group sharing common characteristics Data stratification basic unit of epidemiologic analysis  Without laboratory support:  demographic (age, sex, race, country of origin,... salmonellosis could be traced to specific hospitals in the city  Plasmid profile analysis facilitated a more refined study design— case-control? ?to identify specific risk factors E coli O157:H7 surveillance

Ngày đăng: 25/08/2021, 17:23

Tài liệu cùng người dùng

Tài liệu liên quan