ࡱ> :<9)` Qbjbj .& 8 <& $F \ ( $h66 4Kp  l : ~| (a0,X66P $  Lab: Experimental Study Design Drinking Water Intervention Read Colford, J. M., Jr., Wade, T. J., Sandhu, S. K., Wright, C. C., Lee, S., Shaw, S., et al. (2005). A randomized, controlled trial of in-home drinking water intervention to reduce gastrointestinal illness. Am J Epidemiol, 161(5), 472 482. [HYPERLINK "http://www.sjsu.edu/faculty/gerstman/eks/Colford2005.pdf"Link to article]. Questions 1. Experimental studies and observational studies both make causal comparisons of exposed and non-exposed groups. Clarify the primary advantage of using a randomized experiment for this purpose. (50 word maximum) 2. Why was blinding employed in the Colford et al. study? Give examples of improper influences that would have occurred without blinding? (100 words maximum) 3. Colford employed a number of procedures to blind installers, participants, and investigators. Describe three of the procedures employed for this purpose. (100 words maximum) 4. Use the data in Table 4 of the article to calculate the following crude rates of highly credible gastrointestinal illness (HCGI). Show your calculations in each instance. [Recall: rate = the number of episodes divided by person-time] (a) Rate of HCGI during cycle A among active device participants. (b) Rate of HCGI during cycle A among sham device participants. (c) Rate of HCGI in both cycles combined among active device participants. (d) Rate of HCGI in both cycles combined among sham device participants. 5. Use results from the preceding question to calculate the crude rate ratio comparing sham device users to active device users for both cycles combined. Let the active device users serve as the reference (nonexposed) group for the calculation. [Recall: RR = rate in exposed / rate in non-exposed]. 6. How do your calculation in question 5 compare with the results from the authors multivariate RR associated with the device for episodes of highly credible gastrointestinal illness adjusted for additional covariates (reported in Table 5 on p. 479)? 7. Calculate the crude rate difference for HCGI in both cycles combined comparing sham device participants to active device participants. Discuss your findings.     Vic Schoenbach (UNC EPID600 case study) Page  PAGE 1 Modified with permission by B. Gerstman (2007)  FILENAME \p F:\data\eks\Lab_Experimental_Study_Design.doc ;<@A + . 0 1 2 3 w x y Ʒ{{h\Fh5 h-ehN<0JCJOJQJaJ+jhFh-CJOJQJUaJhFCJOJQJaJ%jh-ehN<CJOJQJUaJh-ehN<CJOJQJaJhLCJOJQJaJ"h-eh s6CJOJQJ]aJh-eh sCJOJQJaJh-ehoPCJOJQJaJ"hO'56CJOJQJ\]aJh-eh%OJQJh-eh)5OJQJ\h-eh%CJOJQJaJ;< o p   6  0^`0gd_ 0^`0gd~vgd)gd%gdoPgdN<$a$gd)$a$gd@P sdXsdsIs:sIh-eh)CJOJQJaJh-ehq,CJOJQJaJh>䴳ϴ-r䴳ϴ-䴳ϴ-%䴳ϴ(-%56Cϴ]aO'䴳ϴ->+䴳ϴ-M䴳ϴ-<CJOJQJaJ%jh-ehN<CJOJQJUaJhw#0JCJOJQJaJ   ) - 5 @ [ _ o p w z     = B C p y  ĸĚĚĚĚĚĚĚ|ĚĚ|h-ehoPCJOJQJaJh-ehx'CJOJQJaJh-eh%CJOJQJaJh-ehCJOJQJaJh-eCJOJQJaJh-ehrWCJOJQJaJh-ehq,CJOJQJaJh-ehp@CJOJQJaJh-eh &CJOJQJaJ-    > ? M R _ o q r z   4 8 9 : ; < ǸǸӸ|h-eh~vCJOJQJaJh-ehn CJOJQJaJh-ehCJOJQJaJh-ehoPCJOJQJaJh-eh4CJOJQJaJh>BCJOJQJaJh-eh)CJOJQJaJh-eh%CJOJQJaJh-eh_CJOJQJaJ0 06GMegk} "3;MZǸӸǸǚNjǚǩ|NjNjөөǩǚǩh-ehCJOJQJaJh-ehoPCJOJQJaJh-eh)CJOJQJaJh-ehRACJOJQJaJh-ehn CJOJQJaJh>BCJOJQJaJh-eh),CJOJQJaJh-eh}vCJOJQJaJh-eh%CJOJQJaJ-MNOPQgd@gd@gd?gdigdoPgdRA_KLMӻ|mbUmhKCJaJmHnHuh@hc_-CJaJjh@hc_-CJUaJhC CJaJmHnHujhc_-CJUaJjh!vhc_-CJUaJh!vhc_-CJaJhc_-CJaJhYjhYUh>Bh%CJOJQJ]aJh>Bh%CJOJQJaJh>BhRACJOJQJaJh>Bh-eCJOJQJaJMNOPQh>Bh%CJOJQJaJhYhc_- h@hc_-,1h/ =!"#$% DyK yK http://www.sjsu.edu/faculty/gerstman/eks/Colford2005.pdfyX;H,]ą'c@@@ NormalCJ_HaJmH sH tH V@V @ Heading 3$<@&5CJOJQJ\^JaJDA@D Default Paragraph FontViV  Table Normal :V 44 la (k(No List 6U@6 N< Hyperlink >*B*ph4@4 @Header  !4 @4 @Footer  !>"> x' Footnote TextCJaJ@&1@ x'Footnote ReferenceH*FV@AF )~FollowedHyperlink >*B* phHRH Ctr Balloon TextCJOJQJ^JaJQ &;<op6M N O R (0000000000000000000000000@0%000@0%000@0%000@0%000@0@0@0@0%00` 0;<6R (0000K00K00hK0000@0K0 0K0 0K0 0K0 0@00  MQ Q P 2xQ X<CEx!@  @ 0(  B S  ?H0(  'AHO R : +/1O R 333< ]ppO R <@AAO R s U/$ h hh^h`hH.h 88^8`hH.h L^`LhH.h   ^ `hH.h   ^ `hH.h xLx^x`LhH.h HH^H`hH.h ^`hH.h L^`LhH.s U         PO@op C |Y%NN!w#'O'p@*>+),ù,]--<jz<p@>B>LoP%!SWrW>X%ZL4_, b-e)hikn&GqCtr s}v /z)~M~RA~vG-qtd9eFi`f>[-n Ax'WT|iKBjpFi'? &749bL_l[)@:Q @UnknownG:Ax Times New Roman5Symbol3& :Cx Arial7&@Calibri5& >[`)Tahoma"1h̫f ڦ"HOHO&4d 2QHX ?N<2Lab: Experimental Study Design Bud Gerstman B. Gerstman Oh+'0 ,8 X d p | Lab: Experimental Study DesignBud Gerstman Normal.dot B. Gerstman34Microsoft Office Word@G@@ZN@8,@,|HO՜.+,D՜.+,d  hp  San Jose State University' Lab: Experimental Study Design Title 8@ _PID_HLINKSA9http://www.sjsu.edu/faculty/gerstman/eks/Colford2005.pdfm  !"#$%&'(*+,-./02345678;Root Entry F8|=Data 1TableWordDocument.&SummaryInformation()DocumentSummaryInformation81CompObjq  FMicrosoft Office Word Document MSWordDocWord.Document.89q