tag:blogger.com,1999:blog-3376652857238136457.post2513184013397788269..comments2023-10-18T02:15:03.175-07:00Comments on Informs Data Mining Contest: Questions on data set and nosocomial pneumonia as a responseNickhttp://www.blogger.com/profile/07170357132948522449noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-3376652857238136457.post-22969310910760050752008-07-18T14:58:00.000-07:002008-07-18T14:58:00.000-07:00Apologies for a confusion-inducing mistake in the ...Apologies for a confusion-inducing mistake in the above answer: There is no DRG code recorded in this dataset. In the comments made in response to questions 1 and 3, the statements made about DRG codes are instead true of ICD9 codes.<BR/><BR/>So in answer to Sridhar:<BR/>1. Sorry, there are no DRG codes.<BR/>2. Pretty much, yes.<BR/>3. See 1.Nickhttps://www.blogger.com/profile/07170357132948522449noreply@blogger.comtag:blogger.com,1999:blog-3376652857238136457.post-81632194687417000952008-07-01T04:44:00.000-07:002008-07-01T04:44:00.000-07:00We are still not quite clear about identifying pat...We are still not quite clear about identifying patients with nosocomial infection. <BR/><BR/>1.It is not clear to us which variable in the data sets specifies the DRG code for the patient.<BR/>2.If there is a secondary diagnosis of pneumonia according to icd codes, does that necessarily imply nosocomial pneumonia? <BR/>3.Is it possible that DRG code relates to pneuomonia while the primary icd code does not relate to pneumonia?Sridharhttps://www.blogger.com/profile/00205098975205173296noreply@blogger.comtag:blogger.com,1999:blog-3376652857238136457.post-65612889460359668232008-06-23T10:11:00.000-07:002008-06-23T10:11:00.000-07:00Following is a forwarded response from Patricia:1....Following is a forwarded response from Patricia:<BR/><BR/><BR/>1. Our first question relates to the relevance of the conditions data base. Since there does not appear to be any way to determine if pneumonia for these patients was nosocomial, the conditions data set seems not to be relevant to classification into a "likely to contract nosocomial pneumonia" group. Are we correct in this conclusion?<BR/><BR/>It is difficult but not impossible to determine whether the condition was nosocomial or not. Examine the DRG code for the patient-is it pneumonia or something else? Is pneumonia primary or secondary? If the DRG is for something else, the pneumonia is nosocomial.<BR/><BR/>2. Related to the first question, what exactly does the variable ICD9CODX in the conditions data set represent? How does it relate to the four diagnosis codes in the hospital data?<BR/><BR/>The conditions data set contains all ICD9 codes associated with that patient in the course of a year. There is exactly one code (ICD9CODX) per observation, and many observations per patient.<BR/><BR/>3. Back to the hospital data set. What exactly does SPECCOND represent? How does it help identify the nosocomial pneumonia cases? The definition is vague, and does not correlate directly with RSINNHOS. For example, there are 238 records where SPECCOND is 2, so the hospital stay is not related to the condition. One would suspect that nosocomial pneumonia cases would be included in this number. However, of the 257 records that show pneumonia codes in one of the four ICDx variables and where SPECCOND is not missing, all have SPECCOND = 1.<BR/><BR/>These are flag variables. This is another way of determining nosocomial or not. If it is not related to the condition, then the code will be 2. I would suggest looking at this variable more as a validation of the information from the DRG codes concerning whether or not the condition is nosocomial.<BR/><BR/>4. In the medication data set Rx0304, it is not clear what the nine variables on the Type of Pharmacy Provider (PHARTP1 through PHARTP9) represent. How do these nine variables relate to a single prescription?<BR/><BR/>Don't be overly concerned with the type of pharmacy. Be more concerned with the type of drug. In particular, was an antibiotic used to treat the patient with pneumonia?Nickhttps://www.blogger.com/profile/07170357132948522449noreply@blogger.com