[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Eva Poen <eva.poen@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: RE: Complex Medical Data |

Date |
Sun, 15 Feb 2009 10:32:15 +0000 |

TA, -reshape- does work for you. You simply need to fiddle a bit with the id variable. See here: egen namevisit = group(name visit) bys name visit (drug) : gen drugID = _n reshape wide drug, i(namevisit) j(drugID) gen str drug = drug1 + " " + drug2 + " " + drug3 + " " + drug4 gen GMMT = 0 replace GMMT = 1 if strpos(drug,"A") > 0 & strpos(drug,"C") > 0 replace GMMT = 1 if strpos(drug,"A") > 0 & strpos(drug,"D") > 0 replace GMMT = 1 if strpos(drug,"B") > 0 & strpos(drug,"C") > 0 replace GMMT = 1 if strpos(drug,"A") > 0 & strpos(drug,"B") > 0 & strpos(drug,"A4") == 0 & strpos(drug,"B3") == 0 replace GMMT = 1 if strpos(drug,"A") > 0 & strpos(drug,"B") > 0 & strpos(drug,"C") > 0 And the result is . list name visit drug GMMT +-----------------------------------+ | name visit drug GMMT | |-----------------------------------| 1. | Jen 1 B3 C3 D4 1 | 2. | Jen 2 A4 B3 C3 D4 1 | 3. | Jen 3 A4 B3 0 | 4. | Jen 4 A4 B3 C4 1 | 5. | John 1 A1 B1 C1 1 | |-----------------------------------| 6. | John 2 A1 B1 1 | 7. | John 3 C1 D1 0 | 8. | John 4 B2 C2 D1 1 | +-----------------------------------+ HTH, Eva 2009/2/15 TA Stat <tastat@gmail.com>: > If reshape wide is used, that means "visit" is ignored. However, but > both visit and drug are important in this case. Look at each clinic > visit of John, he received GMMT at visit 1, visit 2 and visit 3, but > not visit 4. GMMT can't be determined by one drug, but it is a > combination of some drugs. In other words, I want to conclude that at > each visit, each patient received GMMT or not. > > I can do Table 2 to Table 3, but not Table 1 to Table 2. > > TA > > > On Sun, Feb 15, 2009 at 4:01 AM, Tom Trikalinos <ttrikalin@gmail.com> wrote: >> TA, see >> >> help reshape wide >> >> in reshaping, j will be your drug >> >> hth, >> >> TAT >> >> >> On Sat, Feb 14, 2009 at 12:22 AM, TA Stat <tastat@gmail.com> wrote: >>> For example, from table 1 to table 2, create a variable name "GMMT" >>> with the conditions explain earlier. John received GMMT for visit 1, >>> visit 2, visit 4, but not visit 3. Jen received GMMT at visit 1, >>> visit 2, visit 4, but not visit 3. >>> >>> Although at visit 4 for Jen, a combination of A4 and B3 is not GMMT, >>> but she also received C4. So she received GMMT. At visit 3 for John, >>> he received C1 and D1 (any combination of drug from Class C and Class >>> D), so he received GMMT. >>> >>> I want to know how to get Table 2 from Table 1. >>> >>> *************************************************************************************** >>> >>> Table 1 >>> >>> Name Visit Drug >>> >>> John 1 A1 >>> John 1 B1 >>> John 1 C1 >>> John 2 A1 >>> John 2 B1 >>> John 3 C1 >>> John 3 D1 >>> John 4 B2 >>> John 4 C2 >>> John 4 D1 >>> Jen 1 B3 >>> Jen 1 C3 >>> Jen 1 D4 >>> Jen 2 A4 >>> Jen 2 B3 >>> Jen 2 C3 >>> Jen 2 D4 >>> Jen 3 A4 >>> Jen 3 B3 >>> Jen 4 A4 >>> Jen 4 B3 >>> Jen 4 C4 >>> >>> *************************************************************************************** >>> >>> Table 2 >>> >>> Name Visit Drug GMMT (Y=Yes, N=No) >>> >>> John 1 A1 Y >>> John 1 B1 Y >>> John 1 C1 Y >>> John 2 A1 Y >>> John 2 B1 Y >>> John 3 C1 N >>> John 3 D1 N >>> John 4 B2 Y >>> John 4 C2 Y >>> John 4 D1 Y >>> Jen 1 B3 Y >>> Jen 1 C3 Y >>> Jen 1 D4 Y >>> Jen 2 A4 Y >>> Jen 2 B3 Y >>> Jen 2 C3 Y >>> Jen 2 D4 Y >>> Jen 3 A4 N >>> Jen 3 B3 N >>> Jen 4 A4 Y >>> Jen 4 B3 Y >>> Jen 4 C4 Y >>> >>> >>> *************************************************************************************** >>> >>> Table 3 >>> >>> Name Visit Drug GMMT (Y=Yes, N=No) >>> >>> John 1 A1 Y >>> John 2 A1 Y >>> John 3 C1 N >>> John 4 B2 Y >>> Jen 1 B3 Y >>> Jen 2 A4 Y >>> Jen 3 A4 N >>> Jen 4 A4 Y >>> >>> >>> On Sat, Feb 14, 2009 at 2:29 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote: >>>> <> >>>> >>>> An example of data would help make this clearer. >>>> >>>> Nick >>>> n.j.cox@durham.ac.uk >>>> >>>> TA Stat >>>> >>>> I have data arranged in the format that each row represents one drug >>>> for clinic visit. So, one patient may have more than one drug (row) >>>> for each visit. Each patient also have several clinic visits. >>>> >>>> There are 4 classes of drugs, called Class A, Class B, Class C and >>>> Class D. Class A has four drugs, called A1, A2, A3 and A4. Class B has >>>> 3 drugs, called B1, B2, B3. so as Class C and Class D (as presented >>>> below). Some combination of drugs is called GMMT, other is not. I >>>> would like to find if each patient has received GMMT at each clinic >>>> visit or not. Basically, I want to identify whether patient receive >>>> GMMT for each visit or not. >>>> >>>> Class A: A1, A2, A3, A4 >>>> >>>> Class B: B1, B2, B3 >>>> >>>> Class C: C1, C2, C3 >>>> >>>> Class D: D1, D2, D3, D4 >>>> >>>> a combination of drug is called GMMT if >>>> >>>> Class A and Class B >>>> >>>> Class A and Class C >>>> >>>> Class A and Class D >>>> >>>> Class B and Class C >>>> >>>> Class A and Class B and Class C >>>> >>>> (However, there are some exception) >>>> >>>> but not for Class C and Class D >>>> >>>> but not for A4 and B3 >>>> >>>> >>>> * >>>> * For searches and help try: >>>> * http://www.stata.com/help.cgi?search >>>> * http://www.stata.com/support/statalist/faq >>>> * http://www.ats.ucla.edu/stat/stata/ >>>> >>> * >>> * For searches and help try: >>> * http://www.stata.com/help.cgi?search >>> * http://www.stata.com/support/statalist/faq >>> * http://www.ats.ucla.edu/stat/stata/ >>> >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/statalist/faq >> * http://www.ats.ucla.edu/stat/stata/ >> > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ > * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: RE: Complex Medical Data***From:*TA Stat <tastat@gmail.com>

**References**:**st: Complex Medical Data***From:*TA Stat <tastat@gmail.com>

**st: RE: Complex Medical Data***From:*"Nick Cox" <n.j.cox@durham.ac.uk>

**Re: st: RE: Complex Medical Data***From:*TA Stat <tastat@gmail.com>

**Re: st: RE: Complex Medical Data***From:*Tom Trikalinos <ttrikalin@gmail.com>

**Re: st: RE: Complex Medical Data***From:*TA Stat <tastat@gmail.com>

- Prev by Date:
**Re: st: Overfitting?** - Next by Date:
**Re: st: event history analysis with years clustered in individuals** - Previous by thread:
**Re: st: RE: Complex Medical Data** - Next by thread:
**Re: st: RE: Complex Medical Data** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |