/* Dear SEER-Medicare user:                                               */
/*   In order to use this program, you must make some modifications.      */
/* All lines which may need modifications have the comment USER next to   */
/* them so that you can search for them more easily.                      */
/*                                                                        */
/*   First, change the filenames to match your own directory structure.   */
/* You need to have an input PEDSF file (the original file sent to you by */
/* IMS when you received your data request) and an output PREP file name. */
/* You may also have an input ANALYSIS file (the file you have created,   */
/* containing your selected cohort and 'cleaned' data) if you wish to use */
/* variables which you have defined/created in the SEER*Stat analyses.    */
/* Your analysis file must contain the SEER Registry/Case number as a     */
/* length 10 character string and the SEER Sequence number for the tumor  */
/* as a length 2 character string.  All analysis variables you wish to    */
/* use in SEER*Stat must be numeric.  You should either have a SAS file   */
/* (libname) or a text file (filename) statement for each analysis file,  */
/* NOT BOTH.  TWO SAMPLE SAS DECLARATIONS ARE PROVIDED, YOU MUST SELECT   */
/* ONE.  These samples are both commented out.  If you wish to use one,   */
/* you must uncomment one by removing the * in the first column and       */
/* change the directory path/file name to match your file.                */
/*                                                                        */
/*   Second, you should review the end of study year and month (EOS_Y and */
/* EOS_M).  These are preset to the last month of the Medicare expense    */
/* files which were sent to you by IMS.  If your analysis ends at an      */
/* earlier time, please adjust these 2 variables accordingly.             */
/*                                                                        */
/*   Third, you should modify the 'ANALYSIS MERGE SECTION', where the     */
/* analysis file is merged to the PEDSF information.  A SAS file is       */
/* assumed here, so you may have to add an input statement and you will   */
/* always have to select which variables you wish to add.  Again, this    */
/* section is preset to be commented out.  To uncomment, delete the 2     */
/* lines 'USER comment out begins' and 'USER comment out ends'            */
/*                                                                        */
/*   Finally, if you have added new variables, you must place them in the */
/* output PREP file where the lenX_n variables are shown.  X (lenX) tells */
/* you the length of a variable that can be placed in that position.      */
/* n (_n) allows you to track how many variables of that length are used. */
/* There are 10 variables of length 1 and 10 of length 2.  There are 5    */
/* variables of length 3, 5 of length 4 and 5 of length 5.  The total     */
/* record length is 674.  Your variables MUST be of numeric type and be   */
/* sure to use an appropriate length variable.                            */
/*                                                                        */
/* ** NOTE:  You can NOT use the 90 characters of lenX_n space in any     */
/*           manner, variables MUST be placed in appropriate sized slots. */
/*                                                                        */
/*  Thank you and please let us know if you have any questions.           */
/*                              IMS support staff                         */


/* input files */
filename PEDSF    '/directory/pedsf.cancer.txt';           /* USER - PEDSF file received from IMS */
*libname  ANALYSIS '/directory/';			   /* USER - SAS type analysis file */
*filename ANALYSIS '/directory/analysis.txt';              /* USER - Text type analysis file */

/* output file */
filename PREP  '/directory/SEERPrep.infile.txd';           /* USER - output file being created.  Must have .txd extension for SEER*Prep */

data prep(drop=ctnon90x cthso90x ctscl90x ctcol90x ctblk90x ctwht90x cthsp90x ctpov90x  
	       cten590x cten690x povrc90x nonrc90x hsorc90x sclrc90x colrc90x zpnon90x  
	       zphso90x zpscl90x zpcol90x zpblk90x zpwht90x zphsp90x zpen590x zpen690x  
	       znonrc9x zhsorc9x zsclrc9x zcolrc9x ctnon00x cthso00x ctscl00x ctcol00x  
	       ctblk00x ctwht00x cthsp00x ctpov00x cten500x cten600x povrc00x nonrc00x  
	       hsorc00x sclrc00x colrc00x zpnon00x zphso00x zpscl00x zpcol00x zpblk00x  
	       zpwht00x zphsp00x zpen500x zpen600x znonrc0x zhsorc0x zsclrc0x zcolrc0x  
	       ctpci90x ctmed90x zppci90x zpmed90x ctpci00x ctmed00x ctden00x medag00x  
	       medrc00x zppci00x zpmed00x zpden00x pos i eos_y eos_m count);  /* ctden90x medag90x medrc90x zpden90x */

  EOS_Y = '2005';	     				   /* USER - End of Study Year */ 
  EOS_M = '12';					           /* USER - End of Study Month */ 

  infile PEDSF lrecl=2273 recfm=f missover pad;
  input 
      @0001  regcase   $char10.
      @0011  fivepct   $char1.
      @0022  dod_flg   $char1.
      @0024  med_dodm  $char2.
      @0026  med_dodd  $char2.
      @0028  med_dody  $char4.
      @0032  birthm    $char2.
      @0036  birthyr   $char4.
      @0040  m_sex     $char1.
      @0041  race      $char1.
      @0042  rsncd1    $char1.
      @0043  cur_ent   $char1.
      @0044  chr_esrd  $char1.
      @0045  med_stcd  $char2.      
      @0047  STATE     $char2.
      @0049  COUNTY    $char3.
      @0052  zip       $char9.        /*Special permission required*/
      @0061  code_sys  $char1.
      @0062  census    $char6.        /*Special permission required*/
      @0068  census2k  $char6.        /*Special permission required*/
      @0074  HSA       $char3.
      @0077  URBRUR    $char1.
      @0078  URBAN     $char2.      
      @0080  ctpci90x  $char10.
      @0090  ctmed90x  $char10.
/*    @0100  ctden90x  $char10.      all missing*/
      @0110  ctnon90x  $char6.
      @0116  cthso90x  $char6.
      @0122  ctscl90x  $char6.
      @0128  ctcol90x  $char6.
      @0134  ctblk90x  $char6.
      @0140  ctwht90x  $char6.
      @0146  cthsp90x  $char6.
      @0152  ctpov90x  $char6.
      @0158  cten590x  $char6.
      @0164  cten690x  $char6.
/*    @0170  medag90x  $char10.      all missing*/
/*    @0180  medrc90x  $char10.     all missing*/
      @0190  povrc90x  $char6.
      @0196  nonrc90x  $char6.
      @0202  hsorc90x  $char6.
      @0208  sclrc90x  $char6.
      @0214  colrc90x  $char6.            
      @0220  zppci90x  $char10.
      @0230  zpmed90x  $char10.
/*    @0240  zpden90x  $char10.     all missing*/
      @0250  zpnon90x  $char6.
      @0256  zphso90x  $char6.
      @0262  zpscl90x  $char6.
      @0268  zpcol90x  $char6.
      @0274  zpblk90x  $char6.
      @0280  zpwht90x  $char6.
      @0286  zphsp90x  $char6.
      @0292  zpen590x  $char6. 
      @0298  zpen690x  $char6.
      @0304  znonrc9x  $char6.
      @0310  zhsorc9x  $char6.
      @0316  zsclrc9x  $char6.
      @0322  zcolrc9x  $char6.
      @0328  ctpci00x  $char10.
      @0338  ctmed00x  $char10.
      @0348  ctden00x  $char10.
      @0358  ctnon00x  $char6.
      @0364  cthso00x  $char6.
      @0370  ctscl00x  $char6.
      @0376  ctcol00x  $char6.
      @0382  ctblk00x  $char6.
      @0388  ctwht00x  $char6.
      @0394  cthsp00x  $char6.
      @0400  ctpov00x  $char6.
      @0406  cten500x  $char6.
      @0412  cten600x  $char6.      
      @0418  medag00x  $char10.
      @0428  medrc00x  $char10.
      @0438  povrc00x  $char6.
      @0444  nonrc00x  $char6.
      @0450  hsorc00x  $char6.
      @0456  sclrc00x  $char6.
      @0462  colrc00x  $char6.      
      @0468  zppci00x  $char10.
      @0478  zpmed00x  $char10.
      @0488  zpden00x  $char10.      
      @0498  zpnon00x  $char6.
      @0504  zphso00x  $char6.
      @0510  zpscl00x  $char6.
      @0516  zpcol00x  $char6.
      @0522  zpblk00x  $char6.
      @0528  zpwht00x  $char6.
      @0534  zphsp00x  $char6.
      @0540  zpen500x  $char6.
      @0546  zpen600x  $char6.
      @0552  znonrc0x  $char6.
      @0558  zhsorc0x  $char6.
      @0564  zsclrc0x  $char6.
      @0570  zcolrc0x  $char6. 
      @0576  reg65     $char2.          
      @0578  S_SEX     $char1.
      @0579  RAC_RECB  $char2.
      @0581  ICD_CODE  $char1.
      @0582  COD89V    $char4.
      @0586  COD10V    $char4.
      @0590  codkm     $char3.
      @0593  SER_DODM  $char2.
      @0595  SER_DODY  $char4.
      @0599  AGE_DX    $char3.
      @0602  sRACE     $char2.
      @0604  ORIGIN    $char1.
      @0605  DX65M     $char2.
      @0607  DX65Y     $char4.
      @0611  RESNREC   $char1.
      @0612  STAT_REC  $char1.
      @0619  plc_brth  $char3.
      @0622  COUNT     $char2.
      @0624  NUMPRIMS  $char2.
      @0626  AGE1DX    $char3.
      @0629  SEQ1OV65  $char1.
      @0630  MULTSITE  $char1.
      @0631  SEQ1DX    $char2.             
      @;

  /* change text fields to numeric */
  if fivepct='N' then fivepct='0';
  else fivepct='1';

  if chr_esrd='Y' then chr_esrd='1';

  if med_dody='    ' then med_dody='0000';
  if med_dodm='  ' then med_dodm='00'; 
  if med_dodd='  ' then med_dodd='00'; 

  /* change . to blank, modify lengths */
  length ctnon90 cthso90 ctscl90 ctcol90 ctblk90 ctwht90 cthsp90 ctpov90 
         cten590 cten690 povrc90 nonrc90 hsorc90 sclrc90 colrc90 zpnon90 
	 zphso90 zpscl90 zpcol90 zpblk90 zpwht90 zphsp90 zpen590 zpen690 
	 znonrc90 zhsorc90 zsclrc90 zcolrc90 ctnon00 cthso00 ctscl00 
	 ctcol00 ctblk00 ctwht00 cthsp00 ctpov00 cten500 cten600 povrc00
	 nonrc00 hsorc00 sclrc00 colrc00 zpnon00 zphso00 zpscl00 zpcol00
	 zpblk00 zpwht00 zphsp00 zpen500 zpen600 znonrc00 zhsorc00 zsclrc00 
	 zcolrc00 $ 5;

  length ctpci90 ctmed90 zppci90 zpmed90 ctpci00 ctmed00 ctden00 medag00 
         medrc00 zppci00 zpmed00 zpden00 $ 6;    /* ctden90 medag90 medrc90 zpden90 */ 

  /* replace . with blank, shorten to length 6 */
  ctpci90=put(input(substr(ctpci90x,5,6),6.),Z6.);       if ctpci90='     .' then ctpci90='      ';
  ctmed90=put(input(substr(ctmed90x,5,6),6.),Z6.);       if ctmed90='     .' then ctmed90='      '; 
  /*  ctden90=put(input(substr(ctden90x,5,6),6.),Z6.);       if ctden90='     .' then ctden90='      ';    */
  /*  medag90=put(input(substr(medag90x,5,6),6.),Z6.);       if medag90='     .' then medag90='      ';    */
  /*  medrc90=put(input(substr(medrc90x,5,6),6.),Z6.);       if medrc90='     .' then medrc90='      ';    */
  zppci90=put(input(substr(zppci90x,5,6),6.),Z6.);       if zppci90='     .' then zppci90='      '; 
  zpmed90=put(input(substr(zpmed90x,5,6),6.),Z6.);       if zpmed90='     .' then zpmed90='      '; 
  /*  zpden90=put(input(substr(zpden90x,5,6),6.),Z6.);       if zpden90='     .' then zpden90='      ';    */
  ctpci00=put(input(substr(ctpci00x,5,6),6.),Z6.);       if ctpci00='     .' then ctpci00='      '; 
  ctmed00=put(input(substr(ctmed00x,5,6),6.),Z6.);       if ctmed00='     .' then ctmed00='      '; 
  ctden00=put(input(substr(ctden00x,5,6),6.),Z6.);       if ctden00='     .' then ctden00='      '; 
  medag00=put(input(substr(medag00x,5,6),6.),Z6.);       if medag00='     .' then medag00='      '; 
  medrc00=put(input(substr(medrc00x,5,6),6.),Z6.);       if medrc00='     .' then medrc00='      '; 
  zppci00=put(input(substr(zppci00x,5,6),6.),Z6.);       if zppci00='     .' then zppci00='      '; 
  zpmed00=put(input(substr(zpmed00x,5,6),6.),Z6.);       if zpmed00='     .' then zpmed00='      '; 
  zpden00=put(input(substr(zpden00x,5,6),6.),Z6.);       if zpden00='     .' then zpden00='      '; 

  /* replace . with blank, shorten to length 5 */
  ctnon90 = put(input(compress(ctnon90x,'.'),6.),Z5.);   if ctnon90 ='    .' then ctnon90 ='     ';
  cthso90 = put(input(compress(cthso90x,'.'),6.),Z5.);   if cthso90 ='    .' then cthso90 ='     ';
  ctscl90 = put(input(compress(ctscl90x,'.'),6.),Z5.);   if ctscl90 ='    .' then ctscl90 ='     ';
  ctcol90 = put(input(compress(ctcol90x,'.'),6.),Z5.);   if ctcol90 ='    .' then ctcol90 ='     ';
  ctblk90 = put(input(compress(ctblk90x,'.'),6.),Z5.);   if ctblk90 ='    .' then ctblk90 ='     ';
  ctwht90 = put(input(compress(ctwht90x,'.'),6.),Z5.);   if ctwht90 ='    .' then ctwht90 ='     ';
  cthsp90 = put(input(compress(cthsp90x,'.'),6.),Z5.);   if cthsp90 ='    .' then cthsp90 ='     ';
  ctpov90 = put(input(compress(ctpov90x,'.'),6.),Z5.);   if ctpov90 ='    .' then ctpov90 ='     ';
  cten590 = put(input(compress(cten590x,'.'),6.),Z5.);   if cten590 ='    .' then cten590 ='     ';
  cten690 = put(input(compress(cten690x,'.'),6.),Z5.);   if cten690 ='    .' then cten690 ='     ';
  povrc90 = put(input(compress(povrc90x,'.'),6.),Z5.);   if povrc90 ='    .' then povrc90 ='     ';
  nonrc90 = put(input(compress(nonrc90x,'.'),6.),Z5.);   if nonrc90 ='    .' then nonrc90 ='     ';
  hsorc90 = put(input(compress(hsorc90x,'.'),6.),Z5.);   if hsorc90 ='    .' then hsorc90 ='     ';
  sclrc90 = put(input(compress(sclrc90x,'.'),6.),Z5.);   if sclrc90 ='    .' then sclrc90 ='     ';
  colrc90 = put(input(compress(colrc90x,'.'),6.),Z5.);   if colrc90 ='    .' then colrc90 ='     ';     
  zpnon90 = put(input(compress(zpnon90x,'.'),6.),Z5.);   if zpnon90 ='    .' then zpnon90 ='     ';
  zphso90 = put(input(compress(zphso90x,'.'),6.),Z5.);   if zphso90 ='    .' then zphso90 ='     ';
  zpscl90 = put(input(compress(zpscl90x,'.'),6.),Z5.);   if zpscl90 ='    .' then zpscl90 ='     ';
  zpcol90 = put(input(compress(zpcol90x,'.'),6.),Z5.);   if zpcol90 ='    .' then zpcol90 ='     ';
  zpblk90 = put(input(compress(zpblk90x,'.'),6.),Z5.);   if zpblk90 ='    .' then zpblk90 ='     ';
  zpwht90 = put(input(compress(zpwht90x,'.'),6.),Z5.);   if zpwht90 ='    .' then zpwht90 ='     ';
  zphsp90 = put(input(compress(zphsp90x,'.'),6.),Z5.);   if zphsp90 ='    .' then zphsp90 ='     ';
  zpen590 = put(input(compress(zpen590x,'.'),6.),Z5.);   if zpen590 ='    .' then zpen590 ='     ';
  zpen690 = put(input(compress(zpen690x,'.'),6.),Z5.);   if zpen690 ='    .' then zpen690 ='     ';
  znonrc90= put(input(compress(znonrc9x,'.'),6.),Z5.);   if znonrc90='    .' then znonrc90='     ';
  zhsorc90= put(input(compress(zhsorc9x,'.'),6.),Z5.);   if zhsorc90='    .' then zhsorc90='     ';
  zsclrc90= put(input(compress(zsclrc9x,'.'),6.),Z5.);   if zsclrc90='    .' then zsclrc90='     ';
  zcolrc90= put(input(compress(zcolrc9x,'.'),6.),Z5.);   if zcolrc90='    .' then zcolrc90='     ';
  ctnon00 = put(input(compress(ctnon00x,'.'),6.),Z5.);   if ctnon00 ='    .' then ctnon00 ='     ';
  cthso00 = put(input(compress(cthso00x,'.'),6.),Z5.);   if cthso00 ='    .' then cthso00 ='     ';
  ctscl00 = put(input(compress(ctscl00x,'.'),6.),Z5.);   if ctscl00 ='    .' then ctscl00 ='     ';
  ctcol00 = put(input(compress(ctcol00x,'.'),6.),Z5.);   if ctcol00 ='    .' then ctcol00 ='     ';
  ctblk00 = put(input(compress(ctblk00x,'.'),6.),Z5.);   if ctblk00 ='    .' then ctblk00 ='     ';
  ctwht00 = put(input(compress(ctwht00x,'.'),6.),Z5.);   if ctwht00 ='    .' then ctwht00 ='     ';
  cthsp00 = put(input(compress(cthsp00x,'.'),6.),Z5.);   if cthsp00 ='    .' then cthsp00 ='     ';
  ctpov00 = put(input(compress(ctpov00x,'.'),6.),Z5.);   if ctpov00 ='    .' then ctpov00 ='     ';
  cten500 = put(input(compress(cten500x,'.'),6.),Z5.);   if cten500 ='    .' then cten500 ='     ';
  cten600 = put(input(compress(cten600x,'.'),6.),Z5.);   if cten600 ='    .' then cten600 ='     ';
  povrc00 = put(input(compress(povrc00x,'.'),6.),Z5.);   if povrc00 ='    .' then povrc00 ='     ';
  nonrc00 = put(input(compress(nonrc00x,'.'),6.),Z5.);   if nonrc00 ='    .' then nonrc00 ='     ';
  hsorc00 = put(input(compress(hsorc00x,'.'),6.),Z5.);   if hsorc00 ='    .' then hsorc00 ='     ';
  sclrc00 = put(input(compress(sclrc00x,'.'),6.),Z5.);   if sclrc00 ='    .' then sclrc00 ='     ';
  colrc00 = put(input(compress(colrc00x,'.'),6.),Z5.);   if colrc00 ='    .' then colrc00 ='     ';
  zpnon00 = put(input(compress(zpnon00x,'.'),6.),Z5.);   if zpnon00 ='    .' then zpnon00 ='     ';
  zphso00 = put(input(compress(zphso00x,'.'),6.),Z5.);   if zphso00 ='    .' then zphso00 ='     ';
  zpscl00 = put(input(compress(zpscl00x,'.'),6.),Z5.);   if zpscl00 ='    .' then zpscl00 ='     ';
  zpcol00 = put(input(compress(zpcol00x,'.'),6.),Z5.);   if zpcol00 ='    .' then zpcol00 ='     ';
  zpblk00 = put(input(compress(zpblk00x,'.'),6.),Z5.);   if zpblk00 ='    .' then zpblk00 ='     ';
  zpwht00 = put(input(compress(zpwht00x,'.'),6.),Z5.);   if zpwht00 ='    .' then zpwht00 ='     ';
  zphsp00 = put(input(compress(zphsp00x,'.'),6.),Z5.);   if zphsp00 ='    .' then zphsp00 ='     ';
  zpen500 = put(input(compress(zpen500x,'.'),6.),Z5.);   if zpen500 ='    .' then zpen500 ='     ';
  zpen600 = put(input(compress(zpen600x,'.'),6.),Z5.);   if zpen600 ='    .' then zpen600 ='     ';
  znonrc00= put(input(compress(znonrc0x,'.'),6.),Z5.);   if znonrc00='    .' then znonrc00='     ';
  zhsorc00= put(input(compress(zhsorc0x,'.'),6.),Z5.);   if zhsorc00='    .' then zhsorc00='     ';
  zsclrc00= put(input(compress(zsclrc0x,'.'),6.),Z5.);   if zsclrc00='    .' then zsclrc00='     ';
  zcolrc00= put(input(compress(zcolrc0x,'.'),6.),Z5.);   if zcolrc00='    .' then zcolrc00='     ';

  pos = 1273;
  if count > '10' then count = '10';
  do i = 1 to input(count,2.);
    input @(pos +  0) seq_num  $char2.
          @(pos +  2) site     $char3.
          @(pos +  5) siterec  $char2.
	  @(pos +  7) siterkm  $char2.
	  @(pos +  9) site2r   $char2.
	  @(pos + 11) hist     $char4.
	  @(pos + 15) hist_2   $char4.
	  @(pos + 19) beh      $char1.
	  @(pos + 20) beh_2    $char1.
	  @(pos + 21) behtrnd  $char1.     
          @(pos + 22) modx     $char2.
	  @(pos + 24) yrdx     $char4.
	  @(pos + 28) agerec   $char2.
	  @(pos + 30) rpt_src  $char1.
	  @(pos + 31) hststga  $char1.
	  @(pos + 32) modajc1  $char2.
	  @(pos + 34) modajc3  $char2.
	  @(pos + 36) ajccstg  $char2.
	  @(pos + 40) frstprm  $char1.
	  @(pos + 41) multprm  $char1.
	  @(pos + 42) marital  $char1.
	  @(pos + 43) registry $char2.
	  @(pos + 45) monrx    $char2.
	  @(pos + 47) yr_rx    $char4.
	  @(pos + 51) radiatn  $char1.
	  @(pos + 58) othertx  $char1.
	  @(pos + 59) radsurg  $char1.
	  @(pos + 60) radbrn   $char1.
	  @(pos + 61) no_surg  $char1.
	  @(pos + 62) grade    $char1.
	  @(pos + 63) dxconf   $char1.
	  @(pos + 64) lateral  $char1.
	  @(pos + 65) ss_surg  $char2.
	  @(pos + 67) sx_prim  $char2.
	  @(pos + 69) num_nd   $char2.
	  @(pos + 71) sx_scop  $char1.
	  @(pos + 72) sx_site  $char1.
	  @(pos + 73) reconstr $char1.
	  @(pos + 74) sxprif   $char2.
	  @(pos + 76) sxscof   $char1.
	  @(pos + 77) sxsitf   $char1.
	  @(pos + 78) tumor1   $char1.
	  @(pos + 79) tumor2   $char1.
	  @(pos + 80) tumor3   $char1.
          @(pos + 81) e4size   $char2.
	  @(pos + 83) e4ext    $char1.
	  @(pos + 84) e4node   $char1.
	  @(pos + 85) e10size  $char3.
	  @(pos + 88) e10ext   $char2.
	  @(pos + 90) e10node  $char1.
	  @(pos + 91) e10pn    $char2.
	  @(pos + 93) e10ne    $char2.
	  @(pos + 95) e10pe    $char2.
	  @(pos + 97) cnty     $char3.
	  @;

    if modx='13' then modx='99';
    if monrx='13' then monrx='99';
    if SER_DODM='13' then SER_DODM='99';
    if sxprif=' 0' then sxprif='00';

  /* create unchanging variables */
  /* Date of Last Contact: SEER DOD or MEDICARE DOD or END OF STUDY & vital status recode */
  if SER_DODY = '0000' then do;
    if med_dody = '    ' or med_dody < yrdx then do;
      SER_YOLC = eos_y;
      SER_MOLC = eos_m;
      stat_rec = '1';
      end;
    else do;
      SER_YOLC = med_dody;
      SER_MOLC = med_dodm;
      stat_rec = '4';
      end;
    end;
  else do;
    SER_YOLC = SER_DODY;
    SER_MOLC = SER_DODM;
    stat_rec = '4';
    end;
  if (SER_YOLC > eos_y) or ((SER_YOLC = eos_y) and (SER_MOLC > eos_m)) then do; /* died after end of study */
      SER_YOLC = eos_y;
      SER_MOLC = eos_m;
      stat_rec = '1';
      end;

    /* create changing variables */
    /* eod coding system */
    if '1988' <= yrdx then eod_sys='4';
    else if '1983' <= yrdx <= '1987' then eod_sys='3';
    else eod_sys=' ';

    /* age at Dx (3 digit) */
    AgeDX = input(yrdx,4.) - input(birthyr,4.);
    if modx <= birthm then agedx = agedx-1;
    if agedx < 0 then agedx=0;

    /* type FUP expected */
    if rpt_src in ('6' '7') then type_fup = '1';  	   /* death certificate/autopsy */
    else if behtrnd='2' & siterec='47' then type_fup='3';  /* cervix insitu      */
    else type_fup = '2';			  	   /* active followup    */
    						  	   /* = 4: SanFrancisco? */
     output;
     pos = pos + 100;
     end;
   input;
     

proc sort data=prep;
  by regcase seq_num;
run;

/* *********  BEGIN ANALYSIS MERGE SECTION ********* */    /* USER - begin edits */
/* USER: keep desired variables.  These variables must be numeric.  if you */
/*       named regcase and seq_num differently, you must rename them and   */
/*       they must be character strings (regcase $char10, seq_num $char2)  */ 

/* USER comment out begins        (delete line if using this section)
data analy;
  set ANALYSIS.analysis(keep=regcase seq_num     );	   
							   
							   
proc sort data=analy;
  by regcase seq_num;  

data prep;
  merge prep(in=p) analy(in=a);
  by regcase seq_num;
  if p;
  if a then cohort='1';
  else cohort='0';
run;

 USER comment out ends */	  /* (delete line if using this section) */

/* **********  END ANALYSIS MERGE SECTION ********** */	   /* USER - end edits */ 

data prep(drop=numrec) temp(keep=regcase numrec);
  set prep;
  by regcase seq_num;
  retain numrec seq_rcd;
  if first.regcase then do;
    numrec=0;
    seq_rcd=0;
    end;
  if not (first.regcase and last.regcase) then seq_rcd=seq_rcd+1;
  numrec=numrec+1;
  output prep;
  if last.regcase then output temp;

data prep;
  merge prep temp;
  by regcase;

data _null_;
  set prep;
  by regcase seq_num;
  file PREP lrecl=674;
  
  put @1   regcase   $char10.
      @11  fivepct   $char1.
      @12  dod_flg   $char1.
      @13  med_dodm  $char2.
      @15  med_dodd  $char2.
      @17  med_dody  $char4.
      @21  birthm    $char2.
      @25  birthyr   $char4.
      @29  m_sex     $char1.
      @30  race      $char1.
      @31  rsncd1    $char1.
      @32  cur_ent   $char1.
      @33  chr_esrd  $char1.
      @34  med_stcd  $char2.
      @36  STATE     $char2.
      @38  COUNTY    $char3.
      @41  zip       $char9.
      @50  code_sys  $char1.
      @51  census    $char6.
      @57  census2k  $char6.      
      @63  hsa       $char3.      
      @66  urbrur    $char1.
      @67  urban     $char2.
      @69  ctpci90   $char6.      
      @75  ctmed90   $char6. 
      /* @81  ctden90   $char6.   all missing */
      @87  ctnon90   $char5.
      @92  cthso90   $char5. 
      @97  ctscl90   $char5. 
      @102 ctcol90   $char5. 
      @107 ctblk90   $char5. 
      @112 ctwht90   $char5. 
      @117 cthsp90   $char5. 
      @122 ctpov90   $char5. 
      @127 cten590   $char5.       
      @132 cten690   $char5. 
      /* @137 medag90   $char6.   all missing */
      /* @143 medrc90   $char6.   all missing */
      @149 povrc90   $char5. 
      @154 nonrc90   $char5. 
      @159 hsorc90   $char5. 
      @164 sclrc90   $char5. 
      @169 colrc90   $char5. 
      @174 zppci90   $char6. 
      @180 zpmed90   $char6. 
      /* @186  zpden90   $char6.   all missing */
      @192 zpnon90   $char5. 
      @197 zphso90   $char5. 
      @202 zpscl90   $char5. 
      @207 zpcol90   $char5. 
      @212 zpblk90   $char5. 
      @217 zpwht90   $char5. 
      @222 zphsp90   $char5. 
      @227 zpen590   $char5. 
      @232 zpen690   $char5. 
      @237 znonrc90  $char5. 
      @242 zhsorc90  $char5. 
      @247 zsclrc90  $char5. 
      @252 zcolrc90  $char5. 
      @257 ctpci00   $char6. 
      @263 ctmed00   $char6. 
      @269 ctden00   $char6. 
      @275 ctnon00   $char5. 
      @280 cthso00   $char5. 
      @285 ctscl00   $char5. 
      @290 ctcol00   $char5. 
      @295 ctblk00   $char5. 
      @300 ctwht00   $char5. 
      @305 cthsp00   $char5. 
      @310 ctpov00   $char5. 
      @315 cten500   $char5. 
      @320 cten600   $char5. 
      @325 medag00   $char6. 
      @331 medrc00   $char6. 
      @337 povrc00   $char5. 
      @342 nonrc00   $char5. 
      @347 hsorc00   $char5. 
      @352 sclrc00   $char5. 
      @357 colrc00   $char5. 
      @362 zppci00   $char6. 
      @368 zpmed00   $char6. 
      @374 zpden00   $char6. 
      @380 zpnon00   $char5. 
      @385 zphso00   $char5. 
      @390 zpscl00   $char5. 
      @395 zpcol00   $char5. 
      @400 zpblk00   $char5. 
      @405 zpwht00   $char5. 
      @410 zphsp00   $char5. 
      @415 zpen500   $char5. 
      @420 zpen600   $char5. 
      @425 znonrc00  $char5. 
      @430 zhsorc00  $char5. 
      @435 zsclrc00  $char5. 
      @440 zcolrc00  $char5. 
      @445 s_sex     $char1.
      @446 RAC_RECB  $char2.
      @448 ICD_CODE  $char1.
      @449 COD89V    $char4.
      @453 COD10V    $char4.
      @457 SER_DODM  $char2.
      @459 SER_DODY  $char4.
      @463 srace     $char2.
      @465 origin    $char1.
      @466 stat_rec  $char1.
      @471 seq_num   $char2.
      @473 site      $char3.
      @476 siterec   $char2.
      @478 siterkm   $char2.
      @480 site2r    $char2.
      @482 hist      $char4.
      @486 hist_2    $char4.
      @490 beh       $char1.
      @491 beh_2     $char1.
      @492 behtrnd   $char1.
      @493 modx      $char2.
      @495 yrdx      $char4.
      @499 agerec    $char2.
      @501 rpt_src   $char1.
      @502 hststga   $char1.
      @503 modajc1   $char2.
      @505 modajc3   $char2.
      @507 ajccstg   $char2.
      @509 no_surg   $char1.
      @510 grade     $char1.
      @511 dxconf    $char1.
      @512 ss_surg   $char2.
      @514 radiatn   $char1.
      @515 lateral   $char1.
      @516 eod_sys   $char1.
      @517 e4size    $char2.
      @519 e4ext     $char1.
      @520 e4node    $char1.      
      @521 e10size   $char3.
      @524 e10ext    $char2.
      @526 e10node   $char1.
      @527 e10pn     $char2.
      @529 e10ne     $char2.
      @531 e10pe     $char2.
      @533 marital   $char1.
      @534 sx_prim   $char2.
      @536 num_nd    $char2.
      @538 sx_scop   $char1.
      @539 sx_site   $char1.
      @540 reconstr  $char1.
      @541 tumor1    $char1.
      @542 tumor2    $char1.
      @543 tumor3    $char1.
      @544 frstprm   $char1.
      @545 multprm   $char1.
      @546 radsurg   $char1.
      @547 radbrn    $char1.
      @548 plc_brth  $char3.
      @551 numprims  $char2.
      @553 registry $char2.
      @555 monrx    $char2.
      @557 yr_rx    $char4.
      @561 othertx  $char1.
      @562 sxprif   $char2.
      @564 sxscof   $char1.
      @565 sxsitf   $char1.
      @566 cnty     $char3.
      

  /* created variables */
      @569 SER_MOLC  $char2.
      @571 SER_YOLC  $char4.
      @575 stat_rec  $char1.
      @576 agedx     Z3.
      @579 type_fup  $char1.
      @580 numrec    Z2.
      @582 seq_rcd   Z2.
      @584 cohort    $char1.

  /* USER - replace lenX_n variables as needed. */
  /* The existing tag must be the appropriate length. */
  /* The variables must be numeric. */
      @585 len1_1    z1.
      @586 len1_2    z1.
      @587 len1_3    z1.
      @588 len1_4    z1.
      @589 len1_5    z1.
      @590 len1_6    z1.
      @591 len1_7    z1.
      @592 len1_8    z1.
      @593 len1_9    z1.
      @594 len1_10   z1.
      @595 len2_1    z2.
      @597 len2_2    z2.
      @599 len2_3    z2.
      @601 len2_4    z2.
      @603 len2_5    z2.
      @605 len2_6    z2.
      @607 len2_7    z2.
      @609 len2_8    z2.
      @611 len2_9    z2.
      @613 len2_10   z2.
      @615 len3_1    z3.
      @618 len3_2    z3.
      @621 len3_3    z3.
      @624 len3_4    z3.
      @627 len3_5    z3.
      @630 len4_1    z4.
      @634 len4_2    z4.
      @638 len4_3    z4.
      @642 len4_4    z4.
      @646 len4_5    z4.
      @650 len5_1    z5.
      @655 len5_2    z5.
      @660 len5_3    z5.
      @665 len5_4    z5.
      @670 len5_5    z5.
      ;
run;