Document npMzBZbXj3v5RwEJX1qjdO5KR

-* J V f\K, U Dear Dr. Knox, Statistical Research Unit, London School of Hygiene and Tropical Medicine (ttnversity of London) Keppel Street (Cover Street) f.C.l. li*-th August 1955* I received your letter a few days ago but I hope you will forgive ns if 1 don't go into it in full, at 1 aa just off on holjdcy and have had so many things to clear up before leaving. I can, hOTever, sake a number of consents now. (1) The disparity between the numbers you first provided (76 men left and 5} still ea*loyeA) and the present ones (52 left and 62 still employed) is not as great as appears at first sight. "Left* according to the first definition meant, I believe, "Left the Scheduled Areas" - we have, therefore to transfer the 16 sen still working at T.B.A. but suspended from the Scheduled Arete from the still employed caterer? to the left category, tfien the comparison becomes* ' Pir*t_estimete 3econd estimate Left Still esployed 76 55 52 16 * 68 62 -- l6 46 If you think that the seeond estimate is likely to have been mere carefully carried out, I an quite ready to accept it and fOTget about the first estimate - though it sust be realised that any defisiency in the estimated population must result in the actual risk of lung cancer being over-estimated, I an, however, rather worried by the second possibility you suggest - namely, that men may have left and died else where or died in Rochdale without an autopsy. 1 can understand that the deaths of such men may have been missed but if the records of employment are adequate for our purpose the names of such men should appear in the list of "men alloyed in the scheduled areas, wto have left T.B.A. and are presumed to be still alive" - at present a Hat of only 15 names. (2) You say that W. Cavanagh, a man who had lung oanoer at autopsy, ns eoployed in the Maintenance Department - yet his name appears on the list as having worked for 29 years in the Scheduled Areas. How did his name come to get recorded theret ix it got there tnrough the Personnel Manager1* search of eoployment cards - i.e. as I presume all tne othernames got there - it is not fair to remove it because of special informa tion you have obtained through interest in the cause of death, unless you make an equally intensive study of all the other men en the list. (5) Presuming that the lists define the total population we are trying to study pretty fully, we can, I think, make something of the data. I suggest we moke our calculations on the position as at 51*7.55 (a pity that the review via not made a month earlier, in the middle of the year, as the working would have been neater - but it is not important), and we should take no acoount of any deaths after that date. One neoessary step will be to check that the 15 men recorded as having left T.8.A. but still alive were still alive on 51*7.55* Addresses are known for 12 of these men - oould you possibly arrange for visits to be mads to them? (4) The data will bs suoh more impressive if we oan extend it to include men with 15 years service in the 3*A. The numbers will then certainly be quite adequate for our purpose, I would suggest we forget bout the 10-14 year exposure group, but that we tried to get information about the 15-19 year group. (5) Can you let me know in detail what has been the Personnel Manager's method for the preparation of these lists of names? It would be s great help In deoiding what credence we can give the results. With best wishes, Yours sincerely Richard Doll, W.D. ,?<.R.C.P.