X-Message-Number: 9468 Date: Sun, 12 Apr 1998 17:44:45 -0400 (EDT) From: Charles Platt <> Subject: Risk Assessment and Cryonics Growth ------------------------------------------------------------- Computer Simulation Shows Future Impact of Growth on Standby, Transport, and Perfusion Services in a Cryonics Organization ------------------------------------------------------------- For 20- and 40-year future periods assuming annual growth rates ranging from 0% to 20% by Charles Platt President, CryoCare Summary Most cryonicists favor growth, but a minority (including myself) have wondered about possible penalties--such as a heavier case load for standby/transport teams. In addition I have wondered about the possibility of two cryonics members dying simultaneously, which would create major logistical problems. To address these concerns I have written a computer program that generates likely future scenarios. The results are reassuring, at least for my own organizaton, CryoCare. During the next 20 years, if there is ANY growth rate ranging from 0 to 10 percent, generally speaking we should expect an average of 1 case per year and a maximum of 4 cases in any one year, with a 1% chance (or less) of simultaneous cases, and some years in which no cases occur at all. I think it's reasonable to expect a single standby/transport/perfusion team to manage this case load. These numbers do not change significantly if the median age of new members increases from 30 to 40. The numbers double, more or less, with a growth rate of 20 percent, which would strain a single team to the limit over the next 20 years and would probably result in 2 simultaneous cases during that period. If a 20 percent growth rate continues over 40 years, this creates a major impact, including a probable maximum of 300 cryopreservation procedures in any one year and a 50-50 chance of any one case being simultaneous with another case. However, with 20 percent growth per year, by the end of a 40- year period the cryonics organization would have expanded from 80 members to about 100,000 members, and would be well placed to afford several standby/transport/perfusion teams to cope with the case load. CONCLUSION: We need not be anxious about the consequences of moderate growth in cryonics over the next two decades at least. Details For practical purposes, bearing in mind the time required to complete a case and redeploy equipment, "simultaneous" cases are those which occur fewer than 3 days apart. Ralph Merkle compares the problem of calculating the probability of this with the well-known "birthday" problem, which asks how many people should be added to a room in order to reach a 50-50 chance of any two people sharing the same birthday. However, our situation is more complicated because elderly people have a higher risk of dying than younger people. Therefore, if the average age of cryonics members increases over time (as is likely, since most people join in their middle years and then remain signed up after that), the risk of simultaneous death also increases with time, even if the number of members in an organization remains constant. Rather than try to figure this probability, I wrote a program that uses actuarial tables and birth dates to figure the chance of _each individual_ dying in this and future years. The program selects a random date for each death and notes if one death is followed or preceded by another within 2 days or less. Since the program depends on weighted random functions, it repeats its projections 100 times and then averages them. Initially the program was designed specifically for CryoCare, but it can be adapted easily for any other organization that is willing to supply member birth dates. I am happy to share the program freely, because I believe longterm planning based on calculated probabilities, rather than guesswork, will be good for cryonics generally. Results of several runs of the program are shown below. Each run assumes that the organization has 80 members as of January 1st, 1999, more than half of them aged between 40 and 60 (the age distribution in CryoCare). The program uses actuarial data from Statistical Abstract of the United States (1997 edition) to figure the chance of death for each person according to his/her age. A random-number function, weighted with the chance of death, determines whether each individual dies or survives. Survivors are moved forward to the next year, new members are added, and the process repeats. This "cohort" technique is also used by federal agencies and the UN to project future population figures based on birth rates, death rates, and immigration. Caveats: Age-related death rates may change in the future, if life expectancy increases. This would diminish the future case load per year. Cryonicists may enjoy greater longevity than the average American population--although I tend to doubt this after seeing the high-fat/low-exercise regime enjoyed by many of my "immortalist" contemporaries. The death-rate figures that I have used are for the U.S. population as a whole, including men, women, whites, and nonwhites; but cryonics members tend to be male and white. Males die younger than females, but whites live much longer than nonwhites. Overall, this means I have slightly overestimated the cryonics caseload per year. I have made arbitrary assumptions about new members joining the organization. The median age is user-selectable, but the age spread either side of the median is always +/- 16 years, conforming with a rough bell curve. It is possible, of course, that I have made erroneous assumptions or math errors that I am unaware of. Despite these caveats, I believe this is the first time that anyone has attempted to model the consequences of growth, and the results are useful, since we can now see the _relative_ consequences of adjusting variables such as membership growth and median age of new members. Note: the tables below will not display legibly unless you use a monospaced font such as Courier. ------------------------------------------------------------- Table 1 Simulation period: 20 years Median age of new members: 30 Situation at the end of the projected period: percent growth rate during whole period 0 3 6 10 20 Living members 57 112 212 464 2779 Cryopreserved 23 24 25 28 45 Avg cases/year 1 1 1 1 2 Averaged MAXIMUM 4 4 4 4 7 cases in any year % chance of the next 1 1 1 1 2 case occurring less than 3 days after the current case ------------------------------------------------------------- Table 2 Simulation period: 20 years Median age of new members: 40 Situation at the end of the projected period: percent growth rate during whole period 0 3 6 10 20 Living members 57 111 208 454 2724 Cryopreserved 23 25 28 33 65 Avg cases/year 1 1 1 2 3 Averaged MAXIMUM 4 4 4 5 11 cases in any year % chance of the next 1 1 1 1 3 case occurring less than 3 days after the current case ------------------------------------------------------------- Table 3 Simulation period: 40 years Median age of new members: 30 Situation at the end of the projected period: percent growth rate during whole period 0 3 6 10 20 Living members 22 142 574 2836 102115 Cryopreserved 58 67 84 135 1139 Avg cases/year 1 2 2 3 28 Averaged MAXIMUM 5 5 7 13 184 cases in any year % chance of the next 1 1 2 4 41 case occurring less than 3 days after the current case ------------------------------------------------------------- Table 4 Simulation period: 40 years Median age of new members: 40 Situation at the end of the projected period: percent growth rate during whole period 0 3 6 10 20 Living members 22 131 523 2651 96573 Cryopreserved 58 74 106 204 2009 Avg cases/year 1 2 3 5 50 Averaged MAXIMUM 5 6 9 20 328 cases in any year % chance of the next 1 1 2 6 55 case occurring less than 3 days after the current case ------------------------------------------------------------- The source code for this program is heavily annotated, easy to understand, and was written in an extended dialect of BASIC that could be adapted very easily for the old Microsoft QuickBASIC compiler. I will email the code to anyone who wants it. Also I can send the compiled program as an .EXE file to anyone who is interested. It will run under any version of MS-DOS. Note that since the program was written within less than a day, its error trapping is rudimentary, and there aren't any cute little buttons or dialog boxes. ------------------------------------------------------------- Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=9468