[Babase] Censoring + Alto's split
Niki Learn
nlearn at princeton.edu
Tue Oct 20 11:02:42 EDT 2009
Sorry, I guess I'm the only one steeped in these biograph dates and a little more explanation is required.
The statdate for any given animal presently is either (1) the assigned deathdate (which, unless the date of death is actually known, is calculated by taking the midpoint between the last date the animal was seen alive and the first date the animal was recorded absent) or (2) the last date the animal appears in the census record. In terms of animals with old, dangling statdates, the second case is true for both males that disappear from a group (but are assumed to have emigrated rather than to have died) and the animals from groups that we stopped monitoring like Proton's and Lodge's (unless they showed up again sometime later in an other group census or by immigrating into a study group). We also have to assign a statdate when entering births and new immigrant males (in this case it is the same as the bithdate or date of entry to the group); the statdate then gets updated automatically by babase when census data is entered.
To assign animals to a status of censored, I have to link the status change to a date, just like I would when changing animals to dead; hence the censordate. Babase will not allow any data to be entered after a statdate linked to a status other than alive. So I am proposing that (1) males who emigrate out of a study group (and therefore have both a date last present and a date first absent from the group) get a censordate that is the midpoint between the date last seen and the date first not seen (just like we would calculate the deathdate) and that (2) animals from groups that we are no longer studying or who stop showing up in other group censuses for a peripherally monitored group and therefore have no date when they were first recorded absent (at least not one that's anywhere near the last date they were recorded present...) be assigned a censordate the day after they last appeared in the census record (since we know they were there that last day but don't know what happened to them after that).
This will produce a database with three possible meanings of statdate instead of two. Only animals seen alive in the last update will have a statdate equal to the last date they were censused. All others will have a statdate equal to either their real or estimated deathdate or their newly assigned censordate. I will have to update that part of the data management protocol and repost on the wiki once it is implemented. Hmm...actually it looks like it's not even on the wiki (only the babase 1.0 version of Princeton's protocol is there) - not sure why that is. I'll put getting that up on my list of things to do.
Is this an agreeable protocol for assigning censordates?
Thanks,
Niki
-----Original Message-----
From: babase-bounces at eeblistserv.Princeton.EDU [mailto:babase-bounces at eeblistserv.Princeton.EDU] On Behalf Of Susan Alberts
Sent: Monday, October 19, 2009 8:54 PM
To: The Baboon Database Project
Subject: Re: [Babase] Censoring + Alto's split
I am not completely sure what the censordate is -- are you referring
to the statdate for an animal that has "censored" as its status? Just
clarifying.
With respect to calculating what the date should be, it sounds like
you are saying that the statdate is calculated differently in the
following two cases:
1. The animal is known or strongly suspected to have died, so you give
it a death in status, versus
2. The animal suddenly disappears, not enough evidence to assign
death, the animal just has an unknown fate, as in a male that leaves
and may have emigrated and we don't know where he goes, and until know
you would have given him a status of "alive" sometime near the last
date he was seen.
Am I correct that you calculate these two types of statdates
differently? There is no information in the babase tech specs about
this, so I'm not sure.
If so, then I think maybe we should revisit this, as I am not sure
there is a reason to calculate them differently. And, if you do them
differently, it probably makes most sense now to calculate the date of
censoring the same way you calculate a death date. Jeanne?
Susan
On Oct 16, 2009, at 4:21 PM, Niki Learn wrote:
> That's fine. I wasn't sure initially how far off the other data was
> in coming but Lacey called me earlier and said it could be a while
> yet. I'll go ahead and censor them all and we'll worry about
> updating the censordates when the other data is ready.
>
> I guess the best date to use as the censordate is a) the midpoint
> between when an animal was last seen and when it was first absent
> (as with the deathdate) or b) in these Alto's cases (and any others
> where animals were rarely censused and have not been seen or in any
> way marked absent in some time), the day after an animal last
> appears in the census record. Does that make sense to everyone?
>
> Thanks,
> Niki
>
> -----Original Message-----
> From: babase-bounces at eeblistserv.Princeton.EDU [mailto:babase-bounces at eeblistserv.Princeton.EDU
> ] On Behalf Of Susan Alberts
> Sent: Friday, October 16, 2009 3:26 PM
> To: The Baboon Database Project
> Subject: Re: [Babase] Censoring + Alto's split
>
> Right, thanks for clarifying your concern. I still think it makes most
> sense to change all the alives to censoreds, and deal with the
> statdate changes at the time they come up. I say this because I think
> it is likely that we will need to use the data in analysis before we
> update the census info for Proton's group, and it will be confusing to
> have those animals alive -- does this make sense to you? YOu have
> more experience with uploading data than I do...
> Susan
>
> On Oct 16, 2009, at 3:00 PM, Niki Learn wrote:
>
>> Susan,
>>
>> When I "kill" a baboon in babase the statdate changes to the
>> deathdate. This would work the same way with censoring them. So
>> the reason I was thinking it would be tricky to censor the baboons
>> in Proton's, for example, is that the census data for them currently
>> ends at a totally random point and if the later data is to be added,
>> yes the individuals would still end up censored but not on the same
>> date. I could go ahead and censor all of them now with the best
>> dates available in the database but we would have to remember to
>> change the censor date before uploading any missing data that fell
>> after that date; otherwise babase will yell at us.
>>
>> I'm happy to try to help sort out Proton's group too. Puzzles are
>> fun :)
>>
>> Niki
>>
>> -----Original Message-----
>> From: babase-bounces at eeblistserv.Princeton.EDU [mailto:babase-bounces at eeblistserv.Princeton.EDU
>> ] On Behalf Of Susan Alberts
>> Sent: Friday, October 16, 2009 2:46 PM
>> To: The Baboon Database Project
>> Subject: Re: [Babase] Censoring + Alto's split
>>
>> Hi Niki and Lacey,
>>
>> I thought I should jump in here to comment.
>>
>> First, with respect to the censored/alive/dead codes in biograph (I
>> think this is what you are asking about when you say "I think we
>> should skip censoring any Alto's animals...") I think you should
>> still
>> change ALIVE to CENSORED for any case in which the statdate is not
>> the
>> date of your most recent update (i.e. where statdate is earlier than
>> Jun 2009). The status (censored, alive, dead) for those animals
>> mostly
>> won't change after the data are uploaded as actual census data (they
>> will still mostly be censored), although the statdate might change in
>> biograph. In other words, implement the censoring code in biograph
>> without skipping any periods for any groups. I think you should be
>> able to apply the rule for all cases without producing any errors.
>> Does this make sense?
>>
>> Second, unfortunately I think that there will always be a lot of
>> confusion about Proton's group in the early 1990's. The messy nature
>> of the data accurately reflects what happens when you gradually,
>> rather than suddenly, stop monitoring a group. The uploading of the
>> census data that Lacey has on her list will probably not clear many
>> of
>> these things up and in fact will probably raise some sticky
>> questions,
>> like how do we deal with the fact Scion clearly had a infant whose
>> birth was entirely missed (I am guessing that we will just censor
>> Scion at an earlier age than we actually know she lived to, just as
>> we
>> do now). The virtue of the current database, frustrating as it can
>> be,
>> is that it simply skirts these issues. We will have to deal with them
>> at some point but I am quite sure that when we go to enter the data
>> we
>> will raise more questions, initially, than we will resolve.
>>
>> And, before we can enter the data we still need to work through some
>> details about how to capture the fission information -- it will
>> require at least one and possibly two new groups to be created in the
>> database I think, because commonly during fissions the animals are in
>> "groups" that don't currently exist (for instance, some animals from
>> what eventually became Nyayo's might have spent some days with some
>> animals from what eventually became Dotty's group, while the rest
>> were
>> in another subgroup that has no name). To capture this information,
>> these groups will have to be given a group identity (presumably
>> something like "temporary fission group") before the data can be
>> entered. I don't think we are quite there yet.
>>
>> Thanks for the offer of helping with catch up on the lagging parts of
>> Alto's. I think this would be great; now that we have extra hands in
>> our lab, we are still finding that the amount of old data we are
>> entering is huge (old darting data, old juv point samples data, old
>> predation data, old demog notes)), and also we have been hindered in
>> the census data by the design issues I raised above.
>>
>> I also have some comments on your email about Allison's infants but I
>> will do that separately; I hope this helps as a start, let me know if
>> you have other questions about the status codes.
>>
>> Susan
>>
>>
>>
>>
>>
>>
>> On Oct 16, 2009, at 11:12 AM, Niki Learn wrote:
>>
>>> Lacey,
>>>
>>> Is the census (other than demog notes you mentioned you are waiting
>>> on), biograph (births, immigrations, and deaths), and pregnancy/
>>> cycling (upon which births rely) data ready to be uploaded?
>>>
>>> If so, what if you go ahead and send those to me for upload. We can
>>> fill in the demog notes later since those are less likely to impact
>>> the statdate. And then I guess you will have interaction data to
>>> upload too (and eventually we’ll do ranks for the years/groups
>>> with
>>> sufficient data?). Even if only some of the years/subgroups are
>>> ready it could be worth going ahead with what we’ve got for now.
>>>
>>> If the census/biograph data is not ready, I guess we should skip
>>> censoring any Alto’s animals whose statdates fall within that
>>> period since we would just have to uncensor most of them later to
>>> upload the missing data. I could go ahead and censor animals like
>>> Scion who were seen after that period though. Presumably most
>>> members of Nyayo’s and Dotty’s would fall into this category.
>>>
>>>
>>> The censoring is something I think I can implement in early
>>> November, even if I have to skip that period of Alto’s for now.
>>> Possibly next year I can help catch up the lagging parts of Alto’s
>>> if there is still a lack of undergraduates? Demog notes are, after
>>> all, much more interesting to enter than old sexskin swelling sizes
>>> and PCS colors – that I would gladly yield to undergrads.
>>>
>>> Thanks,
>>> Niki
>>>
>>> From: Lacey Roerish [mailto:lroerish4 at gmail.com]
>>> Sent: Tuesday, October 13, 2009 8:12 PM
>>> To: Niki Learn
>>> Subject: Re: wounds and pathologies question (white monkeys) + that
>>> crazy Alto's split (and Cocoa, another white monkey)
>>>
>>> Just my little input on the part of this that is pertinent to me...
>>> The Alto's Census backfill and demog notes backfill are both on my
>>> list of things to do. I'm getting there. As for how to deal with the
>>> fission of Alto's Group, this is also on our list, and has already
>>> been looked at once.. we have all of the notes. Am just waiting to
>>> put an undergrad on fishing through demog notes.
>>>
>>> So its all in the works! :)
>>> We'll keep you posted
>>> Lacey
>>>
>>> On Tue, Oct 13, 2009 at 4:30 PM, Niki Learn <nlearn at princeton.edu>
>>> wrote:
>>> I've attached a sheet that began as Alison's infants of interests
>>> file. Most of the infants and a handful of others were used in her
>>> final analysis. I've added the others she used and a couple others
>>> that she maybe could have used but did not. I am not sure how she
>>> concluded that several of them could have had white monkey syndrome
>>> (particularly WAU, LOW, and ENE, all of whom were less than two
>>> months old and died of unrelated causes, and PAW seems to me to have
>>> had something different).
>>>
>>> Alison did locate all the infants with wound sheets indicating they
>>> had white fur (except CAC who was only slightly whitish) and those
>>> indicating a stiff walk (coded as either 10 or 14). She flagged
>>> some early in the process who had code 7 but only ended up using
>>> TOF, HAV, and CYC, none of which I am very confident had the
>>> syndrome (TOF at least is reported to have had weak feet, though his
>>> arms were primarily affected by whatever he had; the other two seem
>>> like unseen injuries to me). None of these have demography notes
>>> indicating any additional useful information (other than to confirm
>>> the dcauses that demonstrate that the questionable infants WAU, LOW,
>>> and ENE did not die because they were sick).
>>>
>>> Code 7 - Limp, no wound visible
>>> Code 10 - Malaise, weakness, stiffness in absence of wound
>>> Code 14 - Other pathology
>>>
>>> I would not advise using code 7 for the stiff walk. This is the
>>> most frequently used code and it already has some irregularities
>>> that should probably be addressed. [Essentially, limp has changed
>>> from time to time on the datasheets, sometimes being listed simply
>>> as limp or even as limp from wound but usually as "limp, no wound
>>> visible" - sometimes it is marked along with wounds that could
>>> obviously be causing a limp, particularly in older data where the
>>> name may not match up with the code's name. Recent data (2000
>>> onward) is fairly consistent in not checking code 7 if there is a
>>> limp-causing wound. This is fine because there is also the column
>>> for checking whether the wound "impairs locomotion" which SHOULD
>>> always be checked when code 7 is used and whenever a limp-causing
>>> wound is evident. It also often marked with code 10 when the
>>> stiffness or weakness impairs locomotion.]
>>>
>>> Code 10 is mostly used in the following cases: To indicate slowing
>>> by older baboons (rheumatism, stiffness, slowing down, etc.), to
>>> indicate infants that are not thriving (can't cling, etc.), and for
>>> cases where a baboon has been badly wounded or is very sick and so
>>> is weak/in a state of malaise. The large outbreak of Coxsackie
>>> mentioned earlier was coded as 10, as are two cases that seem
>>> similar where 3 or 4 baboons simultaneously had stiff walks/stiff
>>> sacral regions for a short period of time.
>>>
>>> Code 14 is mostly used for really bizarre things that nobody knows
>>> where else to put and for more extreme cases where baboons become
>>> very thin (I found several cases of thinning that were flagged "low
>>> body weight" in the notes - I similarly flagged Liberty and can flag
>>> any other individuals that I actually get wounds & pathologies
>>> sheets for).
>>>
>>> That said it sounds like 10 is the best existing code to use.
>>> Possibly if we think they are white monkey cases we could flag them
>>> "possible white monkey syndrome" in the notes? Of course, right now
>>> there are four tables, three of which have notes. I'll have to see
>>> how Karl wants to arrange everything in babase since these sheets
>>> were designed with a different database system in mind; babase
>>> usually designated ids but these have predesignated ids (with many
>>> gaps) to allow the information in the four tables to be aligned with
>>> each other.
>>>
>>> It appears that wounds and pathologies mentioned solely in
>>> demography notes (of which I think there are many) do not appear in
>>> the existing wounds and pathologies tables. For example, surely
>>> there were cases of white monkey syndrome before 2003 but they were
>>> not given wounds sheets and so cannot be easily located. Is this
>>> something we need to remedy?
>>>
>>> On a related note, Catherine mentioned a white monkey case she has
>>> been discussing with you that she discovered in Proton's group in a
>>> demography note for darting of his mother, Scion, on 14 May 1994.
>>> While investigating the case, more and more questions were raised.
>>> The infant is Cocoa who is not in babase (but is on the list of
>>> snames not in babase). I came to find a few problems:
>>>
>>> The demography note indicates that Cocoa was about 9 months old and
>>> then determines (and this data matches what is in the notebook
>>> elsewhere) that the infant would have been born in August 1993. But
>>> this demography note is not in babase, nor is any of the Proton's
>>> data contained in this notebook, which includes census and related
>>> data from Feb 1994 to 2 Sep 1999 (admittedly with a decreasing
>>> amount of information over time as fewer individuals became
>>> identifiable). Records for all of SCI's infants last known to be
>>> living end on 11 June 1993, which is also when SCI's census record
>>> ends, save for a single demog note entry in August 2001. In fact,
>>> the census records for SCI and others in Proton's group (and from
>>> what I can tell all of Alto's fission groups) are not actually in
>>> babase. Instead there is some kind of interpolated data (more like
>>> you would see in the members table) that begins on the split date
>>> (1/1/1989) and goes (for Proton's anyway) through 11 June 1993 and
>>> was apparently carried over from another database system - Scion is
>>> listed as R - "(result of Alto's breakup) The datum is S, E, F, B,
>>> G, T, or L datum which has had locations which were changed from 1.0
>>> to the group in which the animal was censused on 15/4/92. This
>>> change left all R rows as part of a contiguous series of days during
>>> which the animals are located in the Alto's sub-group as censused on
>>> 15/4/92, and the time-adjacent locations were not 1.0." and her
>>> slightly older kid Cake, for example, as L (even though he has a
>>> birthdate in babase) - "(lineage) The group is from the Matgrp on
>>> the old CYCTOT database, either because the animal did not appear in
>>> the DISPERSE database, or because the first location for the animal
>>> in the old DISPERSE database had a Datein and this Datein was after
>>> the birth date of the animal.". Proton's then was not monitored
>>> again until Feb 1994 (maybe they were in Tanzania?).
>>>
>>> Catherine said she recalled at some point that this census data was
>>> supposed to be fixed. She thought by Lacey? Is this still in
>>> progress somewhere? On the backburner? Forgotten? Should this be
>>> brought back from the dead?
>>>
>>> In addition to the lost census information there is clearly other
>>> useful information that can be gleaned from these data, such as
>>> presence of white infants). This wonky split/database conversion
>>> combination also led to inclusion of a bunch of pretty much useless
>>> data in babase.
>>>
>>> Furthermore, it is clear from the darting/white infant demography
>>> note and the dates that Cocoa is SCI5, but SCI5 is the one case
>>> listed as "pregnancy with undetermined outcome" (status 4). Given
>>> the demography note, I'm not sure why that code was invented but
>>> perhaps the demography note was forgotten since it was not in
>>> babase? There may be other cases where this code should be used,
>>> particularly among the data from 1996 - 1999 that has not been
>>> entered into babase and where information becomes increasingly
>>> sketchy over time.
>>>
>>> The demography note [with comments]:
>>> "14 Mar 94 1330 - Scion darted, fitted with radio collar (channel
>>> 6), released at 1700, rejoined group. NOTE: Scion has an infant,
>>> named Cocoa, ~9 months old, that looks unwell. Coat is extremely
>>> pale, almost white and kid spent most of morning riding lying
>>> dorsal. After Scion was darted, it joined other kids and played,
>>> continued after Scion returned to group.
>>>
>>> "There is no record of this kid's birth or sex. RSM/SNS have been
>>> assuming it was Cake (born Jun92) but this is not possible. We
>>> looked at reproductive records; decided that it was likely to have
>>> been conceived in Feb 93 (she's scored as cycling in Feb and Jan93,
>>> then as PPA [flat and black, actually] in Mar) and born Aug93. She
>>> was seen only once, Jun93, in the interim [no data collected in
>>> April or May] and scored P/B at that time, but this was not noted on
>>> the reproductive records in the field notebook [not sure what this
>>> means - she is marked P/B on the sexskin sheet in the 1993 notebook
>>> - 11 June was the only census date that month]. No census of
>>> Proton's done, June93-Jan94. The new infant has been called Cake
>>> since Mar94 [but Cake and other juveniles, aside from infants seen
>>> with their mothers, was not marked present on 1994 census sheets -
>>> Cocoa was, beginning in May, where he is marked as male - he is
>>> sporadically marked present for the rest of 2004 but then either
>>> died or joined the older juveniles in the realm of the
>>> nonidentifiable]; not known if Cake present.
>>>
>>> "Note: Something similar may have happened with Voi, less likely
>>> with Vera, Sunbird."
>>>
>>> Scion's kids:
>>> bioid ¦ sname ¦ name ¦ pid ¦ birth ¦ bstatus ¦ sex ¦
>>> matgrp ¦ statdate ¦ status ¦ dcause
>>> −−−−−−+−−−−−−−+−−−−−−−+−
>>> −
>>> −−−−+−−−−−−−−−−−−+−−−−−−
>>> −−−+−−−−−+−−−−−−−−+−−−−−
>>> −
>>> −−−−−−+−−−−−−−−+−−−−−−−
>>> 67 ¦ CAS ¦ CASEY ¦ SCI1 ¦ 1988-02-05 ¦ 0 ¦ F ¦
>>> 1.00 ¦ 1993-06-11 ¦ 0 ¦ 0
>>> 74 ¦ CHO ¦ CHOMA ¦ SCI2 ¦ 1989-10-23 ¦ 0 ¦ M ¦
>>> 1.00 ¦ 1989-11-26 ¦ 1 ¦ 1
>>> 68 ¦ CAT ¦ CATE ¦ SCI3 ¦ 1990-11-01 ¦ 0 ¦ F ¦
>>> 1.30 ¦ 1993-06-11 ¦ 0 ¦ 0
>>> 63 ¦ CAK ¦ CAKE ¦ SCI4 ¦ 1992-06-05 ¦ 0 ¦ M ¦
>>> 1.30 ¦ 1993-06-11 ¦ 0 ¦ 0
>>> 1015 ¦ ¦ ¦ SCI5 ¦ 1993-06-11 ¦ 1 ¦ U ¦
>>> 1.30 ¦ 1993-06-11 ¦ 4 ¦ 7
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>> Sent: Saturday, October 10, 2009 11:01 AM
>>> To: Jeanne Altmann
>>> Cc: Niki H. Learn
>>> Subject: Re: JA input needed - wounds and pathologies question
>>>
>>> OK, so we're agreed on 2 codes, one for the walk and one for the
>>> white
>>> fur. I don't think it is feasible to have a "stiff walk" code that
>>> is
>>> specific to this particular pathology, simply because our judgment
>>> won't always be right -- we're not making a clinical assessment,
>>> just
>>> an observational one. I think it makes sense to just class it with
>>> other instances of "stiff walk."
>>>
>>> It does seem to me that it will be useful to track down Alison's
>>> work
>>> and compare it to what you have found, Niki. I don't know whether
>>> you'll be able to do that before Jeanne gets home, but it seems like
>>> the next step in the process of figuring out what to do.
>>>
>>> Susan
>>>
>>> On Oct 10, 2009, at 4:05 AM, Jeanne Altmann wrote:
>>>
>>>> Sorry for the delay.
>>>> Thanks for the analysis and thoughtful input Niki.
>>>>
>>>> I agree that we should have 2 codes, 1 for the walk, the other for
>>> the
>>>> white fur and that infants exhibiting both should get both. I am
>>>> not
>>>> clear whether we need a new code for stiff walk or can use the
>>>> existing
>>>> one for locomotor problem? However, Susan rightly raised the
>>> potential
>>>> confusion when we have a localized locomotor problem that seems to
>>> be
>>>> from a fall or wound (e.g. thorn or other injury of that sort).
>>>>
>>>> It would be good to see whether Alison caught all of the candidate
>>>> infants--multiple-limb, non-injury stiff walk or white fur.
>>>>
>>>> jeanne
>>>>
>>>> -----Original Message-----
>>>> From: Niki Learn [mailto:nlearn at princeton.edu]
>>>> Sent: Friday, October 09, 2009 1:48 PM
>>>> To: 'Jeanne Altmann'
>>>> Subject: FW: JA input needed - wounds and pathologies question
>>>>
>>>> Jeanne,
>>>>
>>>> Have you had an opportunity to look at this?
>>>>
>>>> Currently the "stiff walk" associated with white monkey syndrome
>>>> has
>>>> been
>>>> listed using two or three different codes (malaise/weakness, other
>>>> pathology, etc.). I had suggested combining it with the code for
>>>> white
>>>> fur
>>>> but Susan thought it would be better to give the stiff walk its own
>>>> code,
>>>> which is fine too. I think we then should just mark the rare
>>>> individual
>>>> with both white fur and the stiff walk with both codes.
>>>>
>>>> Thanks,
>>>> Niki
>>>>
>>>> -----Original Message-----
>>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>>> Sent: Tuesday, September 29, 2009 8:55 AM
>>>> To: Jeanne Altmann
>>>> Cc: Niki H. Learn; Lacey Maryott
>>>> Subject: Re: JA input needed - wounds and pathologies question
>>>>
>>>> I think there is no hurry or urgency on this at all -- it can
>>> wait. I
>>>> just wanted to make sure that you knew it will need your input.
>>>>
>>>> Susan
>>>>
>>>>
>>>> On Sep 29, 2009, at 8:07 AM, Jeanne Altmann wrote:
>>>>
>>>>> Hi niki & susan,
>>>>> You'll need to give me a few more days to get back to you on this,
>>>>> Sorry,
>>>>> jeanne
>>>>>
>>>>> -----Original Message-----
>>>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>>>> Sent: Monday, September 28, 2009 3:54 PM
>>>>> To: Niki H. Learn
>>>>> Cc: 'Jeanne Altmann'; 'Lacey Maryott'
>>>>> Subject: Re: JA input needed - wounds and pathologies question
>>>>>
>>>>> Hi, see below.
>>>>>
>>>>>
>>>>>> Adding another code for the stiff walking makes a lot of sense.
>>>>>> Since none
>>>>>> of the older cases appear to have received wounds & pathologies
>>>>>> sheets there
>>>>>> are only cases from 2004 onward then in the record so far. Would
>>>>>> previous
>>>>>> cases have been noted in demography notes or something? (There
>>> are
>>>>>> a few
>>>>>> recent cases where wounds and illnesses mentioned in demog notes
>>>>>> that did
>>>>>> not have wounds sheets were added to the database but I don't
>>> think
>>>>>> this was
>>>>>> done for older data and maybe not consistently for the newer
>>> data.)
>>>>>
>>>>> I think Jeanne needs to give input on this. I believe she had an
>>>>> undergrad working on this at one point, and there was an issue of
>>>>> finding these cases if I recall correctly. I think it turned out
>>> that
>>>>> they had not been consistently recorded. But, I also seem to
>>>>> recall
>>>>> that the undergrad scoured the data and found whatever there
>>>>> was. I
>>>>> certainly do feel like there have been cases before 2004 but I
>>> don't
>>>>> remember.
>>>>>
>>>>>> I think
>>>>>> Rhus is the only one that displayed both symptoms during the
>>> 2004-6
>>>>>> period
>>>>>> and Rhus currently has two codes for the one illness. We could
>>>>>> easily have
>>>>>> the two symptoms separate and just assign both to any individuals
>>>>>> that
>>>>>> display both symptoms.
>>>>>>
>>>>>> What about the Coxsackie? Is that something that has shown up
>>> again
>>>>>> since
>>>>>> 1987? Should that have a code of its own if the team is readily
>>>>>> able to
>>>>>> identify it? Or is it really only obvious that it's Coxsackie
>>> if a
>>>>>> bunch of
>>>>>> animals are seen with it at once like in 1987?
>>>>>
>>>>> I don't know. I suspect that the team would not name it. I don't
>>>>> think
>>>>> we've had any outbreaks of multiple individuals since 1987, but we
>>>>> may
>>>>> have had single cases, and I don't think we would necessarily know
>>>>> from the wounds sheets.
>>>>>
>>>>> Susan
>>>>>
>>>>>>
>>>>>> Thanks,
>>>>>> Niki
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>>>>> Sent: Monday, September 28, 2009 3:17 PM
>>>>>> To: Niki Learn
>>>>>> Cc: 'Jeanne Altmann'; 'Lacey Maryott'
>>>>>> Subject: Re: wounds and pathologies question
>>>>>>
>>>>>> Hi Niki,
>>>>>>
>>>>>> Thanks for the detailed input.
>>>>>>
>>>>>> The 1987 outbreak was a different pathology, as far as we know,
>>> from
>>>>>> white monkey syndrome (I am the SA that took the notes with RSM
>>> -- I
>>>>>> didn't know I ever used just SA!). We think this is so because,
>>>>>> unlike
>>>>>> white monkey syndrome, which develops somewhat gradually, the
>>>>>> 1987
>>>>>> cases came on very suddenly and resolved fairly quickly and
>>> affected
>>>>>> all age classes, not just infants. And, there was no white fur.
>>>>>> It
>>>>>> behaved like a previous outbreak of Coxsackie virus (a water-
>>>>>> borne
>>>>>> virus) that Jeanne and Stuart identified in the 1960's. I
>>> guessed at
>>>>>> the time that it was Coxsackie, but have no real way of knowing.
>>>>>> But,
>>>>>> it wasn't the same as the syndrome that the little kids get. It's
>>>>>> true
>>>>>> that the stiffness of the walk is similar, but the onset and
>>>>>> termination were so much more rapid, and other things so
>>> different,
>>>>>> that I think it's a reasonable assumption.
>>>>>>
>>>>>> Regarding how to record it when it seems to be what we are
>>>>>> calling
>>>>>> white monkey syndrome, I would almost favor creating a new
>>> category
>>>>>> just for that (combination of white fur and stiff walking OR
>>>>>> stiff
>>>>>> walking in an infant that appears to be consistent with white
>>> monkey
>>>>>> syndrome). This would potentially create errors if we categorize
>>> the
>>>>>> disease wrong but would have the advantage of making it a lot
>>> easier
>>>>>> to find these cases. I think using 10 is way to nonspecific given
>>>>>> what
>>>>>> we know about the disease.
>>>>>>
>>>>>> How does this sound?
>>>>>>
>>>>>> SUsan
>>>>>>
>>>>>>
>>>>>> On Sep 28, 2009, at 2:52 PM, Niki Learn wrote:
>>>>>>
>>>>>>> There is a sheet of observations from 2006 that specifically
>>> deals
>>>>>>> with
>>>>>>> identified and possible white monkey infants. This refers to
>>>>>>> the
>>>>>>> stiff walk
>>>>>>> as a "walking problem with the back legs". In the regular
>>>>>>> wounds
>>>>>>> and
>>>>>>> pathologies sheets I found references to white-furred and stiff-
>>>>>>> walking
>>>>>>> baboons between 2004 and 2006 - I guess there was a bit of an
>>>>>>> outbreak then.
>>>>>>> For these years the wounds and pathologies sheets almost always
>>>>>>> call
>>>>>>> it
>>>>>>> the/that "stiff walk". Sometimes when white fur is mentioned
>>>>>>> the
>>>>>>> team
>>>>>>> specifically noted that no stiff walk was noticeable. Rhus had
>>>>>>> both
>>>>>>> white
>>>>>>> fur and the stiff walk. Rhus is marked with both codes 15
>>>>>>> (white
>>>>>>> fur) and
>>>>>>> 10 (malaise, weakness, stiffness in absence of wound). Little's
>>>>>>> note
>>>>>>> referred directly to this case: "She is beginning to have that
>>>>>>> stiff
>>>>>>> walk
>>>>>>> like Rhus". This is one of the ones marked with a 14 (other
>>>>>>> pathology)
>>>>>>> during the more recent data entry effort.
>>>>>>>
>>>>>>> It appears there was another outbreak in 1987 where 31 animals
>>> were
>>>>>>> given
>>>>>>> pathology sheets on the same day with the same symptoms. They
>>> were
>>>>>>> all
>>>>>>> reported on one sheet by SA and RSM with the following note
>>>>>>> "Partial
>>>>>>> hind
>>>>>>> limb paralysis with stiff, not fully extended legs seen in many
>>>>>>> baboons of
>>>>>>> Alto group on this same day". These are marked with pathology
>>> code
>>>>>>> 10
>>>>>>> (malaise, weakness, stiffness in absence of wound). There is no
>>>>>>> mention of
>>>>>>> white fur in these notes. Strangely the symptoms were gone five
>>>>>>> days later
>>>>>>> and the original datasheet mentions that "this bout of paralysis
>>>>>>> was
>>>>>>> very
>>>>>>> brief relative to bouts of paralysis in the past". I have not
>>> yet
>>>>>>> found any
>>>>>>> prior references though - perhaps they were not previously
>>> recorded
>>>>>>> on
>>>>>>> wounds sheets? There are no cases of white fur recorded back
>>> then
>>>>>>> either -
>>>>>>> the first record of it (in wounds and pathologies) appears to
>>> be in
>>>>>>> 2004.
>>>>>>>
>>>>>>> Sometimes stiff or stiffly is used when referring to walking or
>>>>>>> general
>>>>>>> movement of other baboons but "stiff walk" is not used and it is
>>>>>>> pretty
>>>>>>> clear from the notes that the baboon is wounded, has stiffness
>>>>>>> restricted to
>>>>>>> one limb though no wound is visible, is sick in some other way,
>>> or
>>>>>>> is just
>>>>>>> plain getting old (this last often includes mention of being
>>>>>>> hunched).
>>>>>>> There are very few cases where "stiff walk" or the above note
>>>>>>> for
>>>>>>> Alto's
>>>>>>> group was not used that cannot be quickly recognized as being
>>> cause
>>>>>>> by one
>>>>>>> of these other causes.
>>>>>>>
>>>>>>> I would definitely say we need to change the ones that are
>>>>>>> listed
>>>>>>> as
>>>>>>> 14 - we
>>>>>>> could go with 10, as in the 1987 cases and even Rhus' case, or
>>>>>>> expand 15 to
>>>>>>> include the stiff walk along with white fur under the umbrella
>>>>>>> of
>>>>>>> white
>>>>>>> monkey syndrome.
>>>>>>>
>>>>>>> Thanks,
>>>>>>> Niki
>>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>>>>>> Sent: Saturday, September 26, 2009 10:14 AM
>>>>>>> To: Niki Learn
>>>>>>> Cc: 'Jeanne Altmann'
>>>>>>> Subject: Re: wounds and pathologies question
>>>>>>>
>>>>>>> Thanks Niki,
>>>>>>>
>>>>>>> It might be helpful if you could compile a list of the different
>>>>>>> ways
>>>>>>> the white monkey syndrome has been recorded in the past, as you
>>>>>>> alluded to below. Then Jeanne and you and I and Lacey can
>>> consider
>>>>>>> options better.
>>>>>>>
>>>>>>> Susan
>>>>>>>
>>>>>>> On Sep 25, 2009, at 1:53 PM, Niki Learn wrote:
>>>>>>>
>>>>>>>> Sure, we'll see what Jeanne says. All the instances I've seen
>>> so
>>>>>>>> far refer
>>>>>>>> to it as "that stiff walk" or "the stiff walking" so it seems
>>> like
>>>>>>>> the team
>>>>>>>> has a specific way they refer to it.
>>>>>>>>
>>>>>>>> Also I found a case this morning where one individual was at
>>> first
>>>>>>>> noted to
>>>>>>>> have white fur and then developed the stiff walk. The team
>>> wrote
>>>>>>>> up a
>>>>>>>> separate wounds sheet for the stiff walking but noted that it
>>>>>>>> was a
>>>>>>>> continuation of the other note. In the wounds database this
>>>>>>>> was
>>>>>>>> combined as
>>>>>>>> one note but given two wounds/pathologies codes, 15 for the
>>> white
>>>>>>>> fur and 10
>>>>>>>> for "Malaise, weakness, stiffness in absence of wound", which
>>>>>>>> seems
>>>>>>>> more
>>>>>>>> appropriate than the 14 (other pathology) that Tabby used
>>> anyway,
>>>>>>>> if
>>>>>>>> we want
>>>>>>>> to keep the stiff walk separate (a code I suspect will get
>>> used a
>>>>>>>> lot in
>>>>>>>> this year's wounds & pathologies entries for those slipping
>>>>>>>> away
>>>>>>>> due
>>>>>>>> to the
>>>>>>>> drought). One of the cases I found yesterday even referred to
>>>>>>>> this
>>>>>>>> individual's case, as in "so and so is developing that stiff
>>> walk
>>>>>>>> like
>>>>>>>> Rhus".
>>>>>>>>
>>>>>>>> In any case there seems to have been some inconsistency in how
>>> the
>>>>>>>> stiff
>>>>>>>> walk has been recorded under different data people. I also
>>> found
>>>>>>>> that
>>>>>>>> sometimes body part code 44 (over entire body) is entered for
>>>>>>>> cases
>>>>>>>> of white
>>>>>>>> fur and in other cases there is no entry for body part (as is
>>> the
>>>>>>>> case with
>>>>>>>> the stiff walk and "limping, no wound visible" if no body part
>>> was
>>>>>>>> specified
>>>>>>>> by the team). So I will wait to see what the consensus is and
>>>>>>>> then
>>>>>>>> implement a uniform policy for it.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Niki
>>>>>>>>
>>>>>>>> -----Original Message-----
>>>>>>>> From: Susan Alberts [mailto:alberts at duke.edu]
>>>>>>>> Sent: Friday, September 25, 2009 1:35 PM
>>>>>>>> To: Niki Learn
>>>>>>>> Cc: 'Jeanne Altmann'
>>>>>>>> Subject: Re: wounds and pathologies question
>>>>>>>>
>>>>>>>> Thanks Niki,
>>>>>>>>
>>>>>>>> I like your idea of modifying the description to read "white
>>>>>>>> monkey
>>>>>>>> syndrome, symptoms = white fur and/or stiff walk". In practice,
>>>>>>>> these
>>>>>>>> two symptoms are different, but we do seem to be converging on
>>> the
>>>>>>>> idea that they are indicating the same thing. And, it is very
>>>>>>>> likely
>>>>>>>> that the team essentially treats them as two different
>>>>>>>> manifestations
>>>>>>>> of the same thing. That is, it seems possible that both would
>>>>>>>> occur
>>>>>>>> but they would record only one at a time, and not refer to an
>>>>>>>> individual with stiff walking already having had white fur,
>>>>>>>> especially
>>>>>>>> if they did a "white fur" sheet a couple of months prior.
>>>>>>>>
>>>>>>>> I think the risk is that stiff walking could come from other
>>>>>>>> things,
>>>>>>>> so we run the risk of attributing some stiff walking to white
>>>>>>>> monkey
>>>>>>>> syndrome when it's not really that.
>>>>>>>>
>>>>>>>> I'd like to hear Jeanne's thoughts. Also she might recall the
>>>>>>>> discussion when Tabby changed them to 14.
>>>>>>>>
>>>>>>>> Susan
>>>>>>>>
>>>>>>>> On Sep 25, 2009, at 9:13 AM, Niki Learn wrote:
>>>>>>>>
>>>>>>>>> Jeanne and Susan,
>>>>>>>>>
>>>>>>>>> I've been sifting through the wounds and pathologies tables
>>>>>>>>> figuring
>>>>>>>>> out
>>>>>>>>> where things left off, how this or that was assigned, and
>>>>>>>>> investigating
>>>>>>>>> bizarre notes, etc. I found two notes from Linda's 2006
>>>>>>>>> wounds
>>>>>>>>> where young
>>>>>>>>> baboons developed "that stiff walk that some Viola's juveniles
>>>>>>>>> get".
>>>>>>>>> Both
>>>>>>>>> were followed for several months and the stiff walk did not go
>>>>>>>>> away
>>>>>>>>> during
>>>>>>>>> that time. These were originally marked with a pathology
>>> code of
>>>>>>>>> 15
>>>>>>>>> (white
>>>>>>>>> fur) and Tabby changed them to 14 (other pathology). I don't
>>>>>>>>> know
>>>>>>>>> who
>>>>>>>>> entered them originally but I remember from Alison's
>>> presentation
>>>>>>>>> at
>>>>>>>>> the
>>>>>>>>> joint meeting that this stiff walk was also a symptom of white
>>>>>>>>> monkey
>>>>>>>>> syndrome, just like the white fur and that baboons often
>>>>>>>>> displayed
>>>>>>>>> only one
>>>>>>>>> of the two symptoms. So wouldn't the original code of 15 be
>>>>>>>>> correct
>>>>>>>>> here?
>>>>>>>>> Perhaps the code description should be updated to read
>>> something
>>>>>>>>> like "White
>>>>>>>>> monkey syndrome, symptoms = white fur and/or stiff walk"?
>>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>> Niki
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> --------------------------------------------------------
>>>>>>>> Susan Alberts, Dept of Biology, Duke University, Box 90338,
>>> Durham
>>>>>>>> NC
>>>>>>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --------------------------------------------------------
>>>>>>> Susan Alberts, Dept of Biology, Duke University, Box 90338,
>>> Durham
>>>>>>> NC
>>>>>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --------------------------------------------------------
>>>>>> Susan Alberts, Dept of Biology, Duke University, Box 90338,
>>>>>> Durham
>>>>>> NC
>>>>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> --------------------------------------------------------
>>>>> Susan Alberts, Dept of Biology, Duke University, Box 90338,
>>> Durham NC
>>>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>> --------------------------------------------------------
>>>> Susan Alberts, Dept of Biology, Duke University, Box 90338, Durham
>>> NC
>>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>> --------------------------------------------------------
>>> Susan Alberts, Dept of Biology, Duke University, Box 90338, Durham
>>> NC
>>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> --
>>> - -
>>> Lacey K. Maryott Roerish
>>> Alberts Lab
>>> Department of Biology
>>> Duke University
>>> ph: 919-660-7306
>>> fax: 919-660-7293
>>> Lacey.Maryott at duke.edu
>>> _______________________________________________
>>> Babase mailing list
>>> Babase at www.eco.princeton.edu
>>> http://www.eco.princeton.edu/mailman/listinfo/babase
>>
>> --------------------------------------------------------
>> Susan Alberts, Dept of Biology, Duke University, Box 90338, Durham NC
>> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>>
>>
>>
>>
>>
>> _______________________________________________
>> Babase mailing list
>> Babase at www.eco.princeton.edu
>> http://www.eco.princeton.edu/mailman/listinfo/babase
>>
>>
>> _______________________________________________
>> Babase mailing list
>> Babase at www.eco.princeton.edu
>> http://www.eco.princeton.edu/mailman/listinfo/babase
>
> --------------------------------------------------------
> Susan Alberts, Dept of Biology, Duke University, Box 90338, Durham NC
> 27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
>
>
>
>
>
> _______________________________________________
> Babase mailing list
> Babase at www.eco.princeton.edu
> http://www.eco.princeton.edu/mailman/listinfo/babase
>
>
> _______________________________________________
> Babase mailing list
> Babase at www.eco.princeton.edu
> http://www.eco.princeton.edu/mailman/listinfo/babase
--------------------------------------------------------
Susan Alberts, Dept of Biology, Duke University, Box 90338, Durham NC
27708, 919-660-7272 (Ph), 919-660-7293 (Fax)
_______________________________________________
Babase mailing list
Babase at www.eco.princeton.edu
http://www.eco.princeton.edu/mailman/listinfo/babase
More information about the Babase
mailing list