Happy Birthday

Tomorrow is my last day! I've asked to be picked up from my bed and breakfast in the morning (so I can bring my suitcase to work with me, and then leave with it for the airport after work is done). I have a meeting with Hasifa and Ruth at noon, hopefully we can discuss how to get a paper out. We'll mainly discuss biases that might be happening and things of that nature that we might want to take into account before writing up a paper. I've been in contact with Tim about the time series analysis, and he's been gracious and kind enough to offer to fit an ARIMA model for me using his $10,000 time series program that I can try to recreate the results of using Stata. 

In tomorrow's meeting I'll also bring up getting a copy of whatever other databases we want to clean and merge. The end goal for the surveillance program is to have a clean database with the same variables for each health center- that way we can look at different regions of Uganda, and ask our questions and hopefully get real answers. The question of "did this cause that" is a tough one to nail down. 

Happy birthday to my sister, from Uganda. I didn't get you a giraffe, but I got you something better from Chile. 

Work Related Update (Might be boring, but at least it shows I'm alive)

Yesterday and today were spent cleaning the database, and just as I thought it was done, it turns out there is more missing information that they want to go back and look up in the paper records- something out of my control. We’re shooting for a complete database and meeting for writing up potential study findings tomorrow.

Yesterday had a lot of rain- Ugandan weather is either nice, or it’s a downpour like you wouldn’t believe (last time I was here it was hailing bullets). Fortunately, I’ve been lucky enough to be inside each time it rained (thunderstorms Monday night, with clearing and sudden showers Tuesday during the day). I ate out yesterday, at an Indian restaurant across the street (I’ve been there before). In general, the largest population after the native Ugandans are Indians, and so Indian food is a pretty good bet. I had a vegetable dish, just because I am once again feeling the lack of vegetables in my diet.

I think when I get home, a lot of eating is in order. I actually am curious what my weight is now too.

In terms of the paper, the intervention seems to have worked pretty well (I mentioned that there was a study going on that ended at the same time the spraying happened so using the number of people showing up with malaria was a bad idea. However, if we use % of positive blood smears as our outcome instead, the percentage of people with positive malaria smears showing up probably won’t be affected, so we have a good outcome measure). I took a look at the % positive malaria lab results today, and you can see the same effect, a drop from about 40% positive to something almost sub 5%.

The major issue to deal with in this study is to account for all the biases, and things that could make the study results flawed. Usually the way this problem is approached is by asking, “what is the ideal way I would have done this experiment” and then look at how much your reality deviated from the ideal- each deviation is a potential source of bias/error which in turn can make your study results flawed and untrue. For example:

Ideal study: Find everyone in the Kihihi district (where this anti-mosquito spraying campaign, hereafter referred to as “IRS,” took place). Measure each person daily and capture every single case of malaria by taking a daily blood sample and looking at it under the microscope, to look for malaria parasites. This way you know exactly how many cases of malaria there are at any point in time. Do your spraying campaign, count malaria cases afterwards.

Now go back in time, to a world where you didn’t do your IRS campaign, so that all the weather conditions, distributions of people, conditions of roads, everything is the same, and instead of doing your spray campaign, don’t do anything, and see how many malaria cases happened afterwards.

Any difference in malaria cases between the first world and the second world (both identical except one world got the IRS, the other didn’t) would be the causal effect of spraying.

Single biggest limitation of the current study:

1) Can’t go back in time (I should work on this one, and make the big bucks).

Why this makes problems: I don’t have good comparison groups- I’d like to compare a sprayed population to a non-sprayed population to prove that IRS works or doesn’t work.

If I could go back in time, I would have 2 groups of identical people to compare, which would be great. The next best thing is to compare 2 groups of nearly identical people in one world, and give IRS to one but not the other, and look at malaria cases. I don’t have that either. The only comparison I get to make is a group of people without spray (before) and the same group of people after IRS (after).

The whole before-after comparison issue is fraught with problems, mainly the problem of all sorts of things changing over time (making things less and less identical between my before and after population). I can go about looking at my before and after population in a number of different ways:

1) Age: Make sure the age distribution of the before population is similar to the after population

2) Gender: Check before vs. after populations on this too

3) Parish: Make sure that people showing up to the clinic are representing their parish the same before and after spraying

4) Do all the above, and in combination (ie: make sure that the percent of children under the age of 5 that are male who showed up before spraying is the same as the percent of children under the age of 5 that are male who showed up after spraying)

5)…

There are a lot of checks I should run, comparing the before and after populations. If I can show that they are mostly identical, and also indicate that all these other variables (rainfall, humidity, road conditions, etc.) either don’t matter or were the same as well, then the paper should be in good shape.

In general, all the limitations that I have to think about now are the same limitations that I’ll have to work on in general if I plan to use surveillance data in the future. It’s nice to know that the work is applicable and has future implications. 

Weekend Post

This weekend was a half interesting, and half not so interesting. Saturday I went out to all the familiar haunts- I took a boda boda (moped) out to Lugogo Mall, where I picked up 4 glasses for drinking port/wine with (I got one last time I was out, and broke it about a month before I left for Uganda, and they’re only about a dollar each). I got a boda back home, and dropped off my purchases for the day, then went across the street to get some water. 

After that I took a matatu down into town (prices are similar, but a bit higher because of the gas crisis- apparently Kenya’s instability is affecting Uganda’s access to fuel) to Sam’s restaurant (where I had alligator tail last trip) and had ostrich this time. Ostrich (the meat, not the bird) looks just like beef actually, and was described as such on the menu. It was a good piece of ostrich- like eating the best parts of a steak (not stringy, very flavorful and well done). They also have wildebeest and some other antelope-like animal meat available, which I’ll probably try if/when I come back to Uganda to do work. Too bad I don't have my camera cable, or I could post a picture to complement the alligator tail posting.

After getting my first good lunch in a while (clif and luna bars are ok, but not the best thing to have every day) I went to garden city, the biggest? mall in the area. I decided to kill the rest of the afternoon by watching “I Am Legend” in the theater there- interestingly enough, they don’t open the doors till 15 min before the show, mainly because they book the theater back to back with very short windows. The showing right before mine had some very disgruntled customers (especially a family of muzungus/white people/foreigners) that complained that the sound system was terrible for this movie… I was a bit nervous, but all the complaining got the showing fixed in time for my showing, so that was a plus. The movie ticket prices are comparable to the States, at around 7 dollars. The theater itself was actually very nice, smaller than the big theaters at home, but larger than some of the independent cinema complexes.

Sunday was not as eventful- I stayed in most of the day, tried unsuccessfully to work, and basically waited out the weekend till today.

 

Today was great actually- I finished reading the 9^th or 10^th of 10 or so documents on time series analysis, and played around with the data that I have (still waiting on Ruth to finish her independent recreation of the dataset- she said she’d have it done by today, but looks like we’ll have to wait till tomorrow to present the dataset to Grant (via email) and Hasifa (in person). That ended up being ok with me, because I got to play around with fitting time series models in STATA, and got some really nice fitting models going on. I still have to figure out some of the details behind the best way to get rid of autocorrelation (I’m just differencing or lagging or both to see if I can get a stationary time series, I’m sure there’s some sort of guideline to which is more appropriate when- I’ll have to read this website by Duke University to clear that up). Playing around with time series and real data is actually rather satisfying, especially when you have something real to explain.

This past March 2007, the area that we have data for had a big drop in malaria cases showing up at the county hospital- before then had been erratic but on average, pretty high. After March there was a nice steep drop in cases, and the average dropped to a pretty low level- this was right around the time of the spraying campaign in the region, so that seemed like a no brainer, at least by eyeballing the data. So before this time point, there was no spraying, but high numbers of cases reporting in, and after that time point there was spraying and lower numbers of cases reporting in. The timeframe looked roughly realistic too, with spraying happening a week to two weeks or so (I still need the exact dates of the spraying campaign) before the precipitous decline. All was well, but I also had dates for a drug efficacy study that they were doing in the area at the time… which was being done since the beginning of surveillance data collection, and which also ended in March 2007! That is a potentially big snafu- if cases were reporting in because a big study was happening, and cases stopped showing up once the study was done, then maybe spraying didn’t cause the decrease in malaria cases. Getting the exact dates of the spraying will be important, and the amount of time needed to do the campaign as well (if the spraying got done very quickly, the entire region in a week, then that might well explain a precipitous decline in malaria cases, but if they took their time and needed a full month, it seems more likely that the study end was the cause of the decline in cases). Likely, it’s a combination of the two, one that may or may not be possible to tease apart. I’ll get input from the rest of the people on the project, and we’ll see what happens.

On a related note, I’ve thought of a good potential paper that I might be able to work on- it seems to be very promising, at least to me. I’ve run the idea by Grant, and if possible, it’ll be a great paper (hopefully I can get a first author off of it). I might be constrained by lack of data though. All in all, a good day at work, and a decent weekend. 

Ebola Update

I was talking to Ben Bellows the other day, another student who works in Uganda in the Mbarara area (where there is Ebola apparently) and he mentioned that one case of Ebola actually came to Mulago Hospital (where I work) in the capital city of Kampala (some stories online say 2 cases of Ebola came to Mulago, and both died). Luckily I'm up in the computer lab region, far away from the wards and where people check in, but it's still a bit close for comfort. Ebola only spreads by fluid contact with infected individuals, so I should be fine. 

Here's an update I got off of ProMED: 

EBOLA HEMORRHAGIC FEVER - UGANDA: (BUNDIBUGYO)

**********************************************

A ProMED-mail post <http://www.promedmail.org>

ProMED-mail is a program of the International Society for Infectious Diseases

<http://www.isid.org>

 

Date: Fri 4 Jan 2008

Source: UN Office for the Coordination of Humanitarian Affairs

(OCHA), ReliefWeb, Government of Uganda report [edited]

<http://reliefweb.int/rw/RWB.NSF/db900SID/EDIS-7AJL5L?OpenDocument>

 

Uganda: Ebola outbreak press statement -- 4 Jan 2008

----------------------------------------------------

The Ministry of Health wishes to update the general public on the Ebola situation in the country. The cumulative total is 149 patients with 37 deaths.

In Bundibugyo District, 5 people are currently admitted, one in Kikyo Health Centre and 4 in Bundibugyo Hospital. Of these, one new suspected case has been registered and is admitted in Bundibugyo hospital in the last 24 hours. One patient was discharged from Kikyo Health Centre. On a good note, 441 out of 771 contacts have completed 21 days of follow up and are considered safe.

The Ministry of Health continues to urge the public to avoid unnecessary movements and gatherings especially in the districts around Bundibugyo, that is, Kabarole, Kasese, Hoima, Kibaale, Kyenjojo, Kamwenge, Ibanda, Bushenyi, and Mbarara.

As we end the festive season, all District Health Officers, in charge of Health Units and the general public should not relax but continue being on alert and immediately report any suspected cases. Similarly, the public is reminded to report to Health Facilities any persons who present with the following symptoms: very high fever of sudden onset, diarrhoea, vomiting associated with red eyes, and a measles-like rash. Ebola is spread by close contact with body fluids of infected persons or people who have died of Ebola. Anybody handling suspected cases must use appropriate protective wear and must wash his/her hands thoroughly with soap and water or JIK [bleach] and water.

The Ministry assures the general public that the epidemic is still confined in Bundibugyo district and is getting contained. The public is urged to be compassionate and provide care and support to persons that have been discharged. I wish to thank all those who have provided support to the discharged patients. Let us continue. These people are not infectious and therefore should not be discriminated against.

The public should continue to be vigilant but without creating unnecessary fear and panic. The Central Alert Team will continue to respond to all alert cases brought to our attention. The numbers to contact remain the same:

Police Mobile 111 or 999

Police control room, Ministry of Health, 0772 587567, 0414 345108.

[Dr Sam Zaramba, Director General of Health Services]

 

--

Communicated by:

ProMED-mail

<promed@promedmail.org>

 

[The cumulative number of Ebola hemorrhage fever cases in the Bundibugyo District of Uganda is now given as 149 and the number of deaths as 37. These figures are considerably greater than the 113 cases and 27 deaths declared by the Ugandan Health Ministry on 13 Dec 3007. No doubt some of this increase may be due to enhanced surveillance and retrospective diagnosis, but it is unlikely that the outbreak has been extinguished. Indeed some patients remain in hospital and at least one is a new suspected case. - Mod.CP

In view of the fact that Ebola virus is only spread by contact with the body fluids of an infected monkey or ape, or a human case who is likely to be too sick to attend any gathering, the requirement to avoid movements and gatherings seems excessive. - Mod.JW]

Delays

Today was not as productive as it could have been- I was supposed to be picked up at 9:30 to move my belongings to the new bed and breakfast, but didn’t get picked up till 12 noon. Part of the reason was that Marx asked someone else to drive me, and that person needed gas (and there is a big gas crisis in Uganda, something on the order of 5x more expensive to buy gas here than in the States, and availability is lower too).

I exchanged some money to be able to pay for my stay at the bed and breakfast up front (something about since I got a discounted price at the bed and breakfast, the landlady wanted me to pay up front) dropped off my suitcase, took my mefloquine tablet for the week, and walked to work. This new bed and breakfast is right across the street from Kisamente, which I was very familiar with back when I was in Uganda last time- it’s very convenient since I both know the area well and know the transport in the area well. It’s overall made me feel less stress concerning my belongings and the commuting. I’ll actually look forward to wandering around this weekend doing late Christmas shopping and such. The only downside is that I don’t have a key for my room- apparently I can lock from the inside of my room, but when I leave, I have to leave it unlocked (something about the lock not being completely functional). The caretaker there assured me that it would be ok since she would be around and the front door guard would also be there, so I’m relatively confident that it’ll be ok.

I’ve stuck mainly to my own food so far, and not really eaten a lot of Ugandan food this trip- I tried pretty much all of the types of food last trip, and was rewarded with GI distress, so I don’t feel to bad living on Clif/Luna bars, ramen packs, dried fruit, almonds, and various snacky foods (Pringles, Cheez-its, etc.) that I packed from home. I do plan on going out to eat, and learned from Edwin how to potentially prevent diarrhea- eat a pepto-bismol tablet before the meal. They sell tablets now, which are much more convenient than drinking a cupful of the liquid stuff (and it tastes better too). I’ve eaten 1 Ugandan meal and at a Thai place with Grant, and before both I took a tablet, and so far my stomach has been ok. Who knows what would have happened sans-tablets.

The most productive part of today was going through 2 time series analysis primers, and I feel a lot happier and more confident about time series and ARIMA in general. I’ve downloaded a few other documents that should help me out as well, and I’m looking forward to using tomorrow and this weekend to really play around with my data and see what I can do with my newfound knowledge. Oh, and I have to do my lit reviews tomorrow so I have lots to read over the weekend.

Last note: I’ve been reading Kim Stanley Robinson’s Mars trilogy (Red Mars, Green Mars, and Blue Mars) and will probably finish the last book by the end of this trip- I’m halfway through. It’s a good series, very technical, but plausible sounding- the tech descriptions seemed well researched. After reading this series, I found it surprising that only today I thought of Uganda as a Mars-like place… but only with regards to the color of the soil. 

New Day, New Lodgings

Today was a bit slower work wise: it involved cleaning up the database and getting it to a final point (I think I've got the data in perfect shape now). I'm working on this database with Ruth, a Ugandan woman also involved with the surveillance project. She is going to try and merge the databases and clean it on her own as well, and we'll see if we get identical results, and then go to Hasifa with the final result. Once that is done, we can proceed to think about writing the introduction (meaning I'll have to do a lit review of indoor residual spraying, the intervention that we're studying) and the methods. I'll also have to think hard about the analysis section. 

As far as analysis goes, I've been emailing back and forth with Tim Bruckner, a UCB graduate who does a lot of work with time series analysis, and he's a great great help. I'm also downloading a lot of tutorials and documents on how to do stuff like removing autoregression, and figuring out how to do ARIMA models and such. If I can figure out the analysis stuff out, I'll be in great shape for my dissertation I think, since the data I have to work with are all surveillance data, which are collected over time (hence why "time series" analysis is useful). 

In terms of living situations, I've decided to move closer to Mulago hospital, which will be a huge relief. I've been living in the dorms at the Red Chilli Hideaway (advantages: very cheap, 2 cute goats and dogs to look at. disadvantages: very far, cheap means cheap lodging, additional stress because of unsecured belongings and 40 min-1hr commute with valuables), and after getting eaten by mosquitoes (they ran out of nets for the dorms) I bought a net of my own. They don't have any single rooms either, so I've been living in the dorms (they house 34 in a series of rooms) and carrying all my valuables on me, which makes me a bit paranoid about being out late, and also makes it tedious to carry around so much stuff. Marx, one of the drivers for the program here, helped me find a new place that Grant and some others had stayed at in the past, and it's a much nicer bed and breakfast within walking distance of the hospital. It's 30k shillings  a night (got a small discount from 35k) which is roughly 20 dollars US a night- it's more than 3x more expensive than Red Chilli's, but it's way closer (walking distance means just a few minutes, versus the 40 min to an hour otherwise), and I get my own locked room, and it comes with a bednet! (Now I have a bednet that I'll only be using one night, but I suppose I'll take it home with me and bring it with me to Uganda whenever I come back). The Edge house, which was full yesterday, suddenly called me today to let me know that the people who were going to stay there cancelled, and so they had room for me again... but this was after I put down a deposit on the other place, so I had to turn them down. In order to pay for all the new expenses, I'll have to get a bit more cash exchanged, but thankfully I brought enough with me. 

Having my own room in a locked compound will take a lot of stress off me- I still worry about leaving my suitcase in a room that sleeps 34 people that come in and out depending on where they are on in their touring around the country. I'll feel better about leaving my stuff in a locked room and locked compound and then going out to buy some christmas presents (haha) and traveling around the city for shopping etc. 

So far, my goals for the trip are 2-fold: 1) Figure out a plausible area for my dissertation, and flesh out a few potential papers I might be able to write (big task, I'll have to do some lit review searches) and 2) Get Hasifa's paper/project/database well on the way to getting into a publishable state by the end of the month (big, but realistic). 

First day back in Uganda

Hey all, looks like I’m back in Uganda again. After a long series of flights, I got into Uganda’s Entebbe airport at 10:30, and then passed through immigration and customs without any problem. One thing of note is that the visa entry fee is almost twice as much now, at $50 as opposed to $30 the last time I came through. Thank goodness they didn’t look through my luggage- not that I had anything illicit, but I did pack a whole lot of food so that I wouldn’t have to rely only on Ugandan food for the whole trip. If they took that away, I’d probably be on Pepto-Bismol for most of the trip and pretty uncomfortable (I never really adjusted to Ugandan food the last time I came out, and GI distress for the most part of 2-3 months was a bit much.

I was picked up at the airport by a driver from the hostel that I’m staying at currently, the Red Chilli Hideaway. The UCSF house that I stayed at last time has closed down, as has the Blue Mango, which was the great hostel that was right across the street. It’s a shame, especially since it turns out that the Red Chilli Hideaway is really really far away from Mulago hospital, all the way on the other side of the city (I left at 8 am, and between walking, riding a matatu, and exchanging money, it took me 1.5 hrs to get to Mulago). I’m considering moving to a place in Makerere University called the “Edge House,” where I know that Troy stayed at last time I was in Uganda. That’d cut down a lot on transport time and costs, and it would make me feel more comfortable about commuting later in the day. I was bitten a lot by mosquitoes last night at Red Chilli despite them having screens on the windows, but it was midnight by the time I checked in, so I couldn’t even get a bednet. Good thing I’m on antimalarials, and also that Kampala has a low level of malaria compared to the rest of the country. It’s just a bit annoying to be itchy and sticky when you wake up in the morning.

In other news, my cell phone is on the fritz, and refuses to stay on/work except under very specific conditions, namely me not moving or touching it too much. My SIM card from last time doesn’t work anymore, and I have to buy a new one, so hopefully someone can run errands with me at lunch or so (and maybe I can move into the Edge house too, depending on if I can contact anyone/arrange that to happen).

In terms of work, there is a lot to be done. The head of the surveillance project here, Hasifa, has a database that needs cleaning (I did most of that work on the plane) and after it’s clean, a paper needs to come out of it. This means that I/we can start on the intro, and then do the methods at least. However, it’s up to me to figure out the time series analysis part of the paper- I have some rainfall data that needs to be factored in, and since rainfall is a continuous time series thing just like number of malaria cases, it seems like some time series analysis of some sort needs to be done. I think I’ll try to predict malaria cases with rainfall, and use the residuals as the outcome variable for an analysis that looks at certain interventions that they did during the calendar year (spraying, etc.). That way, the residuals will represent all the malaria cases that can’t be accounted for by rainfall patterns and malaria seasonality. I’ll have to run this by Tim (a graduate from UCB who does a lot of time series work) and hopefully that will be the right thing to do.

Other work that needs to be done also includes taking all the old databases that they used to use and merging them into one big comprehensive clean database- I don’t know how hard/easy that will be, since I don’t have a long history with these databases. I also have to do probably 2-3 lit searches, one for Hasifa’s paper and the introduction, a second one on time series analysis and surveillance data (to get ideas for what I might get a first author paper out of) and then a third one for just surveillance data and malaria/other infectious disease papers (to get an idea of what other people have done with surveillance data, and see if that will be interesting to apply to the data that I’ll have). Hopefully I’ll come up with some good ideas (I already feel like controlling for climate/rain factors is a good thing to be able to do).

That’s it for now, time to see if I can arrange for some transport around town to do errands. 

Revival

I'll be back in Uganda from 1/7/08 - 1/18/08, so vector-borne.blogspot.com is revived for this short trip. Since I'm not there, here are a few Uganda highlights that I'm keeping an eye on:

From ProMED:

Sleeping sickness hit Mayuge
----------------------------
A total of 12 people were diagnosed with sleeping
sickness at Mayuge Grade Three health centre last
month [November 2007], the LC5 vice-chairman,
Peter Tasibula, has said. He warned that
residents were prone to the disease because the
Farming in Tsetse Controlled Areas project
stopped operating in the area. Sleeping sickness
is a vector-borne parasitic disease. The
parasites are transmitted to humans by tsetse
flies which acquire the virus from human beings or animals harbouring it.

[Mayuge District is located in the Southeastern
part of Uganda, and was already several years ago
classified as in a "pre-endemic" state regarding
sleeping sickness (trypanosomiasis) (see:
Berrang-Ford L et al. Spatial Analysis of
Sleeping Sickness, Southeastern Uganda,
1970­2003. Emerg Infect Dis 2006;12:813-20
(<http://www.cdc.gov/ncidod/EID/vol12no05/05-1284.htm>).
The cases reported here are therefore not a
surprise, and just confirm that trypanosomiasis
continues to be a problem in this part of Uganda.
General information on African trypanosomiasis
can be found at the WHO website
(<http://www.who.int/topics/trypanosomiasis_african/en/>). - Mod.EP]

*~*~*~*~*~*

Also from ProMED:

EBOLA HEMORRHAGIC FEVER - UGANDA (10): (BUNDIBUGYO)

[This death toll from Ebola hemorrhagic fever in western Uganda
continues to rise, having increased from 25 on Sat 8 Dec 2007 to 30
on Tue 11 Dec 2007. During the same period, the number of confirmed
cases has risen from 104 to 116. The number of people under
surveillance is not stated but was 328 on Sat 8 Dec 2007. - Mod.CP]

{Note that 30/116 deaths to confirmed cases is a much lower
case-fatality rate than in previous Ebola outbreaks -- see Karl
Johnson's comment in (09) in this thread. - Mod.JW]

*~*~*~*~*~*

There is a surveillance site in the southwestern corner of Uganda that I'm doing some data analysis for, so we'll see if I decide to go out there or not with an Ebola outbreak taking place. Probably not, but it depends on how big/widespread the outbreak is, and how close it is to the surveillance site.

I'll be traveling to Chile from today till I go to Uganda, so no posts until I start on my way to Uganda.