Think of it. Often while driving, you see a personalized license plate with some odd combination of characters and it takes several seconds before you can figure out what it says. Wouldn’t it be fun to have a book full of these encoded messages? The Polk Company would be the perfect source for this data and I figured it would only take me a few weeks to put together a book and shop it around to a publisher.

Unfortunately, it turns out it had already been done at least once. Within a couple days several of my coworkers told me they had already seen such a book and one of them brought one in and gave it to me. Alas, the book has since been lost. (Sorry, Carol.) A search on Amazon turns up thousands of possible replacements, so maybe there would still be room for one more. Oh well.

One of many fun fact books on license plates. Image from Amazon

I hadn’t thought about that project in quite a while, but within the past two weeks artwork made from personalized license plates has caught my eye twice.

The first was a piece called Preamble by Mike Wilkins, an artist the same age as me. He recreated the preamble of the U.S. Constitution using license plates from all 50 states and Washington, DC. The work was completed in 1987 in celebration of the bicentennial of the Constitution. I saw it while visiting the American Art Museum in Washington.

He uses many of the abbreviations common to personalized license plates. I wonder what the censors at the department of motor vehicles thought of some of them:

WE TH | P PUL | OF TH | UNI | DIDD | ST8S

INNOR | DUR 2 | 4M A | MOR PUR | FEC UNE | NONE

S TAB | LISH | JUSTIZ | N SURE | DOME

ESTIK | TRAN | KWILI | T PRO | VIDE 4 | TH COM

UN DE | FENZ | PRO MOT | THE JEN R | L WEL

FARE N | C CURE | TH BLES | NGS OF | LIBBER | T 2 R

SELVES | N R POS | TERI T | DO R | DANE N

S-TAB | LISH | THIS | CON STI | 2 10 | 4 TH

U NI | TID | ST8S | OF AH | MARE | E CUH

The 51 plates are arrayed in a grid, in alphabetical order. (I think it would have been even more clever to lay them out to form the shape of the continental U.S. with the plates placed close to their geographical location. But that’s me.) You can see a photograph of this wonderful piece at http://www.americanart.si.edu/collections/search/artwork/?id=27722. (More on the reason for my visit to DC, an fMRI scan of my brain, in a future blog post.)

The second piece of license plate art I saw recently was in connection with the famous typographer and designer Jessica Hische. Her work is featured at the cleverly named website, http://jessicahische.is/awesome/.

The Society of Design in Pennsylvania recently invited Ms. Hische to speak at an event. In order to create the invitation, they crafted a heartfelt message. Then 35 members of the society each bought a personalized license plate with a piece of the message. The text was broken up as follows, notice the lack of cute abbreviations:

DEAR JES | SICA PLE | ASE CONS | IDER VIS

ITING SO | CIETY OF | DESIGN I | N PENNSY

LVANIA A | ND SHARI | NG CAPTI | VATING A

ND AMAZI | NG TYPOG | RAPHIC W | ORK THAT

WILL AMA | ZE ASTON | ISH MOTI | VATE AND

PROVE TO | BE BENEF | ICIAL TO | AN ENORM

OUSLY LA | RGE CROW | D THANKS

You can read about the project and see a photo of the actual invitation at http://invitinghische.com/.

Estimating page views per month

Estimating peak server requests per second is a four step process. First, we must estimate page views per month. Next, we estimate average page views per second during the heaviest or prime viewing period. Then we estimate peak page views per second during the prime viewing period. Finally, we estimate peak server requests per second during the prime viewing period.

Average page views per month can be obtained by looking at server logs. If server logs are not available or you are creating a new site, then it can be estimated using logs from similar sites. For this blog post, we will assume the website generates an estimated 2.6 million page views per month. (By comparison, this blog generates about 2,000 page views per month, mostly from bots, I think.)

Estimating average page views per prime viewing second

Assume your website that generates 2.6 million page views per month has traffic that is fairly steady all day and all night on every day. That is, of the 730 hours in a month (= 365 days per year / 12 months per year * 24 hours per day) , all of them will be prime viewing hours. In that case, we can calculate the mean page views per second by doing some simple arithmetic.

page views per second = 2.6 million page views per month / 730 hours per month / 3600 seconds per hour = approx. 1 page view per second

But what if traffic to the website isn’t steady. What if people only visit it during work hours? Well, there are about 168 work hours per month compared to about 730 actual hours per month, a ratio of about 4.3 to1. So during prime viewing hours there will be about 4.3 page views per second and 0 page views per second during non-work hours. (This assumes everyone works the same days and same hours regardless of time zone.)

The prime viewing hours for a website can be even more compressed. Let’s say you run a website for NBC and it has a blog that contains a synopsis of the television show Grimm and an update is posted immediately after each new episode airs. In that case, perhaps all of the page views will occur during a 4 hour period starting at 10:00 pm every Friday. Thus, there will be 16 prime viewing hours per month during which there will be 45 page views per second and 0 page views per second during the rest of the month.

The chart below shows the page view distribution for the three cases described above. This model is quite simplified. It can obviously be made more complex by assuming that the prime viewing hour is dependent on time zone, that page views do not drop to zero during the non-prime viewing hours, and having multiple variables that affect page views during a particular hour.

Three ways to achieve 2.6 million page views per month. Image by George Taniwaki

Let’s look at the distribution of page views in more detail. In the four-hour prime viewing period case we said that there were 0 page views per second before and after the prime viewing period and an average of 45 page views per second during the prime viewing period. If the number of page views is constant throughout the prime viewing period, then the distribution curve is rectangular as shown by the blue line in the chart below.

But it is unlikely that the change in page view rate is so abrupt. It is more likely that page views rise steadily to a peak and then fall. If the distribution is triangular and spread across four hours, then the average page views at the maximum point will be 90 per second (=45*2) as shown by the brown line below. If the distribution is bell shaped, called the normal distribution, then the average peak page views at the maximum point will be somewhere in between as shown by the green line below.

Three ways to achieve 625,000 page views in an evening. Image by George Taniwaki

One caveat, I tried to draw the curves in the chart above so that all three would have similar variance but didn’t actually do the calculations to verify it.

Estimating peak page views per prime viewing second

All of the work above was to find the average number of page views per second during the prime viewing time. However, visitors to the website will arrive randomly. So we can expect that there will be some fluctuation in the number of page views during a second. Some seconds during the prime viewing time will have fewer than the average number of visitors and some seconds will have more. We can model this random arrival of visitors using the Poisson distribution.

Since the arrival of visitors will be random, we cannot estimate the maximum number of visitors the website will ever receive in a second. That number is actually infinite. But we can estimate it for a variety of confidence levels, such as 90%, 99%, and even 99.999% (the so-called five 9s availability level). In this case confidence level indicates the proportion of one second intervals that will be below the peak.

Using Excel’s Poisson distribution function we can estimate the ratio between peak page views per second to average page views per second at various confidence levels. The results are shown in the three tables below. Notice that although the average page views per second can be a fraction, the peak page views per second is always an integer.

*Assumes 168 prime viewing hours per month with uniform distribution

Also notice that when the average page views per second is low, the peak page views per second can have two solutions, 0 or 1. These cases occur when the average page views per second is below 1- confidence level. For instance, if all you care is that your web server can handle all the traffic 99% of the time, and your average traffic is less than 0.01 page views per second you don’t need a web server at all! That’s because 99% of the time (during the prime viewing period), there is no traffic to your website.

However, if your goal is to be able to serve 99% of your visitors during the peak viewing time, then you need a web server than can deliver at least one page view per second. And if you provide a web server to deliver one page every 100 seconds, your ratio of peak to average will be 100. Providing a complete web server ready to serve the rare visitor results in tremendous overhead costs, which is why cloud computing, where resources are shared among many websites, is becoming so popular.

Finally, notice that when the average page views per second is high (say 1,000) , then the ratio of peak to average is close to 1 and does not grow very much even at high availability (or confidence) levels. At high levels of average page views, the error in estimating the average number of page views per second is likely to be much greater than the error introduced by ignoring the random distribution of page views per second.

Estimating peak server requests per prime viewing second

Our last step is to estimate the number of server calls generated by a single web page request from a user. A typical web page consists of a static html file plus one or more images, videos, ads, and JavaScript widgets displayed on the page. (In the case of dynamic pages, the content will be generated on the server, usually as a jsp file, or a aspx file if you are using the Microsoft .NET Framework.) If the page is not cached, sending all of the page contents may take over 100 requests to the server.

Assume your website consists of a single page that contains 100 items and all of those items reside on a single server. Now assume a single user calls for that page and you don’t want the user to have to wait more than one second before being able to interact with any part of it. That means the web server will need to be able to handle at least 100 requests per second per page view. (There are other potential bottlenecks in rendering the web page including Internet traffic, ISP speed, and the speed of the client computer, but we’ll ignore these for purposes of this blog post.)

Using the five nines confidence level, the final results for page views and server calls are shown in the table below. For our website with an expected 2.6 million page views per month, we need a web server that can handle 1,600 requests per second.

*Assumes 100 server calls per page view
**Assumes 168 prime viewing hours per month with uniform distribution
***Assumes goal is to satisfy 99.999% of visitor requests, not 99.999% of time

*****

Update: If your website is one of many hosted on a single server, then you should skip the calculations for estimating peak page views per prime viewing second. That’s because traffic to your site will be combined with traffic to other sites. In that case, it is up to the company hosting the sites to combine the average traffic from all the sites first, then calculate the peak page views based on their promised availability.

The house already had a laundry room, but we demolished it and the wall between it and a guest bedroom. The combined rooms will become the new master suite bath. (Yes it will be a large bathroom.)

Meanwhile, we demolished the old master bathroom. Most of it will be converted into a walk-in closet. (Yes, that’s a large walk-in closet. And we’re combining the two reach-in closets in the master bedroom and the guest bedroom to make another walk-in closet. So we will have a pair of walk-in closets, a hers and hers set.)

The end of the old master bathroom, where the bathtub was, will be walled off and converted into the laundry room.

The before and after floor plans are shown below.

Original floor plan with old laundry room and master bathroom highlighted

Proposed floor plan with new smaller laundry room and larger master bathroom highlighted

Demolition

The first step in creating the laundry room was removing the old bathtub. It is cast iron and really heavy. It took four guys to move it down the stairs into the new dining room until I make room for it in the basement (where we will be adding a full bathroom).

At this point, we discovered a problem. The gasket between the shower faucet handle and the tub surround wasn’t sealed properly and one or two drops of water would land on the subfloor every day. In Seattle that means the subfloor would stay wet and was rotted. Also, when the plumber originally installed the drain for the tub, he cut through the joist. Now the floor was sagging. So we had to rip out the ceiling on the first floor underneath the tub, sister the existing joist with a new 2×10” and replace a section of the subfloor.

Framing

You will notice that the new laundry room has a bi-fold door where a wall in the old bathroom was. This is a load bearing wall, so it needed to be replaced with a beam and two columns. Normally, when you do this, you install temporary “crib” walls on both sides of the load bearing wall before removing it. But I cheated. Instead, I used the existing hallway wall as one support and built the new wall for the back of the laundry room and used that as the other temporary support. Then I removed the load bearing wall and installed two 4×4′” columns and a 2.75”x9.75”x7’ glulam beam. (Actually, I hired a framing contractor to do this since it’s too heavy to lift alone.)

Rough-in

The next step was to complete the plumbing, mechanical, and electrical work. The new plumbing includes a hot and cold water outlet box with single handle shut-off valve and  a pair of anti-hammer arresters, washer drain line and trap, and a floor drain and trap with drip valve (to prevent the floor trap from drying out and creating a stink). The mechanical work includes a 1/2” gas line, an in-wall dryer exhaust that vents through the roof, and a room exhaust fan that vents through the roof. This last item was not my choice, it’s required by building code. Electrical work includes two GFCI circuits (one each for washer and dryer), two IC recessed lights, the exhaust fan, and a wall switch.

Walls

Once all rough-in inspections were passed, I filled the wall with R-13 glass fiber insulation for sound control and covered it with 5/8” green wall board held with ceramic coated screws. I followed that with tape, mud, three coats of plaster, primer, and paint.

Laundry room getting a coat of primer. Photo montage by George Taniwaki

Floor

The laundry room has a tile floor. First, I cut a piece of 4×4” pressure-treated lumber to build a curb. I originally wanted to put the curb entirely inside the laundry room so that it would not be visible from the hall. But that would not leave enough room for the dryer and its duct. So I cut notches in the lumber and fit it around the opening. (Note to self, next time make the laundry room at least 35” deep.)

Because the washer and dryer will vibrate a lot, I wanted to add a water-proof isolation membrane between the underlayment and the tile. I used Schluter Ditra and Kerdi-Band. Since the laundry room will rarely (hopefully, never) have running water on the floor, I decided not to slope the floor toward the drain. This also allows me to use larger tile on the floor.

Here’s where I learned (the hard way) that there are several different kinds of mortar. Mortar used to lay bricks tends to have coarse sand in it. Mortar for tile has fine sand. Mortar for tile also contains a small amount of acrylic latex to make it smoother. This is called thin-set mortar. If you add even more acrylic latex (rather than water) when mixing the mortar, it then is called modified thin-set mortar.

You use modified thin-set mortar to apply the membrane to the floor. You use unmodified thin-set mortar to apply the tile to the membrane. (Don’t confuse brick mortar with unmodified thin-set mortar.) You need to use unmodified thin-set mortar because latex-modified mortar requires air to cure and since the membrane and tile are both nonporous, there won’t be enough air contact for the mortar to cure properly.

Laundry room with waterproof membrane. Photo montage by George Taniwaki

After waiting an hour for the membrane mortar to set, it’s time to lay the tiles. In the picture below, notice the arrangement of tiles around the floor drain. Normally, you want to the tiles to intersect over the hole so that you can use a tile saw to cut a series of notches and nibble away to the edge of the hole. However, that wasn’t possible in the small laundry room. The hole had to be in the center of the tile.

Cutting a large hole in a floor tile is difficult. It requires enormous patience and arm strength if you cut it by hand using a rod saw. I gave up and took the tile to Tile For Less and they cut it for me for $35 using a heavy-duty handheld angle saw. This is another reason (besides fitting the slope) for using small tiles in a shower stall. All of the other tiles in the laundry room have straight cuts or bevel cuts and I was able to make them using my inexpensive ($39 on sale from Harbor Freight Tools) 4”wet tile saw.

When installing tile, always use the little cross-shaped rubber tile spacers. You’ll never get the tiles aligned nicely by eye. Also use the right sized notched trowel when applying the mortar, bigger notches for bigger tiles. And back butter large tiles (like the 12×12 tiles in this project) to ensure good adhesion. Finally, use a dead blow hammer to tap the edges of the tiles to get them all to the same height.

Laundry room with tile. Notice the tile spacers, notched trowel, and dead blow hammer. Photo montage by George Taniwaki

After waiting 20 minutes, I washed the tiles with a sponge to get the stray mortar off. Then I waited a day and applied grout. The local Home Depot carries two kinds of grout, sanded (for use with tiles spaced 1/4” or wider) and unsanded (softer but can fill smaller gaps, used for tiles set under 1/4” apart). All of the grout is pigmented. There are 30 different colors, all with cryptic names like sandstone, fawn, and bone. I didn’t have a tile with me, so I called my wife and asked her to pick up a spare tile from the laundry room and open a web browser compare it to the color swatches on the Home Depot website.

Being an expert on color theory, Windows GDI, LCD display technology, and color management didn’t give me any confidence that this would work. But the alternative would be to compare a tile to the colors printed on the bags of grout. Neither seemed ideal. She picked haystack, a grout color that seemed darker than the darkest color in the tile and I bought a bag of the sanded grout. I mixed it up and used a plastic mud knife and a hard foam pad to apply it.

If you are doing this, wait twenty minutes, then wash off the excess using a scrubbing sponge. Afterwards, a thin haze will appear. Don’t worry. Wait two hours until it hardens, then use a polishing sponge to wipe it off. Wait three days, then seal the tile and grout with a liquid sealant. Touch up the wall paint and you are done.

It turns out the grout color is lighter than we expected. However, it is exactly the same color as the body of the tiles. This made it blend in around the walls and the curb which is actually even better than planned.

Laundry room with wet grout (top), with a coat of haze after washing excess grout (middle), and the finished floor after polishing, sealing, and touch-up painting of walls (bottom). Photo montages by George Taniwaki

Finish

The final step is installing shelving, a clothes rod, and connecting the appliances. The shelf brackets are designed for wire shelves. I snipped off the little hooks that hold the wire shelves so that I could use solid shelves. The shelves are made from 3/4” MDF with white melamine laminate. I cut notches in each shelf to fit around the vertical posts. The clothes rod is hollow and sounded cheap when a hanger would click against it. So I filled the rod with polyurethane spray foam insulation to deaden the sound.

When connecting the dryer vent to the exhaust, use plenty of metal foil tape (not duct tape) to seal the connections. This will keep lint out of the laundry room and reduce the amount of carbon monoxide (assuming you have a gas dryer) released into the house.

Mostly finished laundry room. Photo montage by George Taniwaki

In total, this project took me over 100 hours to complete (everything described above except heavy framing and plumbing) and probably cost over $3,000 in labor (split between the framing team and plumber) and $1,000 in materials. And it still isn’t done. I still need to add tile and grout to the exterior of the curb and I need to build and install custom bi-fold doors. Neither project can be done until I install the floor in the hall. And that project is in the distant future.

Before going into the specifics of the errors in The Fiscal Times article, let’s consider how one could compare whether cooking at home is more expensive than eating at restaurants. The typical cost-benefit analysis for eating in versus dining out goes something like as follows. Cooking a meal at home isn’t free. From a classical economic point of view one should include the opportunity cost of the time needed to buy groceries , drive it home, store it,  prepare a meal, and clean up afterwards. Further, one should include the implicit rental value of the automobile used to transport the groceries and the kitchen and dining room used to prepare and serve the meal.

However, shopping, cooking, and cleaning are not just chores that one is required to do. They are a form of entertainment, social interaction, and a way to share your skills with others as Nathan Myhrvold insightfully states in this Dec 2011 Slate interview. The cook receives utility from hosting a meal, even if it is a regular daily event. Naturally, if one hates to shop, cook, or clean, then there can be disutility as well. When deciding whether to eat at home or dine out, a person will want to maximize the expected utility from the decision.

Examining the wrong factors

Write a letter

One of the most successful methods for finding a living donor is to send a letter to your social network. It is often hard to ask for a donation for yourself. In this case, it may be helpful to first find a friend or family member who cannot be a donor herself to be the sender of the letter. We will call this person the “matchmaker”.

The letter should include the following:

1. Introduction of patient and matchmaker and why you are sending this letter
2. Short medical history explaining why patient needs a kidney
3. Explanation why matchmaker cannot be donor
4. Request to spread the word
5. Request for potential donors to get a blood test (mention type O, if patient is type O)
6. Contact info for the living donor transplant coordinator at the transplant hospital where surgery will take place (or the living donor advocate, depending on the process at the transplant hospital). If the letter will be sent to people who live in another state, discuss this with your transplant team first. There may be options available for the potential donor to start evaluation at another hospital
7. Most transplant centers will only evaluate a single potential donor at a time. Thus, to reduce the chance that a strong candidate is kept waiting, you may want all potential donors to contact your matchmaker (not the transplant center’s transplant coordinator or donor advocate) first, and pick the most determined potential donor to go through the evaluation first

The letter shown below is based on an actual letter sent out by a person on behalf of a patient. Some details have been changed to maintain anonymity.

<Date>

<Letter_Recipient_Address>

Dear <Letter_Recipient_Name>,

    I am writing to share with you the latest news about my dear friend and your former accountant, <Patient_Name>. As you know, <Patient_Name> closed his business in 2010, after thirty years of dedicated service to the community. What you may not know is that <Patient_Name> also has chronic kidney disease and will soon need a transplant.

    In 1996, <Patient_Name> received a transplant from a deceased donor. He took excellent care of this precious gift and it lasted until this summer. Today, he undergoes dialysis therapy three times a week for four hours per treatment while waiting for a kidney. This time around, the chances of finding a kidney from a deceased donor are quite low since he has developed antibodies caused by the first transplant. It pains me to see my former boss who I love and respect being disabled like this.

    Several of <Patient_Name>’s friends have offered a kidney but for various reasons none have turned out to be acceptable candidates for donation. <Patient_Name> has no other family in the Washington area. For that reason, I have taken it upon myself to help him locate as many people as possible with blood type O who are willing to have a blood test to see if they would be medically suitable to be <Patient_Name>’s donor. There is absolutely no expense or obligation. If you or someone you know is willing to take this first step toward saving <Patient_Name>’s life, please contact me, <Matchmaker_Name> at <Matchmaker_Phone_Number>.

    As much as I would like to, I am unable to donate a kidney to <Patient_Name> because <Matchmaker_Reason>.

    If you have any questions about being a donor, you may contact the Independent Living Donor Advocate at the University of Washington Medical Center, where <Patient_Name> is a patient. The advocate’s role is to look out for the best interests of donors.  Her name is Paige Kayihan and her phone number is 206-598-3627 .

Thank you for reading this heart-felt request.

Sincerely,

<Matchmaker_Name>

For another example letter, see http://lkdn.org/letter_from_friend.html.

Thanks to Harvey Mysel of Living Kidney Donors Network for providing me with example letters that I use in my patient counseling. Thanks also to the patient who wishes to remain anonymous, whose letter is used in the example above. For more ideas on finding a donor, check the LKDN Resource Guide, on pages 30-33.

Run a classified ad

Some of you may be familiar with the story of Chaya Lipschutz. In 2005, she donated her kidney to a stranger after reading an ad in the newspaper. Afterwards she started a free kidney matchmaker service. Among the ways she helps her clients find a donor is through the use of classified ads. Below is a typical Craigslist ad she ran for one of her clients.

A typical kidney matchmaker Craigslist ad. Image from Chaya Lipschutz

The key points that make this simple posting a winning ad are:

1. A catchy headline
2. Emphasis that donating a kidney may save a life
3. Description of the problem (what your life as a patient is like)
4. Statement that live donation is generally safe
5. If the patient has O-blood type, request a matching donor (otherwise this isn’t important, unmatched blood type donors can enter a kidney exchange)
6. Call to action (remember to take it one step at a time, you want the reader to commit to getting tested at the transplant center, do not ask to commit to the actual donation)
7. A photo of the patient (in this case of the matchmaker) that is well-lit, in focus, and showing a sincere smile

To learn more about Chaya Lipschutz and her effort to match kidney patients with potential donors, visit KidneyMitzvah.com. Ms. Lipschutz is very enthusiastic about helping patients and about recruiting kidney donors and it shows in the design of her website.

Thanks to Harvey Mysel of Living Kidney Donors Network for pointing out this classified ad to me. For more ideas on finding a donor, check the LKDN Resource Guide, on pages 30-33.

The previous blog post argued that switching from opt-in to opt-out could increase the number of people on the organ donor registry but could actually reduce the number of organs recovered.

That is because an opt-out process creates ambiguity about the intent of those on the organ donor registry. This would make the Uniform Anatomical Gift Act (UAGA) harder to enforce.

One way to avoid this problem is to couple the use of an opt-out-donor registry with increased training of clerks at the DMV to inform each customer that they will be added to the registry unless they opt-out. In addition to training costs, there will be increased labor costs since each customer transaction may be about a minute longer as the clerk explains what the organ donor registry is and sells the benefits of organ transplantation to the customer.

This extra effort to educate the public is needed to get implicit consent from the driver. Unlike presumed consent where the customer is never told that a decision is being made for them, implicit consent creates a true decision. Unlike mandated choice where the customer is forced to make a cognitively complex choice in a short time span, implicit consent relies on framing to make the default option (the one most people will pick) the one that is most beneficial for society.

Using the same hypothetical data presented in the prior blog posts, I have created a table showing the organ recovery rate when combining opt-out with implicit consent. Assume that an opt-out registration system results in 88% of drivers registering to be organ donors (same rate as in table 2 of that blog post). Of these, the OPO is able to get 99% of families to cooperate (same as in table 1). The OPO does not approach the families of patients who were on the opt-out list (same as in table 2). The overall organ recovery rate is 87%, significantly higher than the 81% rate in the opt-in case or the 79% rate for opt-out without implicit consent. This appears to be a big win.

By combining opt-out with implicit consent, 88% of drivers register and 87% of organs are recovered

What impact could the combination of opt-out and implicit consent make in the United States? That is difficult to predict since no state has attempted to implement them together. Legislation was introduced to implement opt-out and presumed consent in New York last year by an assemblyman whose daughter had received two kidney transplants. But the bill never made it out of committee. (See debate in New York Times May 2010.) Similar legislation was introduced in Colorado earlier this year but was withdrawn after public protests and consultation with Donor Alliance, the local OPO.

Why isn’t the combination of opt-out and implicit consent gaining political traction in the U.S.? Most likely it is because the training required to implement implicit consent correctly would be expensive. Even with training, at least one unwilling donor family will probably request an injunction against the OPO. The potential result of this litigation was described in the last blog post. The resulting media coverage and lobbying would likely put pressure on the state legislature to eliminate the opt-out nature of the registry. It could also cause them to revoke the UAGA. This could make it harder for OPOs to recover organs than before the switch to opt-out since they currently can recover organs without consent of the family under opt-in.

Finally, if the driving public feels it is being coerced into becoming donors, it may result in falling donation rates (higher opt-out rates) and reduce trust in the healthcare system. Overall, the combination of opt-out and implicit consent just doesn’t seem like a winning strategy to increase organ recovery rates.

Much thanks to thank Alexandra Glazier, Vice President & General Counsel at The New England Organ Bank, for clearly explaining that adopting an opt-out registration process does not automatically result in adopting presumed consent. Each issue needs to be analyzed separately.

Now I will discuss how opt-out can result in uncertainty in the composition of drivers listed in the registry. This uncertainly can impact the behavior of organ procurement coordinators and family members.

Role of certainty in interactions between counselors and family

In states that maintain a donor registry, they share the list of names on the registry with the organ procurement organization (OPO) that is responsible for recovery and distribution of organs for transplant. If a patient dies under conditions that allow the organs to be recovered, an organ recovery coordinator at the OPO will see if the patient’s name is on the organ registry.

Under opt-in, if the patient’s name is on the registry the coordinator can be fairly certain the deceased patient wanted to be a donor (categories 1a and 1b as defined in the previous blog) and can confidently tell the family this and proceed with recovery. Under the Uniform Anatomical Gift Act enacted in most states, a gift by a donor cannot be revoked by the family.

If the patient’s name is not on the registry, the intent of the patient isn’t known. Perhaps the patient wanted to donate (category 3a), didn’t want to donate (2b or 3b), or wanted the family to decide (2a or 4). The coordinator can say the patient’s wishes were not known and politely request the family to make an organ donation on behalf of the deceased patient.

Under opt-out, there are more categories of drivers included in the registry. This reduces the certainty in the composition of the donor registry. This is true even if no drivers are misclassified (i.e., no drivers fall into categories 3a, 3b, or 4), This uncertainty will have an impact on the behavior of the coordinators.

Specifically, if the deceased patient’s name is on the registry, the coordinator cannot be certain the deceased patient wanted to be a donor. She must rely on presumed consent. However, if the family complains that it was not the deceased patient’s intent to be a donor, then the ambiguous nature of the composition of the registry may lead to a delay, which will make recovery impossible. If the OPO pushes the issue, eventually, a court case may resolve the issue, but if the ruling is in favor of the patient’s family, then the entire registry is placed at risk.

Conversely, if the patient’s name is not on the registry, then having the coordinator approach the family to request a donation is also problematic since a donation would require the family to override the wishes of the deceased. If that is allowed, then the wishes of the deceased should be allowed to be overridden if she is on the organ donor registry as well. Again, if the OPO pushes the issue, the organ donor registry is placed at risk.

A hypothetical example of outcomes

Let’s look at some hypothetical numbers to illustrate a possible outcome. In the first table below, the state has an opt-in registration system and has a 64% registration rate. (This is very high, but is achieved in Washington, the state where I live.) The OPO approaches the family of every patient who dies under conditions that allow the organs to be recovered. For patients on the registry it works to enforce the UAGA and gets 99% of families to cooperate in time. For patients not on the registry, it works hard to persuade the family to donate and gets half to cooperate. Overall 81% of organs are recovered.

Under opt-in, 64% of drivers register to be donors and 81% of organs are recovered

Now suppose that the state switches to an opt-out registration system and the registration rate rises to 88%. However, the cooperation rate among families drops from 99% to 90%. Also, the OPO does not approach any of the families of patients who were on the opt-out list. Overall, the organ recovery rate drops to 79%, lower than it was before the switch. Naturally, I set the numbers to make my case, but it illustrates that switching from opt-in to opt-out will not on its own automatically ensure that donation rates will increase.

Under opt-out, 88% of drivers register to be donors but only 79% of organs are recovered

Mandated choice

As mentioned in the previous blog entry, there is another option besides opt-in and opt-out called mandated choice. Under mandated choice, the state wants to eliminate the last categories 3a, 3b, and 4 (driver choice undeclared or driver undecided) that create ambiguity. Thus, the law requires the DMV clerk to ask every driver to declare a choice. (It’s not clear what happens if the driver refuses to make a choice or if the clerk forgets to ask or forgets to record the choice.) Several states have tried it, but have given up and returned to opt-in. Currently, only California is experimenting with it, see Jun 2010 blog post.

Texas, which had about a 15% registration rate with opt-in, increased it to about 20% with mandated choice. Unfortunately, I can’t find any data to show if overall organ recovery rate rose or fell after this change. However, the state has abandoned mandated choice, so my guess is the OPOs in that state either saw a drop in donation rates or feared one would occur and lobbied for the return to opt-in.

How opt-out and mandated choice may reduce donation rates

Why has mandated choice failed, and why could opt-out cause donation rates to fall? I think a lot of it may be because of people’s fear of death. Signing up to be an organ donor while applying for a driver’s license is an admission by the registrant that she may die in an accident and needs to make a decision about the disposition of her organs in the event that happens.

Under the current opt-in process, those who are not afraid of death opt-in. Those who are afraid don’t state their preference. For those who don’t opt-in, the decision to donate is still available later to the family. Under opt-out, people who are willing to donate (or let their family decide) but are not willing to admit they may die will opt-out. This is a firm decision, precluding the family from making the donation later.

In the next blog post we will explore ways to make opt-out compatible with individual choice and consent.

One of the frequent proposals to increase the number of deceased organ donors is to switch the registration of donors from an opt-in system to an opt-out system. In an opt-out system, all drivers are presumed to want to participate (default=yes) and people who do not wish to participate must state their desire not to be listed.

Let’s look at the logical and ethical issues this change would present.

Not just a framing problem

Several well-known behavioral economists have stated that switching from opt-in to opt-out is simply a framing problem. For instance, see chapter 11 of Richard Thaler and Cass Sunstein’s book Nudge and a TED 2008 talk by Dan Ariely using data from papers by his colleagues Eric Johnson et al., in Transpl. Dec 2004 and Science Nov 2003 (subscription required).

The basic argument is that deciding whether to donate organs upon death is cognitively complex and emotionally difficult. When asked to choose between difficult options, most people will just take the default option. In the case of an opt-in donor registration, this means they will not be on the organ donor registry. By switching to an opt-out process, the default becomes being a donor. Thus, any person who refuses to make an active decision will automatically become a registered organ donor (this is called presumed consent). This will increase the number of people in the donor registry without causing undue hardship since drivers can easily state a preference when obtaining a driver’s license.

However, these authors overlook two important practical factors. First, switching from opt-in to opt-out doesn’t just reframe the decision the driver must make between two options. It will actually recategorize some drivers.

Second, it changes the certainty of the decision of those included in the organ registry, which affects the interaction between the organ recovery coordinators at the organ procurement organization (OPO) and the family member of a deceased patient.

There are more than two states for drivers regarding their decision to donate

Note that the status of a driver’s intent to be an organ donor is not just a simple two-state Boolean value (yes, no). There are actually at least three separate states related to the intension to be an organ donor. First, upon the driver’s death, if no other family members would be affected, would she like to be an organ donor (yes, no, undecided). Second, has she expressed her decision to the DMV and have it recorded (yes, no). Finally, would she like her family to be able to override her decision (yes, no, undecided). The table below shows the various combinations of these variables.

*No or Undecided options make no sense in this context

Opt-in incorrectly excludes some drivers from the donor registry

Now let’s sort these people into two groups, one that we will call the organ donor registry and the other not on the registry.

Under the opt-in process, only drivers in categories 1a and 1b are listed on the organ registry. These drivers have given explicit consent to being on the registry. Drivers in categories 2a, 2b, 3a, 3b, and 4 are excluded from the registry. Thus, we can be quite certain that everyone on the registry wants to be a donor. (There is always a small possibility that the driver accidentally selected the wrong box, changed their mind between the time they obtained their driver’s license and the time of death, or a computer error occurred.)

In most states the drivers not on the organ registry are treated as if they have not decided (i.e., as if they were in the fourth category). When drivers not on the registry die under conditions where the organs can be recovered, the families are asked to decide on behalf of the deceased.

Under an opt-in process, drivers in category 2a are miscategorized. They don’t want to be donors and didn’t want their family to override that decision, but the family is still allowed to decide. The drivers in categories 3a and 3b are miscategorized as well. The ones who don’t want to be donors (3b) are also forced to allow their families to decide. The ones who want to be donors (3a) are now left to let their families decide.

Opt-out incorrectly includes some drivers in the donor registry

Under an opt-out process, drivers in categories 1a, 1b, 3a, 3b, and 4 are grouped together and placed on the organ registry. If the donor registry is binding and the family is not allowed to stop the donation, then the process is called presumed consent. (Note that many authors use opt-out and presumed consent interchangeably. However, they are distinct ideas. Opt-in is a mechanical process of deciding which driver names are added to the registry. Presumed consent is a legal condition that avoids the need to ask the family for permission to recover the organs.)

Drivers in category 3a who wanted to be registered are now correctly placed on the registry. But any drivers in category 3b who don’t want to be on the registry are now assumed to want to be donors, a completely incorrect categorization. Similarly, all drivers in the fourth category who were undecided are now members of the definite donor group and the family no longer has a say.

Only drivers in category 2a and 2b are excluded from the registry. We can be quite certain these people do not want to be donors. But some (category 2a) were willing to let the family decide. Now they are combined with the group of drivers who explicitly do not want to donate.

The distribution of categories into the registry under the opt-in and opt-out process and how they are treated are shown in the table below.

Ethical implications of misclassification

If there are no drivers in categories 3a, 3b, and 4, then switching from opt-in to opt-out will have no impact on the size of the donor registry. However, if there are any drivers in these categories, then some will be incorrectly categorized regardless of whether opt-in or opt-out is used. This miscategorization will lead to some ethical problems.

Under opt-in, there may exist cases where the drivers has made a decision to donate (category 3a) or not (categories 2a or 3b) but family members overrules it. These errors are hard to avoid because they are caused by the lack of agreement between the drivers and other family members.

However, under opt-out combined with presumed consent, there may exist cases where neither the driver (category 3b) nor the family want to donate, but cannot stop it. Similarly, the driver may want to let the family choose whether to donate (category 4) and the family does not want to donate but cannot stop it.

It appears that from an ethical perspective, opt-in is less likely to create a situation where the respect for individual’s right to make decisions about how the body should be treated is denied. For further discussion of the ethical issues see  J. Med. Ethics Jun 2011, and J. Med. Ethics Oct 2011 (subscription required).

Next we will look at the impact switching from opt-in to opt-out will have on the interaction between the organ recovery coordinator and the family. See Part 2 here.

[Update: This blog post was significantly modified to clarify the “decision framing” issue.]

Rob Rushing of the Love Hope Strength Foundation at Red Rocks Amphitheater. Photo by Hyoung Chang for The Denver Post

I’m fascinated by the group’s motto of “Saving lives one concert at a time.” Here is an organization that has found a way to attract the young people who make the best bone marrow donors. It makes itself highly visible at rock concerts.

By doing so, Love Hope Strength generates publicity for its cause. Recruiting people at the event itself is very useful. But by doing it in a public venue it makes joining the registry a socially acceptable activity. And it does more. It generates interest in donating money and in becoming a volunteer. Finally, it educates the public about the important role adult stem cells play in fighting cancer and other diseases. Overall, this is a wonderful model to engage young adults in a healthcare related activity.