32) A complicated picture: getting accurate and meaningful outcome measures- weight changes, etc.

I started this blog on July 2 to track and report the changes (if any) of my son's response to oxytocin.  In my excitement of starting the new hormone replacement, I blogged like crazy to keep up with my observations about the experiment and other related thoughts related to the topic of hypothalamic obesity, hormone replacement, and other miscellaneous topics inspired by the experiment.  I have since stumbled into some serious confounding problems when we found that his thyroid hormones were very low and when we discovered that Sasha was engaging in secret chocolate binging.  I have also realized that relying on Sasha's reports on his own observations during the experiment is highly confounded by certain biases and conflicts of interest including (1) his desire to make the experimenter (me, his mother) happy by reporting what he believes is a desirable response (called "social desirability bias") and (2) the agenda of the hyperphagic subject to seek food at all costs including at the cost of honesty. These discoveries have taught me that getting meaningful results of the experiment are highly dependent on keeping certain variables constant and on choosing only reliable data to measure.

 As I wrote in the post on July 25 (#27), my research questions remain the same. In order to answer the research questions, only objectively observed data should be admissible as reliable.  Unfortunately, this makes some of the data are very difficult or inconvenient to measure given the nature of the limitations of our lack of controls in our (home and beyond) "laboratory." 

Objective 1: to see if oxytocin will help reduce Sasha's intense hunger and food obsession and includes these possible data to measure:

• Weight reduction/stabilization (as long as hormone levels and general health status are kept stable and WNL)
• Reduction of food-related tantrums/arguments down from baseline
• Reduction of the mention of food during daily conversations down from baseline
• Reduction of time spent in kitchen cooking/being near food down from baseline
• Delayed time or absence of snack or meal requests down from baseline

Objective 2: to see if oxytocin will help to reduce incidence of food seeking/stealing and includes these possible data to measure:

• Reduction of discovered incidents of food seeking from baseline including reduced home incidence and reduced reports of incidence from teachers,  camp counselors, etc.  Sasha's current environment will ensure food security which includes locked cabinets and fridge, guardian supervision while exposed to food at home/school/camp, intermittent and random searches of room, pockets and backpack, alarmed doors
• Weight reduction/stabilization over time if/when food sources are to be made available by the gradual loosening of food security (this variable will be measured only if and when Sasha is able to demonstrate that his hunger and food obsession are reduced with food security in place)

Objective 3: to see if oxytocin will improve his social motivation and increase his interest in having social interactions/friendships with peers and includes these possible data to measure:

• Increase of Sasha's friendship network with peers as evidenced by his requests for "play dates" with other kids his age from baseline

Some of the variables that can be measured are much more feasible in an ideal world (contained lab setting) but not really very easy or convenient in the real world (our home and beyond). Do I really want to count the number of minutes Sasha spends in the kitchen/with food per day or how many times he brings up the subject of food in conversation or the exact time he requests his snacks on a daily basis? Ugh. I will have to give these variables more consideration in order to select which ones are most optimal to study. Even Sasha's weight, the easiest measurable variable, has its challenges as an outcome measure to help answer Objective #1. Consider these facts as complications to the reasons for his weight gains and losses:

1. Sasha made steady weight gains (half kilo per week) from May 15 until June 18.  Between June 18 and June 25, he lost an entire kilo (this is the week we started noticing the appetite reducing effects of the oxytocin).  
2. Sometime after July 7, Sasha's weight precipitously increased two kilos and July 11, we learned that his thyroid was very low. On July 13, we started him on an increased dose of his T4 and T3. 
3. On July 15, we learned that he had a huge chocolate stash and was probably eating excessive amounts of chocolate. We do not know when the chocolate binging began but we are guessing (based on his precipitous weight gain between July 7 and July 11) that it started sometime in the first week of July.
4. On August 5, we received his lab results that both his T4  and T3 levels have returned to being in the high-normal range and that he has lost 1.2 kilos (2.6 pounds) in one week.

Low thyroid can easily affect one's weight and energy (as can binging on large amounts of chocolate!) so it is not possible to know the exact portion of responsibility the thyroid levels versus the chocolate consumption played in the weight gains and losses.  All I know is that going forward, I will be prudent in ensuring that his thyroid levels are kept at an optimal and constant level in order for us to know if the weight gains/losses can be credited to food consumption or to metabolism efficiency.

I didn't exactly set out to conduct flawless research when I decided to do this experiment but I certainly want it to be meaningful and helpful to my son and others.  In regards to maintaining the rigors of research in my imperfect and error-prone "lab" environment, endocrinological management and experimentation sure keeps me on my toes!