####################################################################################### ############## To Do ############## ####################################################################################### height, weight and fat % will be the independent variables, but fat will be constrained by (the explanation below) >>NP - FFM range for men is 18.5/1.15*h^2 to 30/1.15*h^2, for women 18.5/1.23*h^2 to 30/1.23*h^2. >>This comes from assuming average % fat for people in this range (15% for men, 23% for women). The lower limits above assume a very low FFM with average % fat - you can go below this limit by pushing down % body fat. Note that the lower limit is around a BMI of 16. At this point, you will be very low % fat and can begin removing FFM to drop weight, but you are now in the danger zone. -"Just right" (green) body fat should be different for men / women. >>SR: This will be converted to heart strength/strain anyways (which may be gender-specific) ####################################################################################### ############## To Do/Noah ############## ####################################################################################### I am not sure that I entirely agree with your calorie expenditures for physical activity. They seem a bit high and in many cases are absolute values. The total caloric cost for weight bearing activities should be based upon a per body weight value. What was your source for this information? ____________________________________________________________________________________________ WEIGHT DEPENDENT EXERCISE >>>> Figure the given values for exercise are for an average person. >>>Model for exercise calories burned: Caliories_burned = Cal * [1 + beta (W - W_0)/W_0] ...where Cal is the base calories for a person of weight W_0 (say 160 lb), W is the person's weight, and beta is 0 or 1 depending on whether the activity is weight dependent (1) or not (0). Beta could also be used to make activities more or less weight dependent (but for now leave it 0 or 1). Cal and beta are adjustable for each exercise, W_0 is a single constant for all exercises. >>>Fix text for "simulation age out of range" message Come up with definitions for different lifestyles ____________________________________________________________________________________________ ADD TO FOODS LIST Add the following to foods list: >>1 large size fast food value meal 1/4 pounder with cheese: 510 cal, 26 g fat, 40 g carb, 29 g protein Large fries: 500 cal, 25 g fat, 63 g carb, 6 g protein Large soft drink: 310 cal, 0 g fat, 86 g carb, 0 g protein Total: 1320 cal, 51 g fat, 189 g carb, 35 g protein Source http://www.mcdonalds.com/app_controller.nutrition.index1.html >>Coffee shop, sugary drink Large sugary coffee blended drink with whipped cream: 590 cal, 22 g fat, 96 g carb, 9 g protein Source http://www.starbucks.com/retail/nutrition_beverage_detail.asp ____________________________________________________________________________________________ CALCULATING BODY FAT BASED ON HEIGHT, WEIGHT, ACTIVITY LEVEL The sim will automatically calculate an approximate % fat based on height, weight, activity level, and sex. The range will be based on the "healthy" BMI (18.5-25) for a person with healthy body fat (Men 15%, Women 22%). >>>Model for calculating body fat: First calculate LBM: LBM_men = BMI_0 * height^2 / 1.15 LBM_women = BMI_0 * height^2 / 1.22 %_fat = (weight - LBM)/weight BMI_0 = 18.5 (very sedentary) 20.0 (sedentary) 22.5 (moderately active) 25.0 (active) % fat is updated whenever someone manually changes height, weight, activity level, and/or sex. Or maybe this should only happen on startup and when someone does "Reset All"?? Otherwise, these parameters are all independent. >>Old notes on this: How would a student know their % body fat to put in the simulation? NP – most people don’t know this. It is done either with a skin fold test (calipers), electric current (this is how these scales do it, not entirely accurate), or a submersion in water test (need a special facility). With skin fold, all the numbers are entered in a spreadsheet and it spits out a %. Electronic scales use weight, height, and conductivity (I don’t know the equation. I know it depends on how hydrated you are, and how long your legs are.) SR - Noah will come up with an equation for this based on other parameters >> SR - Problem: Fat % should be based on height and weight and gender, and update whenever height or weight or gender changes. Fat free mass depends on height & gender only, not on weight, thus BMR depends only on height and gender Noah P will provide these functions We could have the little question mark by the heart pop up a window with the list of risk factors for the heart from the American Heart Association. (smoking, etc) Description Women Men Essential fat 12–15% 2–5% Athletes 16–20% 6–13% Fitness 21–24% 14–17% Acceptable 25–31% 18–25% Obese 32%+ 25%+ source: http://en.wikipedia.org/wiki/Body_fat_percentage I found this page somewhat helpful in thinking about the above. http://weightloss.about.com/od/eatsmart/a/blcalintake.htm Franny said on 6/13/2008: I have been playing with the Simulation this morning and had a question for you. If in this simulation it appears that the change in percent body fat is at a high rate when making even the smallest change. I had a diet that I let run for a few years that was 50 cal above my energy expenditure and i watched my % fat go up by about 15% in a few years and then when I added a walking exercise (one of the lowest calorie expenditures) in a matter of 1 year I went from 30% body fat to about 10% body fat. I was wondering if there was a way to slow down the change in % body fat. i am not sure if it is equation based and has a time function built in, but that portion is not very realistic, so i thought i'd point it out. ____________________________________________________________________________________________ MUSCLE AND FAT MASS LOSS NP - weight lost should be both muscle and fat. There should be some ratio of FFM to Fat mass that is probably constant whenever someone loses weight. Article: Weight Loss-Induced Skeletal Muscle Loss: Accurate Estimation by Bioimpedance Analysis (BIA) http://www.halls.md/fat/muscle.htm Authors: Nuñez C, Kovera A, Wang ZM, Heymsfield SB. Presented at : 8th International Congress on Obesity, Paris, August 29-September 3. 1998. Abstract Objective: Loss of skeletal muscle (SM) is a characteristic body composition change associated with weight reduction treatment. Prevention of dieting-SM atrophy with appropriate food composition or exercise is an important research goal. Design: This study evaluated the accuracy of BIA (50kHz) in monitoring leg SM changes in a group of 71 obese women (X±SD, age, 40.0±7.3 yrs; BMI, 31.1±2.8kg/m2) undergoing 16-week weight loss treatment on conventional low calorie diet. Results: ***Subjects lost mean of 4.9 kg body mass, 4.3 kg as fat and 0.6 kg as fat-free mass.*** >>> So based on this, muscle loss is maybe 12% of total weight loss when dieting. Seems reasonable. >>> Model: when weight is lost due to lower calorie intake than expenditure, 12% of calories come from lean body mass, 88 % come from fat mass. Remove 4 g for every LBM calorie lost, 9 g for every fat calorie lost. So for example, if you take in 100 calories less than you burn, 12*4=48 g of LBM and 88*9=792 g of fat are lost. >>>Model for weight loss: LBM_lost (g) = 0.12 * 4 * Cal_burned Fat_lost (g) = 0.88 * 9 * Cal_burned >>>Starvation mode: When starving (<2%/4% for men/women) make the ratio switch, with buffering as follows: %fat 2-4% (men), 4-6% (women): LBM loss 50%, Fat loss 50% %fat < 2% (men), < 4% (women): LBM loss 95%, Fat loss 5%. (this might become 100% LBM...not yet sure). ____________________________________________________________________________________________ MUSCLE GAINED FROM EXERCISING - there should be some muscle gained from exercising, to a point. This should be limited (you can only get so muscular). I had trouble finding anything on this, so I made up a model that seemed reasonable. >>>Model: For every calorie exercised, all of the calories are burned. In addition, your body tries to make 10% of that number of calories into muscle, at a rate of 1/4 g of muscle per calorie. So for example, if you do 1000 calories of exercise, all of those calories are burned. In addition, 100 calories of what you eat are are turned into muscle (or 25 g). These calories are in addition to those burned to do the exercise itself. At this rate, one would build about 1 lb of muscle in 18 days, which seems reasonable. There is also feedback so that the more muscle you have, the less you can build, and at some point you cannot get any more muscular. >>>Model for muscle building from exercise: Muscle_mass_gained = 0.1 * Cal_exercise * (LBM_0 - LBM) ...where LBM is lean body mass, LBM_0 is 0.96 * sqrt(height / 30), or the lean body mass of a person with 4% body fat and a BMI of 30. >>>These two features above will effectively make BMR vary according to your caloric intake (if you undereat, your BMR should drop some to compensate, vise versa for eating more and exercising). >>> all other weight gained is added fat mass. ____________________________________________________________________________________________ STARVATION AND HEART ATTACK Starvation mode should be Gender specific: what are the threshould %fat values for male & female? NP - let's say 2% for men, 4% for women. >>>Model for starvation: If you drop below 2%/4% (men/women) fat, you can live for 2 months and then death occurs. If you go above this level, the clock resets. See MUSCLE AND FAT MASS LOSS section above for LBM and Fat loss when in starvation mode. (Basically you burn more muscle since you have so little fat.) >>>Model for heart attack: If you go above 25%/32% (men/women) fat, you begin to have a probability of heart attack each day, p_attack. Below these %fat thresholds, p_attack = 0. p_attack = p_0 * (%fat - %fat_0) ...where p_0 is a constant we adjust to make heart attack fairly likely (within a couple of years) for $fat > 50, %fat_0 = 25%/32% for men/women. ####################################################################################### ############## To Discuss ############## ####################################################################################### Add Eating / Exercise labels to each side of the screen? We removed the Edit Diet/Exercise buttons, which were partially serving as labels. Shouldn't be able to set more than 24 hours/day >>SR: Is this a real problem? If each icon is one hour, then the screen will be too overloaded at 24 to make any sense. I'm not sure this problem is worth addressing. Performance: consider profiling or changing clock rate >>How severe is this problem? Time readouts should be Years:Months, not years in decimal >>Wait for interviews Consider a 2-layer human graphic to show a "layer" of fat over a muscular interior How should health indicators be wired up to heart health? What are parameters for health indicators? Add popup-question mark button for BMI readout near scale >>Not a great place for this near the scale, should it go below the scale? Just appear when help is pressed? >>Ingrid said her students didn't use the near-the-heart-and-bmi version during class use. -The adjustable balanced diet works nicely, but I think the text box for calories is too small (on my screen anyway, it cuts off the edges of the number.) >>I tried to address this is 0.00.51. Is this resolved? -On the heart attack, the message also does not come up for this right away (above even 50% fat). >>NP My bad - I had p_0 set to 0. In that case, the warning did not come up. When I set p_0 to non-zero it worked (but there was a slight delay for %fat above 30%). >>SR: I can't reproduce this delay. Can you confirm the steps to repeat this problem? Prehaps the problem is just that thresholds are 25/32% (male/female) Need explanations for different activity levels from Noah