Holidays & weight gain: what the science suggests

Originally published on 12.22.2010. 

Many people have the perception that they are likely to gain 5 or 10 pounds during the holiday season (Thanksgiving to after New Year’s Day).  This myth has been propagated by media (4), partially explaining why it exists.  But since it is untrue, it does not mean it isn’t important; holiday weight gain still may be an important factor in the obesity epidemic.  After all, the average weight gain per year is only ~0.2 to 0.8 kg per year (4).  Holiday weight gain could contribute a large part of this gain.

Several weeks ago, I described 2 studies that looked at the Thanksgiving holiday itself on weight, 1 of the studies finding an average increase in weight over about 2 weeks of 0.5 kg.  Perhaps the most interesting finding is that when classified by weight at the start of the study (BMI), the overweight/obese group had gained on average 1.0 kg, whereas the “normal” weight group only gained 0.2 kg, suggesting that those already overweight may be at a higher risk for holiday weight gain, so, this is a good time to suggest precautions that can be taken.

First, let’s look at the other holiday research to see if it corroborates that on Thanksgiving, and if there are hints of differing effects in subgroups like in that study.  If you want to skip the details, i’ve provided a summary and crude spreadsheet with results from each study at the end.

Note: I focused on weight change from these studies, and noted if they measured other variables like cholesterol concentrations but omitted results on these because I don’t think the short term measurements of these are as important (and again not the focus).

The research appears to start in with the first paper published by Rees, Holman, and Turner in 1985 (1) who tracked weight and several metabolic parameters over the Christmas holiday.  It should be noted that the subjects were fully aware of the purpose of the study, so this may not represent normal populations.

22 (8 men, 14 women) normal subjects and 13 diabetics (5 men, 8 women) (4 of which were receiving treatment by diet counseling and 9 with sulphonylurea as part of another study) completed the study which consisted of reporting for weight measurements 5 times over ~2 month period: during the 1st week of December, 3-4 days before Christmas, and 3-4 days, 10-11 days, and 1 month after Christmas.

And who says research papers have to be dry and boring:

“Each subject was rewarded with a hot drink, a mince pie, and the convivial atmosphere reminiscent of air-raid shelters in the blitz. Any subject sufficiently foolhardy to attend on all five occasions had his or her name entered for a ”draw” with the promise of substantial prizes.”

On average, normal subjects gained 0.9 kg and diabetics 0.7 kg (7/11 or 64% gained over 1 kg; keep in mind this group was on average about 15 kg heavier at baseline).  One non-diabetic woman gained 4.3 kg, the most of any of the subjects.  Importantly, the weight gain was maintained until the last measurement.

However, degree of weight gain was not correlated to degree of obesity in the normal subjects, but did in diabetics (proportion of ideal weight to weight change).

The authors estimate an average consumption of 6160 additional kcal over the 5 days covering before, during, and after Christmas.

In 1992, Andersson and Rössner (2) published the results of a study on 46 obese patients who have lost weight, now in maintenance therapy compared to 76 controls.  The control subjects gained 0.4 kg with little variation in weight over Christmas (2-3 weeks) compared to 0.6 kg in the obese subjects who differed by over 16 kg in weight change.  This suggests that the holiday period causes different challenges to those trying to maintain or lose weight.

Next, we jump to 1999 when Reid and Hackett (3) published the results of a similar experiment as that of the first discussed stud by Rees et al.  In this one, 26 subjects (11 men, 15 women) completed the study and had their height, weight, skin folds and waist circumference, blood pressure, and total cholesterol measured on the 18th or 21st of December, then once again as close to the 4th of January as possible.  The average number of days between the 2 measurement sessions was 15.5 days.  Over this time, average weight gain was 0.93 kg.  The authors noted that this happened even though 5 subjects were ill and lost weight (excluding these it was 1.2 kg).  12 gained more than 1 kg, 4 more than 2 kg, and the maximum gain was 4.4 kg.  Biceps skinfold and waist circumference were significantly increased as well.  Based on the average weight gained, the authors estimated the average excess kcal consumption at 6500 kcal over the Christmas holiday, which is very similar to the results of the first study (1) at 6160 kcal.

In 2000, Yanovski et al. (4) published the results of a larger study on 195 subjects comparing weight gain over the holiday period to that during the rest of the year.  The subjects reported for measurements 4 times during September/October, mid-November, mid-November to January, and January to February/March (optionally, they could come in 2 additional times in June and September/October to complete a full year analysis).  Beside weight, pulse, blood pressure, and various questionnaires were given at each period.

The average weight gain for the holiday period (mid-November to January) was 0.37 kg, which was significant.  During the preholiday period (September/October to mid-November), weigh gain was not significant at 0.18 kg, and during the postholiday subjects on average lost 0.18 kg.  At the end of the 5 month period, subjects gained on average 0.48 kg total. Importantly, over 50% of the subjects did not gain or lose more than 1 kg during the study.  Less than 10% gained more than 2.3 kg over the holiday period.

The authors noted that 29 subjects (15%) tried to lose weight during the holiday period, and their average weight gain was 0.13 kg over the holiday period, compared to those who did not try to lose weight who gained 0.42 kg.

When subjects were classified according to BMI (at the beginning of the study), there was a clear pattern of risk of more weight gain during the holiday period as BMI increases. This again corroborates other results.

Weight change also correlated well with self-reported activity level and hunger:

During the last reporting period for the subjects in February/March, they were invited to come back for 2 more measurements in June and September/October.  Between this period, these 165 subjects only gained on average 0.21 kg, yielding a total of 0.62 kg during the entire year.  Keep in mind that this 0.21 kg gain in this ~7 months was less than the 0.37 kg gain over the ~2 month holiday period. A large amount of weight gain during the year can be explained by that gained during the holiday period, and this weight is not lost throughout the rest of the year.

In 2006, Hull, Nester, and Fields published a paper again on a college population (5).  Unlike the Thanksgiving study that I discussed in my previous post, this one measured weight change from 2 weeks before Thanksgiving to 1-3 weeks after New Year’s Day, at 3 time points.  They also assessed fat by using dual energy X-ray absorptiometry instead of only weight.  Of the 82 subjects completed the study, 31 gained weight, 32 lost weight, and 19 had no change (less than 0.5 kg change either way; unsure why they didn’t consider anything less clinically significant considering a 0.5 kg change is still a significant change considering the prospective study (4) just discussed).  12 of the subjects (15%) gained 2.0 kg or more.  Since they didn’t consider a change of more than 0.5 kg a gain or loss, on average, body weight did not significantly increase in this study.  As expected, correlations were observed between change in body weight and percent fat, total fat mass, total fat free mass, and trunk fat mass.  Of the 19 “weight stable” subjects, 17 of them had an increased fat mass, suggesting that (IMO) these subjects did gain weight.

The study also classified by BMI (< 25 = normal, > or = 25 = overweight/obese).  The normal weight actually on average lost weight but gained body fat %, while the overweight/obese category gained both weight and body fat %.  The body fat increase was observed in leg and trunk fat (my guess is more processed food consumption/glycemia etc), suggesting that even without a change in weight, the holiday season may result in an addition of fat in concerning areas.

In 2007, Costa, Moreira, and Teixeria published the results of another study on female college students (6).  They measured fat mass using bipedal bioimpedance and weight during 4 periods: preholiday (September/October), holiday (early January), postholiday (late February), and the next September/October for 1 year of measurement.  These periods are similar to study (4).  54 subjects finished the study, and the average weight gain between the preholiday and postholiday was 1.39 kg, and average fat mass increased 1.57 kg.  1 year from the start of the study, the subjects were on average 0.74 kg heavier and had 0.86 kg more fat mass.  These results again suggest that fat mass should be looked at independently of weight.

Seasonal (non-holiday-specific) Evidence

Other studies show seasonal differences in caloric and nutrient intake- here and here, for example.  This adds additional evidence that people on average increase energy consumption over holiday periods.

What can be done to prevent weight gain?

Not all hope is lost for the holidays and weight gain.  The key is planning and careful monitoring.

A study by Phelan et al. (7) looked at behavioral differences over the holiday period between people who have lost weight (and kept it off for a long period of time) and those with no history of obesity to see if there are differences in how they maintain their weights.  In short, the people who have lost weight and maintained are much more strict in their efforts over the holidays; they make more specific plans before the holiday to control food intake and exercise,  practice greater eating restraint, eat breakfast more, and exercise more.  However, they still seem more vulnerable to weight gain over the holidays (0.7 kg vs 0.2 kg), and less likely to lose it later on.

So beside the overweight and obese, the formerly overweight/obese who have successfully lose weight seem to be at an increased risk of weight gain over the holidays (as supported by ref. (2) and (7)).  It seems that most struggle with weight gain at least a little, but some much more than others, for reasons yet unknown.  I would speculate that some are simply more genetically resistant (we do know things like non-exercise thermogenesis can impact energy expenditure between people by as much as 1000% per day, and natural activity level would be somewhat influenced by genetics) against various factors that could influence weight gain at this time (e.g. variety of foods, constant food exposure, highly palatable and nutrient dense foods, cultural and social factors, low activity etc).

A couple studies (here and here) show that self-monitoring is key to minimizing weight gain (or even losing weight) for subjects who are on a long term weight loss program.

For the best results, I would suggest the following as paramount to minimizing holiday damage (keeping in mind that there isn’t a lot of research yet unequivocally confirming these things for the holiday period):

  • Plan meals and exercise routines more rigidly than you do during the rest of the year at least a couple of weeks before Thanksgiving
  • Carefully track calorie intake and weight
  • Set reasonable goals for both eating and exercise
  • Eat breakfast so you are less tempted by energy dense foods that inevitably become available
  • Keep protein intake high to maintain satiety
  • Depending on family dynamics (you’ll have to get creative), try to replace at least some of the normal sweets baking with other foods
  • Try to find others to share your efforts with

Please help me expand/alter this list in the comments!

A more dramatic change would be to consume meal replacements during the holiday period to strictly control intake.  In a poster session abstract, Silverstein et al. describe results of a study that found success in replacing 2 meals with liquid replacements.  But I wouldn’t suggest doing this unless under supervision of a professional and only bring it up out of interest for the research.

Summary/Conclusion

The studies I discussed use varying populations and collectively suggest that holiday weight gain is small (the majority of studies finding less than 1 kg on average, and the best data suggesting more likely even less than 0.5 kg or ~1 pound), but still may represent a large portion of average total yearly weight gain.  Because yearly weight gain is generally small and cumulative, holiday weight gain may represent a significant amount of of this gain.  In effect, holiday weight gain could be a significant contributer to current weight epidemics.

Unique challenges exist for those more susceptible to holiday weight gain; the research suggests that current overweight and obese subjects and formerly overweight/obese subjects are those at the greatest risk for weight gain or struggle most to prevent it.  Stringent efforts need to be in place prior to the start of the holidays for maximum success in weight gain avoidance; planning and monitoring seem likely to be the most effective measures to take during the holiday period.

Table of Key Study Results

Link to table

Study
Subjects Analyzed
Population
Subject Age Range
Time
Weight gain
Hull et. al. (previous post)94College~13 days0.5 kg (average), 1.0 kg (BMI >/= 25), 0.2 kg BMI < 25)
Rees et al. (1)35 (22 normal, 13 diabetic)UKUnknown~2 months0.9 kg (normal), 0.7 kg (diabetics)
Andersson & Rössner (2)46 obese, 76 controlsObese patients in maintenance therapy2-3 weeks0.6 kg obese, 0.4 kg controls
Reid et al. (3)26Staff in the School of Educational and Community Studies (UK)17-59Ave. 15.5 days0.93 kg (average)
Yanovski et al. (4)195National Institutes of Health campus; Bethesda, Maryland19-82 (mean 39)5 months (or 1 year)0.37 kg (average over holiday period), 0.48 kg (average total over 5 months), 0.62 kg (165 subjects who reported for 1 year), BMI correlated with weight gain
Hull et al. (5)82College~2 monthsIn BMI <25 no weight change but body fat % increase, BMI >/= 25 weight increase and body fat % increase
Costa et al. (6)54 femalesCollege5 months and 1 yearOver holiday period: 1.39 kg weight gain, 1.57 kg fat mass gain; over 1 year: 0.74 kg heavier, 0.86 kg fat mass
Phelan et al. (7)252National Weight Loss Control Registry0.7 kg for formerly overweight subjects who have successfully kept off weight, 0.2 kg for normal weight with no history of obesity

 

References

1. Rees, S., Holman, R., & Turner, R. (1985). The Christmas feast. BMJ, 291 (6511), 1764-1765 DOI: 10.1136/bmj.291.6511.1764

2. Andersson I, & Rössner S (1992). The Christmas factor in obesity therapy. International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity, 16 (12), 1013-5 PMID: 1335971

3. Reid, R., & Hackett, A. (1999). Changes in nutritional status in adults over Christmas 1998 Journal of Human Nutrition and Dietetics, 12 (6), 513-516 DOI: 10.1046/j.1365-277x.1999.00205.x

4. Yanovski, J., Yanovski, S., Sovik, K., Nguyen, T., O’Neil, P., & Sebring, N. (2000). A Prospective Study of Holiday Weight Gain New England Journal of Medicine, 342 (12), 861-867 DOI: 10.1056/NEJM200003233421206

(Note: Obesity Panacea also covered this study here)

5. Hull HR, Hester CN, & Fields DA (2006). The effect of the holiday season on body weight and composition in college students. Nutrition & metabolism, 3 PMID: 17192197

tr5mBOePHnY_Th-5Qem1_HA

6. Costa C, Moreira P, & Teixeira V (2007). HOLIDAY WEIGHT GAIN IN UNIVERSITY STUDENTS International Journal Of Obesity And Related Metabolic Disorders Link

7. Phelan S, Wing RR, Raynor HA, Dibello J, Nedeau K, & Peng W (2008). Holiday weight management by successful weight losers and normal weight individuals. Journal of consulting and clinical psychology, 76 (3), 442-8 PMID: 18540737