Obesity levels in adults are commonly attributed to a small prolonged discrepancy in energy intake and/or energy expenditure that results in a gradual yearly weight gain. In young adults this has been estimated at approximately 0.2 to 0.8 kg per year [4]. However, there is little available evidence to identify whether annual body weight increases are the resultant effect of the perceived continual daily discrepancy in energy balance or are due to more discrete periods of weight gain such as holiday periods. The holiday season is a time of the year suggested to present an increased risk of weight gain and obesity development [5]. This is postulated to be caused by stress associated with the holidays, increased caloric intake, and/or a decline in physical activity [6].

Participants visited the human body composition laboratory the week prior to the Thanksgiving holiday break and then returned 5 to 7 days following the holiday break. The days between the first and second visit were as follows: the minimum days between visits were 5 days, the maximum days between visits were 17 days and the mean and standard deviations for the group was 13 ± 3 days. During each visit, demographic data for each participant was obtained and body weight was measured using the Detecto Manual Physician scale while participants were dressed in light clothing (i.e. no shoes, sweaters, jackets, or belts) and height was assessed using a stadiometer (Accu-Hite Wall Stadiometer, Seca Corp., Hanover, MD) with shoes removed. A plastic flexible Gulick tape measure was used to assess waist and hip circumferences. The waist measurement was assessed at half the distance between the bottom of the xiphoid process and the umbilicus, and the hip measurement was taken at the largest anterior protrusion. The ratio between waist and hip was computed. All testing was completed with subjects fasted for 6 hours from food (i.e. no solid food or liquids) and physical activity.

There was a significant increase in body weight (BW) between visit 1 (pre-Thanksgiving) and visit 2 (post-Thanksgiving) for the entire group (72.1 kg vs. 72.6 kg; P < 0.05) and when the group was classified as male (80.2 kg vs. 80.8 kg; P < 0.05) and female (65.0 kg vs. 65.4 kg; P < 0.05) (Table ​(Table3).3). When participants were classified based on class standing (undergraduate vs. graduate) graduate students increased body weight between visits (76.3 kg vs. 77.1 kg; P < 0.05), while no significant increase was observed for undergraduates (70.3 kg vs. 70.7 kg; P = 0.07).

To attempt to assess changes in fat patterning, waist circumference, hip circumference and waist/hip ratio was measured. A significant (P < 0.05) decline in waist circumference was found for the group, males, females, undergraduates and normal BMI participants. Similar declines were also found in waist/hip ratio for males, females, undergraduates and normal BMI participants. A summary of these values are listed in Table ​Table22.

Additionally, participants were categorized based on BMI to determine if a BMI classification of normal (<25 kg/m²) vs. overweight/obese (≥ 25 kg/m²) impacted the change in body weight. No significant increase in body weight was found in normal BMI participants (64.7 kg vs. 64.9 kg; P = 0.27); however, overweight/obese participants increased body weight significantly (85.2 kg vs. 86.2 kg; P < 0.05). A summary of these results are listed in Table ​Table33.

Correlations were calculated to assess the impact of body weight (r = 0.17, P = 0.11) and BMI (r = 0.15, P = 0.16) from the pre-Thanksgiving visit on changes in body weight. No significant correlations were found (Figure ​(Figure1).1). Additionally, correlations were calculated to determine if baseline BMI was related to the weight change between visit 1 and visit 2. In females, a significant correlation (r = 0.42; P < 0.01) was found between baseline BMI and weight change between visits however no significant correlation was observed in males (r = -0.03; P = 0.43).

Although a great deal of publicity is given to holiday weight gain, few research studies have been done to examine weight changes during the holiday season and only one research study was performed in college students [4,10-12]. Two studies have examined body weight changes that occurred over the entire holiday season: pre-Thanksgiving to post-New Year's. Yanovski et al. studied 195 adults and found the holiday season resulted in a significant (P < 0.001) increase in body weight of 0.37 kg [4]. Baker et al. examined weight changes during holiday compared to non-holiday weeks, over the entire holiday season, in a group of participants enrolled in an obesity prevention program. Participants gained 500% more weight (specific increase in body weight not reported) during the holiday weeks versus the non-holiday weeks [12].

Other research has specifically focused on either Thanksgiving or Christmas during the holiday season. Reid and Hackett studied 26 participants over the Christmas holiday and found an increase in body weight of 0.93 kg though the results may be confounded by a small number of participants and five participants reporting being ill during data collection [10]. A second study examined the effect of the Thanksgiving weekend on body weight in college students by assessing dietary records before, during and after the holiday [11]. They found significantly increased dietary intake over the Thanksgiving holiday with males consuming more calories than females and non obese consumed more calories than obese.

The holiday season is a time when cultural and social influences combine to create a high risk environment conducive to weight gain. A number of factors particularly prevalent during the holiday celebrations that encourage over consumption include: longer eating durations, easy access to food, eating in the presence of others and increase portion sizes [13]. De Castro (1995) reported that meals eaten in the presence of others were 44% larger than when meals were eaten alone, while Rolls et al. (2002) and Levitsky 2002 demonstrated that increased portion size is associated with increased consumption [14-16]. Given these factors it is hardly surprising that Drapkin et al. (1995) reported than out of four hypothetical high risk situations participants perceived a family mealtime celebration (which includes a holiday meal) as the time they would be most likely to overeat [17]. Further, Boutelle (1999) reported that both their intervention and comparison groups found it difficult to effectively manage their weight during the Christmas to New Year holiday period [5].

Several limitations to the current study should be noted. First, since participants were derived from a volunteer convenience sample a self-selection bias may have occurred such that subjects had a personal interest in monitoring their weight. Second, participants were aware the primary aim of the study was to investigate changes in body weight during the Thanksgiving period. This may have resulted in participants intentionally or unintentionally altering their patterns of behavior, consequently masking true changes in their body weight during the measurement period. In addition, participants may have intentionally tried to reduce their body weight in the days preceding their second measurement. Third, no information is available on changes in body weight during the periods preceding or following the Thanksgiving holiday.