Successfully achieving target weight loss influences subsequent maintenance of lower weight and dropout from treatment
Disclosure: The authors declared no conflict of interest.
Author contributions: T. Yamada., K.H., and T.K. conceived the idea of the study and were responsible for the design of the study. T. Yamada and K.H. were responsible for undertaking the data analysis and produced the tables and figures. T. Yamada and K.H. provided input into the data analysis. The initial draft of the manuscript was prepared by T. Yamada and K.H. and then circulated repeatedly among all authors for critical revision. T. Yamada was responsible for acquisition of the data, and T. Yamada, K.H., N.S., A.K.S., J.H., M.I., T. Yamauchi, and T.K. contributed to interpretation of the results.
Abstract
Objectives
The influence of the amount and rate of weight loss on subsequently regaining weight and dropout from treatment in severely obese patients targeting 5% weight loss was investigated.
Methods
A total of 120 consecutive hospital patients with severe obesity (BMI: 42 ± 9 kg/m2) participated in an inpatient program targeting 5% weight loss that involved goal setting, charting weight four times daily, and diet and exercise. They were followed after discharge to assess subsequent regaining of weight and dropout.
Results
Mean weight loss was 4.9 ± 2.4% after a mean of 19 days in the hospital, and 43% of the patients achieved the target weight loss (>5%). Over the median 2-year follow-up period, greater than 5% in-hospital weight loss was associated with a significantly lower risk of regaining weight after adjustment for various factors (>5% to ≤7% loss: hazard ratio 0.30 [0.11-0.85] for regaining all of the lost weight and 0.32 [0.13-0.78] for regaining half of the lost weight). No significant relation between the amount or rate of weight loss and dropout from subsequent outpatient treatment was seen.
Conclusions
Successfully achieving the target weight loss in a comprehensive program predicts subsequent maintenance of lower weight without increasing the risk of dropout. Successful in-hospital weight loss might increase the motivation of obese patients.
Introduction
Obesity has rapidly become a worldwide epidemic (1) and it is a major risk factor for various disorders (2). Obesity and type 2 diabetes both increase the risk of cardiovascular disease and other comorbidities (3). A J-shaped relationship between body mass index (BMI) and average life expectancy was demonstrated by a study of 1.46 million white adults (4). Among non-smoking Caucasian males and females in the United States, the relative risk of death after 10 years was lowest for those with a BMI of 22.5-24.9 (reference hazard ratio [HR] of 1.0), but the risk increased along with BMI. Thus, the HR was 1.44 for both women and men with a BMI of 30-34.9 (obesity class I), 1.88 for women and 2.06 for men with a BMI of 35-39.9 (obesity class II), and 2.51 for women and 2.93 for men with a BMI of 40-49.9 (obesity class III). Therefore, it is important to treat severe obesity, which markedly increases the risk of death.
Even modest weight loss (3%-5% of body weight) can result in clinically meaningful benefits with respect to reducing triglycerides, blood glucose, and glycated hemoglobin, as well as decreasing the risk of type 2 diabetes, while greater weight loss (>5%) reduces blood pressure, further improves the lipid profile (both low-density and high-density lipoprotein cholesterol), and decreases the need for medications to control blood pressure, blood glucose, and lipids (5).
Although many obese people successfully lose weight by dieting and/or behavioral therapy (using their own methods or whatever is popular at the time), most of them subsequently regain the weight that they lost. Accordingly, new strategies are required to reduce weight and maintain weight loss. Bariatric surgery can be a highly effective option for adults with a BMI greater than 35 kg/m2 and type 2 diabetes, especially if diabetes or associated comorbidities are difficult to control by lifestyle modification and pharmacotherapy (3). However, such surgery has not become popular in Japan for reasons related to the technique and indications (6). According to the OECD report published in 2014, the rate of obesity (BMI > 30) among Japanese adults is only 3.6%, which is one quarter of that in Germany (14.7%) and one tenth of that in the United States (35.3%) (7).
A recent systematic review and meta-analysis (8) showed that behavioral interventions dealing with both diet and physical activity provide a small, but significant, benefit for the maintenance of weight loss. However, little is known about the optimal amount and rate of weight loss for preventing subsequent weight gain and dropout from weight loss programs in patients with obesity and diabetes, although a few studies have shown that initial achievement of weight loss is related to better adherence and success with maintaining a lower weight (9, 10). Initiation or intensification of insulin or sulfonylurea therapy commonly results in weight gain in patients with type 1 or type 2 diabetes (11, 12). In addition, it was recently reported that antidepressant therapy increases the risk of regaining weight in overweight individuals with impaired glucose tolerance (13).
Clarifying the relationship between weight loss and regaining weight or dropout from weight loss programs is useful to explore more effective new strategies for obesity. In the present study, we assessed the influence of inpatient weight loss on the risk of subsequently regaining weight and dropout from treatment among patients with severe obesity who completed a comprehensive weight loss education program and did not receive bariatric surgery.
Methods
Subjects
First, we screened 2,178 consecutive patients who were admitted to the University of Tokyo Hospital between 2009 and 2012. Among them, we selected patients with severe obesity who had a BMI ≥35 kg/m2 and who participated in a comprehensive weight loss program. We excluded patients with severe cardiovascular disease, heart failure, infectious diseases, stroke, or peripheral artery disease, as well as patients with type 1 diabetes, pregnant women, patients with dementia, patients who had orthopedic diseases that could interfere with exercise (walking), perioperative patients, patients taking anti-obesity medications, patients who had undergone bariatric surgery, patients without pertinent data, patients who were transferred to another hospital immediately after discharge, patients who were readmitted, and those under 18 or over 80 years old. The remaining patients were followed after discharge to assess subsequent weight gain and dropout.
Comprehensive inpatient obesity treatment program targeting 5% weight loss
The details of this program have been reported previously (14). Briefly, it consisted of three main components, which were behavioral modification (goal setting and charting weight four times daily), diet, and exercise (patients with diabetes received appropriate anti-diabetic therapy together with this weight loss program, and the treatment of diabetes was decided by each attending physician).
Behavioral modification (goal setting and charting weight four times daily)
At admission, all patients were given the goal of achieving 5% weight loss while in the hospital. They also weighed themselves four times daily (immediately after waking, immediately after breakfast, immediately after dinner, and immediately before going to bed) and charting the data on a weekly graph. Daily charting of weight fluctuations reveals irregular intake of food and fluid that reflects dysfunctional eating patterns and other behavioral abnormalities, and assists in achieving weight loss (15, 16). Patients were recommended to continue charting their weight after discharge. Patients were weighed by using AD-6107NW™ scales (A and D Co. Ltd., Tokyo, Japan).
Diet
A balanced low-calorie diet (20-24 kcal/day/kg of ideal body weight, e.g., the diet for a person with a height of 1.7 m would be calculated for the ideal body weight based on a BMI of 22 as (1.7 × 1.7 × 22 = 63.58 kg) × 20 = 1,271 kcal or ×24 = 1,525 kcal) was provided to the patients in the hospital, consisting of 50%-60% carbohydrate, 20% protein, and 20%-30% fat. Hospital dietitians used food samples and a food exchange table (17) to educate patients about nutritional guidelines. The dietician initially gave each patient information for 1 hour, with subsequent 30-minutes sessions being held at least twice a week until discharge.
Exercise
All patients were given a pedometer and were instructed to walk more than 10,000 steps/daily (a distance of 5-7 km) over approximately 1.5 hours. Using a map of the University of Tokyo Hospital (site area: 540,000 m2), each patient marked out a route around the hospital compound and walked it as part of their regular exercise routine. The exercise program was tailored to accommodate health problems (e.g., morbid obesity, hypoglycemia, joint pain, or diabetic retinopathy) and specific needs (e.g., exercise by walking or training on a bicycle ergometer). The target pulse rate and schedule for each exercise session were set.
Outcome measures
Patients attended our hospital outpatient department every 2 months after discharge to continue their weight loss program and for treatment of other diseases (diabetes, dyslipidemia, and hypertension). Body weight was measured at each visit. The outcomes of this study were regaining weight and dropout from the weight loss program after discharge. Regaining all of the lost weight (100% regain) was defined as returning to the baseline body weight at the time of admission. Regaining half of the lost weight (50% regain) was defined as returning to a body weight midway between that at admission and that at discharge, while 25% and 75% regain were defined similarly. Dropout from the program was defined as missing outpatient appointments. (If the patient presented again within 6 months of the specified appointment, this was not defined as dropout.) Patients were defined as having diabetes if the medical record listed a diagnosis of type 2 diabetes and they were using oral hypoglycemic agents and/or insulin. If a 75-g oral glucose tolerance test was performed, a diagnosis of diabetes, impaired glucose tolerance, or impaired fasting glucose was made according to the American Diabetes Association criteria (18).
Antidepressant medications were classified as selective serotonin reuptake inhibitors, serotonin and norepinephrine reuptake inhibitors, tricyclic antidepressants, tetracyclic antidepressants, serotonin receptor antagonists and reuptake inhibitors, monoamine oxidase inhibitors, and noradrenergic and specific serotonergic antidepressants. All demographic and clinical data were collected from secure electric medical records. The accuracy of the body weight measurements of each subject was confirmed by nurses or physicians.
Statistical analysis
Continuous variables are presented as the mean ± SD. The chi-square test, t-test, or ANOVA was used to assess differences among weight loss categories (≤3% loss; >3% to ≤5% loss; >5% to ≤7% loss; >7% loss), quartiles of weight loss achieved (%) during hospitalization, or the rapidity of weight loss after admission (rapid loss was defined as >3% loss within 7 days after admission and non-rapid loss was ≤3% loss after 7 days) among patients who achieved greater than 3% weight loss during hospitalization. Multivariate regression analysis using the Cox proportional hazards model was employed to determine the adjusted HR and 95% confidence interval (CI) for regaining weight, dropout in relation to the duration of hospitalization, age, sex, BMI, smoking, type 2 diabetes, insulin therapy, sulfonylurea therapy, antidepressant medication, and hospital diet (calories per day). We used STATA statistical software version 12.0 for all analyses. All P values were two-sided and significance was accepted at P < 0.05. This retrospective cohort study complied with the Declaration of Helsinki and it was approved by the Research Ethics Committee of the University of Tokyo Hospital.
Results
Patient selection
Figure 1 is a flow chart showing the disposition of the patients. A total of 2,178 consecutive adults with severe obesity were admitted to our hospital between January 2009 and December 2012. Among them, 2,010 were excluded because their BMI was less than 35 kg/m2. Among the remaining 168 patients, 48 were excluded (21 because of readmission to our department to treat diabetes and obesity, 11 due to emergency admission, 11 due to transfer to another hospital immediately after discharge, and 5 for lack of pertinent data). The remaining 120 patients were included in this study.
Flow chart of patient disposition.
Baseline characteristics
Table 1 shows the baseline characteristics of the subjects. They had a mean age of 48 ± 14 years, a mean BMI of 41.7 ± 8.8 kg/m2, and 58% were women. In addition, 74% had diabetes, 21% had impaired glucose tolerance and/or impaired fasting glucose, and the mean HbA1c was 7.6 ± 2.2%. After a mean of 19 days (19 ± 10) in the hospital, the mean weight loss was 4.9 ± 2.4% (5.3 ± 3.0 kg) and 43% of the patients achieved the target weight loss of greater than 5%. There were no significant differences of these parameters among groups stratified by the extent of weight loss (≤3% loss; >3% to ≤5% loss; >5% to ≤7% loss; >7% loss) (Table 1). There were also no significant differences among patients stratified into quartiles based on weight loss achieved (%) during hospitalization (Table S1). Furthermore, there were no differences between two groups based on the rapidity of weight loss after admission (>3% loss within 7 days vs. ≤3% loss) (Table 2).
| Extent of weight loss | ||||||
|---|---|---|---|---|---|---|
| Variable | All subjects | ≤3% loss | >3% to ≤5% loss | >5% to ≤7% loss | >7% loss | P value |
| N | 120 | 20 (17) | 48 (40) | 35 (29) | 17 (14) | |
| Sex (female) | 69 (58) | 12 (60) | 24 (50) | 23 (66) | 10 (59) | 0.55 |
| Age (years) | 48 ± 14 | 46 ± 18 | 48 ± 13 | 50 ± 13 | 47 ± 12 | 0.76 |
| Body weight (kg) | 109 ± 22 | 103 ± 25 | 110 ± 23 | 107 ± 19 | 114 ± 22 | 0.42 |
| Height (cm) | 162 ± 10 | 162 ± 11 | 163 ± 10 | 160 ± 9 | 163 ± 8 | 0.65 |
| BMI (kg/m2) | 42 ± 9 | 39 ± 6 | 41 ± 6 | 44 ± 13 | 43 ± 7 | 0.19 |
| Waist circumference (cm) | 125 ± 15 | 121 ± 15 | 125 ± 15 | 124 ± 13 | 130 ± 18 | 0.28 |
| Current smoker | 49 (41) | 9 (45) | 20 (42) | 12 (34) | 8 (47) | 0.79 |
| Type 2 diabetes | 89 (74) | 17 (85) | 38 (79) | 23 (66) | 11 (65) | 0.27 |
| Insulin use | 25 (21) | 7 (35) | 9 (19) | 7 (20) | 2 (12) | 0.33 |
| Sulfonylurea use | 9 (8) | 2 (10) | 3 (6) | 2 (6) | 2 (12) | 0.83 |
| Hospital diet (kcal/day/kg of ideal body weight) | 22.6 (2.4) | 22.8 (2.3) | 22.8 (2.8) | 22.7 (1.9) | 21.8 (1.9) | 0.50 |
| Antidepressant use | 18 (15) | 2 (10) | 6 (13) | 7 (20) | 3 (18) | 0.70 |
- Data are the mean ± SD or number (%). BMI, body mass index.
| Rapidity of weight loss | ||||
|---|---|---|---|---|
| Variable | All | Non-rapid loss (≤3% loss within 7 days) | Rapid loss (>3% loss within 7 days) | P value |
| N | 100 | 39 (39) | 69 (69) | |
| Sex (female) | 57 (57) | 22 (56) | 35 (57) | 0.92 |
| Age (years) | 49 ± 13 | 49 ± 13 | 49 ± 14 | 0.98 |
| Body weight (kg) | 110 ± 22 | 109 ± 21 | 111 ± 22 | 0.68 |
| Height (cm) | 162 ± 9 | 161 ± 9 | 163 ± 10 | 0.40 |
| BMI (kg/m2) | 43 ± 9 | 44 ± 12 | 42 ± 6 | 0.23 |
| Waist circumference (cm) | 125 ± 15 | 124 ± 16 | 126 ± 14 | 0.44 |
| Current smoker | 40 (40) | 15 (38) | 25 (41) | 0.80 |
| Type 2 diabetes | 72 (72) | 29 (74) | 43 (70) | 0.67 |
| Insulin use | 18 (18) | 6 (15) | 12 (20) | 0.59 |
| Sulfonylurea use | 7 (7) | 5 (13) | 2 (3) | 0.07 |
| Hospital diet (kcal/day/kg of ideal body weight) | 22.6 (2.4) | 22.6 (2.4) | 22.6 (2.4) | 0.93 |
| Antidepressant use | 16 (16) | 7 (18) | 9 (15) | 0.67 |
- Data are the mean ± SD or number (%). BMI, body mass index.
- Rapid loss group: patients who achieved >3% weight loss within 7 days after admission among patients with >3% weight loss during hospitalization. Non-rapid loss group: patients who achieved ≤3% weight loss within 7 days after admission.
Risk of weight regain and dropout from treatment after discharge in relation to percent weight loss during hospitalization
Over the median follow-up period of 1.8 years (1.8 ± 1.1 years) after discharge, 32 of the 120 patients (27%) regained all of the weight they had lost (100% regain), 54 patients (45%) regained half of the weight they had lost (50% regain), and 32 patients (27%) dropped out of the program (Table 3). Multivariate regression analysis using the Cox proportional hazards model showed that greater than 5% weight loss was associated with a significantly lower risk of regaining all of the lost weight (100% regain) compared with ≤3% weight loss after adjustment for the duration of hospitalization, age, sex, BMI, smoking, diabetes, use of insulin, sulfonylureas, and antidepressants, and hospital diet (calories) (>7% loss: HR 0.04 [0.004-0.37, P = 0.005]; 5% to ≤7% loss: HR 0.3 [0.11-0.85, P = 0.02]). In addition, greater than 3% weight loss was associated with a significantly lower risk of regaining half of the lost weight (50% regain) (>7% loss: HR 0.18 [0.05-0.59, P = 0.005]; 5% to ≤7% loss: HR 0.32 [0.13-0.78, P = 0.01]; 3% to ≤5% loss: HR 0.41 [0.19-0.92, P = 0.03]). Analysis of the risk of 25% or 75% regain yielded similar results. There was no significant relation between the extent of weight loss and dropout from treatment (Table 3, Figure 2, Figure S1).
| Regaining all of the weight lost (100% regain) | Regaining three quarters of the weight lost (75% regain) | Regaining half of the weight lost (50% regain) | Regaining one quarter of the weight lost (25% regain) | Dropout from outpatient treatment | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No of event/ | 32/120 (27%) | 45/120 (38%) | 54/120 (45%) | 58/120 (48%) | 32/120 (27%) | |||||
| N Variable | Multivariate | P | Multivariate | P | Multivariate | P | Multivariate | P | Multivariate | P |
| Sex (female) | 1.51 (0.61-3.78) | 0.38 | 0.75 (0.35-1.60) | 0.45 | 0.99 (0.49-1.98) | 0.97 | 0.93 (0.48-1.81) | 0.83 | 1.16 (0.51-2.62) | 0.73 |
| Age (years) | 0.98 (0.95-1.00) | 0.09 | 0.99 (0.97-1.02) | 0.64 | 1.00 (0.97-1.02) | 0.82 | 1.00 (0.97-1.02) | 0.74 | 0.97 (0.94-1.00) | 0.07 |
| BMI (kg/m2) | 0.98 (0.92-1.04) | 0.46 | 0.99 (0.95-1.04) | 0.78 | 1.01 (0.97-1.05) | 0.75 | 0.99 (0.95-1.04) | 0.68 | 1.00 0.94-1.05) | 0.89 |
| Current smoker | 1.41 (0.58-3.42) | 0.44 | 1.15 (0.56-2.35) | 0.70 | 0.96 (0.50-1.84) | 0.90 | 1.02 (0.55-1.89) | 0.94 | 0.84 (0.38-1.85) | 0.67 |
| Type 2 diabetes | 0.82 (0.30-2.23) | 0.69 | 1.37 (0.57-3.28) | 0.48 | 1.36 (0.59-3.12) | 0.48 | 1.34 (0.61-2.96) | 0.47 | 0.44 (0.19-1.06) | 0.07 |
| Duration of hospital stay (days) | 1.03 (0.98-1.08) | 0.26 | 1.04 (1.00-1.08) | 0.05 | 1.03 (0.99-1.06) | 0.11 | 1.02 (0.99-1.06) | 0.15 | 1.01 (0.97-1.06) | 0.59 |
| Insulin use | 0.94 (0.34-2.57) | 0.90 | 0.80 (0.34-1.86) | 0.60 | 1.11 (0.51-2.42) | 0.79 | 1.26 (0.61-2.58) | 0.53 | 0.76 (0.22-2.59) | 0.66 |
| Sulfonylurea use | 1.92 (0.36-10.1) | 0.44 | 1.04 (0.27-3.97) | 0.96 | 1.94 (0.64-5.92) | 0.24 | 1.9 (0.65-5.61) | 0.24 | 0.77 (0.09-6.50) | 0.81 |
| Antidepressant use | 0.91 (0.29-2.83) | 0.87 | 1.33 (0.50-3.51) | 0.57 | 1.73 (0.73-4.12) | 0.22 | 2.23 (1.03-4.82) | 0.04 | 1.21 (0.46-3.23) | 0.70 |
| Hospital diet (kcal/day/kg of ideal body weight) | 1.04 (0.87-1.24) | 0.67 | 1.09 (0.94-1.27) | 0.25 | 1.1 (0.97-1.25) | 0.14 | 1.07 (0.95-1.21) | 0.28 | 1.03 (0.89-1.19) | 0.71 |
| Extent of weight loss | ||||||||||
| ≤3% | 1.00 | ― | 1.00 | ― | 1.00 | ― | 1.00 | ― | 1.00 | ― |
| >3% to ≤5% | 0.45 (0.18-1.15) | 0.09 | 0.36 (0.15-0.85) | 0.02 | 0.41 (0.19-0.92) | 0.03 | 0.47 (0.22-1.01) | 0.05 | 1.36 (0.48-3.85) | 0.56 |
| >5% to ≤7% | 0.30 (0.11-0.85) | 0.02 | 0.38 (0.15-0.96) | 0.04 | 0.32 (0.13-0.78) | 0.01 | 0.37 (0.15-0.89) | 0.03 | 0.51 (0.15-1.7) | 0.28 |
| >7% | 0.04 (0.004-0.37) | 0.005 | 0.08 (0.02-0.4) | 0.002 | 0.18 (0.05-0.59) | 0.005 | 0.29 (0.09-0.92) | 0.04 | 0.51 (0.12-2.18) | 0.36 |
- Data are the mean ± SD or number (%). BMI, body mass index. Cox proportional hazards analysis was employed to identify parameters significantly associated with weight gain and dropout from treatment after discharge using the following factors: duration of hospitalization, sex, age, BMI, current smoking, type 2 diabetes, insulin use, sulfonylurea use, antidepressant use, hospital diet (kcal/day/kg of ideal body weight), and extent of weight loss during hospitalization (≤3%, >3% to ≤5%, >5% to ≤7%, and >7%).
- Bold values represents a statistical significant hazard ratio and p value (p < 0.05).
Kaplan–Meier curves for (A) regaining all of the weight lost (100% regain), (B) regaining half of the weight lost (50% regain), and (C) dropout from treatment after discharge in patients with severe obesity stratified by weight loss (%) during hospitalization. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Risk of weight regain and dropout from treatment after discharge in relation to weight loss quartiles
We also performed an analysis of the subjects stratified by quartiles of weight loss achieved during hospitalization (Table S2, Figure 3, Figure S2). The highest weight loss quartile (quartile 4) had a significantly lower risk of regaining all of the weight (100% regain) compared with the lowest quartile (quartile 1) after adjustment for various covariates (HR 0.14 [0.04-0.54], P = 0.004 for 100% regain; HR 0.22 [0.08-0.65], P = 0.006 for 75% regain; HR 0.27 [0.1-0.69], P = 0.007 for 50% regain; and HR 0.37 [0.15-0.93], P = 0.03 for 25% regain).
Kaplan–Meier curves for (A) regaining all of the weight lost (100% regain), (B) regaining half of the weight lost (50% regain), and (C) dropout from treatment after discharge in patients with severe obesity stratified by quartiles of achieved weight loss (%) during hospitalization. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Rapidity of weight loss versus the risk of weight regain and dropout
Among patients who achieved greater than 3% weight loss during hospitalization (N = 100), adjusted analysis showed that rapid weight loss (>3% within 7 days after admission) did not significantly increase the risk of subsequently regaining weight compared with patients who showed slower weight loss (≤3% within 7 days) (HR 1.1 [0.43-2.85], P = 0.84 for regaining all of the lost weight [100% regained], HR 1.11 [0.55-2.25], P = 0.77 for regaining half of the lost weight [50% regained]) (Table 4, Figure 4).
| Regaining all of the weight lost (100% regain) | Regaining three quarters of the weight lost (75% regain) | Regaining half of the weight lost (50% regain) | Regaining one quarter of the weight lost (25% regain) | Dropout from outpatient treatment | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No. of event/ | 21/100 (21%) | 33/100 (33%) | 40/100 (43%) | 43/100 (43%) | 26/100 (26%) | |||||
| N Variable | Multivariate | P | Multivariate | P | Multivariate | P | Multivariate | P | Multivariate | P |
| Sex (female) | 2.60 (0.82-8.29) | 0.11 | 0.87 (0.36-2.10) | 0.76 | 1.06 (0.47-2.37) | 0.89 | 1.09 (0.50-2.35) | 0.83 | 1.36 (0.58-3.15) | 0.48 |
| Age (years) | 0.96 (0.92-1.00) | 0.08 | 0.98 (0.95-1.02) | 0.32 | 0.98 (0.95-1.01) | 0.30 | 0.98 (0.95-1.01) | 0.29 | 0.98 (0.95-1.02) | 0.28 |
| BMI (kg/m2) | 1.00 (0.94-1.08) | 0.90 | 1.01 (0.95-1.07) | 0.80 | 1.01 (0.96-1.07) | 0.60 | 0.99 (0.94-1.05) | 0.72 | 1.00 (0.94-1.06) | 0.98 |
| Current smoker | 1.43 (0.55-3.69) | 0.47 | 1.15 (0.53-2.53) | 0.72 | 1.01 (0.49-2.08) | 0.98 | 1.14 (0.58-2.21) | 0.71 | 1.08 (0.46-2.56) | 0.86 |
| Type 2 diabetes | 1.27 (0.39-4.17) | 0.69 | 2.08 (0.76-5.66) | 0.15 | 1.96 (0.77-4.97) | 0.16 | 1.71 (0.72-4.09) | 0.22 | 0.44 (0.18-1.06) | 0.07 |
| Duration of hospital stay (days) | 0.99 (0.94-1.03) | 0.59 | 1.01 (0.97-1.04) | 0.70 | 1.01 (0.96-1.04) | 0.66 | 1.01 (0.98-1.04) | 0.50 | 0.98 (0.93-1.03) | 0.42 |
| Insulin use | 0.52 (0.12-2.21) | 0.38 | 0.53 (0.17-1.66) | 0.28 | 0.87 (0.33-2.28) | 0.78 | 1.03 (0.42-2.51) | 0.94 | 0.24 (0.03-1.95) | 0.18 |
| Sulfonylurea use | 2.34 (0.36-15.4) | 0.38 | 1.22 (0.28-5.36) | 0.79 | 1.90 (0.51-7.07) | 0.34 | 1.84 (0.50-6.76) | 0.36 | 0.86 (0.10-7.62) | 0.89 |
| Antidepressant use | 0.82 (0.24-2.78) | 0.75 | 1.45 (0.52-4.03) | 0.48 | 1.59 (0.62-4.09) | 0.33 | 2.24 (0.98-5.12) | 0.06 | 1.16 (0.41-3.28) | 0.78 |
| Hospital diet (kcal/day/kg of ideal body weight) | 1.19 (0.98-1.44) | 0.07 | 1.17 (1.00-1.37) | 0.05 | 1.16 (1.01-1.33) | 0.04 | 1.14 (0.99-1.31) | 0.07 | 0.96 (0.80-1.15) | 0.65 |
| Rapidity of weight loss | ||||||||||
| Non-rapid loss (≤3% loss within 7 days) | 1.00 | - | 1.00 | - | 1.00 | - | 1.00 | - | 1.00 | - |
| Rapid loss (>3% loss within 7 days) | 1.11 (0.43-2.85) | 0.84 | 1.08 (0.50-2.33) | 0.85 | 1.11 (0.55-2.25) | 0.77 | 1.26 (0.64-2.48) | 0.51 | 1.21 (0.51-2.85) | 0.67 |
- Data are the mean ± SD or number (%). BMI, body mass index. Cox proportional hazards analysis was employed to identify parameters significantly associated with weight gain and dropout from treatment after discharge using the following factors: duration of hospitalization, sex, age, BMI, current smoking, type 2 diabetes, insulin use, sulfonylurea use, antidepressant use, hospital diet (kcal/day/kg of ideal body weight), and the rapidity of weight loss after admission. Rapid loss group: patients who achieved >3% weight loss within 7 days of admission among patients with >3% weight loss during hospitalization. Non-rapid loss group: patients who achieved ≤3% loss within 7 days of admission.
Kaplan–Meier curves for (A) regaining all of the weight lost (100% regain), (B) regaining half of the weight lost (50% regain), and (C) dropout from treatment after discharge in patients with severe obesity stratified by the rapidity of weight loss after admission. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Discussion
The present study indicated that achieving greater than 5% initial weight loss (the target) during comparatively brief hospitalization was an independent predictor of subsequent maintenance of lower weight by severely obese patients who completed a comprehensive program that involved behavioral modification (charting weight four times daily), even after adjustment for other covariates such as BMI and baseline drug therapy, without increasing the risk of dropout from treatment. Thus, a short-term intensive inpatient program targeting 5% weight loss may be clinically useful for achieving long-term improvement of weight. The 5% target we employed is consistent with the recent AHA/ACC/TOS Guidelines for Managing Overweight and Obesity in Adults (5), which states that greater than 5% weight loss can reduce blood pressure, improve the lipid profile and reduce the use of medications for hypertension, hyperglycemia, and hyperlipidemia in obese adults with cardiovascular risk factors. In addition, even modest weight loss (3%-5% of body weight) is beneficial with regard to improving the levels of triglycerides, blood glucose, and glycated hemoglobin, as well as reducing the risk of type 2 diabetes. Moreover, we showed that rapid weight loss did not significantly increase the risk of subsequently regaining weight and there was no significant relation between the pattern of weight loss (both its extent and rapidity) and dropout from treatment after adjustment for several covariates.
There were two novel elements of our weight loss program. First, the participants were each set a precise weight loss goal (5%) at the time of admission to motivate them. Second, we used charting weight four times daily as a behavioral modification technique (15, 16). The goal setting theory was developed by Locke et al. (19). Goals are a form of motivation in which standards for self-satisfaction with performance are specified. Setting goals can influence the outcome in four ways. (a) Choice: goals focus attention and direct efforts to goal-related activities and away from perceived undesirable and irrelevant actions. (b) Effort: goals can lead to more effort. (c) Persistence: people are more likely to persevere despite setbacks if they are pursuing a goal. (d) Cognition: goals can lead individuals to change their behavior. Frequent self-weighing has previously been suggested to improve self-regulation of behavior (20). It probably assists in maintaining weight loss because frequent charting of weight fluctuations reveals irregular dietary habits that reflect other behavioral abnormalities. Once these irregularities become apparent to patients, they tend to improve their eating habits and/or lifestyle, which assists in progress with weight loss and therefore can help them to reach an important therapeutic milestone. In the general population, Linde et al. (21) found that participants who performed daily self-weighing lost significantly more weight than participants weighed themselves weekly, monthly, bi-monthly, or never. However, published evidence about the effect of very frequent weighing (four times per day) has been limited. Although Raynor et al. (22) reported that self-weighing more than once per week was related to a lower BMI, this has not been previously demonstrated for self-weighing more than once daily, as was done in the present study.
Our findings support those reported by several other authors. Nackers et al. (9) showed that initial weight loss was related to adherence and success with a 6-month weight loss program in subjects without diabetes, while Hemmingsson et al. (10) reported that a high baseline BMI and rapid initial weight loss were both independently associated with greater weight loss after 1 year. However, the following differences between our study and these two studies (9, 10) should be noted. (a) We employed an inpatient program with comparatively brief hospitalization (mean: 19 days) while the other studies were based on outpatient programs over 6 months. (b) Our patients had more severe obesity (mean: BMI 41.7 kg/m2) compared with patients in the other studies [mean BMI: 36 kg/m2 (9) and 30 kg/m2 (10)].
The present study had several limitations. First, it was performed at a single center and assessed a small patient population, which could limit the generalizability of our findings. Although analyses were adjusted for many covariates that were expected to be confounding factors, the small sample might have resulted in a lack of significant differences in our analyses. Therefore, it will be necessary to perform investigations on a larger scale (including well-designed randomized controlled trials) in the future.
In addition, we used the percent weight loss instead of actual weight loss (in kg) for our analyses. This was done because we considered that evaluation based on actual weight loss is inappropriate when comparing individuals with different constitutions (e.g., losing 3 kg has very different implications for a person who weighs 60 kg compared with a person who weighs 120 kg. Moreover, the effect of weight loss is shown as percentages in the AHA/ACC/TOS Guidelines (5). However, there is still controversy as to whether actual weight loss (in kg) or the percent loss should be evaluated.
In conclusion, when severely obese patients completed a comprehensive inpatient treatment program targeting 5% weight loss, successful weight loss during comparatively brief hospitalization predicted subsequent maintenance of lower weight without increasing the risk of dropout, and rapid weight loss did not increase the risk of subsequent weight gain. Successful in-hospital weight loss might increase the motivation of obese patients. Thus, a short-term intensive inpatient program targeting 5% weight loss can be clinically useful for long-term improvement of weight. However, we could not fully monitor adherence to our program after discharge from the hospital, so further studies will be required to confirm the efficacy of this educational approach.
