The standards for success of treatment of obesity listed here are broadly based, arbitrary criteria derived from the current technology for obesity treatment (Table 1). Federal health care and regulatory agencies, insurance companies, and other organizations may find these standards useful, but it should be remembered that these definitions are not meant to be rigid. They must be put into perspective for the type of treatment used and for individual differences, and they must be changed continually as improvements in the diagnosis and treatment of obesity occur.

Standards for Successful Weight Reduction

Measures of Body Weight or Body Composition

Body Weight

Because body weight is the most easily measured variable, alterations in weight will be the criterion used by most patients and even by most health professionals in judging treatment success. The simplest measure is to state weight lost in pounds or kilograms; however, a similar loss in weight among persons of widely differing heights has far different implications for the reduction of obesity. Thus, weight loss alone should be avoided as a standard for success.

Percentage of Excess Body Weight

Tables of "ideal," "desirable," or population-referenced weight are available and provide a measure by which "excess body weight" may be calculated. If "ideal" or "desirable" weight is subtracted from current weight and if the difference is divided by the reference weight, a "percentage of excess body weight" can be calculated. The most widely used weight tables are those assembled by insurance companies. These tables are published by the Society of Actuaries and assess the weight of policy holders at which the least number of deaths occur [1, 2]. There is a J-shaped curve of body weights, with a modest increase in deaths at low weights, and a steady increase with increasing obesity. Using these tables, investigators have defined obesity as a weight that is 20% greater than "ideal". Because these tables reflect a selected population of mainly white, middle-class Americans who buy life insurance, they may not yield an accurate assessment of other populations. The National Health and Nutrition Examination Surveys (NHANES) are large-scale screening assessments of health status of Americans [3]. From these studies, population-based tables have been developed. The 85th percentile is a cut point defining the upper limit of "desirable" weight. Because these tables are not based on health experience, they also may not be the most accurate reflection of optimal weight. Evidence suggests that Americans are fatter than is healthy and are getting fatter [4]; thus, the 85th percentile of a fat population moves higher with each survey. Although other similar tables are in use, these are the two most widely used.

Achieving "ideal" weight is difficult and is probably unnecessary. Obese persons may develop an increased lean body mass and skeletal mass that may not revert completely to that of a continuously lean person when large amounts of weight are lost. Also, many metabolic, behavioral, and life-style variables may make it difficult formerly obese persons to maintain "ideal weight". Kanders and Blackburn [5] have shown that modest weight loss (about 10%) may reduce complications associated with obesity.

Body Mass Index

As noted above, changes in absolute body weight and percentage of excess weight have flaws as criteria for success in the treatment of obesity. The 1985 Consensus Development Conference on Obesity suggested that body mass index (BMI) is a desirable measure of the tendency toward obesity [6]. The BMI is calculated by dividing the body weight measured in kilograms by the square of the height measured in meters (BMI = kg/m²). The BMI is easy to determine, correlates fairly well with body fat, and is therefore thought to be the most desirable clinical measure of obesity short of measurement of body composition. When body composition is not measured, persons with an excess lean body mass, such as professional athletes, may have a high BMI and may be falsely classified as obese.

The 1985 Consensus Development Conference on Obesity concluded that, in young persons, a BMI greater than 27 increases risk for morbidity and death. The lowest mortality rates were associated with a BMI of about 20 to 24.9 [6]. Andres and colleagues [7], however, reported that the BMI associated with the lowest mortality rates appears to increase with age, so the criteria may need to be adjusted upward with time. For example, their data suggested that the optimal BMI was 26.6 for men and 27.3 for persons between ages 60 and 69 years. In contrast, Manson and associates [8] found that the risk for coronary heart disease in women correlated with the degree of overweight, regardless of age, suggesting that adjustment of the BMI with age may be inappropriate.

For persons between 20 and 29 years old, reduction of BMI to less than 27 (range, 25 to 27) is a criteria for intermediate success of obesity treatment, and reduction to a range of 20 to 24.9 could be considered full success. Because modest weight loss decreases risk [5], a criteria of minimal success may be a loss of 2 BMI units. Such a loss corresponds to a decrease in body weight of about 5 to 6 kg (10 to 15 lb) for a typical American woman (height, 1.63 m [64 in]) or man (height, 1.75 m [69 in]).

Body Fat Mass

The most accurate assessment of obesity is the body fat content. Unfortunately, an assay for body fat content is not readily available. Body circumferences, body skinfold measurements, or bioelectric impedance measurements approximate body fat content and are useful for population studies but may be too inaccurate for individual assessment. More definitive assays, such as underwater weighing, computed tomography, nuclear magnetic resonance imaging, or dual x-ray absorptiometry, are not available or are prohibitively expensive. The desirable range for body fat content varies with age, but in women between ages 20 and 29 years is about 27% ± 5% of body weight and for men is about 18% ± 5%. Any loss of body fat is a criterion for minimum success; loss of 50% of excess body fat is a criterion for intermediate success; and loss to the desirable range is a criterion for full success.

Maintenance of Improvement

Obesity is a chronic disease that is currently incurable. Therefore, life-long follow-up by health professionals, similar to that provided to other persons with chronic diseases, is needed. Maintenance of weight loss or improvement in complications for 6 months is a minimum standard; lesser periods have little meaning. Due to the high rate of weight regain, few data exist for periods longer than 2 years, a period that may be considered an intermediate standard for weight maintenance. Very few studies have examined persons treated for obesity for 5 years or longer [9]. Some surgical series have followed patients for 5 years or longer and have found reasonable measures of success (maintenance of at least 50% loss of excess body weight and improved complications) [10, 11]. All medical treatment programs that have followed patients for 5 years or longer have shown very low success rates in terms of weight loss, and all have had high dropout rates [9, 11]. Despite the lack of long-term data, full success is defined as a 5-year period of weight maintenance or improvement in complications.

Reduction of Complications

Obesity is associated with many medical complications [12]. Fortunately, most of these complications are improved or cured with weight loss in most obese patients. Because these diseases and complications differ from person to person, it is difficult to define partial success. Therefore, the criteria for success for the medical complications listed here assume that full success is a return to values within the "normal range". Minimal success is any improvement, and intermediate success is a reduction by 50% of the level above "normal".

Hypertension

Full success is defined as a reduction in systolic blood pressure to less than 140 mm Hg and in diastolic blood pressure to less than 90 mm Hg. Intermediate success is a reduction of elevated pressures to a level halfway between the initial and the reference values. These values tend to increase with age, and patients must be evaluated individually. A blood pressure of 139/89 mm Hg in a young person (20 to 29 years old) is of more concern than the same reading in a person older than 60 years.

Diabetes Mellitus

Diabetes is defined by two fasting plasma glucose levels greater than 7.8 mmol/L (140 mg/dL), two postprandial plasma glucose levels greater than 11.1 mmol/L (200 mg/dL), or a combination of these. Fully successful obesity treatment reduces glucose levels to less than 7.8 mmol/L (fasting) and 11.1 mmol/L (postprandial). These values tend to increase with age, and patients must be evaluated individually. A single value is insufficient because blood glucose varies widely. Consistent improvement must be shown by several improved blood glucose values or improvement in glycosylated hemoglobin values, a measure of long-term diabetic control. Reduction of these results to the normal range is defined as full success, and reduction to 50% of "excess" is considered intermediate success. Any reduction qualifies as minimal success.

Hyperlipidemia

Both cholesterol and triglyceride levels have been shown to be risk factors for atherosclerotic disease. Levels of high- and low-density lipoproteins, as well as levels of some apolipoproteins, have also been shown to be risk factors. Because cholesterol and triglycerides are the most easily obtained and frequently used screening and follow-up tests, they are used as standards. In the reference group of persons 20 to 29 years old, reduction to the normal range (< 5.17 mmol/L [< 200 mg/dL] for cholesterol and 1.69 mmol/L [< 150 mg/dL] for triglycerides) is considered full success; a reduction of 50% of the "excess" is considered intermediate success; and any reduction is considered minimal success. Normal values increase with age; therefore, patients must be considered individually. The "normal range" for both cholesterol and triglyceride levels varies greatly among laboratories; therefore, criteria should be based on the normal ranges determined in the laboratory available to the health professional.

Sleep Apnea

Sleep apnea, the cessation of breathing during deep sleep, is a serious problem in obese persons. Sleep apnea is associated with fatal cardiac arrhythmias. The end-stage of sleep apnea, the Pickwickian syndrome, is associated with a high mortality rate if not treated appropriately [13]. If an episode of sleep apnea is defined as desaturation of oxyhemoglobin to less than 90%, full success is defined as no desaturations less than this level during sleep. Intermediate success is defined as a reduction by 50% in the number of apneic episodes, and minimal success is defined as any consistent reduction.

Psychological and Social Complications

Obesity is associated with a poor self-image; a slightly higher score on assessments of depression; and various social problems including discrimination against the obese person for jobs, promotions, and in marriageability [14, 15]. Overall, obese persons experience a lower quality of life. It is difficult to determine success in treatment of these problems because many require subjective assessment of quality of life. More research is needed to assess the success of weight reduction in improving quality of life for obese persons.

In summary, obesity is a complex, multifactorial disease with many complications. Treatments must be individualized, and success is judged individually on several criteria. Reduction of complications, even without weight loss, is considered treatment success. Persistent weight loss, although modest, is also considered to be a success.