One of the most important health issues in the 21st century is the continuous increase in body weight. Excess body weight is associated with several adverse health outcomesReference Nanchahal, Morris, Sullivan and Wilson1–Reference Klein5 and with an increased mortality riskReference Thomas, Bean, Pannier, Oppert, Guize and Benetos6–Reference Calle, Rodriguez, Walker-Thurmond and Thun9. Therefore, it is of importance to identify subjects that would benefit from weight loss treatment. Furthermore, monitoring secular changes of the need for weight loss treatment is necessary for the development of adequate strategies for weight control.
The Obesity Education Initiative Expert Panel of the National Heart, Lung and Blood Institute, one of the US National Institutes of Health (NIH), proposed a treatment algorithm that was published by NIH with the aim to give health care professionals a guide to assess and treat obesity10, 11. This treatment algorithm, based on BMI, waist circumferences (WC) and CVD risk factors, should help health practitioners to identify individuals in need of weight loss. The Spanish Society of Endocrinology and Nutrition adopted this algorithm for use in the Spanish populationReference Arrizabalaga, Masmiquel and Vidal12. Indeed, early identification of individuals at high risk of morbidity and mortality related to excessive weight is important for health policy. However, to the best of our knowledge there are no data on the prevalence and secular trends of individuals in need of weight loss based on this treatment algorithm. Hence, the aim of the present study was to identify changes in the proportion of the population from a representative Mediterranean region eligible for weight loss treatment according to the NIH algorithm for men and women aged 25 to 74 years.
Subjects and methods
Study population
Data of two independent and representative population-based surveys conducted in 1995 and 2000 in Girona (Spain) were used for this analysis. Recruitment procedures have been described in detail elsewhereReference Masia, Pena, Marrugat, Sala, Vila, Pavesi, Covas, Aubo and Elosua13. From 6925 randomly selected and eligible subjects between 25 and 74 years of age, 4908 (2383 men and 2525 women) agreed to participate. Total participation rate reached 71 %. A two-stage sampling method was used in 1995 and 2000: thirty-three and seventeen towns, respectively, were randomly selected in the first stage. Half of the towns were urban (>10 000 inhabitants) and half were rural (500–10 000 inhabitants). In both studies, the second sampling stage consisted of randomly recruiting the same number of female and male participants, stratifying by 10-year age groups from the closest census. All participants were duly informed and consented to participate in the study. The protocol was approved by the local Ethics Committee.
Anthropometric measurements
A precision scale of easy calibration was used for weight measurement. Readings were rounded to 200 g. Individuals wore underwear. Height was measured in the standing position and measurements rounded to 5 mm. BMI was determined as weight divided by height squared (kg/m2). BMI categories (normal weight, overweight and obesity) were calculated according to the WHO classificationReference Kuczmarski and Flegal14. WC was measured at midway between the lowest rib margin and iliac crest, with the subject standing. Measurements were taken with a tape measure in cm and rounded to 5 mm. Abdominal obesity was defined as WC>88 cm in women and >102 cm in menReference Lean, Han and Morrison15.
Measurements of blood pressures and lipid profile
Two blood pressure determinations were taken by trained personnel using a periodically calibrated Hg sphygmomanometer with strict standard procedures. Measurements were performed in the sitting position after a 5-min rest. Two measurements were taken, at least 20 min apart, and the lower value was recorded for the study. The mean of two measurements was used for analysis.
Blood samples were obtained after a 14-h fast. Serum was immediately frozen at − 120°C in liquid N for transportation and stored at − 80°C for final conservation. Total cholesterol, TAG and HDL-cholesterol were analysed by standardized enzymatic methods (Roche Diagnostic, Basel, Switzerland) adapted to a Cobas Mira Plus autoanalyzer (Hoffmann-La Roche, Basel, Switzerland).
National Institutes of Health algorithm
The NIH algorithm was calculated as follows: BMI ≥ 30 or ((BMI 25–29·9 or WC>88 cm (women), >102 cm (men)) and two or more CVD risk factors). Participants were assigned a CVD risk status based on the presence of two or more risk factors. These included age, ≥ 45 years in men and ≥ 55 years in women, current smoking, HDL-cholesterol ≤ 35 mg/dl, LDL-cholesterol >160 mg/dl, hypertension (systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg or on hypertension drug treatment), glycaemia >125 mg/dl or history of diabetes.
Educational level and lifestyle variables
Leisure-time physical activity was measured by the Minnesota leisure-time physical activity questionnaire. This questionnaire has been previously validated for Spanish men and womenReference Elosua, Marrugat, Molina, Pons and Pujol16, Reference Elosua, Garcia, Aguilar, Molina, Covas and Marrugat17.
Information on smoking habits of the participants was obtained by a structured standard interview. Participants were categorized as non-smokers or current smokers (at least one cigarette per d on average during the past year).
Maximum level of education attained was elicited and recorded as primary school, secondary school and university for purposes of analysis.
Statistical analysis
Data are presented descriptively for each gender with means or proportions. Student t test was used in the univariate analysis.
Adjusted OR were calculated to investigate the association of the need for weight loss treatment, according to the NIH treatment algorithm, with age and educational level. Cohort analysis was added to the model to examine if there were secular changes in the prevalence of recommended weight loss treatment according to age and educational level (PROC LOGISTIC procedure, version 8.0; SAS Institute Inc., Cary, NC, USA).
Results
Table 1 shows the general characteristics of the participants. BMI significantly increased from 1995 to 2000 in both genders. The same trend was observed for WC in men (Table 1). The crude prevalence of men and women meeting the criteria for weight loss treatment according to the NIH treatment algorithm was highest for the oldest male and female age group in 1995 and 2000 (Table 2). The adjusted relative risk of need for weight loss treatment significantly increased with older age in both genders and in both surveys (Table 3). More men and women with only a basic educational level (primary school) would benefit from weight loss treatment as compared with higher educational levels (Table 2). We observed an increasing linear trend across lower educational levels in the need for weight loss treatment in both men and women (Table 3), with the exception of men in 1995. Multiple logistical regression analysis revealed an increased need for weight loss treatment in men and women from 1995 to 2000 (Table 4). Stratifying this analysis for age groups revealed a significant increase among women aged 25 to 39 years and in both genders aged 60 to 74 years (Table 4). More men and women with a primary school education would benefit from weight loss treatment in 2000 than in 1995. A similar trend was observed for women with secondary school education (Table 4).
Table 1 General characteristics, BMI, waist circumference of men and women aged 25–74 years in 1995 and 2000†
Results are expressed as mean (standard deviation or 95 % confidence interval)
LTPA, Leisure-time physical activity; MET, standard metabolic equivalent.
* Age standardized. Standard World Population28.
† For details of subjects and procedures, see Subjects and methods.
Table 2 Proportion (%) and 95 % CI of the male and female population meeting the criteria for weight loss treatment* according to age group and educational level§
* According to National Institutes of Health treatment algorithm: BMI ≥ 30 or ((BMI 25–29·9 or waist circumference >88 cm (women), >102 cm (men)) and two or more CVD risk factors). Risk factors: age ( ≥ 45 years in men and ≥ 55 years in women); current smoking; HDL-cholesterol ≤ 35 mg/dl; LDL-cholesterol >160 mg/dl. Hypertension: systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg or on hypertension drug treatment. History of diabetes or glycaemia >125 mg/dl.
† Age standardized. Standard World Population28.
‡ Significantly different (P < 0·05) between 1995 and 2000.
§ For details of subjects and procedures, see Subjects and methods.
Table 3 OR and 95 % CI of the need for weight loss treatment* according to age group and educational level§
* According to National Institutes of Health treatment algorithm: BMI ≥ 30 or ((BMI 25–29·9 or waist circumference >88 cm (women), >102 cm (men)) and two or more CVD risk factors). Risk factors: age ( ≥ 45 years in men and ≥ 55 years in women); current smoking; HDL-cholesterol ≤ 35 mg/dl; LDL-cholesterol >160 mg/dl. Hypertension: systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg or on hypertension drug treatment. History of diabetes or glycaemia >125 mg/dl.
† Adjusted for educational level and leisure-time physical activity.
‡ Adjusted for age and leisure-time physical activity.
§ For details of subjects and procedures, see Subjects and methods.
Table 4 OR and 95 % CI of secular trends of the recommendation for weight loss treatment* of men and women‖
* According to National Institutes of Health treatment algorithm: BMI ≥ 30 or ((BMI 25–29·9 or waist circumference >88 cm (women), >102 cm (men)) and two or more CVD risk factors). Risk factors: age ( ≥ 45 years in men and ≥ 55 years in women); current smoking; HDL-cholesterol ≤ 35 mg/dl; LDL-cholesterol >160 mg/dl. Hypertension: systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg or on hypertension drug treatment. History of diabetes or glycaemia >125 mg/dl.
† Adjusted for educational status and leisure-time physical activity.
‡ Adjusted for age and leisure-time physical activity.
§ Significantly different (P < 0·05) between 1995 and 2000.
‖ For details of subjects and procedures, see Subjects and methods.
Discussion
Our data indicate that about half of the present population is recommended to lose weight. Most importantly, the need for weight loss treatment increased markedly from 1995 to 2000 in both genders. This trend was more pronounced in older age groups and among the population with only a basic educational level.
Excessive weight is strongly associated with several adverse health outcomes, such as CVDReference Grundy18. Identifying individuals with excessive weight who are at increased risk of obesity-related morbidity and mortality is of importance for the development of promising treatment strategies. Based on National Heart, Lung, and Blood Institute (NHLBI) recommendations, the US NIH proposed an algorithm to identify individuals with excessive weight who would greatly benefit from weight loss treatment10, 11. In a large longitudinal study, Mason and colleagues have shown that the risk of all-cause mortality significantly increased in men meeting these criteria for weight loss treatment compared with normal weight menReference Mason, Katzmarzyk and Blair19. Mason and KatzmarzykReference Mason and Katzmarzyk20 also found that 24 % of Canadians would be eligible for weight loss treatment, according to these criteria. In 1995, about 40 % of the present study population would have been identified for weight loss treatment; by 2000 there was a significant increase in this prevalence in men and women. Indeed, the magnitude and time trend of the prevalence of men and women meeting the NIH weight loss treatment criteria is a serious problem in the present population and should be an alarming signal for public health policy.
Evidence indicates that low education is associated with greater weight gain over time in young and middle adulthoodReference James21–Reference Everson, Maty, Lynch and Kaplan23. Hence, it was not surprising to observe a significant increase in the prevalence of subjects meeting the NIH criteria for weight loss treatment among men and women with a basic educational level between 1995 and 2000 in the present population.
The NIH treatment algorithm included WC, an indirect measure of abdominal obesity. It has been shown that abdominal obesity, clinically defined by WC>88 cm in women and >102 cm in men, is an independent risk factor for CHD, hypertension, increased oxidized LDL-cholesterol, and type 2 diabetesReference Rexrode, Carey, Hennekens, Walters, Colditz, Stampfer, Willett and Manson24–Reference Wang, Rimm, Stampfer, Willett and Hu27. However, these WC cut-offs were developed as an alternative to BMI measures and are strongly correlated with a BMI of 30 or moreReference Lean, Han and Morrison15. Therefore, the question has been raised if the inclusion of these WC cut-offs in the NIH treatment algorithm is useful in increasing the rate of identification of subjects who meet weight loss treatment criteria.
In the present population, excluding the WC cut-offs of >88 cm in women and >102 cm in men would have excluded only 0·3 % of men and 1·0 % of women from those identified for weight loss treatment by the NIH treatment algorithm. In contrast, an additional 3·2 % and 4·2 % of men and women, respectively, would be considered at elevated risk according to the NIH algorithm using WC cut-offs of >80 cm in women and >94 cm in men; these levels are strongly correlated with a BMI of 25 or moreReference Mason and Katzmarzyk20. All of these subjects were normal weight (BMI 18·5–24·9) with elevated WC and two or more CVD risk factors. In view of the increasing evidence of adverse health effects of elevated WC, independent of BMI, it might be reasonable to incorporate these WC cut-offs into the NIH treatment algorithm. This would help to identify considerably more subjects who would benefit from weight loss treatment.
It should be noted that the loss of about 30 % of participants can be considered a certain limitation of the present study.
In conclusion, in 1995 about 40 % of the present population would be identified for weight loss treatment according to the NIH obesity treatment algorithm. This prevalence significantly increased 5 years later, by 6·5 % in men and 5·3 % in women. These results should be a strong indicator of the need to reinforce public policy and patient education efforts to develop weight loss strategies in the present Mediterranean population.
Acknowledgements
We acknowledge the English revision by Elaine Lilly of Writer's First Aid. This research was supported by grant 2FD097-0297-CO2-01 from Fondo Europeo de Desarrollo Regional (FEDER) and by parts of the Ministerio de Sanidad y Consumo, Instituto de Salud Carlos III, Red HERACLES RD06/0009, CIBER: Epidemiología y salud pública CB06/02/0029, CIBER: Fisiopatología de la obesidad y nutrición CB06/03/0028, Fondo de Invesigación Sanitaria ISCIII CP 03/00 115.
