Effects of Age on Validity of Self-Reported Height, Weight, and Body Mass Index: Findings from the Third National Health and Nutrition Examination Survey, 1988–1994

doi:10.1016/S0002-8223(01)00008-6

Journal of the American Dietetic Association

Volume 101, Issue 1, January 2001, Pages 28-34

https://doi.org/10.1016/S0002-8223(01)00008-6 Get rights and content

Abstract

Objective To compare self-reported to measured heights and weights of adults examined in the Third National Health and Nutrition Examination Survey (NHANES III), and to determine to what extent body mass index (BMI) calculated from self-reported heights and weights affects estimates of overweight prevalence compared with BMI calculated from measured values.

Design A complex sample design was used in NHANES III to obtain a nationally representative sample of the US civilian, noninstitutionalized population. During household interviews, survey respondents were asked their height and weight. Trained health technicians subsequently measured height and weight using standardized procedures and equipment.

Subjects The analytical sample consisted of 7,772 men and 8,801 women 20 years old and older.

Statistical analyses performed Only persons with measured and self-reported heights and weights were included in the analysis, and statistical sampling weights were applied, t Tests, Pearson product moment correlation coefficients, sensitivity, and specificity analyses were used to determine the validity of self-reported measurements and prevalence estimates of overweight, defined as BMI of 25 or greater.

Results Age is an important factor in classifying weight, height, BMI, and overweight from self-reports. Statistically significant differences were found for the mean error (measured-self-reported values) for height and BMI that were notably larger for older age groups. For example, the mean error for height ranged from 2.92 to 4.50 cm for women and from 3.06 to 4.29 cm for men, 70 years and older. Despite the high correlation between measured and self-reported data, the prevalence of overweight calculated from measured values was higher than that calculated from self-reported values among older adults. When calculated with self-reported height, BMI was one unit lower than when calculated from measured height for persons >70 years. Specificity was high but sensitivity decreased with increasing age cohorts. Regression equations are provided to determine actual height from self-reported values for older adults.

Conclusion/Applications Self-reported heights and weights can be used with younger adults, but they have limitations for older adults, ages >60 years. In research studies and in clinical settings involving older adults, failure to measure height and weight can result in subsequent misclassification of overweight status. Therefore, registered dietitians are encouraged to obtained a measured weight and height using a calibrated scale and stadiometer. J Am DietAssoc. 2001;101:28-34.

Section snippets

Sample Size

National Health and Nutrition Examination surveys are designed to obtain a nationally representative sample of the US civilian, noninstitutionalized population (20). In NHANES III, 11,029 men and 12,229 women aged 20 years and older were selected as potential respondents for the survey (21). Of these individuals, approximately 80% of men (8,816 of 11,029) and 82% of women (10,009 of 12,229) were subsequently examined in the Mobile Examination Center (MEG). A modified home examination was

Height

Pearson correlation (validity) coefficients between measured and self-reported heights were significant (P<.001) for all of age groups within each gender. Correlation coefficients for the older age groups, 70 to 79 and 80 plus years, were less than those of the younger age groups, especially for women, which suggests that self-reported height was less valid in the older subgroups. The correlations ranged from 0.93 for men aged 20 to 29 years to 0.85 for men 80 years old and older. For women,

Discussion

Our analyses clearly demonstrate that age is an important factor in classifying weight, height, BMI, overweight, and obesity from self-reports. The error associated with self-reported height was greatest in the oldest age groups, and overreporting of height with increasing age is consistent with findings of others (11), (12), (13), (15). Overestimation of height by older adults may occur because greater time has elapsed since height was last measured and because height decreases with age (12).

Applications/Conclusions

This article described the direction and magnitude of the reporting error for self-reported height and weight in a representative sample of the general population of adults in the United States. Errors in self-reporting of height are related to a person's age, and unreliability increases directly with age.

■ Whenever possible, dietitians should measure the height and weight of their older adult clients.

■ Practicing dietitians could estimate actual height from self-reported heights of their older

References (29)

K.S Kubena et al.
Anthropometry and health in the elderly
J Am DietAssoc.
(1991)
G.M Wardlaw
Putting body weight and osteoporosis into perspective
Am J Clin Nutr.
(1996)
J.C Cornoni-Huntley et al.
An overview of body weight of older persons, including the impact on mortality
J Clin Epidemiol.
(1991)
M.L Rowland
Self-reported weight and height
Am J Clin Nutr.
(1990)
A.L Stewart
The reliability and validity of self-reported weight and height
J Chronic Dis.
(1982)
A.J Stunkard et al.
The accuracy of self-reported weights
Am J Clin Nutr.
(1981)
P Pirie et al.
Distortion in self-reported height and weight data
J Am DietAssoc.
(1981)
J Nutr Health Ageing.
(1998)
C.P de Groot et al.
Longitudinal changes in anthro-pometric characteristics of elderly Europeans
Eur J Clin Nutr.
(1996)
A.N Galanos et al.
Nutrition and functionis there a relationship between body mass index and the functional capabilities of community-dwelling elderly
J Am Geriatr Soc.
(1994)

D Hemenway D Feskanich et al.

Body height and hip fracturecohort study of 90,000 women

Int J Epidemiol.

(1995)

J.C Alvarez-Torices et al.

Self-reported height and weight and prevalence of obesityStudy in a spanish population

IntJObes.

(1993)

D.K Dey et al.

Height and body weight in the elderlya 25-year longitudinal study of a population aged 70 to 95 years

Eur J Clin Nutr.

(1999)

N.Y Haboubi et al.

Measurement of height in the elderly

J Am Geriatr Soc.

(1990)

Cited by (809)

Eating disorder-related functional impairment predicts greater depressive symptoms across one semester of college
2024, Eating Behaviors
Eating disorder (ED) behaviors and depression are associated with numerous negative outcomes, including lower quality of life and functional impairment. College women are at elevated risk for both. Prior research indicates ED behaviors, including binge eating, self-induced vomiting, and fasting, predict increases in future depressive symptoms. However, symptom heterogeneity in EDs is common, and all disordered eating, or its associated distress, cannot be captured by the endorsement of behaviors. Impairment that results from ED behaviors may be a comparable, or stronger, predictor of depressive symptoms. We sought to characterize the longitudinal relationship between ED-related functional impairment, ED behaviors, and depressive symptoms. College-aged women [N = 304; 72 % white, mean (SD) age = 18.45 (0.88)] completed an online survey in August (baseline), and then three months later in November (follow-up). Baseline ED-related functional impairment, but not baseline ED behaviors, significantly predicted depressive symptoms at follow-up, controlling for baseline depressive symptoms, negative affect, and body mass index. Findings indicate ED-related functional impairment is a risk factor for increases in depressive symptoms across one semester of college, irrespective of ED behavior engagement, weight status, and dispositional negative affect. Intervening upon ED-related functional impairment may reduce or prevent future depressive symptoms among college-aged women.
Associations between body mass index and smoking behaviour: A cross-sectional study of the German adult population
2024, Physiology and Behavior
Both smoking and high body weight are risk factors for disease, hence, the association between smoking and body weight is an important health issue. Furthermore, concern about weight gain after quitting smoking is for many smokers a barrier to smoking cessation. The present study aims to explore the association between body mass index (BMI) and current tobacco smoking status in the population of Germany, and smoking and quitting behaviour amongst smokers (and recent ex-smokers =<12 months since quitting).
Cross-sectional analysis of two waves of data collected from March through June 2021 through a representative face-to-face household survey in Germany (N = 3 997 respondents aged ≥18). The associations between smoking and quitting behaviours and BMI were analysed through four regression models adjusted for socio-demographic, socio-economic, and smoking characteristics of respondents.
Long-term ex-smokers (>= 12 months since quitting smoking) were more likely to have a higher BMI compared to never smokers (β = 0.64, 95% confidence interval (CI) = 0.10–1.19). There was no statistically significant association between current smoking status or recent ex-smoking status and BMI (β = -0.29,95 %CI = -0.75–0.17 and β = -0.53, 95 %CI = -2.45–1.40). Among current smokers, no statistically significant association was found between BMI and the motivation to stop smoking (OR = 1.01, 95 %CI = 0.99–1.03). Neither number of cigarettes smoked a day nor outcome of most recent quit attempt were related to BMI (β = 0.01, 95 %CI = -0.04–0.05 and OR = 0.41, 95 %CI = 0.05–3.05).
In the German population long-term ex-smoking but not current and recent ex-smoking was associated with increased BMI. Future research should further explore the association between smoking behaviour and abdominal obesity, preferably using a more accurate measure for abdominal obesity than BMI.
Gastroesophageal reflux and snoring are related to asthma and respiratory symptoms: Results from a Nordic longitudinal population survey
2024, Respiratory Medicine
To study if individuals with nocturnal gastroesophageal reflux (nGER) and habitual snoring are more likely to develop asthma and respiratory symptoms (i.e. wheeze, cough, chest tightness, breathlessness) than those without these conditions, and if these associations are additive.
We used data from the population-based prospective questionnaire study Respiratory Health in Northern Europe (RHINE) (11,024 participants), with data from 1999 and 2011. Participants with heartburn or belching after going to bed, at least 1 night/week, were considered to have nGER. Participants reporting loud snoring at least 3 nights/week were considered to have habitual snoring. Participants were grouped into four groups by their nGER and snoring status: “never”; “former”; “incident”; “persistent”. Incident respiratory symptoms were analyzed among participants without respective symptom at baseline.
Snoring and nGER were independently associated with incident asthma and respiratory symptoms. The risk of incident wheeze was increased in subjects with incident or persistent snoring (adjusted odds ratio (95 % CI): 1.44 (1.21–1.72)), nGER (2.18 (1.60–2.98)) and in those with both snoring and nGER (2.59 (1.83–3.65)). The risk of developing asthma was increased in subjects with incident or persistent snoring (1.44 (1.15–1.82)), nGER (1.99 (1.35–2.93)) and in those with both snoring and nGER (1.72 (1.06–2.77)). No significant interaction was found between snoring and nGER. A similar pattern was found for the incidence of all other respiratory symptoms studied, with the highest risk among those with both incident or persistent nGER and snoring.
The risk of developing asthma and respiratory symptoms is increased among subjects with nGER and habitual snoring. These associations are independent of each other and confounding factors. Snoring and nGER together are additive on respiratory symptoms.
Are self-reported anthropometric data reliable enough to meet antibiotic prophylaxis guidelines in orthopedic surgery?
2023, Revue de Chirurgie Orthopedique et Traumatologique
L’infection du site opératoire est une complication grave en chirurgie orthopédique. L’antibioprophylaxie (ABP) avec les autres facteurs de prévention permet de diminuer ce risque à 1 % pour les prothèses de hanche et 2 % pour le genou. La Sfar (Société française d’anesthésie réanimation) recommande de doubler la posologie lorsque le poids du patient est supérieur ou égal à 100 kg et un IMC (indice de masse corporelle) supérieur ou égal à 35 kg/m². De même, dans notre centre, les patients présentant un IMC supérieur à 40 kg/m² ou inférieur à 18 kg/m² sont récusés à la chirurgie. En pratique, ce sont fréquemment les données anthropométriques déclarées par le patient qui vont servir pour calculer l’IMC, mais la validité de cette déclaration n’est pas évaluée dans la littérature orthopédique. Aussi, nous avons mené une étude afin de comparer les données déclarées par le patient à celles vérifiées systématiquement et d’observer l’impact de cette différence sur le schéma d’antibioprophylaxie peropératoire et sur les contre-indications à la chirurgie.
L’hypothèse de notre étude était qu’il existait une différence entre les données anthropométriques déclarées par le patient et celles vérifiées en consultation d’orthopédie.
Il s’agit d’une étude rétrospective à recueil prospectif monocentrique conduite entre octobre et novembre 2018. Les données anthropométriques déclarées par les patients étaient recueillies puis vérifiées par un(e) infirmier(e) de consultation. La précision de mesure pour le poids était de 500 g et pour la taille de 1 cm.
Trois cent soixante-dix patients (259 femmes et 111 hommes) d’un âge médian de 67 ans [17–90] ont été inclus. L’analyse des données a relevé une différence significative entre les valeurs déclarées et mesurées pour la taille 166 cm (147–191) vs. 164 cm (141–191) (p < 0,0001), le poids 72,9 kg (38-149) vs. 73,1 kg (36–140) (p < 0,0005) et l’IMC 26,3 (16,2–46,4) vs. 27 (16–48,2) (p < 0,0001). Au total, 119 (32 %) patients déclaraient une taille exacte, 137 (37 %) un poids exact et 54 (15 %) un IMC exact. Aucun patient n’avait deux données exactes. La sous-estimation maximale du poids était de 18 kg, de la taille 9 cm et de l’IMC 6,15 kg/m². La surestimation maximale du poids était de 28 kg, de la taille 10 cm et de l’IMC 7,2 kg/m². Le contrôle des données anthropométriques a permis de dépister 17 patients présentant une contre-indication chirurgicale (12 avec un IMC > 40 kg/m² et 5 avec un IMC < 18 kg/m²) qui n’aurait pas été détectée en se basant sur les valeurs rapportées par le patient.
Les patients sous-estimaient leur poids et surestimaient leur taille dans notre étude, sans que cela ait impacté le schéma d’antibioprophylaxie peropératoire. Cette mésestimation engendrait toutefois un défaut de détection de potentielles contre-indications à la chirurgie.
IV ; rétrospective à recueil prospectif sans groupe contrôle.
Associations of the Geriatric Nutritional Risk Index with high risk for prostate cancer: A cross-sectional study
2023, Nutrition
The geriatric Nutritional Risk Index (GNRI) is an effective tool to assess the nutritional status of the elderly. However, the relationship between the GNRI and the risk for prostate cancer (PCa) remains uncertain in middle-aged and older men. The aim of this study was to investigate the association between the GNRI and the risk for PCa by analyzing the serum total (tPSA) and free prostate-specific antigen (fPSA) levels (including percent fPSA [%fPSA]).
Data for this study were obtained from 7396 men ≥40 y of age from the 2001–2010 National Health and Nutrition Survey (NHANES). We obtained the tPSA and fPSA and calculated the %fPSA and the GNRI. Participants with %fPSA >25% and tPSA <4 ng/mL were defined as high PCa risk. The relationship between the GNRI and serum PSA levels was investigated using a linear regression model. The odds ratios (ORs) and 95% confidence intervals (CIs) for the association between the GNRI and PCa risk were estimated by a logistic regression model. The non-linear relationship was also characterized by a restricted cubic spline regression model.
The median of tPSA, fPSA, and %fPSA was 0.90, 0.26, and 29%, respectively. The mean of the GNRI was 29. The proportion of participants in the low PCa- and high PCa-risk groups was 93% and 7%, respectively. There was a negative and linear correlation between the GNRI and serum tPSA and fPSA levels in all models. However, no association between the GNRI and the %fPSA was observed. In the adjusted model, lower GNRI was associated with higher PCa risk (OR, 0.570; 95% CI, 0.415–0.784; P_trend = 0.001). The restricted cubic spline regression model showed a non-linear and negative association between the GNRI and PCa risk (P_{non-linearity} = 0.020), with inflection points of 109.148.
The results of this study suggest that nutritional status, as represented by the GNRI, is associated with the risk for PCa.
Are self-reported anthropometric data reliable enough to meet antibiotic prophylaxis guidelines in orthopedic surgery?
2023, Orthopaedics and Traumatology: Surgery and Research
Surgical site infection is a serious complication in orthopedic surgery. The use of antibiotic prophylaxis (AP) combined with other prevention strategies has been shown to reduce this risk to 1% for hip arthroplasty and 2% for knee arthroplasty. The French Society of Anesthesia and Intensive Care Medicine (SFAR) recommends doubling the dose when the patient's weight is greater than or equal to 100 kg, and the body mass index (BMI) is greater than or equal to 35 kg/m². Similarly, patients with a BMI greater than 40 kg/m² or lesser than 18 kg/m² are ineligible for surgery in our hospital. Self-reported anthropometric measurements are commonly used in clinical practice to calculate BMI, but their validity has not been assessed in the orthopedic literature. Therefore, we conducted a study comparing self-reported with systematically measured values and observed the impact these differences may have on perioperative AP regimens and contra-indications to surgery.
The hypothesis of our study was that self-reported anthropometric values differed from those measured during preoperative orthopedic consultations.
This single-center retrospective study with prospective data collection was conducted between October and November 2018. The patient-reported anthropometric data were first collected and then directly measured by an orthopedic nurse. Weight was measured with a precision of 500 g and height was measured with a precision of 1 cm.
A total of 370 patients (259 women and 111 men) with a median age of 67 years (17–90) were enrolled. The data analysis found significant differences between the self-reported and measured height [166 cm (147–191) vs. 164 cm (141–191) (p < 0.0001)], weight [72.9 kg (38–149) vs. 73.1 kg (36–140) (p < 0.0005)] and BMI [26.3 (16.2–46.4) vs. 27 (16–48.2) (p < 0.0001)]. Of these patients, 119 (32%) reported an accurate height, 137 (37%) an accurate weight, and 54 (15%) an accurate BMI. None of the patients had two accurate measurements. The maximum underestimation was 18 kg for weight, 9 cm for height, and 6.15 kg/m² for BMI. The maximum overestimation was 28 kg for weight, 10 cm for height, and 7.2 kg/m² for BMI. The verification of the anthropometric measurements identified another 17 patients who had contra-indications to surgery (12 with a BMI > 40 kg/m² and 5 with a BMI < 18 kg/m²) and who would not have been detected based on the self-reported values.
Although patients underestimated their weight and overestimated their height in our study, these had no impact on the perioperative AP regimens. However, this misreporting failed to detect potential contraindications to surgery.
IV; retrospective study with prospective data collection and no control group.

View all citing articles on Scopus

View full text

Effects of Age on Validity of Self-Reported Height, Weight, and Body Mass Index: Findings from the Third National Health and Nutrition Examination Survey, 1988–1994

Abstract

Section snippets

Sample Size

Height

Discussion

Applications/Conclusions

J Am DietAssoc.

Am J Clin Nutr.

J Clin Epidemiol.

Am J Clin Nutr.

J Chronic Dis.

Am J Clin Nutr.

J Am DietAssoc.

J Nutr Health Ageing.

Longitudinal changes in anthro-pometric characteristics of elderly Europeans

Eur J Clin Nutr.

Nutrition and functionis there a relationship between body mass index and the functional capabilities of community-dwelling elderly

J Am Geriatr Soc.

Body height and hip fracturecohort study of 90,000 women

Int J Epidemiol.

Self-reported height and weight and prevalence of obesityStudy in a spanish population

IntJObes.

Height and body weight in the elderlya 25-year longitudinal study of a population aged 70 to 95 years

Eur J Clin Nutr.

Measurement of height in the elderly

J Am Geriatr Soc.