Validity of self-reported body mass and height: relation with sex, age, physical activity, and cardiometabolic risk factors

ABSTRACT:

Objective:

To evaluate the validity of self-reported body mass and height measurements in adolescents, adults and older adults according to sex, age, leisure-time physical activity level, nutritional status, and cardiometabolic risk factors.

Methods:

The study included 856 subjects, aged 12 years or older, who participated in the São Paulo Health Survey (ISA-2015) and who had their body mass and height measured and self-reported. Based on the Body Mass Index (BMI), a classification of nutritional status was made according to standardized criteria for each phase of life. The validation of self-reported data was examined by the Intraclass Correlation Coefficient, Bland-Altman and paired T-Test. Linear regression models were used to estimate the calibration coefficients, and sensitivity and specificity tests were performed.

Results:

Self-reported body mass and height values tend to be very similar to measured values, with a few exceptions. For the adolescents, an underestimation of height was noted, while for the older adults, an overestimation. There was a consistent underestimation of self-reported body mass among women, and an overestimation of BMI among men who practiced less than 150 minutes of physical activity per week during leisure time. The calibration process of self-reported measures made them more consistent with the values measured, increasing the sensitivity in the classification of nutritional status among women and the specificity among men.

Conclusions:

Self-reported measures of height, body mass and BMI provided valid and reliable measures, presenting a substantial improvement after calibration.

Keywords:
Validation study; Health surveys; Body weight; Body height; Body mass index; Sensitivity and specificity

RESUMO:

Objetivo:

Avaliar a validade da massa corporal e da estatura autorreferidas em adolescentes, adultos e idosos segundo sexo, idade, nível de atividade física no lazer, estado nutricional e fatores de risco cardiometabólicos.

Métodos:

Participaram do estudo 856 sujeitos, com 12 anos ou mais, que responderam ao Inquérito de Saúde de São Paulo (ISA-2015) e que possuíam a massa corporal e a estatura autorreferidas e aferidas. Com base no índice de massa corporal, realizou-se uma classificação do estado nutricional de acordo com critérios padronizados para cada fase da vida. A validade das medidas autorreferidas foi examinada usando o coeficiente de correlação intraclasse, Bland-Altman e o teste t pareado. Utilizaram-se regressões lineares para elaborar os coeficientes de calibração, e realizaram-se testes de sensibilidade e especificidade.

Resultados:

Os principais resultados apontam que os valores de massa corporal e estatura autorreferidas tendem a ser bem similares aos aferidos, apesar de algumas exceções. Para os adolescentes, notou-se uma subestimação da estatura, ao passo que, para os idosos, houve superestimação. Com relação à massa corporal, houve consistente subestimação da medida autorreferida entre as mulheres. Entre os homens que praticavam menos de 150 minutos semanais de atividade física no lazer, notou-se superestimação do índice de massa corporal. O processo de calibração das medidas autorreferidas tornou-as mais concordantes com as medidas aferidas, aumentando a sensibilidade na classificação do estado nutricional entre as mulheres e a especificidade entre os homens.

Conclusões:

As medidas autorreferidas de estatura, massa corporal e índice de massa corporal forneceram medidas válidas e confiáveis, apresentando melhoras substanciais após a calibração.

Palavras-chave:
Estudo de validação; Inquéritos epidemiológicos; Peso corporal; Estatura; Índice de massa corporal; Sensibilidade e especificidade

INTRODUCTION

Self-reported body mass (BM) and height are often used to calculate the body mass index (BMI) in order to quantify excess weight and obesity in epidemiological studies¹, because of the ease of data collection and logistics, as well as low cost and time required - with questionnaires or interviews -, when compared to anthropometric measurements².

However, self-reported values are susceptible to important limitations such as social desirability bias³, memory difficulties and body image perception⁴. Thus, it is essential to assess the extent of error present in the calculation of BMI based on self-reported measures before applying them in epidemiological studies. Otherwise, incorrect information for BMI calculation can produce inaccurate results from associations with other health indicators.

Review studies indicate that, in general, people tend to overestimate their height and underestimate their BM^5,6,7, leading to a potential bias in the estimates of BMI^7,8. Other studies have identified several factors that are associated with inaccurate reporting of height and/or BM, including gender, height and actual BM measured, nutritional status, recent medical appointments, health history^{9,10,11,12,13,14,15,16,17}, and nutritional status^18,19,20. Especially regarding age, the literature does not bring clear evidence. While studies show that differences between actual and self-reported BM tend to increase with age^10,21, others point to a U-shaped relationship, with younger and older people underestimating it¹⁸, or even with age not influencing the precision of the self-reported measure^22,23.

Although there is a substantial body of evidence regarding the behavior of self-reported measures between genders and nutritional status, there is little evidence of how these measures behave as a function of cardiometabolic risk factors, such as physical inactivity, hypertension, dyslipidemia, diabetes^24,25,26, yet these self-reported measures are frequently used in epidemiological studies²⁷. It is justifiable to explore the accuracy of these measures, specifically in these subgroups, as the presence of such conditions can change the precision with which individuals report their information. There could be, for example, greater knowledge due to more frequent medical follow-ups or even greater underreporting of BM due to the stigma associated with these conditions^28,29.

Another gap to be explored involves the calibration process, in which imprecise statements of BM and height are made more accurate through statistical adjustments. Although studies in Brazil^30,31 and in other countries around the world^32,33,34,35 have addressed this type of strategy, the magnitude of this possible improvement is still little explored.

Thus, the objectives of this study were:

• to analyze the relations and validity of self-reported BM and height per biological sex, age, leisure-time physical activity, nutritional status and cardiometabolic risk factors;

• to perform calibration coefficients to adjust BM, height and BMI measurements for each of the mentioned subgroups.

METHODS

This study is part of the São Paulo Health Survey, a cross-sectional population-based study conducted in 2014/15 in the city of São Paulo, Brazil (ISA-2015)³⁶. This dataset is also part of the baseline of the longitudinal study entitled “ISA: Physical Activity and Environment”³⁷, whose objective was to verify the relations between the built environment where people live and work and the practices of physical activity in the period of leisure, while commuting between environments and nutritional status. The city of São Paulo has 12,325,232 million inhabitants, with a population density of 7,398.26 inhabitants per km²³⁸.

A subsample of the ISA-2015 was used in a study called ISA-Nutrition³⁹, which, in addition to being added to the baseline of the survey (n = 4,043), was used in two other subsequent steps: application of dietary recall (n = 1,737) and blood sample collection, blood pressure measurement and anthropometric assessment (n = 901)³⁹. In short, the baseline sampling was carried out by clusters and stratified into two stages (urban census sectors and households), and all residents aged 12 years and over were invited to participate in the study⁴⁰.

Of the 901 subjects who participated in the ISA-Nutrition, 856 had their BM and height self-reported and measured. Self-reported measurements were obtained by the questions: “What is your weight?” and “How tall are you?”.

The BM and height of the participants were measured in duplicate, and in the face of a ≥5% difference, an additional measurement would be collected and the discrepant measurement, discarded.

Data collection was conducted by four trained nursing technicians with previous experience in anthropometric measurements. The training was based on a manual produced for this purpose⁴¹ and on the recommendations by the Ministry of Health⁴².

After the first week of data collection, a new training was carried out to standardize the procedures and solve possible queries. Additionally, there were daily meetings between the anthropometrists and the team of coordinators to verify the data collected, discuss difficulties and clarify queries.

During data collection, participants remained barefoot, wearing light clothing and no adornments that could interfere with the measurement³⁹. To measure the BM, a digital scale (Tanita^®, model HD-313, accurate to 100 g) was used, calibrated and checked daily, supported on a flat, firm, smooth surface, away from the wall. The individuals were positioned in the center of the scale, in orthostatic position, with feet parallel and together, and arms positioned along the body. To measure height, a portable stadiometer (Seca^®, model 208, with 0.1 cm precision) was used, fixed to a smooth wall and without a baseboard, with the subjects’ heads positioned in the Frankfürt plane, with heels, calves, buttocks, shoulders and back of the head touching the wall and the upper part of the head against the stadiometer shaft.

The average time elapsed between the reporting of measurements and actual measurement was 131.9 standard deviations = 118.5 days. The mean values of BM and height of each participant were used to calculate BMI, which was used to classify the nutritional status of the participants into three categories. Adolescents (12 to 17 years old) were classified as not overweight when BMI ≤ +1 standard deviation (SD) of the Z-score for BMI/age, overweight when +1 SD < BMI ≤ +2 SD, and obese when BMI > +2 DP⁴³. For adults and the elderly (18 years or older), the following cutoff points were adopted: BMI < 25 kg/m² = not overweight, 25 kg/m² ≤ BMI < 30 kg/m² = overweight, and BMI ≥ 30 kg /m² = obese⁴⁴.

Participants were considered physically active during leisure time when they performed at least 150 weekly minutes of moderate-intensity physical activity, 75 weekly minutes of vigorous activity or the equivalent combination of moderate and vigorous physical activity. Information about leisure-time physical activities was evaluated using the International Physical Activity Questionnaire, long version⁴⁵.

Participants were categorized into two cardiometabolic risk groups, having three or more of the following conditions as cutoff:

• obesity (BMI ≥ 30 kg/m² for adults⁴⁶ and BMI > +2 SD for adolescents);

• diabetes (fasting plasma glucose ≥ 126 mg/dL or drug treatment for diabetes) or insulin resistance (HOMA-IR ≥ 2.71)⁴⁷;

• hypertension (use of antihypertensive drugs or systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg for adults, or SBP or DBP > 95th percentile of sex, age and height for adolescents aged 12 and 13 years, or SBP ≥ 130 or DBP ≥ 80 for adolescents aged 14 to 19 years)⁴⁸;

• dyslipidemia (drug treatment for dyslipidemia or low-density lipoprotein cholesterol (LDL-C) ≥ 160 mg/dL for adults or LDL-C ≥ 130 mg/dL for adolescents, or high-density lipoprotein cholesterol (HDL-C) ≤ 40 mg/dL for men, ≤ 50 mg/dL for women, HDL-C < 45 for adolescents, or triglycerides ≥ 150 mg/dL for adults or ≥ 130 mg/dL for adolescents)⁴⁸.

After verifying the distribution of adherence to the normality curve, through the analysis of asymmetry and kurtosis, the descriptive parameters were presented as means and 95% confidence intervals.

The validity of self-reported measurements compared to measured BM and height was examined with the intraclass correlation coefficient (ICC), the Bland-Altman analysis and the paired t-test. For ICC classification, values lower than 0.4 were classified as poor agreement, 0.4 ≤ ICC < 0.6 reasonable agreement, 0.6 ≤ ICC < 0.75 good agreement, and e ≥0.75 excellent agreement⁴⁹.

Additionally, linear regression models were used to elaborate the calibration coefficients using the equation y = B₀ + B₁x, with y being the measured measure, x the reported measure, B₀ the linear coefficient and B₁ the angular coefficient. All analyses were stratified by sex, age group, nutritional status, level of leisure-time physical activity, and cardiometabolic risk.

Sensitivity and specificity stratified by sex, overweight and obesity were calculated for both the BMI based on the self-reported measurements and the calibrated BMI, using the measured value as reference. In addition, the proportions of overweight individuals were calculated based on the above-mentioned measures, and a proportion test was performed to verify the difference between groups, adopting a statistical significance level of p < 0.05.

This research was approved by the Research Ethics Committee of the Faculty of Public Health of Universidade de São Paulo (FSP-USP) (processes no. 32344014.3.3001.0086 and 30848914.7.0000.5421) and by the School of Arts, Sciences and Humanities of USP (EACH) (process No. 10396919.0.0000.5390). A written informed consent was obtained from all participants and, in the case of adolescents, from their guardians.

RESULTS

The final sample of the study had 856 individuals, with a mean age of 42.7 years (SD = 23.3 years, minimum = 12, maximum = 93), with 50.2% women. Of 856 participants, 24.6% performed at least 150 minutes/week of physical activity, 22.3% were considered obese, 16.9% for men and 27.7% for women (Table 1).

Table 1.
Mean and standard deviations of measured and self-reported body mass, height and body mass index and sample distribution according to age, physical activity, nutritional status and cardiometabolic risk factors. São Paulo, Brazil, 2015.

	Males (n = 426)		Females (n = 430)		Total (n = 856)
	n	%	n	%	n	%
Age (years)
12-17	107	25.12	110	25.58	217	25.35
18-39	86	20.19	90	20.93	176	20.56
40-59	88	20.66	83	19.30	171	19.98
60+	145	34.04	147	34.19	292	34.11
Leisure physical activity (min/week)
< 150	293	68.78	352	81.86	645	75.35
≥ 150	133	31.22	78	18.14	211	24.65
Nutritional status
No overweight	227	53.29	184	42.79	411	48.01
Overweight	127	29.81	127	29.53	254	29.67
Obesity	72	16.90	119	27.67	191	22.31
Cardiometabolic risk
less than 3 factors	301	70.66	276	64.19	577	67.41
3 or more factors	125	29.34	154	35.81	279	32.59

Table 2 shows the mean BM, height and measured and self-reported BMI according to age, physical activity level, nutritional status and cardiometabolic risk factors, stratified by sex. Significantly lower self-reported BM values were observed for adolescents and individuals with obesity.

Table 2.
Means of measured and reported body mass, height and body mass index according to age, physical activity, nutritional status and cardiometabolic risk factors, stratified by sex. São Paulo, Brazil, 2015 (n = 856).

			Males			Females
			Measured mean	Reported mean	p	Measured mean	Reported mean	p
Age (years)	12-17	BM	57.22	55.85	0.0318	58.02	56.15	0.0005
		Height	1.68	1.64	< 0.0001	1.60	1.58	0.0002
		BMI	20.16	20.73	0.0179	22.53	22.46	0.6875
	18-39	BM	71.77	74.02	0.0016	65.89	63.98	0.0079
		Height	1.76	1.75	0.5923	1.62	1.61	0.0829
		BMI	23.23	24.01	0.0028	25.13	24.84	0.1884
	40-59	BM	80.14	80.74	0.3345	69.74	68.21	0.0150
		Height	1.73	1.73	0.2691	1.58	1.59	0.0013
		BMI	26.82	26.98	0.9455	27.96	26.77	< 0.0001
	60 +	BM	76.28	77.54	0.0063	68.68	69.30	0.0941
		Height	1.69	1.70	0.0143	1.55	1.58	< 0.0001
		BMI	26.71	26.85	0.3333	28.51	27.98	0.0251
Leisure physical activity (min/week)	< 150	BM	73.31	74.39	0.0036	65.71	64.81	0.0078
		Height	1.71	1.71	0.3296	1.58	1.59	0.3589
		BMI	25.05	25.48	0.0056	26.24	25.80	0.0016
	≥ 150	BM	66.73	66.66	0.7551	64.78	63.55	0.0059
		Height	1.70	1.69	0.0006	1.58	1.59	0.1915
		BMI	22.84	23.23	0.0560	25.73	25.10	0.0125
Nutritional Status	No overweight	BM	60.26	62.35	< 0.0001	52.96	53.12	0.4715
		Height	1.70	1.69	0.0001	1.59	1.58	0.003
		BMI	20.68	21.71	< 0.0001	20.86	21.31	0.0035
	Overweight	BM	77.75	77.46	0.4140	67.09	65.86	0.0097
		Height	1.70	1.71	0.3333	1.58	1.59	0.0009
		BMI	26.68	26.45	0.0651	26.95	25.98	< 0.0001
	Obese	BM	94.01	90.59	0.0018	82.96	79.95	0.0017
		Height	1.73	1.71	0.1812	1.57	1.59	0.0444
		BMI	32.00	30.88	0.0083	33.48	31.88	< 0.0001
Cardiometabolic risk	Less than 3 factors	BM	65.93	66.94	0.0046	59.45	58.80	0.0075
		Height	1.71	1.70	0.0030	1.58	1.58	0.5210
		BMI	22.56	23.09	0.0001	23.63	23.51	0.1781
	3 or more factors	BM	84.61	84.12	0.2964	76.25	74.92	0.0250
		Height	1.71	1.71	0.4850	1.58	1.59	0.0007
		BMI	28.71	28.76	0.6735	30.64	29.61	< 0.0001

• BM: body mass; BMI: body mass index.

Among men, there was an overestimation of BM by individuals aged 18 to 39 years and 60 years or more, among those who practiced less than 150 minutes/week of physical activity, without excess weight and with less than three cardiometabolic risk factors. In general, BMI showed a similar behavior to the BM, except among male adolescents, who showed an underestimation of BM and an overestimation of BMI. For women, in all situations in which there was a significant difference between measured and self-reported BM, such difference pointed to an underestimation of self-reported BM. In both sexes, there was an overestimation of BMI measures reported among those classified as not overweight and underestimation among those with obesity. Regarding height, adolescents reported lower values than the measured, while the elderly reported higher values, in both sexes.

Men with less than three cardiometabolic risk factors overestimated BM and BMI and underestimated height, while women with three or more risk factors had the opposite behavior.

The Bland-Altman analyzes show excellent agreement rates between all reported and measured measures, for both sexes, with mean differences very close to zero and few cases outside the 95% agreement limits (Figure 1).

Figure 1.
Bland-Altman analysis for the agreement of measured and reported body mass, height and body mass index between men and women. São Paulo, Brazil, 2015.

When evaluating the intraclass correlation, among men, all categories evaluated had excellent ICC values (≥ 0.75) (Supplemental Material 1). As for women, the ICC values (≥ 0.6) for height were considered good among elderly women, among women who practiced less than 150 min/week of leisure-time physical activity, and among overweight women. In contrast, the ICC value of 0.587 for overweight women is considered reasonable (0.4 ≤ ICC < 0.6) (Supplemental Material 1).

As for the calibration coefficients stratified by sex and by other variables of interest in the study, most of the angular coefficients (B1) are close to 1 (Supplementary Material 1), indicating good equivalence with the measured values⁵⁰. Similar values were also found for samples stratified only by sex, being for men B₀ = -0.17961 and B₁ = 0.9935 and for women B₀ = 1.5614 and B₁ = 0.9576. After applying these last two equations, individuals were classified using two independent cutoff points: the first referring to people with excess weight (overweight + obesity) and the second referring to people with obesity.

After calibrating the measurements, the mean BMI of the measured and calibrated values remained statistically similar to each other for all subgroups evaluated (Supplementary Material 2). Without the calibrations, several differences were found in the means of BMI for most of the groups evaluated.

Based on the BMI data, according to each of the three methods (measured, self-reported and calibrated BMI), the proportions of overweight and obesity were evaluated, as well as the sensitivity and specificity of each measure (Table 3). There was no significant difference between the proportions obtained based by self-report and calibrated values when compared to the ratings based on actual measured values. However, for all cases presented, the proportions found using the calibrated measures tended to be closer to the measured value compared to the prevalence based on the reported values.

Table 3.
Proportion of overweight and obese individuals, based on measured, self-reported and calibrated measurements, plus sensitivity and specificity analyses of body mass index stratified by sex. São Paulo, Brazil, 2015 (n = 856).

		Overweight (overweight+obesity) (%)		Obesity (%)
Females	Measured	57.72		28.22
	Reported	53.24		25.06
	Calibrated	57.05		26.40
Males	Measured	47.62		16.67
	Reported	50.11		18.37
	Calibrated	46.70		16.33
		Sensitivity	Specificity	Sensitivity	Specificity
Females	Reported	0.841	0.888	0.760	0.950
	Calibrated	0.881	0.846	0.784	0.940
Males	Reported	0.876	0.840	0.819	0.947
	Calibrated	0.837	0.870	0.736	0.955

After calibrating the above-mentioned measures, there was an increase in the sensitivity in the classification of women for overweight (from 0.84 to 0.88) and for obesity (from 0.76 to 0.78) (Table 3). These results indicate that a greater number of women were correctly classified in these two categories (true positive). In contrast, specificity values decreased after calibration (from 0.89 to 0.85 and from 0.95 to 0.94, respectively) (true negative). Conversely, for men, the calibrated values decreased sensitivity and increased specificity.

DISCUSSION

The main results of this study show that self-reported BM and height values tend to be similar to the measured values, however the calibration process makes them even more consistent. Furthermore, by using the BMI results categorically, the calibration process increased the sensitivity of nutritional status classification among women and increased specificity among men.

Specifically for height, regardless of gender, an underestimation was observed among adolescents, which corroborates other studies involving boys^51,52,53, but contradicts studies carried out with girls^54,55. Among the main reasons for these inconsistencies among adolescents, Enes et al. highlight that adolescents may not be aware of their current measurements, as they tend to measure themselves infrequently and may only remember outdated values⁵⁵ or be affected by the rapid morphological and psychosocial changes of this stage of life⁵⁶. Finally, the social desirability bias in this group may fluctuate, since body image dissatisfaction can vary according to nutritional status among adolescents⁵⁷.

Among the elderly, both men and women overestimated their height in accordance with other studies involving the elderly^58,59. Probable reasons for this finding would be the lack of knowledge of a possible natural reduction in height with aging^60,61 and the tendency of the elderly to report heights measured during their adult/youth life¹². Furthermore, the higher values reported by older women, when compared to older men, may also be related to a possible osteoporosis, which tends to affect women more than men⁶².

Regarding BM, for women, in all situations where there was a significant difference between measured and self-reported MC, there was an underestimation in self-reported BM. This consistent underestimation of BM was also pointed out by a systematic review study⁸ that evaluated 60 studies and, in 51 of them, women underestimated BM, with mean differences ranging between 0.1 and 3.4 kg. Among men, 39 studies reported underestimation of BM, with mean differences ranging from 0.1 to 2.2 kg. Part of this phenomenon can be explained by social desirability, when subjects report a BM (or height) value that complies with a social norm, even if the reported value is imprecise^3,63.

From the perspective of BMI, there was an overestimation of the BMI measures reported in both sexes among those classified as not overweight and underestimation among those with obesity. This underestimation is supported by several other studies in different populations^7,8,64. However, it is noteworthy that these gross differences in BMI based on self-reported and measured values remained small for both sexes (1.6 kg/m² for women and 1.12 kg/m² for men).

In a study involving 1,061 university students, those who reported BM more accurately also had a significantly higher metabolic equivalent (MET)-minute/week of physical activity, when compared to those who overestimated it²⁶. When stratified by sex, men who overestimated BM had lower levels of vigorous activity and MET-minute/week, and among women, those who overestimated BM had lower levels of moderate activity and MET-minute/week. This study also identified that, based on the prediction of participants’ cardiorespiratory fitness, there is a tendency to underestimate the BM and overestimate self-reported levels of physical activity. On the other hand, in this study, men who practiced less than 150 minutes of leisure-time physical activity per week overestimated their BMI, while women, regardless of the level of physical activity, underestimated it.

As for the differences according to cardiometabolic risk factors, there was greater precision in the information provided by men with three or more risk factors compared to men with less than three. This data may be a consequence of greater monitoring of health by those at higher cardiometabolic risk (consequently, greater awareness of their BM and height), since it is common that men have low demand for health services in Brazil^65,66. Among women, the result was the opposite: all underestimated their BM, but those with three or more cardiometabolic risk factors did it to a greater extent, in addition to overestimating height, which resulted in an underestimation of BMI in this group. This may have been a consequence, among other factors, of a report influenced by greater social and health professional pressure for women to have a lower BMI⁶⁷.

Although no significant differences were found in the proportions of individuals classified as overweight or obese, the proportions found after measures were calibrated tended to be closer to the measured value, in comparison with the prevalence based on reported results. In this sense, it is important to highlight that, in epidemiological studies involving large samples, a small difference in proportions can have great impact.

After calibrating the above measures, there was an increase in sensitivity in the classification of women, both for overweight and obesity, and a reduction in specificity. For men, the calibrated values decreased sensitivity and increased specificity. The implications of these results must take into account the reasons and motivations for using the BMI variable. For example, if BMI is used as an outcome and the main objective of the study is to investigate factors associated with overweight or obesity, measures with greater sensitivity should be chosen. However, if the objective is to assess positive health habits to maintain adequate nutritional status, measures with greater specificity should be preferred.

Finally, this study has some limitations that are worth mentioning. The first is related to the time elapsed between reported measures and actual measurements (131.9 SD = 118.5 days), as there could have been changes between the two moments. However, intervals longer than 30 days between measurements did not result in greater differences in BMI (-0.03 95%CI -1.13 - 0.66 vs 0.10 95%CI -0.26 - 0.47). Similar results were also found among adolescents, who would be the most susceptible to this type of bias. Another limitation of the study is the assessment of physical activity through questionnaires, which are susceptible to response bias, which may even vary according to the respondent’s nutritional status⁶⁸. However, to mitigate this issue, the analyses were performed in a stratified manner according to nutritional status and time of physical activity analyzed in a dichotomous way.

Based on these facts, it is possible to conclude that the self-reported measurements of height and BM and, consequently, the BMI are valid and reliable measurements, showing substantial improvements after calibration.

The results of this study also suggest that the calibration equations for self-reported values tend to approximate the proportions of people classified as overweight and obese to the proportions based on the measured values. Finally, the calibrated measures increased the sensitivity for nutritional status classification among women and specificity among men living in the city of São Paulo. Thus, given the easy logistics, cost and time reduction for data collection, the use of self-reported height and BM measurements is recommended for epidemiological studies; there is also indication for application of calibration coefficients for each specific group to improve the accuracy of measures.

ACKNOWLEDGMENTS

The authors are grateful to everyone who, directly or indirectly, was involved with the São Paulo Health Surveys, especially the ISA-2015 Nutrition Study Group (Marcelo Macedo Rogero, PhD; Flávia Mori Sarti, PhD; Chester Luis Luis Galvão Cesar, PhD; Moises Goldbaum, PhD; Maria Cecília Goi Porto Alves, PhD) and Marilisa Berti de Azevedo Barros, PhD.

REFERÊNCIAS

• ¹
  The Global BMI Mortality Collaboration. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet 2016; 388(10046): 776-86. https://doi.org/10.1016/S0140-6736(16)30175-1
  » https://doi.org/https://doi.org/10.1016/S0140-6736(16)30175-1
• ²
  Peixoto MRG, Benício MHD, Jardim PCB. Validade do peso e da altura autorreferidos: o estudo de Goiânia. Rev Saúde Pública 2006; 40(6): 1065-72. https://doi.org/10.1590/S0034-89102006000700015
  » https://doi.org/https://doi.org/10.1590/S0034-89102006000700015
• ³
  Burke MA, Carman KG. You can be too thin (but not too tall): Social desirability bias in self-reports of weight and height. Econ Hum Biol 2017; 27(Parte A): 198-222. https://doi.org/10.1016/j.ehb.2017.06.002
  » https://doi.org/https://doi.org/10.1016/j.ehb.2017.06.002
• ⁴
  Madrigal H, Sánchez-Villegas A, Martínez-González MA, Kearney J, Gibney MJ, De Irala J, et al. Underestimation of body mass index through perceived body image as compared to self-reported body mass index in the European Union. Public Health 2000; 114(6): 468-73. https://doi.org/10.1038/sj.ph.1900702
  » https://doi.org/https://doi.org/10.1038/sj.ph.1900702
• ⁵
  Anai A, Ueda K, Harada K, Katoh T, Fukumoto K, Wei CN. Determinant factors of the difference between self-reported weight and measured weight among Japanese. Environ Health Prev Med 2015; 20: 447-54. https://doi.org/10.1007/s12199-015-0489-8
  » https://doi.org/https://doi.org/10.1007/s12199-015-0489-8
• ⁶
  Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr 2002; 5(4): 561-5. https://doi.org/10.1079/phn2001322
  » https://doi.org/https://doi.org/10.1079/phn2001322
• ⁷
  Gorber SC, Tremblay M, Moher D, Gorber B. A comparison of direct vs. self-report measures for assessing height, weight and body mass index: A systematic review. Obes Rev 2007; 8(4): 307-26. https://doi.org/10.1111/j.1467-789x.2007.00347.x
  » https://doi.org/https://doi.org/10.1111/j.1467-789x.2007.00347.x
• ⁸
  Maukonen M, Männistö S, Tolonen H. A comparison of measured versus self-reported anthropometrics for assessing obesity in adults: a literature review. Scand J Public Health 2018; 46(5): 565-79. https://doi.org/10.1177/1403494818761971
  » https://doi.org/https://doi.org/10.1177/1403494818761971
• ⁹
  Gillum RF, Sempos CT. Ethnic variation in validity of classification of overweight and obesity using self-reported weight and height in American women and men: The Third National Health and Nutrition Examination Survey. Nutr J 2005; 4: 27. https://doi.org/10.1186/1475-2891-4-27
  » https://doi.org/https://doi.org/10.1186/1475-2891-4-27
• ¹⁰
  Kuczmarski MF, Kuczmarski RJ, Najjar M. 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. J Am Diet Assoc 2001; 101(1): 28-34. https://doi.org/10.1016/S0002-8223(01)00008-6
  » https://doi.org/https://doi.org/10.1016/S0002-8223(01)00008-6
• ¹¹
  Merrill RM, Richardson JS. Validity of self-reported height, weight, and body mass index: Findings from the national health and nutrition examination survey, 2001-2006. Prev Chronic Dis 2009; 6(4): A121.
• ¹²
  Rowland ML. Self-reported weight and height. Am J Clin Nutr 1990; 52(6): 1125-33. https://doi.org/10.1093/ajcn/52.6.1125
  » https://doi.org/https://doi.org/10.1093/ajcn/52.6.1125
• ¹³
  Sahyoun NR, Maynard LM, Zhang XL, Serdula MK. Factors associated with errors in self-reported height and weight in older adults. J Nutr Heal Aging 2008; 12: 108-15. https://doi.org/10.1007/BF02982562
  » https://doi.org/https://doi.org/10.1007/BF02982562
• ¹⁴
  Villanueva E V. The validity of self-reported weight in US adults: A population based cross-sectional study. BMC Public Health 2001; 1: 11. https://doi.org/10.1186/1471-2458-1-11
  » https://doi.org/https://doi.org/10.1186/1471-2458-1-11
• ¹⁵
  Cozma I, Kukaswadia A, Janssen I, Craig W, Pickett W. Active transportation and bullying in Canadian schoolchildren: A cross-sectional study. BMC Public Health 2015; 15: 99. https://doi.org/10.1186/s12889-015-1466-2
  » https://doi.org/https://doi.org/10.1186/s12889-015-1466-2
• ¹⁶
  Craig BM, Adams AK. Accuracy of body mass index categories based on self-reported height and weight among women in the United States. Matern Child Health J 2009; 13: 489-96. https://doi.org/10.1007/s10995-008-0384-7
  » https://doi.org/https://doi.org/10.1007/s10995-008-0384-7
• ¹⁷
  Wen M, Kowaleski-Jones L. Sex and ethnic differences in validity of self-reported adult height, weight and body mass index. Ethn Dis 2012; 22(1): 72-8. https://doi.org/10.13016/0sok-jed1
  » https://doi.org/https://doi.org/10.13016/0sok-jed1
• ¹⁸
  Stommel M, Schoenborn CA. Accuracy and usefulness of BMI measures based on self-reported weight and height: Findings from the NHANES & NHIS 2001-2006. BMC Public Health 2009; 9: 421. https://doi.org/10.1186/1471-2458-9-421
  » https://doi.org/https://doi.org/10.1186/1471-2458-9-421
• ¹⁹
  Lin CJ, Deroo LA, Jacobs SR, Sandler DP. Accuracy and reliability of self-reported weight and height in the Sister Study. Public Health Nutr 2012; 15(6): 989-9. https://dx.doi.org/10.1017%2FS1368980011003193
  » https://doi.org/https://dx.doi.org/10.1017%2FS1368980011003193
• ²⁰
  May AM, Barnes DR, Forouhi NG, Luben R, Khaw KT, Wareham NJ, et al. Prediction of measured weight from self-reported weight was not improved after stratification by body mass index. Obesity 2013; 21(1): E137-E142. https://doi.org/10.1002/oby.20141
  » https://doi.org/https://doi.org/10.1002/oby.20141
• ²¹
  Großschädl F, Haditsch B, Stronegger WJ. Validity of self-reported weight and height in Austrian adults: Sociodemographic determinants and consequences for the classification of BMI categories. Public Health Nutr 2012; 15(1): 20-7. https://doi.org/10.1017/S1368980011001911
  » https://doi.org/https://doi.org/10.1017/S1368980011001911
• ²²
  Finardi P, Nickel CH, Koller MT, Bingisser R. Accuracy of self-reported weight in a high risk geriatric population in the emergency department. Swiss Med Wkly 2012; 142: w13585. https://doi.org/10.4414/smw.2012.13585
  » https://doi.org/https://doi.org/10.4414/smw.2012.13585
• ²³
  Dahl AK, Hassing LB, Fransson EI, Pedersen NL. Agreement between self-reported and measured height, weight and body mass index in old age-a longitudinal study with 20 years of follow-up. Age Ageing 2010; 39(4): 445-51. https://doi.org/10.1093/ageing/afq038
  » https://doi.org/https://doi.org/10.1093/ageing/afq038
• ²⁴
  McAdams MA, Van Dam RM, Hu FB. Comparison of self-reported and measured BMI as correlates of disease markers in U.S. adults. Obesity 2007; 15(1): 188. https://doi.org/10.1038/oby.2007.504
  » https://doi.org/https://doi.org/10.1038/oby.2007.504
• ²⁵
  Xie YJ, Ho SC, Liu ZM, Hui SS-C. Comparisons of Measured and Self-Reported Anthropometric Variables and Blood Pressure in a Sample of Hong Kong Female Nurses. PLoS One 2014; 9: e107233. https://doi.org/10.1371/journal.pone.0107233
  » https://doi.org/https://doi.org/10.1371/journal.pone.0107233
• ²⁶
  Wilson OWA, Bopp CM, Papalia Z, Bopp M. Objective vs self-report assessment of height, weight and body mass index: Relationships with adiposity, aerobic fitness and physical activity. Clin Obes 2019; 9(5): e12331. https://doi.org/10.1111/cob.12331
  » https://doi.org/https://doi.org/10.1111/cob.12331
• ²⁷
  Keith SW, Fontaine KR, Pajewski NM, Mehta T, Allison DB. Use of self-reported height and weight biases the body mass index-mortality association. Int J Obes 2011; 35: 401-8. https://doi.org/10.1038/ijo.2010.148
  » https://doi.org/https://doi.org/10.1038/ijo.2010.148
• ²⁸
  Brasil. Ministério da Saúde. DATASUS. Sistemas e Aplicativos Epidemiológicos. HIPERDIA - Sistema de Cadastramento e Acompanhamento de Hipertensos e Diabéticos [Internet]. Brasil: Ministério da Saúde [acessado em 1º dez. 2020]. Disponível em: Disponível em: http://datasus1.saude.gov.br/sistemas-e-aplicativos/epidemiologicos/hiperdia
  » http://datasus1.saude.gov.br/sistemas-e-aplicativos/epidemiologicos/hiperdia
• ²⁹
  Phelan SM, Burgess DJ, Yeazel MW, Hellerstedt WL, Griffin JM, Van Ryn M. Impact of weight bias and stigma on quality of care and outcomes for patients with obesity. Obes Rev 2015; 16(4): 319-26. https://doi.org/10.1111/obr.12266
  » https://doi.org/https://doi.org/10.1111/obr.12266
• ³⁰
  Carvalho AM, Piovezan LG, Castro Selem SS, Fisberg RM, Marchioni DML. Validação e calibração de medidas de peso e altura autorreferidas por indivíduos da cidade de São Paulo. Rev Bras Epidemiol 2014; 17(3): 735-46. https://doi.org/10.1590/1809-4503201400030013
  » https://doi.org/https://doi.org/10.1590/1809-4503201400030013
• ³¹
  Ferriani LO, Coutinho ESF, Silva DA, Faria CP, Molina MDCB, Benseñor IJM, et al. Subestimativa de obesidade e sobrepeso a partir de medidas autorrelatadas na população geral: prevalência e proposta de modelos para correção. Cad Saude Publica 2019; 35(6): e00065618. https://doi.org/10.1590/0102-311X00065618
  » https://doi.org/https://doi.org/10.1590/0102-311X00065618
• ³²
  Neermark S, Holst C, Bisgaard T, Bay-Nielsen M, Becker U, Tolstrup JS. Validation and calibration of self-reported height and weight in the Danish Health Examination Survey. Eur J Public Health 2019; 29(2): 291-6. https://doi.org/10.1093/eurpub/cky187
  » https://doi.org/https://doi.org/10.1093/eurpub/cky187
• ³³
  Bolton-Smith C, Woodward M, Tunstall-Pedoe H, Morrison C. Accuracy of the estimated prevalence of obesity from self reported height and weight in an adult Scottish population. J Epidemiol Community Heal 2000; 54: 143-8. https://doi.org/10.1136/jech.54.2.143
  » https://doi.org/https://doi.org/10.1136/jech.54.2.143
• ³⁴
  Hayes AJ, Clarke PM, Lung TWC. Change in bias in self-reported body mass index in Australia between 1995 and 2008 and the evaluation of correction equations. Popul Health Metr 2011; 9: 53. https://doi.org/10.1186/1478-7954-9-53
  » https://doi.org/https://doi.org/10.1186/1478-7954-9-53
• ³⁵
  Kuskowska-Wolk A, Rössner S. The “True” Prevalence of Obesity: A comparison of objective weight and height measures versus self-reported and calibrated data. Scand J Prim Health Care 1989; 7(2): 79-82. https://doi.org/10.3109/02813438909088651
  » https://doi.org/https://doi.org/10.3109/02813438909088651
• ³⁶
  Prefeitura de São Paulo. Inquérito de Saúde do Município de São Paulo (ISA-Capital) [Internet]. São Paulo: Prefeitura de São Paulo; 2016 [acessado em 1º dez. 2020]. Disponível em: Disponível em: http://www.prefeitura.sp.gov.br/cidade/secretarias/saude/epidemiologia_e_informacao/isacapitalsp/index.php?p=216392
  » http://www.prefeitura.sp.gov.br/cidade/secretarias/saude/epidemiologia_e_informacao/isacapitalsp/index.php?p=216392
• ³⁷
  Florindo AA, Teixeira IP, Barrozo LV, Sarti FM, Fisberg RM, Andrade DR, et al. Study protocol: health survey of Sao Paulo: ISA-Physical Activity and Environment. BMC Public Health 2021; 21: 283. https://doi.org/10.1186/s12889-021-10262-5
  » https://doi.org/https://doi.org/10.1186/s12889-021-10262-5
• ³⁸
  Instituto Brasileiro de Geografia e Estatística (IBGE). IBGE cidades [Internet]. 2020 [acessado em 1º dez. 2020]. Disponível em: Disponível em: https://cidades.ibge.gov.br/brasil/sp/sao-paulo/panorama
  » https://cidades.ibge.gov.br/brasil/sp/sao-paulo/panorama
• ³⁹
  Fisberg RM, Sales CH, De Mello Fontanelli M, Pereira JL, Alves MCGP, Escuder MML, et al. 2015 health survey of São Paulo with focus in nutrition: Rationale, design, and procedures. Nutrients 2018; 10(2): 169. https://doi.org/10.3390/nu10020169
  » https://doi.org/https://doi.org/10.3390/nu10020169
• ⁴⁰
  Alves MCGP, Escuder MML, Goldbaum M, Barros MBA, Fisberg RM. Sampling plan in health surveys, city of São Paulo, Brazil, 2015. Rev Saude Publica 2018; 52: 81. https://doi.org/10.11606/s1518-8787.2018052000471
  » https://doi.org/https://doi.org/10.11606/s1518-8787.2018052000471
• ⁴¹
  Fisberg RM, Marchioni DML. Manual de avaliação do consumo alimentar em estudos populacionais: a experiência do Inquérito de Saúde de São Paulo. São Paulo: Faculdade de Saúde Pública, Universidade de São Paulo; 2012. v. 1. 197 p.
• ⁴²
  Brasil. Ministério da Saúde. Orientações para a coleta e análise de dados antropométricos em serviços de saúde: norma técnica do Sistema de Vigilância Alimentar e Nutricional - SISVAN. Brasil: Ministério da Saúde ; 2011. 76 p.
• ⁴³
  Organização Mundial da Saúde. Growth reference data for 5-19 years [Internet]. Organização Mundial da Saúde; 2007 [acessado em 1º dez. 2020]. Disponível em: Disponível em: https://www.who.int/growthref/en/
  » https://www.who.int/growthref/en/
• ⁴⁴
  Organização Mundial da Saúde. Obesity: preventing and managing the global epidemic. Organização Mundial da Saúde; 2000
• ⁴⁵
  Hallal PC, Gomez LF, Parra DC, Lobelo F, Mosquera J, Florindo AA, et al. Lessons Learned After 10 Years of IPAQ Use in Brazil and Colombia. J Phys Act Heal 2010; 7(Suppl. 2): S259-S264. https://doi.org/10.1123/jpah.7.s2.s259
  » https://doi.org/https://doi.org/10.1123/jpah.7.s2.s259
• ⁴⁶
  World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO Consultation. WHO; 2000.
• ⁴⁷
  Sociedade Brasileira de Diabetes. Diretrizes da Sociedade Brasileira de Diabetes 2019-2020 [Internet]. Sociedade Brasileira de Diabetes; 2019 [acessado em 1 dez. 2020]. Disponível em: Disponível em: https://www.diabetes.org.br/profissionais/images/DIRETRIZES-COMPLETA-2019-2020.pdf
  » https://www.diabetes.org.br/profissionais/images/DIRETRIZES-COMPLETA-2019-2020.pdf
• ⁴⁸
  Précoma DB, Oliveira GMM, Simão AF, Dutra OP, Coelho OR, Izar MC de O, et al. Updated cardiovascular prevention guideline of the Brazilian society of cardiology - 2019. Arq Bras Cardiol 2019; 113(4): 787-891. https://doi.org/10.5935/abc.20190204
  » https://doi.org/https://doi.org/10.5935/abc.20190204
• ⁴⁹
  Cicchetti DV. Guidelines, Criteria, and Rules of Thumb for Evaluating Normed and Standardized Assessment Instruments in Psychology. Psychol Assess 1994; 6(4): 284-90. https://doi.org/10.1037/1040-3590.6.4.284
  » https://doi.org/https://doi.org/10.1037/1040-3590.6.4.284
• ⁵⁰
  Kynast-Wolf G, Becker N, Kroke A, Brandstetter BR, Wahrendorf J, Boeing H. Linear Regression Calibration: Theoretical Framework and Empirical Results in EPIC, Germany. Ann Nutr Metab 2002; 46: 2-8. https://doi.org/10.1159/000046746
  » https://doi.org/https://doi.org/10.1159/000046746
• ⁵¹
  Jansen W, van de Looij-Jansen PM, Ferreira I, de Wilde EJ, Brug J. Differences in Measured and Self-Reported Height and Weight in Dutch Adolescents. Ann Nutr Metab 2006; 50: 339-46. https://doi.org/10.1159/000094297
  » https://doi.org/https://doi.org/10.1159/000094297
• ⁵²
  Rasmussen M, Holstein BE, Melkevik O, Damsgaard MT. Validity of self-reported height and weight among adolescents: The importance of reporting capability. BMC Med Res Methodol 2013; 13: 85. https://doi.org/10.1186/1471-2288-13-85
  » https://doi.org/https://doi.org/10.1186/1471-2288-13-85
• ⁵³
  Fortenberry JD. Reliability of adolescents’ reports of height and weight. J Adolesc Heal 1992; 13(2): 114-7. https://doi.org/10.1016/1054-139X(92)90076-N
  » https://doi.org/https://doi.org/10.1016/1054-139X(92)90076-N
• ⁵⁴
  Aasvee K, Rasmussen M, Kelly C, Kurvinen E, Giacchi MV, Ahluwalia N. Validity of self-reported height and weight for estimating prevalence of overweight among Estonian adolescents: The Health Behaviour in School-aged Children study. BMC Res Notes 2015; 8: 606. https://doi.org/10.1186/s13104-015-1587-9
  » https://doi.org/https://doi.org/10.1186/s13104-015-1587-9
• ⁵⁵
  Enes CC, Fernandez PMF, Voci SM, Toral N, Romero A, Slater B. Validity and reliability of self-reported weight and height measures for the diagnoses of adolescent’s nutritional status. Rev Bras Epidemiol 2009; 12(4): 627-35. https://doi.org/10.1590/s1415-790x2009000400012
  » https://doi.org/https://doi.org/10.1590/s1415-790x2009000400012
• ⁵⁶
  de Castro IRR, Levy RB, Cardoso L de O, dos Passos MD, Sardinha LMV, Tavares LF, et al. Body image, nutritional status and practices for weight control among Brazilian adolescents. Ciênc e Saúde Coletiva 2010; 15(Suppl. 2): 3099-188. https://doi.org/10.1590/s1413-81232010000800014
  » https://doi.org/https://doi.org/10.1590/s1413-81232010000800014
• ⁵⁷
  Marques MI, Pimenta J, Reis S, Ferreira LM, Peralta L, Santos MI, et al. (In)Satisfação com a imagem corporal na adolescência. Nascer Crescer 2016; 25(4): 217-21.
• ⁵⁸
  Gunnell DJ, Berney L, Holland P, Maynard M, Blane D, Frankel S, et al. How accurately are height, weight and leg length reported by the elderly, and how closely are they related to measurements recorded in childhood? Int J Epidemiol 2000; 29(3): 456-64. https://doi.org/10.1093/intjepid/29.3.456
  » https://doi.org/https://doi.org/10.1093/intjepid/29.3.456
• ⁵⁹
  Niedźwiedzka E, Długosz A, Wądołowska L. Validity of self-reported height and weight in elderly Poles. Nutr Res Pract 2015; 9(3): 319-27. https://doi.org/10.4162/nrp.2015.9.3.319
  » https://doi.org/https://doi.org/10.4162/nrp.2015.9.3.319
• ⁶⁰
  Brasil. Ministério da Saúde. Envelhecimento e saúde da pessoa idosa. Brasil: Ministério da Saúde ; 2006. Cadernos de Atenção Básica.
• ⁶¹
  Sorkin JD, Muller DC, Andres R. Longitudinal change in the heights of men and women: Consequential effects on body mass index. Epidemiol Rev 1999; 21(2): 247-60. https://doi.org/10.1093/oxfordjournals.epirev.a018000
  » https://doi.org/https://doi.org/10.1093/oxfordjournals.epirev.a018000
• ⁶²
  Wright NC, Looker AC, Saag KG, Curtis JR, Delzell ES, Randall S, et al. The recent prevalence of osteoporosis and low bone mass in the United States based on bone mineral density at the femoral neck or lumbar spine. J Bone Miner Res 2014; 29(11): 2520-6. https://doi.org/10.1002/jbmr.2269
  » https://doi.org/https://doi.org/10.1002/jbmr.2269
• ⁶³
  Lu S, Su J, Xiang Q, Zhou J, Wu M. Accuracy of self-reported height, weight, and waist circumference in a general adult Chinese population. Popul Health Metr 2016; 14: 30. https://doi.org/10.1186/s12963-016-0099-8
  » https://doi.org/https://doi.org/10.1186/s12963-016-0099-8
• ⁶⁴
  Flegal KM, Ogden CL, Fryar C, Afful J, Klein R, Huang DT. Comparisons of Self-Reported and Measured Height and Weight, BMI, and Obesity Prevalence from National Surveys: 1999-2016. Obesity 2019; 27(10): 1711-9. https://doi.org/10.1002/oby.22591
  » https://doi.org/https://doi.org/10.1002/oby.22591
• ⁶⁵
  Figueiredo W. Assistência à saúde dos homens: um desafio para os serviços de atenção primária. Ciênc Saúde Coletiva 2005; 10(1): 105-9. http://dx.doi.org/10.1590/S1413-81232005000100017
  » https://doi.org/http://dx.doi.org/10.1590/S1413-81232005000100017
• ⁶⁶
  Pinheiro RS, Viacava F, Travassos C, Brito AS. Gênero, morbidade, acesso e utilização de serviços de saúde no Brasil. Ciênc Saúde Coletiva 2002; 7(4): 687-707. http://dx.doi.org/10.1590/S1413-81232002000400007
  » https://doi.org/http://dx.doi.org/10.1590/S1413-81232002000400007
• ⁶⁷
  Paim MB, Kovaleski DF. Análise das diretrizes brasileiras de obesidade: patologização do corpo gordo, abordagem focada na perda de peso e gordofobia. Saúde Soc 2020; 29(1): e190227. http://dx.doi.org/10.1590/s0104-12902020190227
  » https://doi.org/http://dx.doi.org/10.1590/s0104-12902020190227
• ⁶⁸
  Warner ET, Wolin KY, Duncan DT, Heil DP, Askew S, Bennett GG. Differential Accuracy of Physical Activity Self-Report by Weight Status. Am J Health Behav 2012; 36(2): 168-78. http://dx.doi.org/10.5993/AJHB.36.2.3
  » https://doi.org/http://dx.doi.org/10.5993/AJHB.36.2.3

• Financial support: 1. Study supported by Process nº 2017/17049-3, São Paulo Research Foundation (FAPESP). 2. São Paulo Municipal Health Department (process no. 2013-0.235.936-0). 3. National Council for Scientific and Technological Development - CNPq (concession number 472873/2012-1 and process number 306635/2016-0). 4. São Paulo Research Foundation (FAPESP) (Processes 2020/01312-0, 2017/17049-3, 2012/22113-9 e 2017/05125-7).

Publication Dates

• Publication in this collection
  09 Aug 2021
• Date of issue
  2021

History

• Received
  11 Feb 2021
• Accepted
  12 Apr 2021