SciELO - Scientific Electronic Library Online

vol.91 issue4Predictors of quality of life change after an acute coronary eventSlow breathing test increases the suspicion of white-coat hypertension in the office author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



  • text in Portuguese
  • English (pdf) | Portuguese (pdf)
  • Article in xml format
  • How to cite this article
  • SciELO Analytics
  • Curriculum ScienTI
  • Automatic translation


Related links


Arquivos Brasileiros de Cardiologia

Print version ISSN 0066-782XOn-line version ISSN 1678-4170

Arq. Bras. Cardiol. vol.91 no.4 São Paulo Oct. 2008 



Prevalence of arterial hypertension and associated factors in adults in São Luís, state of Maranhão



José Bonifácio Barbosa; Antonio Augusto Moura da Silva; Alcione Miranda dos Santos; Francisco das Chagas Monteiro Júnior; Márcio Mesquita Barbosa; Marcelo Mesquita Barbosa; José Albuquerque de Figueiredo Neto; Nivaldo de Jesus S. Soares; Vinicius José da Silva Nina; José Nicodemo Barbosa

Faculdade de Medicina da Universidade Federal do Maranhão (UFMA)/UDI Hospital, São Luís, MA - Brasil

Mailing Address




BACKGROUND: Little is known about the prevalence of arterial hypertension (AH) and its risk factors in the less developed regions of Brazil.
OBJECTIVE: To estimate the prevalence of arterial hypertension and its associated factors in the population > 18 years in São Luís, state of Maranhão according to the Seventh Report of the Joint National Committee (JNC 7) criteria.
METHODS: A cross-sectional study was conducted in São Luís, MA, from February to March 2003, with 835 individuals >18 years who completed a structured household questionnaire. Measurements of arterial pressure (AP), weight, height and waist circumference were taken, and other risk factors for cardiovascular disease were assessed. The Poisson regression method was used for the identification of factors associated with AH, with an estimate of the prevalence ratio (PR) and its corresponding 95% confidence interval.
RESULTS: Age varied from 18 and 94 years (mean age was 39.4 years), 293 (35.1%) individuals were normotensive and 313 (37.5%) were pre-hypertensive. The AH prevalence was 27.4% (95% CI - 24.4% to 30.6%), and it was higher among men (32.1%) than among women (24.2%). In the adjusted analysis, the following remained independently associated with AH: male gender (PR 1.52, 95% CI, 1.25-1.84), age >30 years, with PR=6.65, 95% CI, 4.40-10.05 for > 60 years of age, overweight (PR 2.09 95% CI 1.64-2.68), obesity (PR 2.68, 95% CI, 2.03-3.53) and diabetes (PR 1.56, 95% CI, 1.24-1.97).
CONCLUSION: These findings suggest the need to control overweight, obesity and diabetes, especially among women and individuals > 30 years of age in order to reduce the prevalence of arterial hypertension.

Key words: Hypertension / epidemiology; risk factors; Brazil.




Among the modifiable causes of early cardiovascular morbidity/mortality, Arterial Hypertension (AH) is one of the most important worldwide and an independent risk factor for cardiovascular disease1,2. Epidemiological studies point out that high levels of AP raise the risk for Encephalic Vascular Disease (EVD)2, Coronary Artery Disease (CAD)3,4, Congestive Heart Failure (CHF)5, Chronic Renal Failure (CRF)6,7 and Vascular Involvement (VI)8. In Brazil, in 2003, 27.4% of the deaths were due to cardiovascular diseases, and the leading cause was EVD among women9.

Kearny et al10 estimated the worldwide prevalence of AH to be 26.4% in the year 2000, which corresponded to 972 million hypertensive individuals. The SBH (Brazilian Society of Hypertension) estimates that there 30 million hypertensive people in Brazil (30% of the adult population). Among individuals above 60 years of age, 60% are hypertensive11. Epidemiological studies on AH conducted especially in Brazil's south and southeast regions indicate a rate above 25%12,13. In 2005, in Campo Grande, state of Mato Grosso do Sul, the prevalence of AH was 41.1%14. The risk factors associated with hypertension more consistently found in the Brazilian studies were overweight and/or obesity, low level of schooling and black skin color15.

Reliable information on hypertension prevalence in the different regions of Brazil is essential for the development of prevention and control policies. Moreover, it is important to know the risk factors for hypertension which can vary according to the sites studied. This study, which was conducted in 2003 in São Luís, state of Maranhão, in northeast Brazil, intended to estimate the prevalence of AH and its association with other cardiovascular risk factors in subjects 18 years of age or over.



With an area of 828.01 sq km, the city of São Luís has a population of 870,028 inhabitants living in 202,231 permanent private households. Women compose the majority of the population (463,628 females - 53.2%); most inhabitants are in the 10-19 age bracket (208,309 - 23%); population concentration is greater in cities (837,584 inhabitants - 96.2%); and the literacy rate among individuals aged 10 years and over is 93.1% (Censo 2000 - IBGE)16.

The research, which was approved by the Research Ethics Committee of the Hospital Universitário da Universidade Federal do Maranhão, consisted of a population-based cross-sectional study, conducted with the adult population of São Luís, Maranhão, aged 18 years or over, from February to March 2003, through representative cluster sampling. Age ranged from 18 to 94 years. The inclusion criteria were to be 18 years of age or over and be a resident of the selected household.

In order to determine the size of the sample, we considered 25% to be the estimated prevalence of the disease, with a 5% relative accuracy and a 99% confidence level, which resulted in 497 subjects. In order to assess the risk factors, the power of the study was considered to be 80%, the type-I error probability was 5%, the exposed/non-exposed ratio was 1:1, the prevalence in those non-exposed was 10%, and the prevalence ratio was 2, which resulted in a minimum sample size of 438 subjects. As we chose to work with a cluster sampling technique, the design effect was approximately 2. The final sample consisted of 835 subjects.

Initially, 70 census sectors were drawn and maps of these sectors were obtained with IBGE; next, the sectors were subdivided into smaller sub-sectors with more than 20 households each, from which two households were randomly drawn. Seven other sectors were used as a sample to conduct a pilot study.

In the first sub-sector, the survey was started at one of the corners. Systematic sampling was used, and the first five households at the right-hand side of the interviewer who was facing the street were visited. When it was not possible to complete the five households in the first sub-sector, visits to the second sub-sector were started following the same procedure. In the five households visited, we expected to find 12 people aged > 18 years, in each sector. A closed house was considered a loss when, after 2 visits at different times of the day, no information could be obtained. Empty houses were excluded. Household census was performed at each house, and all members aged 18 years or over were interviewed irrespective of kinship.

The survey team consisted of an epidemiologist, a cardiologist and two medical students. The interviews were conducted by 40 medical and/or nursing students, who had been previously selected and trained under the supervision of the researchers. All selected individuals were asked to sign the informed consent form to confirm they were in agreement. A structured questionnaire on coronary risk factors (arterial hypertension, sedentarism, diabetes, smoking status, alcohol consumption) and socio-demographic characteristics (sex, age, skin color, schooling, family income, marital status and occupational status) was applied to all adults aged 18 years or over living in the selected household. Interviewed individuals had their weight, height, waist circumference and arterial pressure measured by the interviewers.

The classification proposed by JNC717 was used to analyze data relative to systemic arterial hypertension. Blood pressure was considered normal when the SAP (systolic arterial pressure) value was lower than 120 mmHg, and the DAP (diastolic arterial pressure) value was lower than 80 mmHg. Pre-hypertension was considered when SAP ranged from 120 to 139 mmHg, and DAP from 80 and 89 mmHg; stage I hypertension when SAP was between 140 and 159 mmHg, and DAP between 90 and 99 mmHg; and stage II hypertension when SAP was equal to 160 mmHg and DAP greater than or equal to 100 mmHg.

Duly calibrated aneroid tensiometers and stethoscopes were used. Measurements were performed with the patient seated, with the right forearm on a table level with the heart, checking to see whether his/her bladder was full, whether he/she had performed physical exercises, had drunk any alcoholic beverages, coffee, eaten or smoked up to 30 minutes before the measurement. The brachial artery was located by palpation. The cuff was firmly attached approximately 2 to 3 cm above the antecubital fossa, centering the rubber bulb over the brachial artery. The criterion to determine the systolic pressure was the moment the first Korotkoff sound (Korotkoff's Phase I) appears, which becomes stronger with the increasing speed at which the cuff is deflated, whereas the determination of the diastolic pressure was the disappearance of the sound (Korotkoff's Phase V). As the heart beats continued until level 0, the diastolic pressure was determined at the muffling of the sounds (Korotkoff's Phase IV). Two measurements were performed: one right after the identification of the patient, and the other at least fifteen minutes after the questionnaire had been applied. The lowest arterial pressure was recorded.

The height and weight were measured using a Filizola PL-180 Adult Scale (scale with anthropometer). Patients were barefoot, standing on both feet and wearing minimum clothing.

The waist circumference was measured half-way between the iliac crest and rib margin using a non-elastic measuring tape. Regular physical activity was defined as the practice of physical exercises at least three times per week, during at least 30 minutes per day. A smoker was defined as the individual who declared himself a smoker at the moment of the interview, irrespective of the quantity of cigarettes smoked. Consumers of alcohol beverages were those who reported drinking, irrespective of the type and quantity of drinks. As to the frequency of alcohol consumption, the following situations were considered: patient does not drink; drinks rarely, weekly and daily. Waist circumference > 102 in men and > 88 in women were considered as altered, according to NCEP ATP III (National Cholesterol Education Program - Adult Treatment Panel III)18. Skin color was recorded as per the interviewer's observation. Family income was classified as >5, 3 to 5, 2 to 3, >1 to <2, and < 1 minimum wages. Schooling was evaluated according to the number of years of schooling and categorized as > 12, 9 to 11, 5 to 8, and < 4 years. As to occupation, individuals were classified as unemployed, salaried workers, self-employed and businesspeople. Marital status was classified as married, consensual union and without a partner. The subject was asked to inform about his/her family history of hypertension. Diabetic individuals were considered as those who used oral antidiabetic agents and/or insulin.

Data were processed and analyzed using EPI INFO 2000 and Stata 9.0 software. Poisson's regression was initially used to evaluate the association between the study risk factors and the prevalence of AH19. Studies have shown that when the prevalence of the event is greater than 10%, the use of the logistic regression model to estimate the odds ratio results in overestimation of the risk19. The prevalence ratios were calculated according to the robust method and its respective 95% confidence interval. The level of significance used was 5%. All variables presenting p<0.20 in the non-adjusted analysis were selected for the multivariable analysis. For the adjusted analysis, we used the method of stepwise variable selection with backward elimination. Only those variables associated with a P<0.10 value remained in the final model.



Eight hundred and thirty-five inhabitants of São Luís, state of Maranhão, were evaluated, 340 of them men (40.7%) and 495 women (59.3%); their mean age was 39.4 (± 16.6) years.

AH prevalence was 27.4% (95% CI - 24.4% to 30.6%), and this prevalence was higher among men (32.1%) than among women (24.2%).

Table 1 shows the percentages of the sample in each level of classification of arterial pressure, per gender. Among the individuals evaluated, 293 (35.1%) presented pressure levels within normal ranges. Pre-hypertension levels were detected in 313 (37.5%) of the respondents. Eighty-seven individuals (10.4%) were classified within the Stage I hypertension category. Thirty-eight individuals (4.6%) were classified within the Stage II hypertension category. One hundred and four individuals (12.5%) were on antihypertensive medication. These individuals were hypertensive, but could not have their hypertension classified as Stage I or II. A higher percentage of women used antihypertensive medication, and a higher percentage of men within the pre-hypertension and Stages I and II arterial hypertension was observed (P<0.001 as per the chi-square test).

In the non-adjusted analysis (Table 2), the following items were associated with arterial hypertension: male gender, age equal to or over 30 years, being married, daily consumption of alcohol, overweight/obesity, altered waist circumference, greater physical activity and diabetes. Low family income, unemployment and rare consumption of alcohol were protective factors against arterial hypertension. The highest prevalence of hypertension was detected among those individuals who went to school for > 12 and <4 years, whereas the lowest prevalence was observed among groups with intermediate levels of schooling. Skin color, smoking status and family history of hypertension were not associated with hypertension.

In the adjusted analysis (Table 3), the following items remained independently associated with hypertension: male gender (PR 1.52, 95% CI, 1.25-1.84), age >30 years, overweight (PR 2.09, 95% CI, 1.64-2.68), obesity (PR 2.68, 95% CI, 2.03-3.53) and diabetes (PR 1.56, 95% CI, 1.24-1.97).




It is estimated that approximately 1 billion people worldwide have hypertension. In the United States, an estimated 50 million people suffer from high blood pressure13. According to the 5th Brazilian Guidelines on Arterial Hypertension11 (2006) and using the current criterion for diagnosing arterial hypertension (SAP >140 and/or DAP >90 mmHg), the prevalence rates among the Brazilian adult urban population in selected studies ranged from 22.3% to 44%, with values estimated to be between 15.0% and 47.8% among men, and 15.0% and 41.1% among women. (Araraquara 1990 - 43%; São Paulo 1990 - 22%; Piracicaba 1991 - 33%, Porto Alegre 1994 - 26%, Cotia 1997 - 44%, Catanduva 2001 - 32%, Cavange 2003 - 36.5% and Rio Grande do Sul 2004 - 33.7%)11. In this study, 27.4% of the individuals evaluated were found to be hypertensive, and the prevalence among men was 32.1% and 24.2% among women.

Classification of AP values can be made according to several published national or international guidelines. We chose the classification system proposed by the JNC7 (Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure)17 which divides patients into four categories: normal, pre-hypertension, stage I hypertension, and stage II hypertension. The establishment of the new "pre-hypertension" category is important because physicians already recommend lifestyle changes according to the JNC717, since cardiovascular morbidity/mortality starts at this stage or at even lower AP values. In our study, the greatest prevalence was detected among pre-hypertensive patients (37.5%), showing that there is an indication to treat patients with an AP between 130x139 mmHg and 80x89 mmHg, thus delaying the onset of hypertension.

In Salvador, in 1999/2000, the assessment of 1,439 adults over 20 years of age and considering hypertension as SAP >140 and/or DAP >90 mmHg, revealed that the prevalence of hypertension was 29.9%, lower among men (27.4%) than among women (31.7%)13, whereas in São Luís the prevalence was higher among men (32.1%) than among women (24.2%). When age is taken in consideration, the prevalence in São Luís was 70.7% for patients over 60 years, whereas in Salvador it was 72.5% and 73.3% for men and women, respectively. In São Luís, the highest prevalence was among black and mulatto women, whereas in Salvador the prevalence did not vary according to skin color. It is possible that the different ways used for assessing skin color may account for the difference observed, as skin color was observed by the interviewer in São Luís, and self-reported in Salvador.

Age over 30 years and overweight/obesity were considered risk factors for hypertension, similar to the data recorded by Lessa et al20 In Salvador, abdominal obesity was a risk factor for arterial hypertension in women, whereas in São Luís, this factor did not remain associated with hypertension in the adjusted analysis. In São Luís, diabetes was a risk factor for arterial hypertension, whereas in Salvador, it was associated with hypertension only among women20.

In São Luís, in the non-adjusted analysis, hypertension prevailed at the extreme ends of schooling, i.e., lower than or equal to 4 years (39.1%) and higher than or equal to 12 years (39.2%) of schooling. In Salvador, a high level of schooling was a risk factor for hypertension only among men after the adjustment. The highest prevalence of obesity at the extreme ends of schooling may indicate the start of an inverse association between hypertension and higher schooling levels, which has already occurred in developing countries21.

In Pelotas, during the year 2000, the assessment of 1,968 adults aged 20-69 years and considering hypertension as SAP >160 and/or DAP >95 mmHg, showed that the prevalence of hypertension was 23.6%. Using the same criterion of São Luís (SAP >140 and/or DAP>90 mmHg), the prevalence of hypertension was 37.2%, thus higher than that recorded in São Luís. As in São Luís, in Pelotas, being >30 years of age and obese were risk factors for hypertension22.

In Campo Grande (state of Mato Grosso do Sul), in 2005, 892 people > 18 years of age were assessed. The same criterion for diagnosing hypertension used in São Luís (SAP >140 and/or DAP >90 mmHg) was used in Campo Grande. The prevalence of AH was 41.4%. As in São Luís, pressure levels rose with age; the highest prevalence was among men, in those overweight and obese14.

By comparing risk factors in four Brazilian cities (São Luís-MA, Salvador-BA, Pelotas-RS, and Campo Grande-MS), some risk factors were observed in all cities, especially older age, overweight and/or obesity, whereas other risks, such as diabetes and gender, were different among the cities.

Many lifestyle factors seem to directly influence AP levels, the most important of them obesity, alcohol consumption, physical activity and eating patterns11. In our study, alcohol consumption did not determine an increase of arterial hypertension in the adjusted analysis. Therefore, the greatest risk for hypertension associated with daily consumption of alcohol beverages disappeared in the adjusted analysis due to the confounding nature of these variables. The altered abdominal circumference was associated with hypertension in the non-adjusted analysis, but the association was not significant after the adjustment.

In our study, unemployment was a protective factor against hypertension in the non-adjusted analysis. Unemployed individuals had a lower prevalence of obesity (7.1%) than those self-employed (18.9%) and those who operated their own businesses (12.6%) (P<0.001), were younger (P<0.001) and were predominantly women (P=0.002), so they were less exposed to these risk factors for hypertension. This explains why the lower risk for hypertension among unemployed individuals disappeared in the adjusted analysis. The association between physical activity and hypertension observed in the non-adjusted analysis may indicate that hypertensive individuals are more engaged in physical activities, probably following medical recommendations. Family history was not a risk factor for hypertension. It may well be the case that the family history of hypertension was underestimated, as it was based on the individual's report.

Low family income was a risk factor for hypertension in the non-adjusted analysis. However, this association disappeared in the analysis adjusted for confounding variables. Individuals with higher family incomes (>5 minimum wages) had a higher prevalence of obesity (14.8%) than those with lower family incomes (8.1%), P<0.001, besides being older (P<0.001). Thus, the disappearance of association in the adjusted analysis was due to the fact that the individuals with higher family incomes had a higher prevalence of hypertension because they were older and had a higher prevalence of obesity.

This study had a few limitations as it did not analyze other variables that may be associated with hypertension, such as eating patterns. The possibility of reverse causality is another limitation that may restrict the interpretation of cross-sectional study findings.

It is important that further epidemiological studies on AH be conducted in our country in order to detect if the risk factors differ according to locations, which is suggested by this study, and also to plan more effective interventions. The results suggest the need to control overweight, obesity and diabetes, especially among women and individuals >30 years of age in order to reduce the prevalence of arterial hypertension.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This study is not associated with any graduation program.



1. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R, for the Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002; 360: 1903-13.         [ Links ]

2. Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004; 364: 937-52.         [ Links ]

3. Mohr JP. Lacunes. Neurol Clin North Am. 1983; 1: 201-21.         [ Links ]

4. Hennekens CH, Albert CM, Godfriend SL, Gaziano JM, Buring JE. Adjunctive drug therapy of acute myocardial infarction - evidence from clinical trials. N Engl J Med. 1996; 26: 292-4.         [ Links ]

5. Smith SC Jr, Blair SN, Criqui MH, Fletcher GF, Fuster V, Gersh BJ, et al. The Secondary Prevention Panel: prevention heart attack and death in patients with coronary disease. J Am Coll Cardiol. 1995; 26: 292-4.         [ Links ]

6. Wachtell K, Papademetriou V, Smith G, Gerdts E, Dahlof B, Engblom E, et al. Relation of impaired left ventricular filling to systolic midwall mechanics in hypertensive patients with normal left ventricular systolic chamber function: The Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) study. Am Heart J. 2004; 148: 538-44.         [ Links ]

7. Registro Brasileiro de Diálise, 1997. In: 19º Congresso Brasileiro de Nefrologia; Porto Alegre/RS, 11-14 de outubro de 1998. [acesso em 2005, mar 24]. Disponível em         [ Links ]

8. D'Ávila R, Guerra EMM, Rodrigues CIS, Fernandes FA, Cardaval RAM, Almeida FA, et al. Sobrevida de pacientes renais crônicos em diálise peritoneal e hemodiálise. J Bras Nefrol. 1999; 21: 13-21.         [ Links ]

9. Pannier BM, Avolio AP, Hoeks A, Mancia G, Takazawa K. Methods and devices for measuring arterial compliance in humans. Am J Hypertens. 2002; 15: 743-53.         [ Links ]

10. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worlwide data. Lancet. 2005; 365: 217-23.         [ Links ]

11. Sociedade Brasileira de Cardiologia, Sociedade Brasileira de Hipertensão, Sociedade Brasileira de Nefrologia. V Diretrizes Brasileiras de Hipertensão Arterial. Rev Bras Hipertens. 2006; 13 (4): 260-312.         [ Links ]

12. Lotufo PA. Stroke in Brazil: a neglected disease. São Paulo Med J. 2005; 123 (1): 3-4.         [ Links ]

13. Lessa I. Epidemiologia da hipertensão arterial sistêmica e da insuficiência cardíaca no Brasil. Rev Bras Hipertens. 2001; 8: 383-92.         [ Links ]

14. Souza ARA, Costa A, Nakamura D, Mocheti LN, Stevanato Filho PR, Ovando LA. Um estudo sobre hipertensão arterial sistêmica na cidade de Campo Grande, MS. Arq Bras Cardiol. 2007; 88 (4): 441-6.         [ Links ]

15. Gus I, Harzheim E, Zaslavsky C, Medina C, Gus M. Prevalência, reconhecimento e controle da hipertensão arterial sistêmica no estado do Rio Grande do Sul. Arq Bras Cardiol. 2004; 83 (5): 424-8.         [ Links ]

16. Instituto Brasileiro de Geografia e Estatística. (IBGE). [acesso em 2003 jun 3]. Disponível em         [ Links ]

17. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment oh High Blood Pressure. Hypertension. 2003; 42: 1206-52.         [ Links ]

18. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001; 285: 2486-97.         [ Links ]

19. Barros AJ, Hirakata VN. Alternatives for logistic in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol. 2003; 3: 21.         [ Links ]

20. Lessa I, Magalhães L, Araújo MJ, Almeida Filho N, Aquino E, Oliveira MMC. Hipertensão arterial na população adulta de Salvador (BA) - Brasil. Arq Bras Cardiol. 2006; 87: 747-56.         [ Links ]

21. Dalstra JAA, Kunst AE, Borrell C, Breeze E, Cambois E, Costa G, et al. Socioeconomic differences in the prevalence of common chronic diseases: an overview of eight European countries. Int J Epidemiol. 2005; 34: 316-26.         [ Links ]

22. Costa JSD, Barcellos FC, Sclowitz ML, Sclowitz IKT, Castanheira M, Olinto MTA, et al. Prevalência de hipertensão arterial em adultos e fatores associados: um estudo de base populacional urbana em Pelotas, Rio Grande do Sul, Brasil. Arq Bras Cardiol. 2007; 88: 59-65.         [ Links ]



Mailing Address:
José Bonifácio Barbosa
Departamento de Ensino e Pesquisa UDI Hospital
Av. Carlos Cunha, 2000, Jaracati
65076-820 - São Luís, MA - Brasil

Manuscript received November 29, 2007; revised manuscript received January 23, 2008; accepted March 24, 2008.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License