Sports events and Acute Coronary Syndrome: possible confounding factors and bias

LETTER TO EDITOR

Sports events and Acute Coronary Syndrome: possible confounding factors and bias

Mauro Felippe Felix Mediano; Andrea Silvestre de Sousa; Alejandro Marcel Hasslocher-Moreno

Instituto de Pesquisa Clínica Evandro Chagas - Fundação Oswaldo Cruz, Rio de Janeiro, RJ - Brazil

^{Mailing Address}

Keywords: Acute coronary syndrome; Football; Alcohlism; Tobacco; Substance- Related Disorders.

Dear Editor,

The relationship between sports events and ACS (Acute Coronary Syndrome) has been the subject of some studies in the literature, with controversial results, and was recently addressed by Borges et al¹ in the manuscript entitled "World Soccer Cup as a Trigger of Cardiovascular Events," published in volume 6 of 2013 of ABC. The authors observed a higher incidence of AMI (Acute Myocardial Infarction) in match days of the FIFA World Cup and, in particular, on Brazilian games, concluding that this sporting event can act as a "trigger" of AMI in Brazilians. However, considerations of possible confounding factors and bias should be made to present to readers a better interpretation of the data presented. Firstly, the literature describes that, during sporting events, viewers tend to have numerous risk behaviors for cardiovascular disease, such as the intake of large amounts of alcohol, consumption of fatty food, and use illicit drugs and smoking^2,3. These factors have also been described as possible triggers for ACS^4,5, making the association between FIFA World Cup and ACS established in the research of Borges et al¹ subject to a number of confounding factors and hindering the interpretation of results. Defining, as the authors intend, if the World Cup could be seen as a direct "trigger" for ACS is not possible or merely if it entails a number of risk conditions already well described in the literature. Furthermore, the use of research findings by observing groups of people to infer causal relationships in individuals may result in ecological fallacy⁵, since it is not possible to state that all individuals who attended the event were exposed the same way. Therefore, the study design presented by Borges et al¹ does not allow the establishment of a direct association between the occurrence of ACS and the FIFA World Cup, and the interpretation of the study results should be performed carefully, taking into account the potential confounding factors and bias of ecological study.

References

Manuscript received June 22, 2013, revised June 26, 2013, accepted July 25, 2013.

Reply

We would like to thank our colleagues who sent this letter, which has allowed further discussion on this quite intricate subject¹. Firstly, we agree to the assertions that, over sport events, viewers tend to have a number of risk behaviors for cardiovascular diseases, such as the intake of great amounts of alcoholic beverages, eating of fatty food, use of illicit drugs and smoking, given that the limitations of our research have been addressed in our discussion. However, it is important to note that the measurement of all possible confounding factors referred to is difficult (illicit drugs, for instance), even in prospective studies, as this may be subject to a biased observation². We stress the limitations of our data, retrieved from a public database, which restricted the availability of data to protect the individual's secrecy rights. Another point to be raised is that these potential confounding factors not available in our study may present collinearity with the statement of interests (watching the World Cup and Brazilian matches) and would require a high-cost prospective design to remove all the factors raised^3,4.

Regarding the second remark about the problems associated with ecological studies, we believe that they are also properly pointed out in our discussion, as one can see that "...the exposure may not be uniform (a part of the population may not be watching the match)".

However, even with the considerations raised and clarified above, although we are not able to definitively confirm a causal association between the games and the occurrence of cardiovascular events, we still can assume that these two factors are related, despite the correction for confounding factors available. Moreover, a simple association, easily identifiable and which may be used immediately can be more useful for health planning than establishing the cause and effect relation from the pathophysiological point of view. Until further information is available and considering everything that has been studied in the literature, this information seems consistent with most other studies, and we believe that our findings should be maintained.

Sincerely,

Antonio Pazin Filho

Daniel Guilherme Suzuki Borges

Rosane Aparecida Monteiro

André Schmidt

References

Correspondência:

Mauro Felippe Felix Mediano

Rua Antônio Basílio, 519, cobertura 01, Tijuca

CEP 20511-190, Rio de Janeiro, RJ, Brasil

E-mail: [e-mail]

Artigo recebido em 22/06/13; revisado em 26/06/13; aceito em 25/07/13.

Carta-resposta

Gostaríamos de agradecer aos colegas que enviaram essa carta, que nos permitiu possibilidade de maior discussão sobre um assunto tão complexo¹. Primeiro, concordamos com as afirmações de que, durante eventos esportivos, espectadores tendem a apresentar inúmeros comportamentos de risco para doenças cardiovasculares, como ingestão de grandes quantidades de álcool, consumo de comidas gordurosas, uso de drogas ilícitas e fumo, sendo que as limitações associadas à análise que realizamos contemplam essa limitação em nossa discussão. No entanto, é importante destacar que a mensuração de todos os possíveis fatores de confusão apontados é difícil (drogas ilícitas, por exemplo), mesmo em estudos prospectivos, pois podem estar sujeitas à viés de observação². Salientamos as limitações de nossos dados, que foram obtidos de uma base pública, o que limitou a disponibilidade dos dados ou inexistência dos mesmos para proteger o sigilo do indivíduo. Outro aspecto a ser levantado é que esses possíveis fatores de confusão não disponíveis no nosso estudo podem apresentar colinearidade com a exposição de interesse (assistir a Copa do Mundo e aos jogos do Brasil) e seria necessário um desenho prospectivo de elevado custo para excluir todos os fatores levantados^3,4.

Com relação ao segundo comentário, que diz respeito aos problemas associados a estudos ecológicos, acreditamos que estão também devidamente apontados em nossa discussão, pois inclusive se pode observar que "... a exposição pode não ser uniforme (uma parte da população pode não estar assistindo ao jogo)".

No entanto, mesmo com as considerações levantadas e esclarecidas acima, embora não possamos confirmar definitivamente uma associação causal entre os jogos e a ocorrência de eventos cardiovasculares, ainda podemos assumir que esses dois fatores estão associados, apesar da correção dos fatores de confusão disponíveis. Além disso, uma associação simples, facilmente identificável e que pode ser utilizada de imediato pode ser mais útil para o planejamento em saúde do que o estabelecimento de relação-causa e efeito do ponto de vista fisiopatológico. Até que se disponha de informações mais detalhadas e considerando tudo que já foi estudado na literatura, essa informação nos parece consistente com a maioria dos demais estudos e julgamos que nossas conclusões devem ser mantidas.

Atenciosamente,

Antonio Pazin Filho

Daniel Guilherme Suzuki Borges

Rosane Aparecida Monteiro

André Schmidt

Referências

• 1. Borges DG, Monteiro RA, Schmidt A, Pazin-Filho A. Copa do mundo de futebol como desencadeador de eventos cardiovasculares. Arq Bras Cardiol. 2013;100(6):546-52.
• 2. Leeka J, Schwartz BG, Kloner RA. Sporting events affect spectators' cardiovascular mortality: it is not just a game. Am J Med. 2010;123(11):972-7.
• 3. Merlo LJ, Hong J, Cottler LB. The association between alcohol-related arrests and college football game days. Drug Alcohol Depend. 2010;106(1):69-71.
• 4. Roerecke M, Rehm J. Irregular heavy drinking occasions and risk of ischemic heart disease: a systematic review and meta-analysis. Am J Epidemiol. 2010;171(6):633-44.
• 5. Mittleman MA, Mostofsky E. Physical, psychological and chemical triggers of acute cardiovascular events: preventive strategies. Circulation. 2011;124(3):346-54.
• 1. Borges DGS, Monteiro RA, Schmidt A, Pazin-Filho A. World soccer cup as a trigger of cardiovascular events. Arq Bras Cardiol. 2013;100(6):546-52.
• 2. Levine M, Walter S, Lee H, Haines T, Holbrook A, Moyer V. How to use an article about Harm. JAMA.1994;27(20):1615-9.
• 3. Sorlie P, Wei GS. Population-based cohort studies: still relevant? J Am Coll Cardiol.2011;58(19):2010-3.
• 4. Sauerbrei W, Royston P, Binder H. Selection of important variables and determination of functional form for continuous predictors in multivariable model building. Stat Med.2007;26(30):5512-28.

Publication Dates

• Publication in this collection
  13 Dec 2013
• Date of issue
  Nov 2013