Critical Essay Appraisal: Physical Activity & Risk Of Coronary Heart Disease In India.

Effect of Non-Causal Factors on the Observed Association between Exposure and Outcome

Discuss about the Critical Article Appraisal for Physical Activity.

Rastogi, Vaz, Spiegelman, Reddy, Bharathi, Stampfer, Willett, and Ascherio (2003) seek to address the association between leisure time exercise and sedentary lifestyle and the risk of coronary heart disease in India. Increased mechanization and economic changes has contributed to the rising physical inactivity in the urban areas of India to levels comparable to the west. This study is unique since it is the first to examine association between the risk of coronary heart disease and physical inactivity in India. The study design was a hospital based case control study.The exposure of the study was developed from data collected from 8 Indian hospitals in New Delhi and Bangladesh of 350 patients with acute myocardial infarction and 700 patients without heart diseases as the control group. The researchers collected data on socio economic status, smoking history, history of hypertension, family history of coronary heart disease, dietary intake, supplement use, physical activity of the participants of the study. A physical activity questionnaire was designed for collecting data on occupational and non-leisure activities. participants’ data was grouped into 2; sedentary and leisure time activity depending and evaluated using conditional logistic regression to evaluate the relation between the risk of coronary heart disease and leisure time exercise and non-work activities.

The findings of the study were that more participants in the control group (48 percent) engaged in leisure time exercise than in case group (38 percent) which creates a strong and dose dependent inverse relationship between non-fatal coronary heart disease and leisure time exercise. The risk for coronary heart disease decreased across the levels of leisure time exercise.

People in the highest level of leisure time activity have a relative risk of 0.45 compared to non-exercisers. In addition, there was a positive relation between the risk of coronary heart disease and non-work sedentary activity.  There is a relatively strong association between increased sedentary activity and the risk of coronary heart disease. People with 3.6 hours of sedentary activity had a higher risk of coronary heart disease of 1.8. In addition, increased duration of time at work insignificantly elevated the risk of coronary heart diseases.

The main purpose of an epidemiology is assessing the cause of a disease. Since epidemiological studies are by nature rather than experiments, many explanations for an observed observation have to be examined in order to infer the existence of a cause-effect relationship. An observed association may be in fact as a result of random or systematic errors and confounding, which is not the case for this study. When examining the relationship between the explanatory factor, physical activity, and the outcome, risk of coronary heart disease, the study seeks to establish factors that may modify levels the effect of physical activity on the risk of coronary heart disease. However, this study did not report confounding because the researchers conducted a conditional logistic regression through statistical analysis software to determine the extent to which other variables such as smoking and dietary choices distort the observed observation. More so, the study is accurate because the researchers correctly classified the control participants across all levels of leisure time exercise, sedentary activity and work related activities. In addition, the study categorized continuous covariates to avoid linear associations’ assumptions with the outcome as well as minimize the effect of outliers. Differential misclassification results in over or underestimation of the effect between the exposure and outcome.

Causal Association Between Exposure and Outcome

While the use of population-based controls is ideal, the study addressed potential selection bias by obtaining control subjects from seven-outpatient clinic and inpatient wards in the hospitals. This implies that bias would be diluted is an association exists between the exposure of interest and the disease status of a control group. Secondly, the study employed a systematic method of choosing the control group participants over arbitrary selection of the control group. A systematic selection reduces bias that would attenuate the observed relation between CHD risk and physical activity. Selection bias for choosing case subjects did not exist because participants included in the study were those that survived past the interview. The study is credible because the overall participation for the study was high. However, there was bias in

Selection of participants of the control group because they were slightly more educated and earned lower incomes than the cases. However, controlling for socio economic factors minimized the bias and did not alter the results of the study.

There was no measurement errors because the researchers ensured the case subjects reported their activity levels and exercise before they were diagnosed with acute myocardial infarction. However, there is low potential for recall bias since health consciousness and awareness of heart disease prevention is not as high in India as in the western nations. The researchers reduced recall bias by exclusion of people with previous heart disease history from the study. As much as it cannot be assumed that interviewers reported all physical activities of the patients, the interviewers were trained to conduct thorough interviews of all the activities of the subjects done during the day. While future research may address potential biases that may arise in developing association between the risk of CHD and other factors, the hospital based case control study was appropriate because it was resource, time and cost efficient.

The odds ratio is the measure of causal inferences that explain the strength of an association between exposure and outcome. A relative risk demonstrates a greater probability that a relationship is causal. A temporal relationship must exist in order to demonstrate causation. for a study to demonstrate a temporal relationship, the exposure must precede the outcome. There is a temporal relationship between physical activity and the risk of coronary heart disease because in the selection of the control group, the researchers excluded participants who had prior heart conditions in order to avoid protopathic bias (Greenland, 2011). The study gave the right attention to the effect of physical activity on the risk of coronary heart disease by drawing conclusions about the causation in a situation where the risk factor, physical activity, preceded the risk of heart disease.

Consistency of the Study

The stronger an association is between a risk factor and outcome, the more likely the relationship is causal. In this study, the risk for coronary heart disease decreased across the levels of leisure time exercise with people in the highest level of leisure time activity having a relative risk of 0.45 compared to non-exercisers. A dose response between the exposure and outcome is an important demonstration of the influence of physical inactivity on the coronary heart diseases. There dope- response between the exposure and outcome that assumes the intensity of the level of physical exercise on heart diseases. The intensity of the exposure of physical inactivity to the participant is relatively high which implies a greater risk for a coronary heart disease. The dose response makes it less likely that the factor to which physical inactivity and coronary heart disease are related is an underlying risk factor-disease relationship. The results show a strong inverse relationship between leisure time activity and coronary heart disease. The results of the study are consistent with the study because they agree with the postulated theoretical that regular physical activity reduces the risk of blood lipid abnormalities, non-insulin dependent diabetes mellitus, obesity, and hypertension.

The outcome of this study is consistent with other studies in the examining the risk factors to coronary heart disease. Soares Miranda, Siscovicks, Psatty, Longstrth and Mozaffarian (2015) conducted a prospective cohort analysis among 4207 participants of a mean age of 73 years between 1989 and 1999 that examined the effect of habitual walking, leisure time activity and exercise intensity on cardio vascular disorders. The results indicated that greater physical activity was inversely associated with coronary heart disease, even for people over the age of 75 years.  Prospective cohort studies provide more evidence and tend to be more expensive than case control studies. However, prospective cohort studies are longitudinal observation over time where results are collected at intervals which helps to reduce the possibility of recall errors.

Sattelmair, Pertman, Ding, Kohl, Haskell and Lee (2011) conducted a meta analysis study of epidemiological studies that investigated primary prevention of coronary heart disease and physical activity in the united states. The researchers reviewed 3194 abstract and included 33 studies to estimate pooled dope-response estimates. Nine of the 33 studies showed that people who engaged in leisure time physical activity have 14 percent lower coronary heart disease risk and the risk reduced as a person engaged in higher levels of physical activity. This study also unveiled a significant association of the response and outcome by sex (p=0.03) where the association was stronger among females than males. Consistency of this study allows it to be replicated in other studies and serves as persuasive argument for causality. Despite having a hospital based case control design, the findings of the study identify the same relationship between physical activity and the risk of coronary heart disease with recent meta analysis and prospective cohort epidemiological studies thus adding weight to the causal inference.

The results of the study are plausible with regard to biological mechanisms. The cause and effect interpretation for the association between physical activity and increased risk of heart disease in urban towns in India is synonymous to the effect of sedentary lifestyle and physical inactivity in the western nations on the increasing number of people with coronary heart ailments.

Research findings are more convincing when they are externally valid, that is, they are applicable to different populations. The study was limited to a relative small sample size of patients in 8 hospitals located in New Delhi and Bangladesh. However, the study is free from bias since it takes into account socio economic factors and uses conditional logistic regression to control matching factors such as gender and age. Therefore, the findings are generalizable to the Indian population as well as specific populations as women or old people. The study can be used as background information for further epidemiological study between association between the risk of coronary heart disease and physical activity within specific populations. In addition, this case control study allows for clarity of temporal sequence, calculation of incidences, facilitates study of rare exposure to the subjects, allows examination of multiple effects of a single exposure, and prevents selection bias at enrollment of the subjects, which makes the study viable for use in different populations.

Soares-Miranda, L., Siscovick, D. S., Psaty, B. M., Longstreth, W. T., & Mozaffarian, D. (2015). Physical activity and risk of coronary heart disease and stroke in older adults: the Cardiovascular Health Study. Circulation, CIRCULATIONAHA-115.

Sattelmair, J., Pertman, J., Ding, E. L., Kohl, H. W., Haskell, W., & Lee, I. M. (2011). Dose response between physical activity and risk of coronary heart disease. Circulation, CIRCULATIONAHA-110.

Greenland, S. (2011). Causation and causal inference (pp. 216-221). Springer Berlin Heidelberg.