This study reports a high prevalence of snoring (40%) and daytime sleepiness (59%) in the normal- weight urban South Indian population, with a mean BMI of 23.5 ± 3.8. These values are higher than the study from Western India by Udwadia and colleagues,20
who reported habitual snoring in 26% and daytime sleepiness in 22% of their study subjects.20
These figures were found to be consistent with levels in others populations.21
Our study results are in agreement with that of Ohayan and colleagues, who reported snoring in 40% of the study population.22
However, there are studies that report an even higher prevalence (51.6%) of snoring in Asian populations such as the Chinese population.23
The snoring measure used in our study was found to be associated with a host of anthropometric and cardiometabolic factors. In line with earlier literature, male sex and age both showed an association with the snoring measure.1
In our study, snorers were 3 years older than nonsnorers among both females and males. As our males were generally older than the female snorers, this could be another explanation for the higher incidence of snoring among males, although abdominal obesity could also be a contributing factor.
Similar to results shown by other population-based studies,24–26
the snorers showed higher values of BMI, waist circumferences, systolic/diastolic pressure, fasting blood glucose, and serum triglycerides than nonsnorers. Despite the small number of smokers, they too had a higher frequency of snoring.
Daytime sleepiness did not have nearly the same degree of association with cardiometabolic risk factors as snoring and indeed was not associated with MS. This is likely a methodological issue, as the term “daytime sleepiness” is a rather ambiguous measure, open to misinterpretation in a self-report questionnaire. Nevertheless, the daytime sleepiness measure showed a significant association with BMI and waist circumference. However, Bixler and colleagues have shown an association of daytime sleepiness with depression and metabolic factors.27
Both snoring and daytime sleepiness showed a significant relationship with impaired glucose metabolism. However, whether the sleep abnormalities had an independent association with MS is questionable. In our study we found that snoring, but not daytime sleepiness, is associated with MS.
The daytime sleepiness measure retained a correlation with glucose intolerance when sex and age were controlled for, but this, again, was weakened when measures were adjusted for obesity.
The most significant finding in our study was the high prevalence of snoring and daytime sleepiness in a normal-weight Indian population. It is important to note that of those who reported snoring, 90% had a BMI <30, the cut point for obesity in Europeans.17
Thus, while the snorers in our study had higher BMIs, these individuals were not obese from the Western standpoint. This suggests that sleep abnormalities are also found in those who are slightly overweight. Also our data suggest that the prevalence of sleep abnormalities may be higher in our population. A study of Indians in Singapore showed that Indians had higher snoring prevalence rates than Chinese and Malays.9
It is possible that sleep abnormalities could be another feature of the so-called “Asian Indian phenotype,” which makes this ethnic group more susceptible to diabetes and premature coronary artery disease.7,28
Both snorers and those with daytime sleepiness had significantly higher IDRS scores compared to those who did not snore or have daytime sleepiness. Thus this study shows another use of the IDRS score—to identify those with sleep abnormalities such as snoring and daytime sleepiness.
This study has certain limitations. While the study used self-report to measure sleep quality, the presence of sleep abnormalities, such as snoring, arousals, and apneas, are best detected through overnight polysomnography, which is the gold standard measure for detecting sleep apnea.4
Further, the simplicity of the sleep questions did not elicit specifics of the subjects’ sleep patterns, thus restricting the study to an over-simplified analysis of the sleep abnormalities. Also, the confounding effect of central nervous system depressants, enlarged tonsils, and retrognathia, a familiar pattern could not be ruled out, as these parameters were not measured in this study. The possible implications of a mixed male–female cohort is not reported, as further stratifying data based on gender will lower the sample size. Hence, the analysis includes all subjects as a whole, which is one of the limitations of the study. Finally, being a cross-sectional study, no inferences can be drawn about causality. However, one of the strengths of this study is that it is a population-based sample, unlike most earlier sleep studies from India, which were hospital-based and subject to referral bias.20,29
It is worth noting that despite the risk of over or under reporting, the self-report methodology of our study still shows compelling associations between at least two sleep abnormalities and cardiometabolic factors in our population. Future longitudinal work can aid the understanding of the sleep-cardiometabolism relationship.
In summary, this is the first Indian population-based study to investigate the prevalence and risk factors of sleep abnormalities and their relationship to cardiometabolic phenomenon in a representative population of South India. Given the rising epidemic of non-communicable diseases worldwide and the fact that South Asia in general—and India in particular—bears the brunt of the diabetes and CVD burden, further longitudinal, population-based research in this important area of sleep health is essential.