Estimating the hidden burden of obstructive sleep apnoea: challenges and pitfalls

Adam Benjafield and colleagues should be congratulated for theirefforts to estimate the global burden of obstructive sleep apnoea. They did a systematic review of population studies in which obstructive sleep apnoea had been objectively assessed, and extrapolated data to the entire country population as well as to the populations of other countries matched by age, sex, body-mass index, ethnicity, and geographical proximity. However, not all the studies included had done stratified random sampling rather than convenience sampling, and some reported a large dropout of participants. Caution is needed when dealing with data from non-randomly selected populations. Furthermore, the matching done, even if based on established determinants of obstructive sleep apnoea, fails to recognise anthropological (such as cultural and religious) differences existing even among neighbouring countries, which could affect healthseeking behaviors, one of the main sources of underestimation. Underestimation includes, indeed, underascertainment, in which patients with obstructive sleep apnoea might not be aware of their status, or, if aware, do not seek physician’s advice due to stigma, cultural or religious factors, or legal or administrative barriers. Another component of underestimation is underdiagnosis, in which a misclassification occurs due to measurement errors, the physician’s lack of experience or knowledge about testing or interpretation of results, and, especially in developing countries, lack of health-care facilities and instrumentation. Finally, underestimated cases include also undernotified cases. Both overestimation and underestimation affect the management efficiency of health-care systems and compromises the possibility of making informed, evidence-based decisions. Indeed, it is crucial to have an accurate estimate of the number of cases undiagnosed or untreated. Benjafield and colleagues applied a conversion algorithm to reduce the problem of underdiagnosis. However, issues pertaining to underascertainment and undernotification remain unaddressed. Apart from these methodological considerations, the most important message of the study1 is that the global burden of obstructive sleap apnoea is similar, if not even higher, to the burden generated by other chronic diseases such as hypertension or diabetes. For these disorders, but not for obstructive sleep apnoea, targeted prevention and early diagnosis programmes are currently being implemented in several countries. Lessons that we can learn and capitalise on include the need to improve the awareness that obstructive sleep apnoea can be diagnosed and treated; the need for high-quality epidemiological investigations, including nationwide surveys; and the need to establish a global registry of obstructive sleep apnoea, given its clinical and socioeconomic burden.