Spinal muscular atrophy linked to chromosome 5q (SMA) is a recessive, progressive, neuromuscular disorder caused by bi-allelic mutations in the SMN1 gene, resulting in motor neuron degeneration and variable presentation in relation to onset and severity. A prevalence of approximately 1–2 per 100,000 persons and incidence around 1 in 10,000 live births have been estimated with SMA type I accounting for around 60% of all cases. Since SMA is a relatively rare condition, studies of its prevalence and incidence are challenging. Most published studies are outdated and therefore rely on clinical rather than genetic diagnosis. Furthermore they are performed in small cohorts in small geographical regions and only study European populations. In addition, the heterogeneity of the condition can lead to delays and difficulties in diagnosing the condition, especially outside of specialist clinics, and contributes to the challenges in understanding the epidemiology of the disease. The frequency of unaffected, heterozygous carriers of the SMN1 mutations appears to be higher among Caucasian and Asian populations compared to the Black (Sub-Saharan African ancestry) population. However, carrier frequencies cannot directly be translated into incidence and prevalence, as very severe (death in utero) and very mild (symptom free in adults) phenotypes carrying bi-allelic SMN1 mutations exist, and their frequency is unknown.More robust epidemiological data on SMA covering larger populations based on accurate genetic diagnosis or newborn screening would be helpful to support planning of clinical studies, provision of care and therapies and evaluation of outcomes.Electronic supplementary materialThe online version of this article (doi:10.1186/s13023-017-0671-8) contains supplementary material, which is available to authorized users.