WHO-global stroke initiative-Mumbai stroke registry: a report on incidence of first-ever-stroke during re-survey 2009


Correspondence: Praful. M. Dalal, LKMM Trust Research Centre, At Lilavati Hospital, A/791, Bandra Reclamation, Mumbai 400050, India.

E-mail: pmdalal_in@yahoo.com

The rehabilitation of stroke survivors places a heavy socioeconomic burden on India. The incidence of first ever stroke (FES) in a prospective, well-defined, population-based survey in Mumbai (Study I 2005–06) has been in the range of 145/100 000 persons per year [1]. This high incidence finding is very similar to that reported in developed nations. The Mumbai study is considered comparable to 55 other studies in the world; all suggest that FES incidence is rising in developing countries [2]. Therefore, we re-surveyed in the same district (Study II 2009) by the same team using the identical and well-defined protocol initially used [1]. 156,861 persons between the age of 25 and 94+ were screened in Study I and 174,398 persons in Study II. There were 456 FES in Study I (2005–2006) and 223 in Study II (2009). The overall age specific incidence of FES in Study I was: 145/100 000/year (95% CI 120–170) and in Study II 127·8/100 000/year (95% CI 110–150) (Fig. 1), which indicates no statistically significant difference in FES incidence in both studies. Furthermore, analysis of various parameters in stroke profile also showed no significant change. In low- and middle-income countries, stroke surveillance is essential to define the magnitude of burden and to plan effective preventive strategies. This is difficult in the absence of computerized hospital-based systems and vital statistics [3-5]. Our study possibly suggests a more cost-efficient approach to stroke surveillance would be to re-survey the same population every 5 or 10 years to understand changing trends.

Figure 1.

Annual incidence of first-ever stroke per 100 000 and total population distribution (%) 2005–06 (Study I) and 2009 (Study II).

Annual standardized incidence rates by direct method (Segi) using the 1996 world population: Study I: 152/100 000 persons; CI 95%:132–172; Study II: 137/100 000 persons; CI 95% 119–155.