Geography of hospital admissions for multiple sclerosis in Greece.

BACKGROUND

Multiple sclerosis (MS) is the 2 most common, disability causing neurological disorder in young adults, known for its differences in prevalence according to geographical position. Data on MS epidemiology is lacking in Greece. In this study, we aimed to examine the pattern of MS related hospital admissions in Greece and analyze their spatial distribution. Data for hospital admissions due to MS was obtained from the Hellenic Statistical Authority (ELSTAT) database and they were available from 1999 to 2012.

METHODS

We used the proportional hospitalization ratio (PHR) which was age and sex adjusted according to hospitalizations of Greek population, in order to calculate the ratio between hospitalizations in each regional unit (RU) and hospitalizations of national population. PHR was calculated for each RU, which is categorized as level 3 based on the Nomenclature of Territorial Units for Statistics (NUTS level 3). The classification of MS was based on International Classification of Diseases, Ninth Revision (ICD-9), code 340. Descriptive analysis was conducted to understand the epidemiological characteristics and time series analysis was used to investigate the trend of annual PHR's values during the study period. Global spatial autocorrelation analysis was conducted to assess spatial homogeneity of MS across the country. Moreover, local spatial autocorrelation analysis was performed using the LISA statistic to detect any potential clusters of similar values. Finally, multiple linear regression was conducted to examine correlations between PHR and latitude.

RESULTS

Over the 14-year period the number of hospitalizations increased while the proportion of hospitalizations remained higher for females compared to males, as well as for the 25-44 year age group. RUs that appeared to be most affected included Evros with a sex- and age- adjusted PHR of 2.00 (95%CI: 1.89-2.10), Larisa with a PHR of 1.74 (95%CI: 1.67-1.81) and Chios with a PHR of 1.77 (95%CI: 1.60-1.95). The least affected RU was Arkadia with a PHR of 0.29 (95%CI: 0.24-0.36).

CONCLUSION

In this study we present a rise in hospital admissions related to MS over a 14-year period, possibly indicating a parallel rise in incidence. A combinational analysis of the number of hospitalizations along with incidence studies could be further performed to be used to design public health interventions.