Standardized mortality ratios in multiple sclerosis: Systematic review with meta-analysis.

Abstract

To perform a meta-analysis of all-cause, cause-specific and gender-specific standardized mortality ratio and crude mortality rate for people with multiple sclerosis. We also examined the temporal trends in this data. Medline, Cochrane Library and Scopus were searched. Keywords were "multiple sclerosis" and "standardized mortality ratio" or "Standardized Mortality Ratio". We included longitudinal studies with available data on the number of deaths, follow-up period, person years and reports of standardized mortality ratio (SMR). Crude mortality ratio (CMR) was calculated and SMR was extracted. CMRs and log-SMR were pooled by the method of inverse variance. Meta-regression models were used to investigate temporal trends. Fifty-seven articles were screened. Fifteen studies were included covering a period 1949-2013 (160,000 patients; 21,225 deaths). The all-cause SMR for people with MS was 2.61 (95% CI 2.58 to 2.65). For men this was 2.47 (95% CI 2.42 to 2.52) and for women 2.57 (95% CI 2.53 to 2.61). The CMR was 13.45/1000 person years. Cause-specific SMR was 1.74 (1.67 to 1.81) for CVD, 4.70 (4.45 to 4.87) for respiratory disease and infection, 1.81 (1.64 to 2.0) for accident and suicide and 0.99 (0.93 to 1.06) for cancer. Meta-regression analysis of the SMR compared to midpoint follow-up year revealed no relationship (co-efficient 0.001, p=.98). People with multiple sclerosis (MS) have reduced overall survival and increased risk of death from cardiovascular, respiratory and infectious disease as well as accidents and suicide. This does not appear to have changed over the last 65years.