WASHINGTON — The prevalence of multiple sclerosis in the United States population appears to be nearly double previous estimates. Among servicemembers, however, the rate of new MS diagnoses has dropped significantly in recent years.
Mitchell Wallin, MD, director of the VA’s MS Center of Excellence-East in Washington presented results of a groundbreaking MS prevalence study at the 2017 European Committee for Treatment and Research in Multiple Sclerosis/Americas Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS/ACTRIMS) annual meeting. Using a new method to estimate the number of individuals affected by the disease, the research team calculated that about 727,000 Americans have multiple sclerosis, Wallin said. Previous estimates had put the number at 400,000.1
“Many people have said, “Listen, it has got to be more than 400,000 people.” And, in fact, it is. We have been using the number of 400,000 for probably 25 years, because it is very difficult to get at the accurate numbers,” said Kathleen Costello, MS, ANP-BC, associate vice president of Healthcare Access of the National Multiple Sclerosis Society in a video discussing the study.
”Based on a revised analysis of the ECTRIMS data, the 2010 prevalence for MS cumulated over 10 years was between 265.1 and 309.2 per 100,000,” Wallin told U.S. Medicine in March. The team calculated the current rate using a three-year cumulative prevalence, which they standardized to the 2010 U.S. population. The rate is more than triple the 85 per 100,000 prevalence estimated for the 1989 to 1994 period.
“The U.S. national cumulative MS prevalence rates for 2008-1020 are the highest reported to date and provide a contemporary understanding of the disease burden,” according to Wallin and his colleagues.
Using a consistent MS case definition, the researchers developed and validated an algorithm to identify individuals with MS using three independent databases. They then applied the algorithm across six large U.S. healthcare databases that together represented more than one-third of the national population. The databases included the VA, Medicare, Medicaid, Kaiser Permanente of Southern California, Optum and Truven Health Market Scan.
“Our rigorous algorithm-based approach to estimating prevalence is novel, efficient and has the potential to be used for other chronic conditions,” Wallin said.
The higher prevalence rate determined by Wallin and his colleagues does not necessarily mean that more people are diagnosed with MS each year. The multiyear method, algorithm and large databases they employed might have captured more individuals than previous estimates and more people might have had healthcare visits explicitly for MS treatment as new and more effective therapies for the disease have come online.
Military Cases Drop
A study in Medical Surveillance Monthly Report that specifically looked at new diagnoses of MS offers some good news for the military: From 2007 to 2016, the unadjusted incidence rates of MS declined 25.4% among active component servicemembers. Active duty females had a much higher rate of new diagnoses than men throughout the study period but also experienced a significantly sharper decline in incidence rates.2
During the 10 years studied, 2,031 active duty servicemembers received an initial diagnosis of MS, resulting in an unadjusted incidence rate of 14.9 new cases per 100,000 person years. Overall, women had three times the unadjusted rate of men, 34.4 per 100,000 vs. 11.5 cases per 100,000. The highest rates were seen in non-Hispanic black female servicemembers, who had a rate of 38.4 cases per 100,000 and were lowest among Hispanic males and males of “other/unknown” race or ethnicity at 8.8 and 7.4 cases per 100,000, respectively.
Annual rates of new MS diagnoses for women plummeted 43.2% over the study period but just 16.6% for men. The steep drop in rates for women significantly changed the difference in the rate of diagnoses between women and men. In 2007, 3.7 times as many women had incident diagnoses as men. By 2016, the sex incidence ratio had declined to 2.5. Rates among non-Hispanic black service members fell 40.8% over the 10 years.
Incidence rates varied substantially across the services. The Air Force had the highest rate at 20.3 per 100,000. That was 2.2 times higher than the lowest service, the Marines, which had a rate of 9.3 per 100,000. The Navy”s rate was 13.3 per 100,000 and the Army”s 14.6 per 100,000. The reserve/guard component had an overall rate of 6.9 cases per 100,000.
Officers had 50% higher rates than enlisted members, 9.7 vs. 6.4 per 100,000, respectively. And healthcare workers had nearly triple the rate of combat-specific occupations, 11.4 vs. 4.0 per 100,000.
While rates dropped in nearly all groups during the study period, the crude rates reported were higher than the 12.9 cases per 100,000 reported in 2011 for the 2000 to 2009 period in the same publication using a similar case definition. They were also much higher than the 9.6 per 100,000 seen in a study of Gulf War-era veterans led by Wallin.3
The authors note that the MSMR studies used diagnostic codes for one inpatient encounter, two outpatient encounters or one inpatient and outpatient encounter for MS. They did not verify the diagnoses using clinical records, which the Gulf War veteran study did.
“I would expect there to be a fairly high false positive rate using these ICD-9/10 encounter scenarios, so the incidence rates are likely inflated. Because two encounters were used for MS in a given year, the sensitivity and specificity should be improved versus a single encounter for MS,” Wallin said.
1Wallin M. The Prevalence of Multiple Sclerosis in the United States: A Population-Based Healthcare Database Approach. ECTRIMS Online Library. October 26, 2017. Abstract P344.
2Williams VF, Stahlman S, Ying S. Multiple Sclerosis Among Service Members of the Active and Reserve Components of the U.S. Armed Forces and Among Other Beneficiaries of the Military Health System, 2007-2016. MSMR. August 2017;24(8):2-11.
3Wallin MT, Culpepper WJ, Coffman P, et al. The Gulf War era multiple sclerosis cohort: age and incidence rates by race, sex and service. Brain. 2012;135(Pt 6):1778–1785.