1. MULTIPLE SCLEROSIS INCIDENCE IN TUSCANY FROM ADMINISTRATIVE DATA
Bezzini D1,2, Policardo L3, Profili F3, Meucci G4, Ulivelli M5, Bartalini S5, Francesconi P3, Battaglia MA1,2
[1] Department of Life Sciences, University of Siena, Siena, Italy, [2] Fondazione Italiana Sclerosi Multipla (FISM), Genova, Italy , [3] Agenzia regionale di sanità della Toscana, Florence, Italy, [4] Unit of Neurology, USL6, Livorno, Italy, [5] Department of Neuroscience, Neurology and Clinical Neurophysiology Section, University of Siena, Siena, Italy.
INTRODUCTION: Italy is a high-risk area for multiple sclerosis with 109,000 prevalent cases estimated at January 2015 [1] and 3,400 annual incident cases [2]. Regarding incidence, latest published papers report 5.5 cases per 100,000 in Padua for the period [3], 6.7 cases in Genoa for the period [4] and 9.7 cases per 100,000 in Sardinia for the period [5]. To monitor disease epidemiology, comorbidities and care pathways, but also to describe the disease burden and to plan its prevention, treatment and management strategies and resource allocation, population-based studies are preferable [6,7]. For these reasons, a national population registry has been founded in 2015 and it will represent, in the next future, a valid tool to study MS epidemiology. Another way to obtain data on entire population is from administrative sources. In a previous study, in the Tuscany region, we created and validated a case-finding algorithm, based on administrative data, to identify MS patients and to calculate the prevalence as at 2011 [8]. The algorithm had a sensitivity of 98% and a specificity of 99.99%. We identified 6,890 prevalent cases as at 31/12/2011, with a resulting rate of cases per 100,000, per 100,000 in females and per 100,000 in males [8]. Also in other Italian Regions, administrative data have been used to calculate MS prevalence, as in Lazio with 131 cases per 100,000 as at 2011 [9], in Apulia with 183 cases as at 2012, in Veneto with cases as at 2015, in Sicily with 110 cases as at 2010 and Sardinia with 360 cases as at 2016 [2]. In this study, our aim is to estimate the incidence of multiple sclerosis in Tuscany using a case-finding algorithm based on administrative data, from 2011 to 2015.
METHODS: We created a case-finding algorithm in which we linked hospital discharge records, disease-specific exemptions and drug dispensation records for at least one of the eight drugs that are specific for MS (Glatiramer acetate, Interferon beta 1A and 1B, Fingolimod, Natalizumab, Dimethyl fumarate, Peginterferon beta-1a, Teriflunomide). Incident cases were identified as a subset of prevalent cases among those patients not traced in the previous years. The first record in the health administrative databases was considered as the date of MS diagnosis. From this cohort of incident cases, we selected patients with an age ≤60 years, who lived in Tuscany at the date of MS diagnosis and who were in the beneficiaries of the regional health system registry from at least 10 years (or patients born in Tuscany if were <10 years old). Then, we calculated crude and standardized (for Tuscan population as at 01/01/2015) incidence for the years , and the confidence interval at 95%.
Table 1. Incidence rates – crude and standardized for age and sex (STD) – and confidence interval for the years (standard population: Tuscan population at 1/1/2015).
RESULTS: We identified 1,147 incident cases (380 males and 767 females) from 2011 to The total crude incidence obtained ranged from 5.60 per 100,000 in 2011 to 6.58 per 100,000 in 2015 (Table 1) with a mean of 6.09 for the entire period. In females, standardized incidence rate was around two times much high than in males (Table 2). Analyzing standardized rates by each of the twelve LHA of residence, we observed that the LHAs 8, 3, 11, 10 and 6 had a rate higher than the mean rate of the region, whereas the LHAs 1, 2 and 12 had the lowest rate (Graph 1).
YEAR
CASES
CRUDE RATE
STD RATE
CI 95%
2011
210
5.60
5,46
3,85-7,06
2012
216
5.89
5,82
4,16-7,48
2013
224
6.07
6,00
4,40-7,60
2014
250
6.67
6,62
4,79-8,44
2015
247
6.58
6,58
Table 2. Incidence rates – crude and standardized for sex (STD) – and confidence interval for the years in females and males (standard population: Tuscan population at 1/1/2015).
YEAR
MALES
FEMALES
INCIDENCE CRUDE (STD)
CI 95%
2011
4.04 (3.84)
7.04 (7.05)
2012
3,98 (3.82)
7.65 (7.78)
2013
4.18 (4.00)
7.81 (7.98)
2014
4.89 (4.69)
8.32 (8.50)
2015
4.16 (4.03)
8.83 (9.08)
Graph 1. Mean standardized incidence rates in the twelve LHAs in the period (standard population: Tuscan population at 1/1/2015).
CONCLUSIONS: The estimate, based on the national average of incidence published by the Italian MS patient society, indicates an incidence of 1025 new cases of disease in Tuscany in the analyzed period [2]. This estimate do not differ so much from our result of 1147 incident cases. We found a high incidence rate, similarly to other Italian areas [3,4], especially in women, that may explain the increasing prevalence in Tuscany. To confirm this data and to calculate the possible bias caused by our inclusion method, we will validate our algorithm for incident cases. Anyhow we have described an alternative and economic method for estimate incidence rate using administrative data. We have shown that it can be used to estimate MS incidence in a region. The monitoring of chronic diseases with cost-effective and routinely methods, as administrative data, could be very useful in helping healthcare providers.
REFERENCES:
1.Battaglia MA, Bezzini D (2016) Estimated prevalence of multiple sclerosis in Italy in Neurol Sci 38(3):
2.Associazione italiana sclerosi multipla. Barometro della sclerosi multipla. (2016, accessed 20 March 2017).
3.Puthenparampil M, Seppi D, Rinaldi F et al (2013) Multiple Sclerosis Epidemiology Veneto Study Group. Increased incidence of multiple sclerosis in the Veneto region, Italy. Mult Scler 19(5):
4.Solaro C, Ponzio M, Moran E et al (2015) The changing face of MS: prevalence and incidence in an aging population. Mult Scler 21(10):
5.Cocco E, Sardu C, Massa R et al (2011) Epidemiology of multiple sclerosis in south-western Sardinia. Mult Scler 17(11):
6.Lix LM, Yogendran MS, Shaw SY et al (2008) Population-based data sources for chronic disease surveillance. Chronic Diseases in Canada 29(1):31-38
7.Di Domenicantonio R, Cappai G, Arcà M et al (2014) Occurrence of inflammatory bowel disease in central Italy: a study based on health information systems. Dig Liver Dis 46(9):
8.Bezzini D, Policardo L, Meucci G et al (2016) Prevalence of Multiple Sclerosis in Tuscany (Central Italy): A Study Based on Validated Administrative Data. Neuroepid 46:37-42
9.Bargagli AM, Colais P, Agabiti N et al (2016) Prevalence of multiple sclerosis in the Lazio region, Italy: use of an algorithm based on health information systems. J Neurol 263(4):