Decrease in rate of multiple sclerosis-related hospitalizations in Portugal

We sought to investigate the rate of multiple sclerosis (MS)-related hospitalizations in Portugal and assess whether there have been temporal changes as described in other countries. Using data from the Portuguese National Discharge Registry, we observed that between 2008 and 2013 the rate of MS-related hospitalizations decreased by 44%, from 15.9/100 person-years (95% confidence interval (CI: 14.9-16.9) in 2008 to 8.9/100 person-years (95% CI: 8.2-9.6) in 2013. The change in hospitalization rates is in accordance with what has been observed in other countries, and coincides with the release of new therapies for MS in Portugal.


Introduction
Multiple sclerosis (MS) is a chronic, neurodegenerative autoimmune disorder of the central nervous system 1 . With a prevalence of 1/800 in North America and Northern Europe, MS is the most common acquired neurological disorder in young adults 1 , posing a substantialburden on healthcare systems. Recent studies have, however, suggested that MS-related hospitalization rates have been declining over the last decades 2,3 .
Portugal is considered a region of medium MS prevalence 4 , but epidemiological data are limited. In the present study, we sought to investigate the difference in the rate of MS-related hospitalizations in Portugal between 2008 and 2013. The incidence of hospitalizations (MS-related or all-cause hospitalizations) was calculated by dividing the number of hospitalizations in each year by the number of patients at risk in the same year. The Wald method was used to calculate 95% confidence intervals (CI). Two different definitions of MS-related hospitalizations were used: 1) where MS was the primary reason for admission -the MS ICD-9 diagnostic code was reported in the field of the primary diagnosis, and 2) where MS was reported as either a primary or a secondary reason for admission -the MS ICD 9 diagnostic code was in any position on the diagnoses fields (20 fields available). The proportion of MS-related hospitalizations in each year was obtained by dividing the number of MS-related hospitalizations in each year by the total number of hospitalizations in the same year. Demographic and clinical characteristics of patients with MS hospitalized in each year were described using number and percentage of patients for categorical variables, and mean and standard deviation (variables with a normal distribution) or median and interquartile range (variables with a skewed distribution) for continuous variables. Differences between characteristics in 2008 and 2013 were estimated using t-tests (variables with a normal distribution) or Mann-Whitney test (variables with a skewed distribution) for continuous variables, and Chi square test for other categorical variables. P values < 0.05 were considered to be statistically significant. Statistical analysis was performed using STATA software version 13.
The age of the patients with MS hospitalized and the length of stay in the hospital increased significantly (Table 1) between 2008 and 2013, where MS was either only a primary diagnosis or a diagnosis anywhere on the patient record. There was an increase in the proportions of females admitted to hospital but this was only significant when MS was recorded in any of the diagnosis fields (Table 1; p value = 0.062 for MS as main diagnosis and p value = 0.006 for MS as main or secondary diagnosis).
No differences were observed in the proportion of patients that died during hospitalization (Table 1). Where MS diagnosis was anywhere on the patient record, the proportion of MS patients with other diagnoses during hospitalization increased significantly between 2008 and 2013 for all groups of diseases considered, except for infectious and parasitic diseases, neoplasms, diseases of the nervous system and sense organs (excluding MS), and congenital anomalies (Table 1).

Discussion
Here we show a decrease in MS-related hospitalizations in Portugal from 2008 to 2013. Where MS was the primary diagnosis the decrease was substantial -approximately 44%. The rates of hospitalization observed in Portugal appear to be similar to that documented in Canada 2,3 , suggesting perhaps that the thresholds for admission are similar between the two countries. The change in hospitalization rates cannot be explained entirely by a general change in admissions in Portugal, although this did decline by 4%. The change in hospitalization coincides with the release of new therapies for MS, the first of which, natalizumab (Tysabri ® ), became available in June 2007 in Portugal 7 . It is plausible that these newer therapies prevented some hospital admissions for patients with The raw data of incidence rate of multiple sclerosis related hospitalization between 2008 and 2013 are reported.  MS. The average age of MS patients admitted in 2013 was older as compared to 2008. As MS prevalence has increased over the period, the decrease in hospitalization rate may reflect a diluting by more newly diagnosed patients with lower disease severity. There may also be a change in the way healthcare is delivered, with a possible shift to more outpatient services as seen in other countries. This may explain why the average length of stay for MS-related admissions has increased while the rate of admissions has decreased, although this change in disease management has generally preceded the time period we have investigated in other countries.
Limitations of this study include the lack of clinical data and information on potential confounders. There are also the uncertainties associated with the prevalence data we had available to us. Despite these limitations, our results are in accordance with previous results on hospital admissions in patients with MS. Our findings provide further epidemiological data on MS in Portugal, healthcare resource use in these patients, and impetus to investigate other efforts to reduce hospitalizations in this population.

Data availability
The Portuguese Central Administration of the Health System provided the database and gave permission to use the data to describe in the article. The data are not available online, but are available upon request after approval of the study objectives. Author contributions MP performed the analysis. SVR contributed to the analysis and interpretation of the data. MP wrote the first draft and all authors contributed to subsequent drafts and the final paper.
Competing interests MP, DL and SVR are employees of Evidera.

Grant information
The author(s) declared that no grants were involved in supporting this work. This is an interesting short report that documents a significant fall in the number of hospitalisations for MS in Portugal following the release of higher potency disease modifying drugs. The observation is interesting and adds some support to the presumed greater efficacy of these therapies. Although causality cannot be determined from such an observational study.

Open Peer Review
The finding that the age of admissions increased is also interesting as it may indicate that those with longer duration of disease (directly correlated with age) may have acquired more complications of MS for example urosepsis and have required greater admission rates.
To directly answer the question significantly more detail and prospective acquisition of data is required. however this report does provide a strong background rationale to undertake these studies and to acquire this type of data which is invaluable in assessing the benefits or otherwise of expensive and potentially higher risk therapies in MS.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. There are no methodological errors but, in my opinion, some changes should be introduced: About Introduction when the authors say that the epidemiological data in Portugal are limited I can´t agree. Portugal is a small country of about 10 million people and 3 prevalence studies and 1 incidence study published bring data enough for a knowledge of the basic epidemiological picture 1.
incidence study published bring data enough for a knowledge of the basic epidemiological picture of MS in the country. There is of course other missing data (for instance mortality data/ burden of the disease for the society) but these are often missing in other countries. I considered changing saying for instance that though prevalence and incidence in MS in Portugal are now known, other epidemiological data about the disease are still lacking.
The author stay that they have found a "decrease in MS related hospitalizations..... from 2008 to 2013". In fact they have only these two dates for comparison. They have no data among these two years so that some variability can not be excluded.
In Portugal there are a great diversity of neurological care. Central Hospitals in Lisbon, Oporto and Coimbra treat MS in ambulatory but in other Towns: Évora,Beja, Leiria, Covilhã, Santarem, for example patients are admitted for treating relapses with Methyl, and for Natalizumab infusion because Day-Hospital units are not available. This should be taken in consideration in discussion and could certainly explain why there are so many short-time admissions I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Competing Interests: