Polypharmacy in outpatients with relapsing-remitting multiple sclerosis: A single-center study

Abstract

Background: Multiple sclerosis (MS) is an immune-mediated disease of the central nervous system. Given the chronic and heterogenous nature of the disease, treatment with various therapies is a frequent scenario in clinical practice. In persons with chronic morbidity such as MS patients, polypharmacy can give rise to considerable health problems.

Objectives: The aim of the present study was to examine the frequency of polypharmacy among relapsing-remitting (RR) MS patients as well as to analyse sociodemographic and clinical factors, which might be associated with polypharmacy (use of five or more medications). Differences in medication between MS patients with and without secondary illnesses (PwSI and Pw/oSI), between men and women and between patients with and without polypharmacy (PwP and Pw/oP) were examined.

Methods: For 145 RRMS outpatients, we prospectively collected data by means of anamnesis, patient records, clinical examination and a structured patient interview. This was followed by comparative analyses of various patient subgroups (PwP vs. Pw/oP, PwSI vs. Pw/oSI, men vs. women).

Results: The proportion of included MS patients with polypharmacy (use of ≥5 medications) was 30.3%. PwP were significantly older than Pw/oP (45.9 vs. 41.7 years), had a lower level of education and showed a significantly higher median EDSS score (3.0 vs. 2.0). Comorbidities (p<0.001; odds ratio [OR] = 6.293) and higher EDSS scores (p = 0.029; OR = 1.440) were associated with a higher risk of polypharmacy. The proportion of polypharmacy among PwSI was approximately four times higher than among Pw/oSI (46.8% vs. 11.8%). Particularly in the use of antihypertensives, gastrointestinal drugs and dietary supplements, there were differences between Pw/oP and PwP.

Conclusion: We found a high burden of polypharmacy in patients with RRMS. This particularly applies to more severely disabled MS patients who suffer from comorbidities.

Fig 1. Distribution of the total number of medications taken by the examined MS patients.

Only few MS patients (16.6%, N = 24) in this study took more than five drugs. Most frequently, MS patients took two medications (26.9%, N = 39). MS, multiple sclerosis.

Fig 2. Comparison of EDSS scores between Pw/oP and PwP.

Boxplot of the degree of disability of the patients stratified by the presence of polypharmacy according to the total number of medications taken (Pw/oP: N = 101; PwP: N = 44). The boxes mark the upper and lower quartiles of the EDSS scores per group. The medians are indicated by horizontal lines. The whiskers extend to the minimum and maximum values. PwP had, on average, a significantly higher level of disability than Pw/oP, as evaluated by EDSS (Mann-Whitney U test: p<0.001). EDSS, expanded disability status scale; N, number of patients; p, p-value; PwP, patients with polypharmacy; Pw/oP, patients without polypharmacy.

Fig 3. Comparison of polypharmacy rates between patients with and without comorbidities at different levels of disability.

The patients were split into four groups according to EDSS score and the presence of comorbidities. (1) Pw/oSI with EDSS ≤ 2.0 (that is, below the median of the total population): Three of the 35 patients had polypharmacy (8.6%). (2) Pw/oSI with EDSS > 2.0: Five of the 33 patients had polypharmacy (15.2%). (3) PwSI with EDSS ≤ 2.0: Ten of the 29 patients had polypharmacy (34.5%). (4) PwSI with EDSS > 2.0: Twenty-six of the 48 patients had polypharmacy (54.2%). Considering all four polypharmacy rates, the highest proportion of polypharmacy occurred in MS patients with comorbidity and high EDSS scores. Using both factors in a logistic regression model of polypharmacy yielded an overall prediction accuracy rate of 71.7%. EDSS, expanded disability status scale; MS, multiple sclerosis; Pw/oSI, patients without secondary illnesses; PwSI, patients with secondary illnesses.

Fig 4. Correlation plot for the associations between variables and polypharmacy status.

The symmetric correlation matrix was created using the “corrplot” R package. The colours represent the degree of pairwise correlation regarding Spearman’s rank correlation coefficient (rho). The crosses indicate absence of correlation (asymptomatic t approximation p-values > 0.05). For example, EDSS and the number of symptomatic drugs as well as age and comorbidities correlated with each other. DMD, disease-modifying drug; EDSS, expanded disability status scale; OTC, over-the-counter; PRN, pro re nata.