Session Information
Session Type: ACR Poster Session A
Session Time: 9:00AM-11:00AM
Background/Purpose: Low medication adherence in lupus has been associated with increased hospitalizations, more severe disease activity, and irreversible multi-organ damage. While lower socioeconomic levels, education, depression and polypharmacy have been identified as common determinants of non-adherence among patients with chronic diseases, few studies have investigated the rates of medication adherence with lupus.
Purpose: To compare medication adherence among patients with lupus at high risk for multiple admissions with all admissions for lupus.
Methods: We previously identified 171 lupus patients with a confirmed diagnosis that were admitted to Strong Memorial Hospital between July 1st of 2013 and June 30th of 2015. We then classified a high risk group of 28 lupus patients who had required ≥ 3 admissions/ year over the 2 years. For this study we linked the database of all lupus patients with pharmacy claims database for the same period to calculate the medication possession ratio (MPR), an indicator of whether a patient had adequate medication supply in a given time frame. For the bivariate analysis, we used t-tests and chi-square tests to check for the differences in distribution of patient demographics and MPR across the high-risk and non-high risk group. For the multivariate analyses, we estimated hierarchical linear regression models and controlled for the clustering of refills by patient and medication. We also controlled for patient demographics and medication details. We used two-tailed hypothesis tests and p-value<0.05 to indicate statistical significance.
Results: The high-risk group was significantly younger (mean age 39.64 years [SD: 19.09] as compared to mean age of 47.57 years in non-high risk group, (p=0.03), 82% were females compared to 92% in the non-high risk group and the group had significantly higher proportion of African Americans 61% as compared to 41% in the non-high risk group (p=0.05) see Table 1. Complete pharmacy data was available for 102 patients. The mean MPR was lower among the high-risk group (73.40% as compared to 79.93% in the non-high risk group). Our multivariate analysis showed that after controlling for relevant confounders, on average high-risk patients had 10 percent point lower MPR as compared to non-high risk patients (Estimate: -10.41, 95% CI: [-21.36 to 0.54], p=0.06).
Conclusion: Medication non-adherence is a major cause of increased risk for admissions among patients with lupus. Targeting measures to improve medication adherence is an important component of the management of patients with lupus.
|
All lupus patients (n-143) |
High risk lupus (n=28) |
Total (n=171) |
p-value (Fischer/ chi-square) |
|
|
Age: mean (SD) |
47.57 (17.32) |
39.64 (19.09) |
46.27 (17.81) |
0.03 |
|
Females: n (col%) |
132 (92.31) |
23 (82.14) |
155 (90.64) |
0.09 |
|
Race: n (col%) |
||||
|
Asian |
2 (1.40) |
2 (7.14) |
4 (2.34) |
0.13 |
|
African-American |
58 (40.56) |
17 (60.71) |
75 (43.86) |
0.05 |
|
Caucasian |
76 (53.15) |
8 (28.57) |
84 (49.12) |
0.02 |
|
Hispanic |
7 (4.90) |
1 (3.57) |
8 (4.68) |
1.00 |
|
Average number of medications: mean (SD) |
2.12 (1.02) |
2.74 (1.15) |
2.24 (1.06) |
0.02 |
|
Average number of fills per medication: mean (SD) |
6.64 (5.78) |
5.29 (4.82) |
6.39 (5.62) |
0.35 |
|
Average MPR: mean (SD) |
79.93 (25.02) |
73.40 (22.80) |
78.71 (24.64) |
0.30 |
|
Average MPR for top 4 medications: mean (SD) |
||||
|
Prednisone |
78.51 (36.42) |
69.31 (29.02) |
76.67 (35.06) |
0.38 |
|
Hydroxychloroquine |
78.78 (22.58) |
76.19 (25.39) |
78.29 (22.98) |
0.71 |
|
Mycophenolate |
83.83 (32.15) |
62.22 (38.18) |
77.87 (34.63) |
0.14 |
Table: 1 Descriptive data of lupus cohorts
|
Risk factor |
Confidence interval |
p-value |
|
|
Non high risk High risk |
Reference -10.41 |
[-21.36,0.54] |
0.06 |
|
Age |
0.15 |
[-0.11,0.41] |
0.25 |
|
Female Male |
Reference -8.42 |
[-23.50,6.66] |
0.27 |
|
Caucasian African American Asian Hispanic |
Reference -4.01 27.61 -16.01 |
[-13.29,5.26] [-0.84,56.06] [-39.08,7.06] |
0.40 0.06 0.17 |
|
Hydroxychloroquine Azathioprine Cyclophosphamide Cyclosporine Dexamethasone Leflunomide Methotrexate Methylprednisolone Mycophenolate Prednisone Tacrolimus |
Reference 3.50 -18.23 32.72 33.62 15.95 -0.23 2.54 0.06 -0.18 33.52* |
[-14.33,21.32] [-73.77,37.32] [-21.82,87.26] [-21.54,88.78] [-12.84,44.74] [-16.81,16.35] [-10.94,16.02] [-11.77,11.88] [-8.97,8.62] [1.11,65.94] |
0.70 0.52 0.24 0.23 0.28 0.98 0.71 0.99 0.97 0.04 |
Table 2: Multivariate analysis of medication possession ratio
To cite this abstract in AMA style:
Thirukumaran C, McCarthy K, Patel J, Anandarajah AP. Decreased Medication Adherence Is a Major Cause for Increased Risk of Hospitalizations Among High Risk Lupus Patients [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/decreased-medication-adherence-is-a-major-cause-for-increased-risk-of-hospitalizations-among-high-risk-lupus-patients/. Accessed .« Back to 2017 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/decreased-medication-adherence-is-a-major-cause-for-increased-risk-of-hospitalizations-among-high-risk-lupus-patients/