Session Information
Session Type: ACR Poster Session A
Session Time: 9:00AM-11:00AM
Background/Purpose:
Rheumatoid arthritis (RA) is a significant and well-established risk factor for atherosclerotic cardiovascular disease (ASCVD). Cardiovascular risk stratification is often left to the primary care provider (PCP), who may not be aware of the increased risk, leaving risk inadequately addressed. We aimed to increase documentation and identification of RA patients’ ASCVD risk through the use of an automated phrase in an urban fellows’ clinic.
Methods:
RA patients were seen at a large public hospital serving a predominantly black underserved population. Providers imported an automated phrase into clinic notes for RA patients. The phrase included the 2013 American College of Cardiology/American Heart Association ASCVD 10-year risk score, along with associated risk factors. We compared demographic data, RA disease characteristics, ASCVD risk factors, and documentation of risk factors between the 8-week period before implementation (control) and the 8-week period after (study). All patients with a diagnosis of RA were included. Documentation of ASCVD risk was compared using chi-squared testing.
Results:
A total of 343 RA patient encounters were reviewed (164 in study, 179 in control; 278 patients total). There were no differences between the study and control groups in terms of baseline demographics, RA characteristics, or ASCVD risk factors (Table 1). The automated phrase was used in 93 (56.7%) patient encounters during the study period. ASCVD risk score calculation was attempted in 71 (54.2%) study visits vs. 5 (3.4%) controls (p <0.0001), and was successful in 47 (35.0%) study visits vs. 4 (2.7%) controls (p<0.001). Tobacco use was documented in 99 (60.4%) study visits, vs. 38 (21.2%) controls (p <0.0001). CDAI (clinical disease activity index) scores were documented in 109 (66.5%) study visits vs. 85 (47.5%) controls (p 0.005). Documentation of diabetes status and smoking cessation interventions were not different between the two groups (Table 2).
|
Table 1: Demographic Data and Disease Characteristics Contributing to ASCVD Risk |
|||
|
Study (n=127) |
Control (n=151) |
p value |
|
|
Age – years |
55.7 ± 13.6 |
58.5 ± 12.3 |
0.07 |
|
Female sex – no. (%) |
109 (85.8) |
117 (77.5) |
0.07 |
|
Race – no. (%) |
|||
|
Black or African-American |
97 (76.4) |
119 (78.8) |
0.63 |
|
White |
4 (3.1) |
2 (1.3) |
0.30 |
|
Hispanic |
24 (18.9) |
23 (15.2) |
0.44 |
|
Other |
2 (1.6) |
7 (4.6) |
0.15 |
|
PCP visits – no. (%) |
|||
|
Patients with a PCP |
109 (85.8) |
136 (90.1) |
0.28 |
|
PCP visit within 3 months |
51 (40.2) |
69 (45.7) |
0.29 |
|
PCP visit within 6 months |
86 (67.7) |
98 (64.9) |
0.31 |
|
PCP visit within 12 months |
101 (79.5) |
124 (82.1) |
0.38 |
|
Pre-existing ASCVD – no. (%) |
8 (6.3) |
16 (10.6) |
0.20 |
|
ASCVD 10-Year Risk Score |
|||
|
Risk score (%) |
10.19 ± 8.9 |
12.5 ± 7.5 |
0.62 |
|
No. (%) ≥ 7.5%* |
20 (51.3) |
3 (75) |
0.32 |
|
Blood pressure (BP) |
|||
|
Systolic BP – mmHg |
131.4 ± 14.9 |
132.2 ± 17.4 |
0.41 |
|
Diastolic BP – mmHg |
76.5 ± 11.1 |
81.3 ± 49.9 |
0.30 |
|
Systolic BP – mmHg ≥ 140 |
39 (30.7) |
54 (35.8) |
0.37 |
|
On anti-HTN – no. (%) |
74 (58.3) |
89 (58.9) |
0.91 |
|
Lipids |
|||
|
Lipid panel within 3 years – no. (%) |
89 (70.1) |
100 (66.2) |
0.69 |
|
Total cholesterol – mg/dL |
175.6 ± 43.4 |
177.2 ± 40.3 |
0.83 |
|
Triglycerides – mg/dL |
113.6 ± 75.1 |
119.3 ± 71.0 |
0.56 |
|
High density lipoprotein – mg/dL |
57.9 ± 21.0 |
54.5 ± 18.4 |
0.18 |
|
Low density lipoprotein – mg/dL |
96.4 ± 35.6 |
100.2 ± 32.7 |
0.40 |
|
Statin Use |
|||
|
No. (%) |
46 (36.2) |
53 (35.1) |
0.85 |
|
Statin and risk ≥ 7.5% – no. (%)* |
12 (60.0) |
1 (33.3) |
0.33 |
|
Diabetes |
|||
|
No. (%) |
34 (26.8) |
31 (20.5) |
0.22 |
|
Hgb A1C – % |
7.5 ± 2.1 |
7.3 ± 1.9 |
0.65 |
|
Smoking – no. (%) |
26 (20.5) |
22 (14.6) |
0.19 |
|
Body Mass Index (BMI) |
|||
|
(kg/m2) |
29.8 ± 6.9 |
30.5 ± 7.1 |
0.40 |
|
Underweight (BMI < 18.5) |
3 (2.4) |
1 (0.7) |
0.24 |
|
Normal (≥ 18.5, < 25) |
28 (22.0) |
32 (21.2) |
0.99 |
|
Overweight (≥ 25, <30) |
42 (33.1) |
51 (33.8) |
0.90 |
|
Obese (≥ 30) |
54 (42.5) |
68 (45.0) |
0.76 |
|
Aspirin use – no. (%) |
60 (47.2) |
69 (45.7) |
0.81 |
|
Seropositive – no. (%) |
121 (95.3) |
137 (90.7) |
0.14 |
|
CDAI |
|||
|
Mean ± SD |
10.6 ± 11.7 |
12.4 ± 12.9 |
0.41 |
|
CDAI ≥ 10+ |
26 (36.1) |
29 (50.9) |
0.13 |
|
Use of prednisone |
|||
|
Current use – no. (%) |
88 (69.3) |
108 (71.5) |
0.68 |
|
Current dose – mg/day |
4.7 ± 5.3 |
4.3 ± 4.3 |
0.45 |
|
Dose ≥ 8mg/day – no. (%) |
27 (21.3) |
21 (13.9) |
0.11 |
|
Values are depicted as number (%) of individuals unless otherwise specified. Plus-minus values are means ± SD. P values were calculated using two-sided T test for comparison of means and chi-squared testing for comparison of proportions. Duplicate patient encounters were excluded from analysis. *: Denominators include patients with a successfully calculated ASCVD 10-yr risk score calculation only (n = 39 for study and n= 4 for control). ^: Denominators include patients with pre-existing ASCVD only (n=8 for study and n=16 for control). +: Denominators include patients with a calculated CDAI only (n= 72 in study and n= 57 in control). |
|||
|
Table 2: Implementation of the use of an automated phrase and documentation of ASCVD risk |
|||
|
Study (n=164) |
Control (n=179) |
p value¶ |
|
|
Use of automated phrase – no. (%) |
93 (56.7%) |
0 (N/A) |
N/A |
|
ASCVD risk score documentation * |
|||
|
Attempted – no. (%) |
71 (54.2) |
5 (3.4) |
<0.0001 |
|
Calculated – no. (%) |
47 (35.9) |
4 (2.7) |
<0.0001 |
|
Diabetes documentation |
|||
|
Diabetes documented – no. (%) |
33 (76.7) |
27 (73) |
0.362 |
|
Tobacco use documentation# |
|||
|
Documented (all) – no. (%) |
99 (60.4) |
38 (21.2) |
<0.0001 |
|
Documented (smokers)# – no. (%) |
28 (77.8) |
13 (50.0) |
0.023 |
|
Intervention (smokers)# – no. (%) |
16 (44.4) |
7 (26.9) |
0.416 |
|
CDAI documentation – no. (%) |
109 (66.5) |
85 (47.5) |
0.005 |
|
Values are depicted as number (%) of patient encounters unless otherwise specified. Documentation encounters included 278 patients; 4 patients were seen 3 times, 57 patients were seen twice, and the remaining patients were seen once. ¶P values were calculated with chi-squared testing. *: Denominators include patients eligible for ASCVD 10-yr risk score calculation only (patients ≥ 40 and ≤ 79 years and without pre-existing ASCVD; n = 131 for study and n=146 for control). #: Denominators differ from the totals, and include smokers only (n=36 in study and n=26 in control) |
|||
Conclusion:
ASCVD risk factors were identified in our predominantly black RA population, including uncontrolled hypertension, high mean BMI, high disease activity, and high prednisone use. The use of an automated phrase increased ASCVD risk score, tobacco use, and CDAI score documentation. Although the majority of patients were seen by a PCP within the past 6 months, statins were under-prescribed, and a significant proportion of blood pressures were not at goal. PCPs may underestimate the increased ASCVD risk in RA. Further interventions are needed to increase awareness among PCPs using a multi-disciplinary approach.
To cite this abstract in AMA style:
Cassidy LA, Brandt J, Gizinski A, Khosroshahi A, Bao G, Lim SS. Implementation of an Automated Phrase to Increase Awareness of Cardiovascular Risk in Rheumatoid Arthritis Patients in an Urban Fellows Rheumatology Clinic [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 9). https://acrabstracts.org/abstract/implementation-of-an-automated-phrase-to-increase-awareness-of-cardiovascular-risk-in-rheumatoid-arthritis-patients-in-an-urban-fellows-rheumatology-clinic/. Accessed .« Back to 2018 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/implementation-of-an-automated-phrase-to-increase-awareness-of-cardiovascular-risk-in-rheumatoid-arthritis-patients-in-an-urban-fellows-rheumatology-clinic/