Background/Purpose: Hypertension (HTN) is the most prevalent comorbid cardiovascular disease (CVD) risk factor among adults with rheumatic conditions.  However, we previously found that high blood pressures (BP) were addressed in <1/3 of rheumatology visits, even when severely elevated (≥160/100 mmHg). In this range, only 11 patients need to be treated for HTN to prevent one CVD event.  In primary care, HTN protocols executed by nurses or medical assistants during vital sign assessment have improved control of high BP.  In a prior six-month pilot study, we had tested a theory-based staff protocol intervention to facilitate timely (<4 weeks per a US quality measure) primary care follow-up for patients with high blood pressures at rheumatology visits, and had reported two-fold higher odds of timely follow-up. Here, our objective is to assess the 18-month sustainability of those improvements.

Methods: We conducted a pre-post study in three academic rheumatology clinics. All eligible adult (≥18 years-old) rheumatology visits with BP ≥140/90 mmHg (Dec. 2014-May 2016) were compared to pre-intervention visits (Jan. 2012-Sep. 2014). Our multi-dimensional intervention included (1) education of frontline staff on HTN and CVD risk in rheumatologic diseases, (2) electronic health record (EHR) alerts for staff to re-measure BPs if ≥140/90 and (3) cuing brief patient education and scheduling primary care follow-up if 2nd BP ≥140/90, and (4) monthly audit and feedback with staff about performance. We gave in-person feedback between months 1 and 6, then by email months 6-18.  We assessed timely primary care follow-up for high BPs among patients who received primary care in our system using EHR data. We performed multivariable logistic regression and compared the odds (OR, 95%CI) of timely primary care follow-up before and during intervention, controlling for baseline socio-demographics, comorbidities, and utilization.  

Results: We compared 1,737 intervention period visits to a comparable group of 4,683 pre-intervention visits with BPs ≥140/90. Staff initiated the protocol with BP re-measurement in 57%, 73%, and 69% of eligible visits during intervention months 0-6, 6-12, and 12-18 respectively, compared to <1% pre-intervention, with 66% overall improvement, p<0.0001. Months 4-6 showed peak improvement in BP re-measurement at 80% during monthly in-person audit-feedback. More patients received timely follow-up for HTN during the entire 18-month intervention period (46% vs 29% before, p<0.0001). Multivariable analysis showed that eligible visits during the intervention had two-fold higher odds of timely follow-up compared to pre-intervention (OR 2.1, 1.4-3.0 in months 0-6; OR 2.16,1.72-2.71 across entire intervention), indicating sustained improvement.  

Conclusion: Our intervention with usual staff in rheumatology clinics doubled odds of timely BP follow-up, and these improvements were sustained over 18 months. Future studies should examine this intervention in other rheumatology clinics to evaluate its impact on HTN control and CVD event risk across rheumatology populations.