Effects of 12 mg vs. 6 mg dexamethasone on thromboembolism and bleeding in patients with critical COVID-19 - a post hoc analysis of the randomized, blinded COVID STEROID 2 trial.
Academic Article
Overview
abstract
BACKGROUND: Black Americans have high prevalence of hypertension (HTN), and many have suboptimal blood pressure (BP) control. Black American residents of the rural Southeast face particularly steep challenges to achieving BP control, including historical oppression leading to rampant poverty and poor access to medical care and medications. Peer coaching (PC) can improve self-care for chronic diseases such as diabetes but has not been tested in Black American residents with HTN of the rural southeastern United States. Primary care practices serving residents of this area are often resource constrained and have little training in quality improvement (QI) or population health management. Practice facilitation (PF) is a promising approach to assisting medical practices with population health management and QI but has not been tested in the rural South for HTN control exclusively among Black Americans. OBJECTIVE: To test the hypothesis that PC or PF, alone or in combination, results in better BP control (defined as <140/90 mm Hg) than that with enhanced usual care (EUC) among Black Americans with persistently uncontrolled HTN. Prespecified subgroup analyses were planned by sex, age less than 60 years or 60 years or greater, depressive symptoms, and health literacy. METHODS: This stakeholder-engaged, 4-arm, cluster-randomized controlled trial engaged 69 rural primary care practices in Alabama and North Carolina to test the effectiveness of PF, PC, PF plus PC, and EUC on BP control (percentage with BP <140/90 mm Hg). The primary outcome was change in BP control, and the secondary outcome was change in mean systolic BP (SBP). Each study site enrolled approximately 25 Black American patients with persistently uncontrolled HTN SBP of 140 mm Hg or greater in the year before enrollment plus research-grade BP measurement on enrollment of 140/90 mm Hg or higher. Practices assigned to PC worked for 1 year with a trained community member on a structured telephone-delivered program with content on healthy eating, physical activity, home BP monitoring, stress reduction, social support, and communication with the health care team. Practices assigned to PF worked with a practice facilitator over 1 year to implement at least 4 QI activities focused on improving BP control in each of 4 areas: clinical information systems, team-based care, standardized care processes, and self-management support. All practices received a binder of tools and tips, an algorithm for BP medication titration customized to Black Americans, a laptop computer, and access to a novel patient education system. All participating patients received a home BP monitor. Baseline, 6-month, and 12-month data included research-grade BP measurements, surveys, and medical record data. Main analyses were at the patient level, accounting for clustering of patients within practices and imbalanced characteristics across treatment arms. Secondary analyses examined changes in SBP. RESULTS: Of 248 practices contacted, 39 were ineligible, 99 declined to participate, and 41 were in the recruitment process when the target number of 69 practices was reached (final analytic sample: 16 EUC [342 participants], 12 PF [260 participants], 17 PC [332 participants], 15 PF plus PC [275 participants]). Nearly half (48%) of the practices were federally qualified health centers; the mean number of years in operation was 19, the mean numbers of full-time providers and staff were 4 and 13, respectively; and 21% of their patients lacked health insurance. Of 1592 randomly assigned participants, 1336 (84%) participants completed the 12-month follow-up; their mean (SD) age was 58 (12) years, 62% were women, 22% had less than a high school education, 45% reported an annual household income of $20 000 or less, and their mean BP at baseline was 156/90 mm Hg. Eighty-two coaches (PC) and 6 facilitators (PF) completed training. Of 804 participants receiving PC, 462 (57%) participants completed the intervention. No intervention arm achieved significantly better BP control than with EUC, either overall or by any prespecified subgroups. However, PC resulted in 5.26-mm Hg (95% CI −9.29 to −1.24 mm Hg) lower SBP than with EUC for individuals younger than 60 years; compared with EUC, a greater proportion of participants who received PC had improved medication adherence (43% vs 33%, P = .04). Facilitators succeeded in implementing at least 4 QI activities over the year-long intervention in all PF practices and estimates of monthly practice-level BP control increased from 55% at baseline to 61% over the year (P < .01). Medications were intensified in 38% of participants regardless of intervention arm. CONCLUSIONS: Neither PC nor PF resulted in better BP control among rural Black American patients enrolled in this trial, whether overall or for any prespecified subgroups. However, PC resulted in improved SBP for individuals aged less than 60 years despite a relatively high rate of PC intervention noncompletion. LIMITATIONS: We could not test practice-level outcomes rigorously, because this would have required greater-than-available resources to access clinical information systems; therefore, clinically important practice-level effects may have been missed. Floods, hurricanes, and the COVID-19 pandemic may have negatively influenced participant retention. Physicians were generally unreceptive to nurses implementing algorithm-driven BP medication titration, possibly decreasing the potency of the PF intervention.
