Increasing goals of care conversations in primary care: Study protocol for a cluster randomized, pragmatic, sequential multiple assignment randomized trial.

BACKGROUND: Goals of care conversations explore seriously ill patients' values to guide medical decision making and often inform decisions about life sustaining treatments. Ideally, conversations occur before a health crisis between patients and clinicians in the outpatient setting. In the United States Veterans Affairs (VA) healthcare system, most conversations still occur in the inpatient setting. Strategies are needed to improve implementation of outpatient, primary care goals of care conversations. METHODS: We plan a cluster randomized (clinician-level) sequential, multiple assignment randomized trial to evaluate the effectiveness of patient implementation strategies on the outcome of goals of care conversation documentation when delivered in combination with clinician implementation strategies. Across three VA healthcare system sites, we will enroll primary care clinicians with low rates of goals of care conversations and their patients with serious medical illness in the top 10th percentile of risk of hospitalization or death. We will compare the effectiveness of sequences of implementation strategies and explore how patient and site factors modify implementation strategy effects. Finally, we will conduct a mixed-methods evaluation to understand implementation strategy success or failure. The design includes two key innovations: (1) strategies that target both clinicians and patients and (2) sequential strategies with increased intensity for non-responders. CONCLUSION: This study aims to determine the effect of different sequences and combinations of implementation strategies on primary care documentation of goals of care conversations. Study partners, including the VA National Center for Ethics in Health Care and Office of Primary Care, can consider policies based on study findings.

Effectiveness of the rural transitions nurse program for Veterans: A multicenter implementation study.

BACKGROUND: Veterans are often transferred from rural areas to urban VA Medical Centers for care. The transition from hospital to home is vulnerable to postdischarge adverse events. OBJECTIVE: To evaluate the effectiveness of the rural Transitions Nurse Program (TNP). DESIGN, SETTING, AND PARTICIPANTS: National hybrid-effectiveness-implementation study, within site propensity-matched cohort in 11 urban VA hospitals. 3001 Veterans were enrolled in TNP from April 2017 to September 2019, and 6002 matched controls. INTERVENTION AND OUTCOMES: The intervention was led by a transitions nurse who assessed discharge readiness, provided postdischarge communication with primary care providers (PCPs), and called the Veteran within 72 h of discharge home to assess needs, and encourage follow-up appointment attendance. Controls received usual care. The primary outcomes were PCP visits within 14 days of discharge and all-cause 30-day readmissions. Secondary outcomes were 30-day emergency department (ED) visits and 30-day mortality. Patients were matched by length of stay, prior hospitalizations and PCP visits, urban/rural status, and 32 Elixhauser comorbidities. RESULTS: The 3001 Veterans enrolled in TNP were more likely to see their PCP within 14 days of discharge than 6002 matched controls (odds ratio = 2.24, 95% confidence interval [CI] = 2.05-2.45). TNP enrollment was not associated with reduced 30-day ED visits or readmissions but was associated with reduced 30-day mortality (hazard ratio = 0.33, 95% CI = 0.21-0.53). PCP and ED visits did not have a significant mediating effect on outcomes. The observational design, potential selection bias, and unmeasurable confounders limit causal inference. CONCLUSIONS: TNP was associated with increased postdischarge follow-up and a mortality reduction. Further investigation to understand the reduction in mortality is needed.

Risk of MRSA Infection in Patients with Intermittent versus Persistent MRSA Nares Colonization.

OBJECTIVE: To determine the relative risk of invasive methicillin-resistant Staphylococcus aureus (MRSA) infection among non-colonized (NC) patients, intermittently colonized (IC) patients, and persistently colonized (PC) patients. DESIGN: Observational cohort study of patient data collected longitudinally over a 41-month period. SETTING: Department of Veterans Affairs Eastern Colorado Healthcare System, a tertiary care medical center. PATIENTS: Any patient who received ≥5 MRSA nasal swab tests between February 20, 2010, and July 26, 2013. In total, 3,872 patients met these criteria, 0 were excluded, 95% were male, 71% were white, and the mean age was 62.9 years on the date of study entry. METHODS: Patients were divided into cohorts based on MRSA colonization status. Physicians reviewed medical records to identify invasive infection and were blinded to colonization status. Cox and Kaplan-Meier analyses were used to assess the relationship between colonization status and invasive infection. RESULTS: In total, 102 patients developed invasive MRSA infections, 16.3% of these were PC patients, 11.2% of these were IC patients, and 0.5% of these were NC patients. PC patients were at higher risk of invasive infection than NC patients (hazard ratio [HR] 36.8; 95% CI, 18.4-73.6; P<.001). IC patients were also at higher risk than NC patients (HR, 22.8; 95% CI, 13.3-39.3; P<.001). The difference in risk between PC and IC patients was not statistically significant (HR, 1.61; 95% CI, 0.94-2.78, P=.084). Alternate analysis methods confirmed these results. CONCLUSIONS: The risk of invasive MRSA infection is much higher among PC and IC patients, supporting routine clinical testing for colonization. However, this risk is similar among PC and IC patients, suggesting that distinguishing between the 2 colonization states may not be clinically important.

Dual-energy X-ray absorptiometry and evaluation of the osteoporosis self-assessment tool in men with rheumatoid arthritis.

Males with rheumatoid arthritis (RA) are at risk for osteoporosis but infrequently undergo dual-energy X-ray absorptiometry (DXA). We examined the frequency of DXA in males enrolled in the Veterans Affairs Rheumatoid Arthritis Registry. The Osteoporosis Self-Assessment Tool (OST) index, a formula using age and weight, was calculated for all subjects. DXA was performed on 282 (35.5%) of the males who were younger (p < 0.01), had lower mean OST index score (p < 0.05), and were more likely to have been prescribed prednisone (p < 0.01) than subjects without DXA. Low bone mass (T-score < -1) was present in 73% of subjects with DXA; 37% of subjects with low-risk OST index scores had normal bone mineral density (BMD) compared with 5.6% of those with high-risk OST index scores (p < 0.01). There was a significant but modest correlation between BMD and the OST index (r = 0.17, p < 0.01). No OST score had a sensitivity and specificity of more than 80%. Association between OST index and BMD was strongest in non-Hispanic whites, subjects older than 60 yr, and smokers. DXA was underutilized in males with RA. The OST index correlated with low bone mass but could not reliably predict osteoporosis in this population.

Distal revascularization and interval ligation (DRIL) procedure for the treatment of ischemic steal syndrome after arm arteriovenous fistula.

The objective of this study was to investigate the efficacy of the distal-revascularization-interval ligation (DRIL) technique in alleviating symptoms of ischemic steal syndrome and in preserving hemodialysis access. A retrospective chart review was conducted of all patients receiving the DRIL procedure in a 3-year period. There were 38 DRIL bypass grafts identified in 35 patients, with 36 DRILs with follow-up adequate for analysis. Comparison of preoperative and postoperative digital pulse volume recording (PVR) data was made using the t-test. The majority of patients presented with multiple ischemic symptoms, most commonly coolness, pain, and paresthesias. Six patients presented with frank digital necrosis. The mean interval to DRIL was 4.9 months following fistula construction (range 0.1-24). In 66.7% of patients for whom complete follow-up data were available (24/36), all ischemic symptoms were alleviated by DRIL. Of the remaining 12 patients, 11 experienced partial symptom relief. One patient required digital amputation following DRIL. A comparison of pre- and post-DRIL PVRs illustrated a significant increase in these values following DRIL (P < 0.05). DRIL effectively eliminates ischemic symptoms in the majority of patients and produces a significant increase in flow to the ischemic limb. The data support the usage of DRIL as the procedure of choice in the correction of ischemic steal following arm arteriovenous fistula.