Commentaries on health services research.

PAs deployed in nonprimary (secondary) care roles ABSTRACT The authors investigated the deployment of PAs in secondary care teams in England through the use of a cross-sectional electronic, self-report survey. The findings from 14 questions were presented. Sixty-three PAs working in a range of specialties responded. A variety of work settings were reported, most frequently inpatient units, with work generally taking place during weekdays. Both direct and nondirect patient care activities were reported, with the type of work varying at times, depending on the presence or absence of other healthcare professionals. PAs reported working within a variety of secondary care team staffing permutations, with the majority of these being interprofessional. Line management was largely provided by consultants; however, day-to-day supervision varied, often relating to different work settings. A wide variation in ongoing supervision was also reported. Further research is required to understand the nature of PAs' contribution to collaborative care in secondary care teams in England.1 Commentary by Justine Strand de Oliveira: This article offers a tantalizing glimpse into the activities of PAs in secondary care in England, where the profession is rapidly expanding after years of relative quiescence. First introduced in 2003, with a handful of education programs graduating PAs in the ensuing years, the profession began to take off after the UK National Health Service and other stakeholders developed a newfound interest in it in 2013.2 Like PAs in the United States, PAs in the UK are “healthcare workforce stem cells” whose roles differentiate based on need, whether primary or secondary (hospital) care. This small study surveyed UK PAs to identify those in secondary care and learn more about their activities. Specialty areas included emergency medicine, orthopedics, neurosurgery, cardiology, and internal medicine. Settings reported were inpatient and outpatient hospital, ED, and surgery, and many PAs rotate through different settings. Most PAs reported working in interprofessional, multidisciplinary teams. The authors also asked about supervisory arrangements, and a fourth of the 48 PAs who responded about supervision reported no supervision. Whether this indicates the absence of supervision or the absence of a response is an important question. Further research is needed, and this study is an important pilot toward that end. REFERENCES The cost-effectiveness of PA substitution models of inpatient care in the Netherlands ABSTRACT The cost-effectiveness of substitution in the Netherlands was investigated comparing two models of inpatient care: physician-only and PA-physician models. In total, 34 matched-controlled wards in 23 hospitals participated. The study included 2,292 patients (excluding children, day cases, and those terminally ill) who were followed from admission until 1 month after discharge. Health outcome concerned quality-adjusted life years (QALYs) measured through the EuroQol five dimensions questionnaire (EQ-5D). All direct healthcare costs were included in the analysis. No significant difference in QALYs gained or total costs per patient was found between the physician-only and PA/physician groups. Costs associated with length of stay were significantly lower in the physician model. Personnel costs were significantly lower in the PA/physician model. Supervision costs were significantly lower in the PA/physician model when those wards with medical specialists (that is, without residents) were excluded.1 Commentary by Vari M. Drennan: Cost-effectiveness analysis of PA substitution for physicians based on multicenter data is rare and this is a welcome addition to the evidence base. This study included comparative data on physician supervision time, which was obtained through online questionnaires. This was one aspect that could not be quantified in the economic analysis within a multisite comparative study of substitution by PAs for family physicians in primary care in England.2 Supervision is a nuanced concept, which can include teaching for junior or less-experienced members of the medical team by the seniors or specialists. Understanding context is important not just for judging the generalizability of individual studies but also for judging the transferability of evidence between different healthcare systems. An example of country-specific context is that PAs in the UK do not have legal authority to prescribe medications or ionizing radiation, unlike those in the United States and the Netherlands. Contextual factors such as these influence employing medical directors' viewpoints in considering cost-effective staffing decisions.3 REFERENCES Do physicians take care of sicker patients than NPs or PAs? ABSTRACT Expanded use of PAs and NPs is a potential solution to workforce issues, but little is known about how they can best be used. One question about medical and social complexity of patients is associated with whether their primary care provider (PCP) is a physician, PA, or NP. In this cohort study, the 2012-2013 Veterans Health Administration electronic health record data from 374,223 veterans were examined. The study sought to determine whether PCP type is associated with patient, clinic, and state-level factors representing medical and social complexity, adjusting for all variables simultaneously using a generalized logistical regression. Results indicate that patients with physician PCP are modestly more medically complex than those with a PA or NP as PCP. For the group having a Diagnostic Cost Group (DCG) score greater than 2 compared with the group having a DCG less than 0.5, odds of having a PA or NP as PCP were lower than for having a physician PCP. Social complexity is not consistently associated with PCP type. Overall, minor differences were found in provider type assignment. This study builds on previous work by using a large national dataset that accurately ascribes the work of PAs and NPs, analyzing at the patient level, analyzing providers separately, and addressing social as well as medical complexity. This is a requisite step toward studies that compare patient outcomes by provider type.1 Commentary by Salim S. Virani: Using a large national VA dataset and well-defined algorithms to attribute primary care to a particular provider type, the authors compared whether medical or social complexity of patients with diabetes seeking care from physicians differs from those seeking care from PAs or NPs.1 Using a well-validated measure of global medical complexity DCG, the authors showed that patients with higher comorbidity burden (DCG greater than 2) were modestly (15%) less likely to see PAs or NPs compared with physicians, although this difference was numerically small and clinically not meaningful. Most measures of patients' social complexity (such as race, ethnicity, homelessness, substance abuse, mental health diagnoses) also were numerically comparable. These data combined with other data showing mostly comparable illness burden, effectiveness of diabetes and cardiovascular disease care, and healthcare resource use (even between PAs and NPs).2-4 These observations indicate that it is time to break free of dogmas and identify ways to improve team-based delivery of outpatient chronic disease care. Managers need to identify ways to most efficiently incorporate PAs and NPs in the nation's primary care workforce. REFERENCES