Raising the Bar for Physicians Practicing in Nursing Homes: The Path to Sustainable Improvement

Antibiotic Prescribing Patterns for Urinary Tract Infections and Pneumonia by Prescriber Type and Specialty in Nursing Home Care, 2016-2018

OBJECTIVE

To identify whether differences in antibiotic prescribing practices by prescriber type and specialization in nursing home (NH) care exist for urinary tract infection (UTI) and pneumonia.

DESIGN

Retrospective cohort.

SETTING AND PARTICIPANTS

This national study included antibiotic dispensings to traditional Medicare beneficiaries aged ≥65 years with UTI or pneumonia infections residing long-term (≥100 days) in US NHs between 2016 and 2018.

METHODS

Minimum Data Set assessment data were linked to Medicare data [Part D prescription drug, inpatient hospital (MedPAR), prescriber characteristics, and enrollment]. We compared antibiotic prescribing patterns by prescriber type [physician vs advanced practice practitioner (AP)] and NH specialization (≥90% vs <90% of all associated medication dispensings to NH residents). Antibiotic dispensing measures included the total number of dispensings and duration of therapy (median number of days supplied) by antibiotic class.

RESULTS

There were 264,735 antibiotic dispensings prescribed by 32,437 prescribers for 140,360 residents in 14,035 NHs. NH specialists were less likely to prescribe fluoroquinolones for UTI (22.9% NH specialist physician, 23.9% non-NH specialist physician, 21.3% NH specialist AP, 24.2% non-NH specialist AP), but more likely to prescribe fluoroquinolones for pneumonia (38.9%, 37.8%, 38.8%, 37.3%, respectively). Over time, NH specialists reduced fluoroquinolone prescribing for pneumonia to a greater extent than non-NH specialists. The duration of therapy was similar across prescriber groups for UTI, but longer among non-NH specialist APs for several antibiotic classes for pneumonia, including tetracyclines, glycopeptides and lipoglycopeptides, and metronidazole.

CONCLUSIONS AND IMPLICATIONS

There were differences in antibiotic prescribing patterns by prescriber type and specialization in NH care between 2016 and 2018. Understanding how antibiotic prescribing differs based on prescriber characteristics is essential to inform antibiotic stewardship efforts. Tailoring antibiotic stewardship efforts to prescribers by NH specialization is rational given differences in antibiotic prescribing patterns based on NH specialization.

Qualitative Evaluation of a Pilot Physician Training Program in Long-Term Care Facilities in Switzerland

Increasing demand for long-term care facilities (LTCFs) and increasingly complex medical needs of LTCF residents necessitate recruiting new physician graduates and developing programs which provide them with experience and specific skills in geriatric and palliative medicine. We developed an educational program for physicians in training, which combines 1 year of immersion at 20% full-time equivalent with (1) theoretical teaching, (2) clinical exposure under the supervision of a senior LTCF physician, and (3) structured case presentations and debriefing with experienced geriatricians. This article presents a preliminary qualitative evaluation of this program from the perspectives of the trainees, supervising physicians, and public health stakeholders. The program was well accepted by all parties and showed positive effects on trainees' interest in long-term care medicine. Suggestions for improving future programs and their implementation are provided.

Focusing on Provider Quality Measurement: Continued Consensus and Feasibility Testing of Practice-Based Quality Measures for Primary Care Providers in Long-Term Care

Medical providers in long-term care (LTC) use a unique skillset in delivering comprehensive resident care. Publicly reported quality measures (QMs) do not directly emphasize medical provider competency and their role in care. The impact of providers is understudied and to a large extent, unknown. Our objective was to define, test, and validate QMs to pragmatically measure the practice-based quality of medical providers in a pilot study. We included 7 North American LTC homes with data from practicing medical providers for LTC residents. We engaged in a 4-phased approach. In phase 1, experts rated 95 candidate QMs using 5 pragmatic-focused criteria in a RAND-modified Delphi process. Phase 2 involved specifying 37 QMs for collection (4 QMs were dropped during pilot testing). We created an abstraction manual and data collection tool for all QMs. Phase 3 involved a retrospective chart review in 7 LTC homes on 33 QMs with trained data abstractors. Data were sufficient to analyze performance for 26 QMs. Lastly, in phase 4 results and psychometric properties were reviewed with an expert panel. They ranked the tested measures for validity and feasibility for use by a nonphysician auditor to evaluate medical provider performance based on medical record review. In total, we examined data from 343 resident charts from 7 LTC homes and 49 providers. Our process yielded 10 QMs as being specified for measurement, feasible to collect, and had good test performance. This is the only study to systematically identify a subset of QMs for feasible collection from the medical record by various data collectors. This pragmatic approach to measuring practice-based quality and quantifying select medical provider competencies allows for the evaluation of individual and facility-level performance and facilitates quality improvement initiatives. Future work should perform broader testing and validate and refine operationalized QMs.