Medical Care最新文献

As the 8 patient partners serving on the PCORnet® Steering Committee, we stand at the forefront of a transformative movement in clinical research. PCORnet® Network Partners have been pioneers in integrating patient voices into every aspect of the research process, and we applaud the progress in operationalizing the Patient-Centered Outcomes Research Institute's (PCORI) Framework for Patient Engagement and for leading the way as funders to change how to effectively involve patients and other interested parties in research. However, we believe that now is the time to amplify our efforts and call for a fundamental shift in how health research is conducted across the board. This commentary serves as both a reflection on our journey and a rallying cry for deeper, more authentic patient engagement and partnership in clinical research. The landscape of clinical research has undergone significant changes over the past decade, with patient engagement emerging as a cornerstone of patient-centered outcomes research. This shift is evidenced by major funding agencies now requiring patient engagement and a growing body of literature demonstrating improved study quality, recruitment, and relevance when patients are engaged as partners. As patient partners participating in PCORnet®, we have been at the forefront of this evolution, witnessing firsthand the progress made and the challenges and learnings that remain. Drawing on our experiences and evidence from the literature, we propose strategies to enhance patient involvement across all stages of research. We introduce and explore the concept that clinical research should be "careful, kind, and connected." Our reflections underscore that meaningful patient involvement is essential for advancing health outcomes and achieving a truly patient-partnered research ecosystem.

Objective: The PATHWAYS Study utilized data from the PCORnet® Common Data Model (CDM) at 4 sites participating in the STAR Clinical Research Network to assess the frequency of cardiology encounters for under-represented racial and ethnic minority group people living with Human Immunodeficiency Virus and to evaluate the determinants associated with specialty encounters from 2014 to 2020. This study dealt with several factors that other projects leveraging PCORnet might face. We describe benefits of working with the network, challenges, and recommendations for future study teams.

Methods: PATHWAYS used a mix of queries through the study, including study-specific data quality and analytic queries. A "sidecar" table was created for the PCORnet® Common Data Model to support the inclusion of referral data. Linkage to the National Death Index was incorporated into the study to allow for more comprehensive information on participant deaths.

Results: Data quality assessments identified several issues over the course of the study that needed to be addressed by the data teams at each site. The referral data proved not to be robust enough to support the proposed analyses, so an alternative strategy was required that leveraged encounter information. The National Data Index included information on participant deaths that were not part of each site's PCORnet® CDM.

Conclusion: Incorporating study-specific data characterization into the overall analysis plan is important. When working with new data, or variables not commonly used within studies, teams should include time and effort for site resources to investigate their local clinical workflows and potential mappings to the PCORnet® CDM.

Background: Many researchers want to control for both individual-level demographic/health variables and area-level socioeconomic status (area-SES) when studying health outcomes. However, comparative assessments of area-SES indices across geographic levels and a range of health outcomes are scarce.

Objectives: Compare predictive power for 3 commonly used area-SES indices: the Graham Social Deprivation Index (SDI), the Area Deprivation Index (ADI), and the CDC Social Vulnerability Index (SVI), for a variety of health outcomes, at different geographic levels (county, 5-digit zip-code, census tract, and census block group). Also compare these indices to the simpler Townsend Deprivation Index (TDI) and population percent in poverty (area-Poverty).

Research design: Principal research methods are logistic and ordinary least squares regression.

Subjects: Medicare fee-for-service beneficiaries, COVID-19 decedents, and drug overdose decedents.

Measures: SDI, SVI, ADI, TDI, area-Poverty.

Health outcomes studied: All-cause mortality, diabetes incidence and prevalence, hypertension, renal disease, and 30-day hospital readmission for Medicare beneficiaries; COVID-19 mortality; overdose mortality; Medicare fee-for-service spending.

Results: All measures predict the health outcomes, controlling for age, gender, race/ethnicity, and comorbidities, at zip code, tract, and block-group levels. Predictive power is comparable for SDI, SVI, and a standardized version of ADI, and generally superior to TDI, area-Poverty, and non-standardized ADI. Predictive power is highest at tract level, similar at block-group; reasonably strong at zip code, but weaker at county level.

Conclusions: Across a range of health outcomes, we find similar predictive power for SDI, SVI, and standardized ADI, ideally measured at census tract level. SDI has the value of being more parsimonious, with similar performance. Non-standardized ADI cannot be recommended.

Background: The COVID-19 pandemic exposed differences in health care outcomes, including the prescription of COVID-19 antiviral medications. This analysis aimed to describe prescribing patterns in outpatient COVID-19 treatment and assess factors that contributed to these differences.

Methods: A cross-sectional analysis was conducted using electronic health record data from August 2022 through March 2024 from health care institutions participating in PCORnet,® the National Patient-Centered Clinical Research Network. Descriptive statistics were used to characterize COVID-19 outpatients eligible for treatment, and regression models were used to calculate adjusted prevalence ratios (aPR) of prescribed COVID-19 outpatient treatment. Interaction terms assessed the interactions between race and ethnicity and the combined comorbidity index (CCI), age and sex, and age and race and ethnicity.

Results: Of 1,247,420 patients eligible for COVID-19 treatment, 334,947 (26.9%) were prescribed outpatient treatment. In adjusted analyses, compared with White patients, all other racial and ethnic groups had lower aPR for treatment (aPRs:0.89-0.99), except patients who reported being multiracial (aPR:1.00; 95% CI: 0.93-1.08). Those aged 65-74 were prescribed treatment more often (aPR: 1.13; 95% CI: 1.12-1.13) compared with patients aged 20-49. Patients with a CCI of 1-3 and ≥4 were prescribed treatment less often (aPR: 0.99, 95% CI: 0.97-1.01 and aPR: 0.91, 95% CI: 0.89-0.94, respectively), compared with those with a CCI of ≤0. These differences were sustained when considering the interactions between race and age and race and CCI.

Conclusions: We found differences in recommended outpatient treatment by several sociodemographic variables. Addressing COVID-19 prescription barriers is essential to slow preventable differences from unmet COVID-19 outpatient care.

PCORnet®, a national resource funded by the Patient-Centered Outcomes Research Institute, is designed to enhance the nation's capacity to conduct efficient, patient-centered health research. The robust and adaptable PCORnet infrastructure can be leveraged to support a variety of study designs. Within this framework, PCORnet® Studies, a distinguished subset of research studies that meets specific criteria and approval, serve as exemplary models of patient-centered research, standing out for their rigorous adherence to criteria that elevate the quality and impact of research. Over the past 10 years, over 300 studies have been completed using the PCORnet infrastructure, including 58 studies that have earned the distinction of being a meritorious PCORnet® Study, of which 19 have been completed. We explore several of these efforts, highlighting the study archetypes supported by the PCORnet infrastructure, as well as the clinical therapeutic areas of these studies, funding sources, and breadth of PCORnet institutional users. We also provide lessons learned from our accumulated experience that translates the PCORnet infrastructure into a continuously learning framework and highlight unique capabilities of the PCORnet infrastructure supporting innovation in future trials.

Objective: To examine telehealth care utilization during pregnancy and in the 6 months postpartum during the COVID-19 pandemic.

Research design: We used a repeated cross-sectional design to assess aggregated and standardized electronic health records (EHR) from 28 participating US health systems in the PCORnet®, National Patient-Centered Clinical Research Network.

Subjects: We studied women aged 18-55 years with delivery records from 2018 to 2019 (prepandemic), 2020-2021 (early pandemic), and 2022-2023 (late/postpandemic). Women with hypertension in pregnancy (HTN-PREG) were identified as those with documented hypertension 1 year prepregnancy or during pregnancy (40 wk predelivery).

Measures: We used the International Classification of Disease Clinical Modification/Procedure Classification codes, a normalized naming system for drug names (RXnorm), and the Current Procedural Terminology/Healthcare Common Procedure Coding System to identify deliveries, hypertension, and telehealth visits.

Results: We examined 1,028,426 deliveries, of which 108,606 had HTN-PREG. The proportion of women with 1+ pregnancy telehealth use was higher in the early pandemic (15.8%) and late/postpandemic (16.4%) periods compared with prepandemic (0.2%). Telehealth use among HTN-PREG women was higher than among all women (0.24% prepandemic, 24.9% early pandemic, 26.8% late/postpandemic).. Among HTN-PREG women, telehealth use was lower among Hispanic women (19.2% and 23.2%) compared with NH Black (26.5% and 29.3%) or NH White (24.5% and 26.1%) women during the early and late/postpandemic periods, respectively. We observed similar patterns postpartum. All P values for comparisons were <0.001.

Conclusions: These findings underscore the value of aggregate PCORnet® data infrastructure and other standardized electronic health records in analyzing health care utilization trends and racial and ethnic differences among pregnant women.

Background: A national clinical research network such as PCORnet® must, by design, accommodate myriad approaches to support effective and efficient multisite and multinetwork research activities. Therefore, visualizing the architecture of clinical research networks is imperative to improve their utility for national scale research.

Objectives: (1) To present a methodological approach to apply system mapping methods to characterize the key actors, relationships, and exchanges within the PCORnet infrastructure for data activities. (2) To describe the findings of the PCORnet infrastructure mapping project and characterize the commonalities and heterogeneity between CRN infrastructures based on their data management and sharing activities.

Methods and design: System mapping methods were applied in this paper to provide a more accurate and nuanced methodology to visualize the complex infrastructure provided by health research networks in order to support national-scale health research, specifically patient-centered comparative clinical effectiveness research (CER) using PCORnet.

Results: To make PCORnet more accessible to investigators, patient partners, research funders, and other collaborators, the mapping methods and maps presented in this paper offer a new tool to visualize the design and architecture of the PCORnet infrastructure to fulfill distributed queries.

Conclusions: Critically, the methodology and approach to system mapping offer a new path forward to support Network collaboration as a learning health system (LHS) by improving transparency and dialogue with patient partners regarding the strengths and potential limitations of infrastructure for evidence generation and informing health care decisions.

The PCORnet vision to accelerate progress towards a learning health system by enabling more efficient generation of evidence is shared by the US Food and Drug Administration, where we recently served. One way of achieving this goal is to develop national networks capable of running streamlined clinical trials in the postmarket setting. In this invited commentary, we discuss how PCORnet® and similar networks could facilitate streamlined trials integrated into clinical practice across their sites.

Background: The Person Centered Primary Care Measure (PCPCM) was developed to assess "aspects that contribute to patient perceptions regarding the integrating, prioritizing, and personalizing functions of primary care." Several psychometric issues remain unresolved.

Objectives: We sought to examine the performance of the existing patient-level model, evaluate measurement bias, assess the impact of item-level missingness on reliability, examine the structural validity of creating a clinician-level score, and identify the number of patients needed to achieve a reliable clinician-level score.

Research design: We used confirmatory factor analyses (CFA), item response theory, multilevel CFA, and retrospective survey data.

Participants: Three thousand one hundred ten patients clustered within 32 clinics and 94 clinicians completed the PCPCM.

Results: CFA supported a single-factor patient-level model with 2 sets of correlated errors (RMSEA=0.06; CFI=0.98; TLI=0.98). Item response theory-based marginal reliability curves demonstrated that reliability drops precipitously if fewer than 6 items are answered. Multilevel CFA supported a single factor at the patient level and a single factor at the clinician level, with 2 sets of patient-level correlated errors (RMSEA=0.07; CFI=0.93; TLI=0.91). Scatter plots of clinician-level model-based and response-based scores showed nonlinearity and larger SEs when clinician scores were based on fewer than 5 patients. Reliability was >0.80 with 5 or more patients and 0.90 with 9 or more.

Conclusions: Our study demonstrates the reliability and structural validity of creating a patient-level PCPCM score as the average of answers to at least 6 PCPCM questions and creating a clinician-level score as an average of the PCPCM scores from at least 5 patients within a clinician.

Background: Institutions that participate in PCORnet® transform their local electronic health record (EHR) data into the PCORnet® Common Data Model (CDM), which is then used to generate data extracts for PCORnet® Studies. PCORnet® Studies can also include institutions that do not participate in PCORnet, and for these organizations, the cost of instantiating a PCORnet® CDM can be prohibitive. Fast Health care Interoperability Resources (FHIR) provides an alternative method of obtaining EHR data.

Objective: To determine whether data obtained through FHIR might be a viable study solution for those sites that do not participate in PCORnet.® This mixed-methods project had 2 objectives: (1) survey sites participating in PCORnet on the availability of FHIR (FHIR survey); (2) compare the coverage of a FHIR-based data extract using REDCap with one from the PCORnet® CDM across 3 sites (FHIR extract).

Methods: (1) FHIR survey: A series of questions were asked about the use of FHIR in a production capacity. (2) FHIR extract: REDCap FHIR and PCORnet® CDM extracts were created based on study variables from 2 prior PCORnet® Studies. Data were extracted for 40 patients and concordance measures were computed between the 2 sources.

Results: (1) FHIR survey: Of responding organizations, 73% (n=49) reported that FHIR was deployed in a production capacity. (2) FHIR extract: Results were highly variable. Cohen kappa ranged from 0.01 to 0.76 for certain diagnoses, 0.24 to 0.84 for laboratory results, and 0.1 to 0.87 for medications.

Conclusions: Despite differences in data, certain studies may be well-suited for FHIR-based extracts.