Beyond Hospital-Level Aggregated Data: A Methodology to Adapt Clinical Data From the Electronic Health Record for Nursing Unit-Level Research

BACKGROUND

Studies of nurse staffing frequently use data aggregated at the hospital level that do not provide the appropriate context to inform unit-level decisions, such as nurse staffing.

OBJECTIVES

Describe a method to link patient data collected during the provision of routine care and recorded in the electronic health record (EHR) to the nursing units where care occurred in a national dataset.

RESEARCH DESIGN

We identified all Veterans Health Administration acute care hospitalizations in the calendar year 2019 nationwide. We linked patient-level EHR and bar code medication administration data to nursing units using a crosswalk. We divided hospitalizations into segments based on the patient's time-stamped location (ward stays). We calculated the number of ward stays and medication administrations linked to a nursing unit and the unit-level and facility-level mean patient risk scores.

RESULTS

We extracted data on 1117 nursing units, 3782 EHR patient locations associated with 1,137,391 ward stays, and 67,772 bar code medication administration locations associated with 147,686,996 medication administrations across 125 Veterans Health Administration facilities. We linked 89.46% of ward stays and 93.10% of medication administrations to a nursing unit. The average (standard deviation) unit-level patient severity across all facilities is 4.71 (1.52), versus 4.53 (0.88) at the facility level.

CONCLUSIONS

Identification of units is indispensable for using EHR data to understand unit-level phenomena in nursing research and can provide the context-specific information needed by managers making frontline decisions about staffing.

Postinjury Care and Complications Among U.S. Military Women With Combat-Related Major Limb Traumatic Amputation

BACKGROUND

U.S. servicewomen may be at greater risk of injury in future conflicts as they integrate into combat occupations. More than 1,000 servicewomen were wounded during military conflicts in Iraq and Afghanistan. Some women sustained traumatic amputations, and research on their postinjury health is needed.

OBJECTIVE

To describe acute care, complications, and health care utilization among servicewomen with combat-related amputations, comparing them with injured men.

METHODS

In this retrospective matched-pairs study, women were identified from the Expeditionary Medical Encounter Database between 2003 and 2012 and matched with men on amputation injuries, injury severity, and age. Differences were assessed with nonparametric tests for paired data.

RESULTS

Of 20 women identified for analysis, 13 received tourniquets, three were administered procoagulants, and six had massive transfusions. Women averaged 3.4 (SD = 1.6) postinjury complications, and the most frequent were heterotopic ossification (n = 17), posthemorrhagic anemias (n = 13), and bacterial wound infections (n = 10). Acute care and complications were similar among men. Women averaged more acute care days (M = 49.8, SD = 30.6) than men (M = 46.1, SD = 27.4) but fewer intensive care unit days (women: M = 2.6, SD = 4.0; men: M = 4.4, SD = 8.3). No statistical differences were observed.

CONCLUSION

Postinjury care among servicewomen with combat-related amputations was comparable with servicemen, and complications were common. This information can aid providers and nursing staff in the management of these injuries.

CE: Nursing Orientation to Data Science and Machine Learning

ABSTRACT

Nurses collect, use, and produce data every day in countless ways, such as when assessing and treating patients, performing administrative functions, and engaging in strategic planning in their organizations and communities. These data are aggregated into large data sets in health care systems, public and private databases, and academic research settings. In recent years the machines used in this work (computer hardware) have become increasingly able to analyze large data sets, or "big data," at high speed. Data scientists use machine learning tools to aid in analyzing this big data, such as data amassed from large numbers of electronic health records. In health care, predictions for patient outcomes has become a focus of research using machine learning. It's important for nurses and nurse administrators to understand how machine learning has changed our ways of thinking about data and turning data into knowledge that can improve patient care. This article provides an orientation to machine learning and data science, offers an understanding of current challenges and opportunities, and describes the nursing implications for nurses in various roles.

Barriers and facilitators to using electronic healthcare records in Jordanian hospitals from the nurses' perspective: A national survey

BACKGROUND

Jordan is currently taking significant steps to implement electronic healthcare records (EHRs) in hospitals.

PURPOSE

To explore Jordanian nurses' views on the factors which act as barriers or facilitators to implementing EHRs in the country's hospitals.

METHODS

A descriptive, cross-sectional design was used to obtain a sample of nurses working in Jordanian hospitals. The study utilized a tool used by the American Hospital Association.

RESULTS

The final sample consisted of 2,793 nurses from 26 hospitals. 62.4% of these were working in governmental hospitals, 58.9% were female, and the mean age of the nurses was 30.3 years (SD = 6.2). Incentives to purchase EHRs, the availability of technical assistance, and additional reimbursement for the use of EHRs were the most common facilitating factors identified, while the economic burden of purchasing EHRs, a lack of information technology staff, and disruption to clinical care were the most common barriers.

CONCLUSION

Future projects for implementing EHR systems in Jordan should consider the facilitators and barriers identified.

Abstracting ICU Nursing Care Quality Data From the Electronic Health Record

The electronic health record is a potentially rich source of data for clinical research in the intensive care unit setting. We describe the iterative, multi-step process used to develop and test a data abstraction tool, used for collection of nursing care quality indicators from the electronic health record, for a pragmatic trial. We computed Cohen's kappa coefficient (κ) to assess interrater agreement or reliability of data abstracted using preliminary and finalized tools. In assessing the reliability of study data ( n = 1,440 cases) using the finalized tool, 108 randomly selected cases (10% of first half sample; 5% of last half sample) were independently abstracted by a second rater. We demonstrated mean κ values ranging from 0.61 to 0.99 for all indicators. Nursing care quality data can be accurately and reliably abstracted from the electronic health records of intensive care unit patients using a well-developed data collection tool and detailed training.