Results Summary
PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.
Background
Hospital leaders and doctors—not patients and their families—usually decide what types of care are regarded as safe and high quality. But what do patients and their families think safe, high-quality care looks like?
To learn more about the views of these healthcare customers, researchers are focusing on what happens when patients do not get the care they expect. Hospitals call these nonroutine events. Examples include a patient getting the wrong drug or a patient waiting hours to receive discharge papers.
By studying these examples, researchers can learn about what patients want most from their hospital care. Knowing what patients want can help experts measure the quality and safety of patient care in a way that matches what patients and their families say is most important.
Project Purpose
The research team wanted to
- Identify what patients and families view as nonroutine in their hospital experience
- Understand how patients, families, and healthcare providers decide whether to report nonroutine events to researchers
- Figure out whether collecting information on nonroutine events can help hospitals understand how their care processes fail to work like they should
Methods
The study included four types of hospital patients:
- 144 adults having planned surgery who stayed in the hospital for less than 24 hours
- 129 adults getting treatment for a heart problem
- 93 children with cancer who were getting chemotherapy
- 84 children having planned heart surgery
The research team interviewed these patients, their families, and their healthcare providers. To perform the interviews, the research team made 559 hospital visits.
The research team asked participants, family members, and healthcare providers to describe times when the hospital care they received was not what they expected. The team also asked for information about each participant’s background, such as gender and age.
The research team made a list that described each of the events that participants talked about, and then medical experts reviewed the list. They didn’t know the names of the patients or healthcare providers involved. The experts labeled each event by type, how serious it was, and what could have caused it. They also noted whether the event caused the patient any physical or mental harm.
Findings
Adults having a planned surgery. Among the healthcare providers of patients in this group, 57 percent reported a nonroutine event compared with 40 percent of patients and families. Only about 25 percent of patients reported an event that their healthcare provider also reported.
Among healthcare providers, nurse anesthetists reported a nonroutine event most often (66 percent of their cases). Doctors who were training to be anesthesiologists reported a nonroutine event least often (33 percent of their cases).
The most common nonroutine events reported by patients were problems with diagnosis or treatment. For example, they talked about times when the healthcare provider wasn’t familiar with them as a patient, when a mistake happened, when the patient had to wait a long time for a diagnosis, or when a healthcare provider gave the patient the wrong diagnosis. Patients also reported problems communicating with their provider, like
- Getting wrong information
- Getting too much or too little information
- Getting information at the wrong time
Healthcare providers were more likely to report problems with how they gave care to patients. For example, healthcare providers reported times when patients got care they did not expect, when patients could not get care they needed, when patients had to wait a long time to get treatment, or when the care was interrupted or not consistent.
Adults getting treatment for a heart problem. In this group, 69 percent of patients reported a nonroutine event compared with only 46 percent of healthcare providers. The most common nonroutine events reported by patients were times when they had an unexpected delay in their care. They also talked about times when the treatment they got was different from what they expected.
Children getting cancer treatment. In this group, 32 percent of parents and 20 percent of healthcare providers reported a nonroutine event.
Children getting heart surgery. In this group, 68 percent of healthcare providers reported a nonroutine event compared with 37 percent of parents.
The research team reports that experts and patients are still reviewing the nonroutine events for children with cancer and those getting heart surgery to see what types of events their families and providers reported.
Conclusions
For patients and families, there were many times when their care was not what they expected.
Patients and families often reported different types of nonroutine events than healthcare providers. Healthcare providers also did not know about many of the events that were most important to patients and families.
Patients and families gave information that may help hospitals do a better job of caring for their patients.
Sharing the results
The research team is writing articles for journals about the results of the study.
Professional Abstract
PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.
Background
Care processes are designed to meet quality and safety targets set by clinicians and institutions, not by their “customers” (i.e., patients/families). In fact, little is known about what patients/families consider problematic about care processes. Thus, a key goal of this work was to define more clearly what attributes patients and families consider to be safe, high-quality care. By rigorously studying the application of the nonroutine event (NRE) methodology to identify quality and safety risks to patients, this study addressed a PCORI priority—to develop new patient-centered outcome measures that can be used by multiple stakeholders.
Project Purpose
The aims of this study were to (1) ascertain what aspects of patients’/families’ encounters with a clinical system patients and families consider “nonroutine;” (2) elucidate the factors that influence the reporting of NREs and affect the nature of the reported NREs; and (3) determine whether NREs obtained from patients/families significantly add to evidence about clinical system failure modes.
Study Design
Prospective, staggered entry observational (nonintervention) study of the incidence and nature of NREs reported by patients, families, and their clinicians
Participants, Settings, Interventions, and Outcomes
Participants included patients, family members, and their clinicians in four distinct clinical domains (listed below). Trained research assistants (RAs) monitored patient flow through the study site and worked with the clinical leads to recruit eligible patients.
Participants were selected from four clinical domains: adult elective day surgery with 23-hour stays, adult patients treated in the cardiac catheterization laboratory, pediatric oncology patients receiving chemotherapy, and pediatric patients receiving elective pediatric cardiac surgery.
Each NRE was coded for severity, effect, type, and putative contributory factors. NREs were also assigned a highest level of patient impact (i.e., injury, impact, near miss, or none).
Data Sources
After obtaining written informed consent, participants completed the Comprehensive Open-ended Nonroutine Event Survey (CONES) instrument during either an in-person or phone interview with a trained RA. CONES is a standardized interview designed to capture any event thought by the respondent to deviate from optimal or expected care. Based on the results of 10 focus groups, researchers created a CONES for use with patients and families, known as PCONES, as well as a taxonomy of patient-centered NREs. Researchers concurrently obtained participant demographic and affect data.
Quality of Data and Analysis
Each NRE was reviewed by two clinical site domain experts (SDEs) and/or patient subject matter experts (PSEs), as appropriate. SDEs, blinded to participant and patient identity, confirmed the validity of case data. All reviewers coded each NRE for type, severity, and putative contributory factors using both the clinician and patient-centered taxonomies. The SDEs and/or PSEs also assigned a highest level of patient impact (injury, impact, near miss, or none). An injury was defined as measurable lasting tissue damage or effect on patient quality of life. Impact was defined as any physiological or psychological effect on the patient, even if temporary. SDEs classified each NRE by type (Patient/Family, Course of Illness [e.g., preexisting condition, patient response to correct care], Clinician, Team, Technology, Process, and System) and, if iatrogenic, by deficiency type: Care (e.g., wrong drug given to patient) or Service (e.g., valet lost care keys).
Findings
Researchers studied 383 patients in 559 encounters across the four domains.
Adult Ambulatory Surgery: (144 patients/145 encounters) One patient was studied twice. The average patient age was 58±13 years with 54 percent male. The clinicians reported an NRE more often than patients/families (57 percent vs. 40 percent, respectively). CRNAs more often reported an NRE (66 percent of their cases) relative to anesthesia residents often (33 percent). Only about 25 percent of cases had both patient- and clinician-reported NREs. Patients most often reported NREs that were related to diagnostic and therapeutic issues (e.g., unfamiliar with patient’s condition, mistakes and errors, diagnostic delays, misdiagnoses) and/or communication issues (e.g., wrong amount of information, or wrong content, or mistimed delivery), while clinician NREs were healthcare process deficiencies (e.g., unexpected care, failure to get access, treatment delays, care disruptions or variability).
Adult Interventional Cardiology: (129 patients/130 encounters) One patient was studied twice. Patients were 62±12 years and 69 percent male. Patients more often reported NREs than clinicians (69 percent vs. 46 percent, respectively). The most common patient NREs were related to unexpected delays. Deviations from expected treatment was also commonly reported.
Pediatric Sites: Researchers studied 93 patients (9±5 years, 43 percent male) during 200 mostly outpatient encounters in pediatric oncology (i.e., each patient was usually studied in two or more treatment sessions). On a per-encounter basis, clinicians reported a NRE slightly more often than parents (20 percent and 32 percent, respectively). Researchers studied 84 pediatric cardiac surgery patients (4±5 years and 55 percent male); clinicians more often reported an NRE than did parents (68 percent vs. 37 percent, respectively).
Limitations
This single-site study involved only four clinical domains. There was likely underreporting, especially by clinicians. Potential obstacles to broader implementation include the need for strong institutional support, unit-level leadership buy-in and advocacy, and front-line clinician engagement.
Conclusions
There were three high-level conclusions. First, patients and families reported a large number of NREs in a substantial proportion of clinical encounters studied (more than expected), and many of these NREs appear relevant to patients’ health. Thus, methods used to obtain patient- and family-reported NREs appear to be feasible and relevant.
Second, very few of the patient/family NREs were also reported by their clinicians, and many were actually unknown to the clinicians. However, a substantial number of NREs were deemed important by patients/families to the quality and safety of the care received. Thus, research on how to more widely obtain and analyze patient/family NREs seems warranted as a patient-centered method of identifying healthcare system deficiencies.
Finally, the patient- and family-reported NREs provided useful information about how the care delivery system could be improved. Many of the issues identified are broadly related to deficiencies in communication between clinicians (and others in the delivery systems) and with the patient/family as well as correctable process deficiencies.