Association of Patient and Family Reports of Hospital Safety Climate With Language Proficiency in the US

Key Points

Question  Are differences in the hospital safety climate scores of patients and parents or caregivers of hospitalized children (ie, families) associated with limited English proficiency vs English proficiency?

Findings  In this cohort study of 533 hospitalized pediatric patients and families, compared with participants with English proficiency, those with limited English proficiency had significantly lower odds of freely speaking up, questioning decisions, and being unafraid to ask questions.

Meaning  This study suggests that limited English proficiency is associated with a lower likelihood of speaking up, questioning health care professionals’ decisions, and being unafraid to ask questions; dedicated efforts are needed to enhance communication, improve hospital safety, and reduce disparities for patients and families with limited English proficiency.

Importance  Patients with language barriers have a higher risk of experiencing hospital safety events. This study hypothesized that language barriers would be associated with poorer perceptions of hospital safety climate relating to communication openness.

Objective  To examine disparities in reported hospital safety climate by language proficiency in a cohort of hospitalized children and their families.

Design, Setting, and Participants  This cohort study conducted from April 29, 2019, through March 1, 2020, included pediatric patients and parents or caregivers of hospitalized children at general and subspecialty units at 21 US hospitals. Randomly selected Arabic-, Chinese-, English-, and Spanish-speaking hospitalized patients and families were approached before hospital discharge and were included in the analysis if they provided both language proficiency and health literacy data. Participants self-rated language proficiency via surveys. Limited English proficiency was defined as an answer of anything other than “very well” to the question “how well do you speak English?”

Main Outcomes and Measures  Primary outcomes were top-box (top most; eg, strongly agree) 5-point Likert scale ratings for 3 Children’s Hospital Safety Climate Questionnaire communication openness items: (1) freely speaking up if you see something that may negatively affect care (top-box response: strongly agree), (2) questioning decisions or actions of health care providers (top-box response: strongly agree), and (3) being afraid to ask questions when something does not seem right (top-box response: strongly disagree [reverse-coded item]). Covariates included health literacy and sociodemographic characteristics. Logistic regression was used with generalized estimating equations to control for clustering by site to model associations between openness items and language proficiency, adjusting for health literacy and sociodemographic characteristics.

Results  Of 813 patients, parents, and caregivers who were approached to participate in the study, 608 completed surveys (74.8% response rate). A total of 87.7% (533 of 608) of participants (434 [82.0%] female individuals) completed language proficiency and health literacy items and were included in the analyses; of these, 14.1% (75) had limited English proficiency. Participants with limited English proficiency had lower odds of freely speaking up if they see something that may negatively affect care (adjusted odds ratio [aOR], 0.26; 95% CI, 0.15-0.43), questioning decisions or actions of health care providers (aOR, 0.19; 95% CI, 0.09-0.41), and being unafraid to ask questions when something does not seem right (aOR, 0.44; 95% CI, 0.27-0.71). Individuals with limited health literacy (aOR, 0.66; 95% CI, 0.48-0.91) and a lower level of educational attainment (aOR, 0.59; 95% CI, 0.36-0.95) were also less likely to question decisions or actions.

Conclusions and Relevance  This cohort study found that limited English proficiency was associated with lower odds of speaking up, questioning decisions or actions of providers, and being unafraid to ask questions when something does not seem right. This disparity may contribute to higher hospital safety risk for patients with limited English proficiency. Dedicated efforts to improve communication with patients and families with limited English proficiency are necessary to improve hospital safety and reduce disparities.

Up to 250 000 US patients die annually because of medical errors, making errors a leading cause of death.^1-3 Although hospitalized children have similar rates of medical errors as adults, children have higher rates of potentially harmful errors.⁴ Serious medical errors are frequently caused by communication failures,⁵ which are particularly likely when there are language barriers.^6-8 Approximately 25 million people (8.6% of the US population) have limited English proficiency, reporting that they speak English less than very well.⁹ Children may be increasingly affected by language barriers given that the number of children with 1 or more immigrant parent is rising, and 54% of immigrant children in the US live with a parent with limited English proficiency.¹⁰

Language barriers are associated with increased risk of patient harm,^11-13 readmissions,^14,15 hospital length of stay,¹⁶ hospital-acquired conditions,¹⁷ costs,¹⁸ and decreased adherence to treatment¹⁹ and satisfaction.²⁰ Hospitalized children whose parents have limited comfort speaking English have twice the adverse event rates of children whose parents are comfortable speaking English.²¹ Reasons for higher adverse event rates in patients with limited English proficiency may include communication barriers associated with interpretation,^22,23 differential treatment,^24,25 decreased trust and relationship building,^26,27 structural barriers, systemic racism, bias,^28-31 and discrimination.^32,33 These factors may prevent patients and families with limited English proficiency from being effectively engaged in hospital communication and safety promotion.

Safety climate surveys measure perceptions and attitudes about safety culture—shared organizational beliefs and values promoting behavioral norms that foster patient safety.^34,35 Employee safety climate scores are associated with hospital safety performance.³⁵ However, hospitals generally focus on staff safety climate surveys that measure health care professionals perceptions of safety culture, thereby missing an important stakeholder: the patient and family. Studies of patient and family perceptions of hospital safety climate are limited,^34,36,37 particularly for individuals with limited English proficiency. Families are vigilant partners in care and astute observers of safety.³⁸ They possess key historical patient knowledge, motivation for favorable outcomes, and bedside proximity. Patients and families are also motivated to engage in safety efforts^39,40; their input can help improve patient safety because they frequently identify hospital safety problems, including those related to procedures, hand hygiene, medication administration, and communication.^38,41,42 Thus, collecting data from patients and families about their comfort speaking up about safety (eg, through safety climate surveys) is important to improve patient safety in hospitals.

The association between language proficiency and safety climate has been understudied. To advance equity and patient safety for patients with limited English proficiency, hospitals must understand how language barriers affect care and develop interventions to address these barriers. Interventions that do not account for existing inequities in hospital care risk widening disparities.^43-46 Understanding differences in perceptions of safety climate, particularly relating to communication, among individuals with language barriers might help hospitals target interventions toward improving patient engagement and safety outcomes in disadvantaged groups.

We therefore sought to examine the association between language proficiency and top-box (top most) hospital safety climate scores for patient and family communication in a cohort of hospitalized children. We hypothesized that patients and families with limited English proficiency would be less likely to report top-box (top most) hospital safety climate scores for communication than their counterparts with English proficiency.

This multicenter, prospective cohort study was conducted from April 29, 2019, through March 1, 2020, in pediatric inpatient units of 21 US teaching hospitals (eAppendix in Supplement). We collected data concurrent with data collection for the Patient and Family Centered I-PASS SCORE (Safer Communication on Rounds Every Time) Study, a multicenter dissemination and implementation study of a family-centered rounds communication intervention.⁴⁷ The Boston Children’s Hospital Institutional Review Board approved the study, and participants provided verbal informed consent (verbal instead of written informed consent was obtained because it presents less risk of a confidentiality breach and because the study was minimal risk). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Of the inpatient units at 21 hospitals, 1 was a pediatric neurology unit; the remainder were general pediatric (nonintensive care) units. Sites were freestanding children’s hospitals (n = 4), nested children’s hospitals (n = 14), or general hospitals (n = 3). Nine hospitals were community hospitals; 12 were tertiary care hospitals. Sites were located throughout the US, including the South, Northeast, Mid-Atlantic, Midwest, and West Coast. All sites had in-person, video, or telephonic interpreters.

Patients (aged ≥13 years) and parents or caregivers (hereafter families) of children of all ages hospitalized on the study units who primarily spoke Arabic, Chinese, English, or Spanish (as identified by their clinical care team) were approached and provided verbal consent. Interpreters facilitated consent and survey introduction for participants who did not speak English. Patients aged 13 through 17 years provided assent. Only participants providing both language proficiency and health literacy data were included in the analyses.

We aimed to receive 2 surveys per week per site from randomly selected patients and families before discharge. Therefore, study staff approached 2 to 4 patients and families weekly per site. Staff approached families identified by clinical staff as primarily speaking Arabic, Chinese, or Spanish with in-person, telephonic, or video interpreters. Surveys inquired about hospital experience with health care professionals and communication, hospital safety climate,⁴⁸ language preference, health literacy, and demographic characteristics (including self-reported race and ethnicity). Surveys had a sixth-grade Flesch-Kincaid Grade Level and were professionally translated into Arabic, simplified Chinese, and Spanish. Participants self-completed surveys electronically via REDCap (English) or on paper (all 4 languages).

The primary outcome was the top-box hospital safety climate rating (top-box: top most Likert scale option of 5 of 5; eg, strongly agree). We assessed participants’ responses to 3 hospital safety climate items comprising the Children’s Hospital Safety Climate Questionnaire parent communication openness domain.⁴⁸ Patients and families selected their level of agreement with each item on a 5-point Likert scale (strongly agree, agree, neither agree nor disagree, disagree, or strongly disagree). The 3 items were: (1) “I will freely speak up if I see something that may negatively affect my/my child’s care” (top-box response: strongly agree); (2) “I feel free to question the decisions or actions of health care providers” (top-box response: strongly agree); and (3) “I am afraid to ask questions when something does not seem right” (top-box response: strongly disagree [reverse-coded item]).

Language proficiency, a dichotomous categorical variable, was the primary factor. Limited English proficiency was defined as a response other than “very well” to the question “how well do you speak English?” Participants who answered “very well” were coded as having English proficiency. This item was taken from the Guide to Implementing the Health Literacy Universal Precautions Toolkit.⁴⁹ Health literacy was a dichotomous categorical variable collected through the 3-item screener.⁵⁰ Participants were coded as having limited health literacy if they selected a non–top-box answer for 1 or more of the following 3 questions: (1) “How often do you have someone (like a family member, friend, hospital/clinic worker, or caregiver) help you read hospital materials?” (non–top-box responses: all, most, some, or a little of the time); (2) “How often do you have problems learning about your medical condition because of difficulty understanding written information?” (non–top-box responses: all, most, some, or a little of the time); and (3) “How confident are you filling out forms by yourself?” (non–top-box responses: quite a bit, somewhat, a little bit, or not at all). Otherwise, health literacy was coded as adequate. Respondent-reported age, gender, race and ethnicity, educational level, and income were categorized as shown in Table 1.

We compared respondent sociodemographic characteristics by language proficiency and health literacy status using χ² tests or Fisher exact tests, as appropriate. For each of the 3 safety climate items of interest (freely speaking up, questioning health care provider decisions and actions, being afraid to ask questions), we dichotomized participants’ responses using the top-box method in which the top choice of 5 on the Likert scale (ie, strongly agree or strongly disagree for the reverse-coded items) was considered to be a top-box safety climate score vs non–top-box scores (1 to 4 of 5; eg, strongly disagree, disagree, neither agree nor disagree, agree). We examined bivariate associations between having top-box safety climate scores using logistic regression with generalized estimating equations to control for clustering by site.

We conducted multivariable logistic regression with generalized estimating equations to examine the association of each safety climate outcome with language proficiency and with health literacy, adjusting for potential confounders selected a priori (relationship to patient, age, gender, race and ethnicity, educational level, income). We assumed an exchangeable working correlation matrix and used robust SEs. We also ran an ordinal logistic regression as a sensitivity analysis. In addition, we explored whether health literacy, educational level, race and ethnicity, age, gender, or income level modified the effect of limited English proficiency with our outcomes by fitting a series of models that contained 2-way interaction terms between each characteristic and limited English proficiency. Statistical significance was achieved with a 2-sided P < .05 (except for tests of effect modification, which required P < .01). We used REDCap⁵¹ (Vanderbilt University) for data collection and management and SAS, version 9.4 (SAS Institute) for analyses.

Overall, 813 participants were approached, and 608 completed surveys (74.8% response rate). A total of 533 (87.7%) patients and families (434 [82.0%] female and 95 [18.0%] male individuals [data for 4 participants were missing]) completed both language proficiency and health literacy items and were included in the main analysis (Table 1). Many participants (49.7% [264]) were aged 18 to 34 years and had attended at least some college (64.6% [342]). Self-reported race and ethnicity included 126 (24.0%) Hispanic or Latino, 34 (6.5%) non-Hispanic Asian, 69 (13.1%) non-Hispanic Black, and 254 (48.3%) non-Hispanic White individuals as well as 43 (8.2%) individuals of other races and ethnicities (including American Indian or Alaska Native, Native Hawaiian or Pacific Islander, multiple races, write-in race or ethnicity, or no additional information provided) (data were missing for 7 participants). Nearly half had an annual household income of less than $50 000.

Overall, 75 participants (14.1%) had limited English proficiency, and 132 (24.8%) had limited health literacy. Participants with limited English proficiency and English proficiency considerably differed by race and ethnicity, health literacy, educational level, and income level (Table 2). Individuals with limited English proficiency were more likely to be Hispanic or Latino and have lower levels of health literacy, educational attainment, and income than individuals with English proficiency. Patients with limited and adequate health literacy significantly differed by all demographic characteristics except gender.

Compared with participants with English proficiency, those with limited English proficiency reported being less likely to freely speak up if they see something that may negatively affect their or their child’s care (strongly agreed: 56.7% [95% CI, 47.2%-65.8%] of participants with limited English proficiency vs 82.2% [95% CI, 75.7%-87.2%] with English proficiency; P < .001) (Figure 1). Participants with limited English proficiency were less likely to report feeling free to question the decisions or actions of health care providers (strongly agreed: 37.2% [95% CI, 26.3%-49.6%] of participants with limited English proficiency vs 71.5% [95% CI, 65.7%-76.6%] with English proficiency; P < .001). In addition, 39.3% (95% CI, 30.2%-49.2%) of participants with limited English proficiency strongly disagreed they were afraid to ask questions when something does not seem right compared with 64.1% (95% CI, 57.6%-70.1%; P < .001) of participants with English proficiency (reverse-coded item).

Overall, 72.1% (95% CI, 61.3%-80.8%) of participants with limited health literacy vs 81.0% (95% CI, 73.3%-86.8%; P = .09) of those with adequate health literacy strongly agreed they will freely speak up (Figure 2). Participants with limited health literacy were less likely to feel free to question the decisions or actions of health care providers (strongly agreed: 54.0% [95% CI, 46.9%-60.9%] of those with limited health literacy vs 70.9% [95% CI, 64.6%-76.5%] of those with adequate health literacy; P < .001) and more afraid to ask questions when something does not seem right (strongly disagreed: 49.1% [95% CI, 41.3%-57.0%] of those with limited health literacy vs 64.6% [95% CI, 57.5%-71.1%] of those with adequate health literacy; P < .001).

After controlling for health literacy, relationship to patient, age, gender, race and ethnicity, educational level, and income level, participants with limited English proficiency had lower odds than participants with English proficiency of freely speaking up if they see something that may negatively affect care (adjusted odds ratio [aOR], 0.26; 95% CI, 0.15-0.43; P < .05). Participants with limited English proficiency also had lower odds of questioning decisions or actions of health care providers (aOR, 0.19; 95% CI, 0.09-0.41; P < .05) and lower odds of being unafraid to ask questions when something does not seem right (aOR, 0.44; 95% CI, 0.27-0.71; P < .05) (Table 3). There was also an association between health literacy and education and feeling free to question decisions or actions of health care providers (aOR, 0.66 [95% CI, 0.48-0.91] for limited vs adequate health literacy; aOR, 0.59 [95% CI, 0.36-0.95] for less than some college vs completed college; both P < .05) but no associations with other safety climate items evaluated. The ordinal regression model showed patterns similar to those of the logistic regression (eTable 1 in the Supplement). We did not observe any effect modification by health literacy, educational level, race and ethnicity, age, gender, or income level (eTable 2 in the Supplement).

In this multicenter, prospective cohort study of hospitalized pediatric patients and families, participants with limited English proficiency reported 0.26 odds of freely speaking up, 0.19 odds of asking questions, and 0.44 odds of being unafraid to ask questions when something does not seem right. These items comprise the communication openness domain of hospital safety climate. Given that hospital climate is associated with hospital safety performance,³⁵ the findings of the present study may explain inequities in medical errors and other aspects of patient safety experienced by patients with limited English proficiency. Hospitals may be able to improve patient safety by proactively encouraging patients and families with limited English proficiency to speak up, ask questions, and provide input.

Patient and family input, like ratings of hospitalizations—both formal experience ratings and informal online ratings (eg, Yelp)—are associated with directly measured safety outcomes, including mortality and readmissions.^52-54 Because patients with limited English proficiency are at higher risk of experiencing safety events,^11-13 better capturing their perspectives through safety climate surveys may help identify unrecognized areas to improve safety and communication. Currently, hospitals focus on patient experience surveys to understand patient and family perspectives on communication, safety, and quality. Experience surveys include some safety-related questions but provide an incomplete picture of safety experience. For instance, they do not measure patient and family comfort speaking up when something seems unsafe. Similar to employee safety climate scores,³⁵ patient and family safety climate scores may be a proxy for directly measured safety. Therefore, safety climate may be a complementary measure by which hospitals can more fully capture patient and family experience, particularly for those with limited English proficiency.

Patients with limited English proficiency may feel less comfortable speaking up, questioning health care professionals’ decisions, and asking questions when things do not seem right for many reasons. Patients with limited English proficiency have logistical barriers to communicating because of interpretation challenges. Interpreters are not always available, accessible, or encouraged. Most communications with patients with limited English proficiency occur without an interpreter, including high-risk interactions like medication administration and procedures.^18,55-58 In addition, health care professionals may take communication shortcuts^59,60 for patients with limited English proficiency that make it harder for patients and families to ask questions, question decisions, and speak up. For instance, health care professionals may use basic language skills to “get by” instead of calling certified interpreters.^59,61 Because communication and safety are closely associated,⁵ communication shortcuts may affect communication openness and safety climate as well as patient safety and can make it harder for patients with limited English proficiency to speak up than patients with English proficiency.

Because of time constraints,²⁷ health care professionals may also spend less time with rapport and relationship building with patients with limited English proficiency, perceiving such activities as a less valuable use of time.⁶² In reality, trust and relationship building are essential components of psychological safety^63,64 and may help patients and families with limited English proficiency speak up.⁶⁵ Health care professionals may similarly be less likely to solicit questions from patients with limited English proficiency,⁶² may communicate less with them, and may have fewer touch points.⁶² Patients with limited English proficiency may also be less likely to speak up because they feel intimidated or fear retribution. Discrimination and experiences with systemic racism^32,33 may make patients with limited English proficiency less likely to feel comfortable speaking up or questioning health care professionals. There may also be cultural differences in family expectations of patient–health care professional relationships.^59,66,67 In addition, health care professionals frequently use medical jargon,⁶⁸ which is difficult even for families with English proficiency to understand. As the findings of the present study suggest, health literacy is also associated with patient and family safety climate.

Several clinical implications of this study warrant investigation in future studies. Because health care professionals infrequently use interpreters for patients with limited English proficiency, including for high-risk interactions,⁵⁸ hospitals can mandate interpreter use for all interactions, including ad hoc communications. This policy change can help reduce communication barriers and make it easier and more comfortable for families to speak up and ask their health care professionals questions. Health care professionals can proactively ask patients with limited English proficiency questions and show interest by asking follow-up questions, listening without interruption, validating concerns, and making eye contact.^31,69 They can also invite patients and families with limited English proficiency to partner in care, highlighting family expertise and giving explicit permission to question decisions and speak up. Multiple health care professionals at varied points throughout a hospitalization can welcome patients and families to speak up so that genuine family engagement becomes a core part of the hospital experience. Prioritizing family engagement institutionally can help change institutional culture, increase health care professional willingness to listen to families, increase parent communication openness and experience scores, and in turn improve safety and quality. To better equip patients and families with limited English proficiency to participate in care and speak up, health care professionals can better orient them to hospital processes (eg, rounds, medication reconciliation, interpreter resources) and use question prompt lists to encourage questions.^70,71 Because families identify trust and relationships as prerequisites for speaking up,⁶⁵ health care professionals should invest in relationship building and rapport building in patients with limited English proficiency, instead of taking communication shortcuts.^60,62 Health care professionals should also minimize use of medical jargon for all patients. Hospitals should accordingly invest in health literacy training and training health care professionals in partnering with interpreters, who may act as cultural brokers and patient advocates.^72,73 Hospitals can also hire more bilingual health care professionals to provide language-concordant care, which improves patient experience.^74-77

Interventions to improve safety climate by better engaging patients and families in hospital communication can simultaneously improve safety and equity.⁴³ Such efforts can involve implicit bias and cultural humility training, integrating equity considerations into all efforts, identifying existing disparities, and addressing their root causes. Although efforts to encourage patients to speak up in health care exist^78,79 and some include 1 other language besides English (eg, Spanish),^80,81 efforts should include multiple languages and be culturally tailored to address unique factors that might keep patients with limited English proficiency from speaking up.

This study has limitations. They include generalizability, as the study was conducted at pediatric tertiary care and community teaching hospitals. The experience of patients with limited English proficiency at other hospitals may vary. However, we included 21 hospitals and had a diverse patient sample. Approximately 14% of respondents had limited English proficiency, which is higher than the estimated US population with limited English proficiency (9%) and similar to the estimated percentage of US children who live with parents with limited English proficiency.¹⁰ For ease of administration across study sites, we used the 3-question screener to measure health literacy. The screener was effective for identifying patients with inadequate health literacy but less accurate for patients with marginal health literacy,⁵⁰ and it has not been validated in populations with limited English proficiency. Validating health literacy screeners in populations with limited English proficiency is an important area for future work. Notably, approximately 25% of our sample had limited health literacy, representative of national percentages. In addition, because we surveyed adolescent patients themselves, the health literacy and educational findings may partly reflect age and developmental status. Regardless, adolescents have an important role in their health care, and health care professionals must recognize adolescent health literacy and family language barriers as they plan transitions of care. This study may have been underpowered to detect differences in safety climate by health literacy. In addition, there may be unmeasured confounders that explain our findings.

This cohort study suggests that hospitalized patients and families with limited English proficiency report being less likely to speak up, to question providers’ decisions, and to be unafraid to ask questions when something does not seem right than their counterparts with English proficiency. These factors may contribute to higher hospital safety risk for patients with limited English proficiency. To improve safety and equity, providers and hospitals must enact dedicated interventions to empower patients and families with limited English proficiency.

Accepted for Publication: March 16, 2022.

Published Online: June 13, 2022. doi:10.1001/jamapediatrics.2022.1831

Corresponding Author: Alisa Khan, MD, MPH, Division of General Pediatrics, Department of Pediatrics, Boston Children’s Hospital, 21 Autumn St, Room 230.2, Boston, MA 02215 (alisa.khan@childrens.harvard.edu).

Patient and Family Centered I-PASS SCORE Scientific Oversight Committee: In addition to the listed authors, the committee members included Ellen J. Bass, PhD; Sharon Calaman, MD; April E. Fegley, BA; Andrew J. Knighton, PhD, CPA; Jennifer K. O’Toole, MD, MEd; Theodore C. Sectish, MD; Rajendu Srivastava, MD, MPH; Amy J. Starmer, MD, MPH; Daniel C. West, MD.

Affiliations of Patient and Family Centered I-PASS SCORE Scientific Oversight Committee: Division of General Pediatrics, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts (Sectish, Starmer); Department of Pediatrics, Harvard Medical School, Boston, Massachusetts (Sectish, Starmer); Department of Information Science, College of Computing and Informatics, Drexel University, Philadelphia, Pennsylvania (Bass); Department of Health Systems and Science Research, College of Nursing and Health Professions, Drexel University, Philadelphia, Pennsylvania (Bass); Department of Pediatrics, New York University Grossman School of Medicine, New York (Calaman); New York University Langone Health/Hassenfeld Children’s Hospital, New York (Calaman); Society of Hospital Medicine, Philadelphia, Pennsylvania (Fegley); Healthcare Delivery Institute, Intermountain Healthcare, Salt Lake City, Utah (Knighton, Srivastava); Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio (O’Toole); Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio (O’Toole); Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio (O’Toole); Department of Pediatrics, University of Utah School of Medicine, Salt Lake City (Srivastava); Primary Children’s Medical Center, Salt Lake City, Utah (Srivastava); Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia (West); Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania (West).

Author Contributions: Dr Graham and Mr Trivedi had full access to data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Dr Landrigan served as the senior author.

Concept and design: Khan, Parente, Graham, Spector, Landrigan.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Khan, Graham, Johnson, Knoebel.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Graham, Trivedi.

Obtained funding: Khan, Landrigan.

Administrative, technical, or material support: Johnson, Liss, Landrigan, Fegley.

Supervision: Spector, Landrigan.

Conflict of Interest Disclosures: Dr Khan reported receiving grants from the Patient-Centered Outcomes Research Institute (PCORI) and the Agency for Healthcare Research and Quality (AHRQ) during the conduct of the study and receiving honoraria from AHRQ outside the submitted work. Dr Parente reported receiving grants from the National Institutes of Health (NIH) outside the submitted work. Dr Baird reported receiving grants from PCORI during the conduct of the study and consulting fees from the I-PASS Patient Safety Institute outside the submitted work. Dr Patel reported receiving grants from PCORI during the conduct of the study and holding equity in and serving as a consultant for the I-PASS Patient Safety Institute. Ms Johnson reported receiving grants from PCORI during the conduct of the study. Dr Spector reported receiving grants from PCORI during the conduct of the study; holding equity in and receiving consulting fees from the I-PASS Patient Safety Institute outside the submitted work; and receiving monetary award, honoraria, and travel reimbursement from multiple academic and professional organizations for teaching and consulting on physician performance and handoffs. Dr Landrigan reported receiving grants from PCORI during the conduct of the study; receiving personal fees from the I-PASS Patient Safety Institute and Missouri Hospital Association/Executive Speakers’ Bureau outside the submitted work; receiving monetary awards, honoraria, and travel reimbursement from multiple academic and professional organizations for teaching and consulting on sleep deprivation, physician performance, handoffs, and safety; holding equity in and serving as a consultant for the I-PASS Patient Safety Institute; and serving as an expert witness in cases regarding patient safety and sleep deprivation. Dr Calaman reported receiving grants from PCORI during the conduct of the study and holding stock options and serving as a consultant for the I-PASS Patient Safety Institute. Dr Knighton reported receiving grants from PCORI during the conduct of the study, the Moor Foundation, and the NIH outside the submitted work; owning publicly traded stock in UnitedHealth Group; and receiving personal fees from the American College of Chest Physicians outside the submitted work. Dr O’Toole reported receiving grants from PCORI during the conduct of the study; receiving personal fees from the I-PASS Patient Safety Institute outside the submitted work; and holding stock options in and serving as a consultant for the I-PASS Patient Safety Institute. Dr Sectish reported serving as a consultant for and receiving equity interest in the I-PASS Patient Safety Institute outside the submitted work and eceiving monetary awards, honoraria, and travel reimbursement from multiple academic and professional organizations for teaching and consulting on physician performance and handoffs. Dr Srivastava reported being a physician founder of the I-PASS Patient Safety Institute, with his equity owned by his employer, Intermountain Healthcare, during the conduct of the study; receiving grants from PCORI, NIH, AHRQ, and Centers for Disease Control and Prevention paid to his institution outside the submitted work; and receiving monetary awards, honoraria, and travel reimbursement from multiple academic and professional organizations for teaching about pediatric hospitalist research networks and quality of care. Dr Starmer reported receiving grants from PCORI during the conduct of the study; holding equity in, serving as a consultant for, and receiving personal fees from the I-PASS Patient Safety Institute outside the submitted work; and receiving monetary awards, honoraria, and travel reimbursement from multiple academic and professional organizations for teaching and consulting on handoffs and patient safety. Dr West reported receiving grants from PCORI during the conduct of the study; being a cofounder of, consultant for, and holding equity in the I-PASS Patient Safety Institute; and receiving monetary awards, honoraria, and travel reimbursements from multiple academic, regulatory, and professional organizations for teaching and consulting on handoff safety and competency-based medical education. No other disclosures were reported.

Funding/Support: This project was supported by grant CDR-1306-03556 from PCORI (Dr Landrigan). Dr Khan’s time was supported by grant K08HS025781 from AHRQ.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of authors’ employers or funders, including the AHRQ and the US Department of Health and Human Services.

Meeting Presentations: Data from this study were presented at the Pediatric Academic Societies virtual meeting; May 4, 2021; and the Pediatric Hospital Medicine virtual meeting; August 4, 2021.

Additional Contributions: We thank the Patient and Family Centered I-PASS SCORE Study Group and participating investigators at the Patient and Family Centered I-PASS SCORE Study sites. We also give special thanks to the patients, families, and interpreters who participated in this study.