Presentation
Sociotechnical Analysis to Define a Job Role Dedicated to Improving Equity in Pediatric Care
DescriptionBACKGROUND AND SIGNIFICANCE:
Pediatric health inequities in the United States remain persistent and pervasive. Children of color are more likely to suffer an adverse event during hospitalization and their parents —including those who use a language other than English (LOE) for medical care—report lower trust, poorer communication, and weaker partnerships with healthcare providers. Inequities in the inpatient setting are of particular concern due to high costs, illness severity, and family stress. Given the established links between these inequities and disparate clinical outcomes, interventions aimed at mitigating these inequities are urgently needed. [1]
Patient navigator programs have shown great promise for addressing system-related healthcare disparities. A patient navigator is a nurse or lay health worker who provides culturally competent support to patients and families, via advocacy, support, education, and assistance with care-related logistics. Researchers at Seattle Children’s Hospital (SCH) collaborated with parents/caregivers, staff, and providers to develop and pilot-test a novel program to improve navigation ability, communication, and hospital-to-home transition for a diverse population of children and their families, The Family Bridge Program (FBP). The FBP was designed for a broad population of low-income children of color, is not limited to English-proficient families, and is less time-intensive than traditional navigation.
Our NIH funded grant (R01MD015723-01A1, Lion) led by SCH with the Children’s Hospital of Philadelphia (CHOP) is aimed at using a replicable sociotechnical analysis to map the FBP to a new site (CHOP). This work will result in a comprehensive implementation guide for adaptation to additional sites in the future. Additional aims will test the effect of the FBP on a range of parent reported outcomes, examine whether changes in parent-reported barriers mediate program effects and identify subgroups of parents among whom the FBP is more effective.
METHODS:
Data collection: A sociotechnical analysis at both sites aimed toward: 1) Defining the core FBP Guide job role requirements, 2) Identifying sociotechnical differences between the two sites.
We developed a semi-structured interview guide for a range of clinical staff to elicit sociotechnical details for the six FBP services: 1) Language, 2) Orientation, 3) Communication preferences, 4) Communication coaching, 5) Supportive check in, 6) Discharge follow-up. Participants filled out a questionnaire that included rating their level of proficiency with the six FBP service topics to direct the interview on topics where they indicated experience. Interviews were performed remotely and were recorded. Audio files were transcribed via a transcription service.
Data analysis: A deductive qualitative analysis with codes based on components of the sociotechnical model, Safety Engineering in Patient Safety 2.0 (SEIPS). [2]
We developed a code book based on SEIPS 2.0 and sociotechnical factors addressed in the interview guide. Two team members performed an iterative coding process for interrater reliability using NVivo. Initially two interview transcripts were coded by each team member, then compared using Cohen’s Kappa (K). Any code with a K of < 0.75 was reviewed with team members reaching consensus and updating the code book. This process of coding and reviewing two transcripts at a time was repeated until all coding discrepancies were resolved.
We developed NVivo queries based on the site, six FBG services, and codes representing sociotechnical factors. We extracted task-related coding for each service to define the core tasks of the Guide. We then extracted the remaining codes to map all sociotechnical components (tools, people, environment, organization…) within each service while capturing important differences between the two sites. Emphasis was placed on identifying the most effective communication channels between the Guide and clinical team. Information derived from the queries was applied to a series of tools derived from SEIPS 101 including people maps, task matrix and flow diagrams. [3]
RESULTS:
44 clinicians were interviewed at the two sites. SCH (n=22) 5 bedside nurses, 3 attending physicians, 3 residents, 3 managers, 2 primary care providers, 2 interpreters, 1 social worker, 1 child life specialist, 1 care coordinator, 1 cultural navigator CHOP (n=22), 4 attending physicians, 2 fellows, 3 residents, 3 nurses, 1 care team assistant, 1 social work care coordinator, 1 family visitor service, 3 social workers, 2 case managers, 1 community health navigator, 1 primary care physician.
Participants of every role would receive the Guide favorably and described opportunities for the Guide to contribute meaningfully in the six FBP services. Concerns included duplication of work, communication challenges, and limiting the Guide to appropriate tasks. Participant comments suggest the Guide could take the lead in Language, Orientation, and Follow Up and be a key collaborator to nursing and residents in Communication Preferences and Coaching. For Unmet Needs, participants discussed opportunities for the Guide to contribute to the screening process, but expressed concerns that addressing needs was the responsibility of social work and care managers.
Participants were consistent in ideas for communication within-shift that included the use of EHR-based text messaging and in-person discussions. However, participants lacked consensus on Guide documentation and communication across shifts, with ideas ranging from a separate note authored by the Guide to relying on various clinicians to record and relay issues and concerns. Many participants noted the Guide had the potential to improve communication between families and clinicians, but also between clinicians.
The interviews revealed multiple sociotechnical differences between the two sites. While the patient populations of both sites shared many characteristics and needs, there were differences in the language/cultures of populations, geography impacting transportation needs, and the availability and types of community resources. Organizationally the two sites presented differences including an approach to families who speak a language other than English, format of family centered rounds, practices on communication and documentation, integration of social work, existing programs and roles including coordinators and navigators, and primary care practices associated with the hospital.
While the interviews revealed differences between the two sites, they also indicated important differences by unit within each hospital. This included staffing and rounding models, and even variations in the quality of Wi-Fi for online language interpreter services. Use of the EHR appeared to vary by unit within each hospital with different documentation practices and templates adding challenges to communication and collaboration for the Guide. From the patient-family perspective, one hospital has at least fifteen different unit-based versions of welcome packets for families thereby complicating the capability to provide multilingual resources.
DISCUSSION:
The Agency for Healthcare Research and Quality recommends the application of human factors engineering methods, including methods guided by the SEIPS model, to designing and improving not just medical devices and health information technology, but to “larger process of care” and “creating end-to-end, well-designed processes in which optimal patient safety is at the forefront.” [4] This project demonstrates the capabilities of the SEIPS model to frame complex equity focused process design across two hospitals, and to be a fundamental resource in all phases of data collection and analysis including interview design, deductive coding and themes, analysis, and organization of results.
REFERENCES:
1. Flores G, Research TC on P. Technical Report--Racial and Ethnic Disparities in the Health and Health Care of Children. Pediatrics. 2010;125(4):e979-1020.
2. Holden RJ, Carayon P, Gurses AP, et al. SEIPS 2.0: A human factors framework for studying and improving the work of healthcare professionals and patients. Ergonomics. 2013;56(11). doi:10.1080/00140139.2013.838643
3. Holden RJ, Carayon P. SEIPS 101 and seven simple SEIPS tools. BMJ Qual Saf. Published online May 26, 2021:bmjqs-2020-012538. doi:10.1136/bmjqs-2020-012538
4. Using Human Factors Engineering and the SEIPS Model to Advance Patient Safety in Care Transitions. (n.d.). PSNet. https://psnet.ahrq.gov/perspective/using-human-factors-engineering-and-seips-model-advance-patient-safety-care-transitions
Pediatric health inequities in the United States remain persistent and pervasive. Children of color are more likely to suffer an adverse event during hospitalization and their parents —including those who use a language other than English (LOE) for medical care—report lower trust, poorer communication, and weaker partnerships with healthcare providers. Inequities in the inpatient setting are of particular concern due to high costs, illness severity, and family stress. Given the established links between these inequities and disparate clinical outcomes, interventions aimed at mitigating these inequities are urgently needed. [1]
Patient navigator programs have shown great promise for addressing system-related healthcare disparities. A patient navigator is a nurse or lay health worker who provides culturally competent support to patients and families, via advocacy, support, education, and assistance with care-related logistics. Researchers at Seattle Children’s Hospital (SCH) collaborated with parents/caregivers, staff, and providers to develop and pilot-test a novel program to improve navigation ability, communication, and hospital-to-home transition for a diverse population of children and their families, The Family Bridge Program (FBP). The FBP was designed for a broad population of low-income children of color, is not limited to English-proficient families, and is less time-intensive than traditional navigation.
Our NIH funded grant (R01MD015723-01A1, Lion) led by SCH with the Children’s Hospital of Philadelphia (CHOP) is aimed at using a replicable sociotechnical analysis to map the FBP to a new site (CHOP). This work will result in a comprehensive implementation guide for adaptation to additional sites in the future. Additional aims will test the effect of the FBP on a range of parent reported outcomes, examine whether changes in parent-reported barriers mediate program effects and identify subgroups of parents among whom the FBP is more effective.
METHODS:
Data collection: A sociotechnical analysis at both sites aimed toward: 1) Defining the core FBP Guide job role requirements, 2) Identifying sociotechnical differences between the two sites.
We developed a semi-structured interview guide for a range of clinical staff to elicit sociotechnical details for the six FBP services: 1) Language, 2) Orientation, 3) Communication preferences, 4) Communication coaching, 5) Supportive check in, 6) Discharge follow-up. Participants filled out a questionnaire that included rating their level of proficiency with the six FBP service topics to direct the interview on topics where they indicated experience. Interviews were performed remotely and were recorded. Audio files were transcribed via a transcription service.
Data analysis: A deductive qualitative analysis with codes based on components of the sociotechnical model, Safety Engineering in Patient Safety 2.0 (SEIPS). [2]
We developed a code book based on SEIPS 2.0 and sociotechnical factors addressed in the interview guide. Two team members performed an iterative coding process for interrater reliability using NVivo. Initially two interview transcripts were coded by each team member, then compared using Cohen’s Kappa (K). Any code with a K of < 0.75 was reviewed with team members reaching consensus and updating the code book. This process of coding and reviewing two transcripts at a time was repeated until all coding discrepancies were resolved.
We developed NVivo queries based on the site, six FBG services, and codes representing sociotechnical factors. We extracted task-related coding for each service to define the core tasks of the Guide. We then extracted the remaining codes to map all sociotechnical components (tools, people, environment, organization…) within each service while capturing important differences between the two sites. Emphasis was placed on identifying the most effective communication channels between the Guide and clinical team. Information derived from the queries was applied to a series of tools derived from SEIPS 101 including people maps, task matrix and flow diagrams. [3]
RESULTS:
44 clinicians were interviewed at the two sites. SCH (n=22) 5 bedside nurses, 3 attending physicians, 3 residents, 3 managers, 2 primary care providers, 2 interpreters, 1 social worker, 1 child life specialist, 1 care coordinator, 1 cultural navigator CHOP (n=22), 4 attending physicians, 2 fellows, 3 residents, 3 nurses, 1 care team assistant, 1 social work care coordinator, 1 family visitor service, 3 social workers, 2 case managers, 1 community health navigator, 1 primary care physician.
Participants of every role would receive the Guide favorably and described opportunities for the Guide to contribute meaningfully in the six FBP services. Concerns included duplication of work, communication challenges, and limiting the Guide to appropriate tasks. Participant comments suggest the Guide could take the lead in Language, Orientation, and Follow Up and be a key collaborator to nursing and residents in Communication Preferences and Coaching. For Unmet Needs, participants discussed opportunities for the Guide to contribute to the screening process, but expressed concerns that addressing needs was the responsibility of social work and care managers.
Participants were consistent in ideas for communication within-shift that included the use of EHR-based text messaging and in-person discussions. However, participants lacked consensus on Guide documentation and communication across shifts, with ideas ranging from a separate note authored by the Guide to relying on various clinicians to record and relay issues and concerns. Many participants noted the Guide had the potential to improve communication between families and clinicians, but also between clinicians.
The interviews revealed multiple sociotechnical differences between the two sites. While the patient populations of both sites shared many characteristics and needs, there were differences in the language/cultures of populations, geography impacting transportation needs, and the availability and types of community resources. Organizationally the two sites presented differences including an approach to families who speak a language other than English, format of family centered rounds, practices on communication and documentation, integration of social work, existing programs and roles including coordinators and navigators, and primary care practices associated with the hospital.
While the interviews revealed differences between the two sites, they also indicated important differences by unit within each hospital. This included staffing and rounding models, and even variations in the quality of Wi-Fi for online language interpreter services. Use of the EHR appeared to vary by unit within each hospital with different documentation practices and templates adding challenges to communication and collaboration for the Guide. From the patient-family perspective, one hospital has at least fifteen different unit-based versions of welcome packets for families thereby complicating the capability to provide multilingual resources.
DISCUSSION:
The Agency for Healthcare Research and Quality recommends the application of human factors engineering methods, including methods guided by the SEIPS model, to designing and improving not just medical devices and health information technology, but to “larger process of care” and “creating end-to-end, well-designed processes in which optimal patient safety is at the forefront.” [4] This project demonstrates the capabilities of the SEIPS model to frame complex equity focused process design across two hospitals, and to be a fundamental resource in all phases of data collection and analysis including interview design, deductive coding and themes, analysis, and organization of results.
REFERENCES:
1. Flores G, Research TC on P. Technical Report--Racial and Ethnic Disparities in the Health and Health Care of Children. Pediatrics. 2010;125(4):e979-1020.
2. Holden RJ, Carayon P, Gurses AP, et al. SEIPS 2.0: A human factors framework for studying and improving the work of healthcare professionals and patients. Ergonomics. 2013;56(11). doi:10.1080/00140139.2013.838643
3. Holden RJ, Carayon P. SEIPS 101 and seven simple SEIPS tools. BMJ Qual Saf. Published online May 26, 2021:bmjqs-2020-012538. doi:10.1136/bmjqs-2020-012538
4. Using Human Factors Engineering and the SEIPS Model to Advance Patient Safety in Care Transitions. (n.d.). PSNet. https://psnet.ahrq.gov/perspective/using-human-factors-engineering-and-seips-model-advance-patient-safety-care-transitions
Event Type
Oral Presentations
TimeMonday, March 2510:52am - 11:15am CDT
LocationSalon A-3
Patient Safety Research and Initiatives