Abstract
Objective
To identify the risk factors home healthcare (HHC) clinicians associate with patient deterioration and understand how clinicians respond to and document these risk factors.
Methods
We interviewed multidisciplinary HHC clinicians from January to March of 2021. Risk factors were mapped to standardized terminologies (eg, Omaha System). We used directed content analysis to identify risk factors for deterioration. We used inductive thematic analysis to understand HHC clinicians’ response to risk factors and documentation of risk factors.
Results
Fifteen HHC clinicians identified a total of 79 risk factors that were mapped to standardized terminologies. HHC clinicians most frequently responded to risk factors by communicating with the prescribing provider (86.7% of clinicians) or following up with patients and caregivers (86.7%). HHC clinicians stated that a majority of risk factors can be found in clinical notes (ie, care coordination (53.3%) or visit (46.7%)).
Discussion
Clinicians acknowledged that social factors play a role in deterioration risk; but these factors are infrequently studied in HHC. While a majority of risk factors were represented in the Omaha System, additional terminologies are needed to comprehensively capture risk. Since most risk factors are documented in clinical notes, methods such as natural language processing are needed to extract them.
Conclusion
This study engaged clinicians to understand risk for deterioration during HHC. The results of our study support the development of an early warning system by providing a comprehensive list of risk factors grounded in clinician expertize and mapped to standardized terminologies.
Keywords: home health, qualitative, electronic health record, Omaha System, nursing informatics, natural language processing
BACKGROUND
In the United States, home healthcare (HHC) agencies provide care to more than 3.3 million adults per year.1 Services provided improve physical functioning and allow patients to manage their health conditions in their homes.2 While a HHC episode (period of time using HHC services) can be the result of an acute hospital stay, approximately two-thirds of HHC episode are initiated by other means such as a primary care referral rather than a preceded hospitalization.1 HHC clinicians monitor patients and assess for signs of deterioration, while providing interventions to reduce negative outcomes, such as hospitalizations or emergency department (ED) visits.3–5 Currently, 1 in 5 HHC patients are hospitalized during their HHC episode.1 Several efforts have focused on health conditions in the ambulatory care setting associated with preventable hospitalization,6 but these numbers have not recently improved.7
One way to reduce negative outcomes and alert clinicians’ to deterioration is an early warning system that processes electronic health record (EHR) documentation such as clinical notes.8 There is value in using routine, free-text clinical notes stored in the EHR as a proxy of a clinician’s concern about a patient.9,10 HHC clinicians operate autonomously in the community and communicate with their team remotely, often through documentation in the EHR.10,11 Since the clinicians are seeing the patient independently, detailed documentation is critical to inform the care team of the patient’s status.9,12,13 In a text mining study using HHC clinical notes, 10 risk factor categories (eg, clinical factors, service use, social and environmental factors) were found to be associated with hospitalization or ED visits.9
No previous studies have engaged HHC clinicians to understand their assessment of risk factors for deterioration and how their assessment is translated to their workflow and documentation. Further, to generate a robust early warning system, risk factors should be mapped to a standardized terminology to enable complex concepts and relationships to be integrated into computerized algorithms.14,15 However, previous research in HHC has not focused on this topic. To address this gap, we initiated qualitative interviews with HHC clinicians to identify risk factors for patient deterioration. We also aimed to understand how HHC clinicians respond to risk factors and where this information is captured in EHR documentation. Our results, grounded in HHC clinician expertize and mapped to standardized terminologies, will aid in the development of a HHC early warning system that identifies patients at risk for deterioration through EHR documentation.
METHODS
Study design and sample
We conducted our qualitative study at a large not-for-profit HHC agency located in New York, NY. The agency employs approximately 2300 field clinicians and provides care to more than 75 000 patients annually.16 Clinicians in this study were registered nurses, social workers, physical and occupational therapists, and clinical team managers employed by the HHC agency. To participate in the study, clinicians had to have more than 5 years of experience which previous studies suggest demonstrates clinical expertize.17–19 Clinicians were recruited using electronic flyers circulated by clinical team managers and asking clinicians who participated in the interviews to refer other interested colleagues (snowball sampling). This study was approved by the institutional review board of the participating institutions.
Data collection
Our research team includes members with Bachelor’s degrees or above and are experts in health informatics (KC, SR, MT, KB, and EK) and/or HHC (KB, MM, and MT). The interview guide was discussed by research team members during multiple team meetings. It was amended based on feedback from each member until a final interview guide was created. The final semistructured interview guide included open-ended questions that addressed HHC clinicians’ identification, response, and documentation of risk factors for HHC patient deterioration. Deterioration was defined as risk factors that would lead to a hospitalization or an ED visit. Supplementary Appendix SA includes the full list of interview questions used to guide the discussion.
Study interviews were conducted from January 2021 to March 2021. All clinicians received a $100 electronic gift card for participating in a one-time 45-min interview. The audio from all the interviews were recorded using a video conferencing platform. Consent was given by each participant prior to beginning the interview.20 Research team members (MH, MO, and MT) participated in interviews with HHC clinicians. Reflective field notes were taken to help identify overarching themes discussed in each interview. These notes were then used to validate if the identified themes by the research team were grounded in the interviewer’s comments. We conducted interviews until saturation was achieved (ie, no new themes emerged in subsequent interviews). We used the video conference’s transcription service to transcribe the audio files. We evaluated transcription accuracy by comparing 2-min portions of the interview transcript with the original recordings in a random sample (n = 4) and found transcription accuracy to be high (accuracy >95%).
Data analysis
The transcripts were imported into Dedoose,21 an application used to store and analyze qualitative interviews.22 Based on the interview guide, 3 broad categories were identified: risk factors for deterioration, response to risk factors, and documentation of risk factors. The first category, risk factors for deterioration, was analyzed using directed content analysis. This type of analysis was chosen because of its previous use to inform the development of early warning systems22 and because it maps codes to an existing theory (eg, standardized terminology) which supports the development of early warning systems.23 The latter 2 categories, response to risk factors and documentation of risk factors, were analyzed using inductive thematic analysis. This approach previously used to understand clinicians’ perspectives24,25 was identified as appropriate because we sought to code the data without using a framework to best understand the behaviors of HHC clinicians.26
Standardizing medical terminology helps to describe concepts and relationships between each concept.27 Through mapping interview content to standardized terminologies, complex concepts and relationships can be integrated into an early warning system.15 Therefore, we standardized the risk factors for deterioration by mapping each to a concept in the Omaha System Classification Scheme.21,28,29 The Omaha System was chosen as the primary standardized terminology in this study because its previous use with early warning system demonstrated validity, it is grounded in nursing practice,30 and it is built on a comprehensive framework that covers 4 domains of patient care.31 The Omaha System is recognized by the American Nurses Association as a standardized nursing terminology, and it is a widely used taxonomy of clinical information used in community-based care to describe clinical problems.32 It includes 42 problems (eg, “Cognition,” “Circulation,” and “Neuro-musculoskeletal function”) aligned with 335 unique signs or symptoms (eg, “disoriented to time/place/person,” “edema,” and “decreased muscle strength”).28 Each problem is classified into 1 of 4 domains: physiological, environmental, psychological, and health-related behaviors.28 Our research team discussed the mapping of codes until full consensus was achieved. Alternatively, if the themes did not fit into an Omaha System, they were mapped to other Unified Medical Language System (UMLS) terminologies to support their integration into a future early warning system. An example of how the risk factors were extracted from the interviews is contained in Figure 1.
Figure 1.
Example of how risk factors were extracted from interviews. *Not included in the Omaha System.
For the latter 2 categories, response to risk factors and documentation of risk factors categories, we were more interested in understanding the clinicians’ behavior rather than preparing the information to be integrated into an early warning system. Thus, this information was not mapped to a standardized terminology. We identified emerging themes using Kinger & Varpio’s Guide to thematic analysis guidance as follows:26 (1) we familiarized ourselves with the data through participation in the original interview or listening to the audio recordings (MH and MO). (2) Codes were generated from the transcriptions using Dedoose.21 Both coders (MH and MO) had formal training or previous experience with qualitative analysis.17 (3) Inter-rater reliability was assessed by dual coding the first 3 interview transcripts. Discrepancies were discussed until consensus was met. If consensus could not be met, a third reviewer who has expertize in health informatics and HHC (MT) resolved the conflict. When new codes emerged, they were discussed (MH, MO, and MT) and included in the coding scheme if agreed upon. (4) Themes (eg, a domain that includes several risk factors) were drawn from collated codes. Themes were named and defined based on a standardized terminology and via discussion of the theme’s appropriateness to our research question (MH, MO, and MT).
RESULTS
Fifteen HHC clinicians participated in the semistructured interviews. Five disciplines of HHC clinicians were represented in our sample with most of the clinicians being nurses (53%). Participant characteristics are further described in Table 1.
Table 1.
Clinician characteristics
| Characteristics | Category | No. of clinicians (n = 15) |
|---|---|---|
| Profession (n = 15) | Nurse | 8 (53.3%) |
| Physical Therapist | 3 (20%) | |
| Social Worker | 1 (6.7%) | |
| Occupational Therapist | 1 (6.7%) | |
| Clinical Team Manager | 2 (13.3%) | |
| Highest level of education (n = 15) | Bachelors | 6 (40%) |
| Masters | 5 (33.3%) | |
| Doctorate | 4 (26.7%) | |
| Years in HHC (n = 12) | 0–5 | 1 (8.3%) |
| 6–10 | 5 (41.7%) | |
| >10 | 6 (50%) |
HHC: home healthcare.
Risk factors for deterioration
A total of 79 risk factors were identified by HHC clinicians. Of those 79 risk factors, 60 (76%) could be mapped to 1 of the 4 Omaha System domains: physiological (n = 29), environmental (n = 7), psychological (n = 7), and health-related behaviors (n = 16) (Tables 2 and 3). The common risk factors from the Omaha System problem list were “Skin (60% of clinicians mentioned this risk factor),” “Circulation (53.3%),” “Social Contact (73.3%),” and “Healthcare supervision (73.3%).” Risk factors not found in the Omaha System (Table 4) but frequently discussed were “Patient-clinician interaction (80%)” and “Medical diagnosis (73.3%).”
Table 2.
Physiological risk factors for deterioration
| Omaha problem | Omaha sign/symptom | Risk factors discussed in interviews | No. of clinicians, N (%) |
|---|---|---|---|
| Skin | Delayed incisional healing | Size (increasing), drainage (not granulating), stage (increasing) | 9 (60%) |
| Lesion/pressure ulcer | Pressure ulcers, chronic wounds | ||
| Circulation | Edema | Fluid overload, swelling, weight gain | 8 (53.3%) |
| Abnormal blood pressure reading | Uncontrolled blood pressure | ||
| Syncopal episodes | Dizziness | ||
| Anginal pain | Chest pain | ||
| Neuro-musculoskeletal function | Decreased muscle strength | Weakness | 7 (46.7%) |
| Fractures | Hip dislocation | ||
| Gait/ambulation disturbance | Falls | ||
| Decreased sensation | Sensory impairment, recent surgery | ||
| Other | Mobility | ||
| Nutrition | Does not follow recommended nutrition plan | Salty diet | 5 (33.3%) |
| Hypo/hyperglycemia | Uncontrolled blood sugar | ||
| Cognition | Disoriented to time/place/person | Mental status change | 5 (33.3%) |
| Limited reasoning/abstract thinking ability | Not cognitively intact | ||
| Consciousness | Lethargic | Lethargy | 4 (26.7%) |
| Unresponsive | Unresponsive | ||
| Other | Alertness | ||
| Respiration | Abnormal breath patterns | Shortness of breath, gasping, cannot lie flat | 4 (26.7%) |
| Communicable/Infectious condition | Fever | Fever | 4 (26.7%) |
| Sepsisa | Signs of sepsis |
Not included in the Omaha System.
Table 3.
Environmental/psychological/health-related behaviors risk factors for deterioration
| Omaha problem | Omaha sign/symptom | Risk factors discussed in interviews | No. of clinicians, N (%) |
|---|---|---|---|
| Environmental risk factors | |||
| Income | Low/No income | Income | 4 (26.7%) |
| Uninsured medical expenses | Public insurance | ||
| Sanitation | Soiled living area | Cleanliness, infested environment | 3 (20%) |
| Residence | Cluttered living space | Clutter, lighting | 3 (20%) |
| Inadequate safety devices | No access to proper equipment | ||
| Psychological risk factors | |||
| Social contact | Limited social contact | Lack of family/formal caregiver support | 11 (73.3%) |
| Minimal outside stimulation | Homebound | ||
| Communication with community resources | Transportation barrier | No transportation | 3 (20%) |
| Language barrier | Language barrier | ||
| Caretaking/Parenting | Difficulty providing physical care/safety | Lack of aide hours, living alone | 3 (20%) |
| Insufficient caregiver knowledgea | Caregiver knowledge | ||
| Health-related behavior risk factors | |||
| Healthcare supervision | Inadequate/inconsistent source of health care | No primary care provider | 11 (73.3%) |
| Fails to seek care for symptoms requiring evaluation/treatment inadequate treatment plan | Refusing care | ||
| Other | Additional disciplines consulted patient involvement in care HHC, weekend staff, no provider response to HHC clinician | ||
| Low health literacya | Patient health knowledge | ||
| Medication regimen | Inadequate medication regimen | Frequent medication changes | 5 (33.3%) |
| Inadequate system for taking medications | Many medications, follows medication instructions | ||
| Other | Medication knowledge | ||
| Personal care | Unwilling/unable/forgets to complete personal care activities | Bed bound, ability to complete activities of daily living | 4 (26.7%) |
| Family planning | Fears other’s reaction regarding family planning choices | Relationship with caregivers, code status, palliative care conversation | 3 (20%) |
Not included in the Omaha System.
HHC: home healthcare.
Table 4.
Other risk factors for deterioration not included in the Omaha System
| Problem | Risk factors discussed in interviews | No. of clinicians, N (%) |
|---|---|---|
| Patient–clinician interaction | Increased communication with provider (SNOMED-CT, A9414839) | 12 (80%) |
| Unplanned provider visit (HL7V3.0, A19720524) | ||
| Increased use of HHC agency call center (MSH, A27191876) | ||
| Caregiver/patient concern (MSH, A28394190) | ||
| Past hospitalizations (HL7V3, A19721666) | ||
| Past HHC episode (LNC, A28739331) | ||
| Increased frequency of care (SNOMED-CT, A2931514) | ||
| Interaction with Adult Protective Services (MEDCIN, A13853820) | ||
| Medical diagnosis | Heart disease (SNOMED-CT, A2876047) | 11 (73.3%) |
| Diabetes (SNOMED-CT, A2928669) | ||
| Depression (SNOMED-CT, A16967729) | ||
| Dementia (SNOMED-CT, A2880500) | ||
| COPD (SNOMED-CT, A3007198) | ||
| Cancer (SNOMED-CT, A2972175) | ||
| Chronic kidney disease (MEDCIN, A13608525) | ||
| Sepsis (SNOMED-CT, A91302008) | ||
| Clinical profile | High number of comorbidities (LNC, A28289184) | 7 (46.7%) |
| High number of chronic conditions (SNOMED-CT, A21626116) | ||
| Older age (SNOMED-CT, A2873259) | ||
| Change over time | Change in baseline— symptoms and vitals (SNOMED-CT, A3207577) | 7 (46.7%) |
| Lack of symptom improvement (SNOMED-CT, A11738700) |
UMLS terminologies—HL7V3.0: Health Level Seven Version 3; LNC: Logical Observation Identifiers Names and Codes terminology; MSH: Medical Subject Headings; SNOMED-CT: Systematized Nomenclature of Medicine-Clinical Terms; HHC: home healthcare.
Response to risk factors
HHC clinicians most frequently responded to risk factors by communicating with the prescribing provider (86.7%) or following up with patients and caregivers (86.7%). Alternatively, HHC clinicians would schedule a Pro Re Nata (PRN) visit with the patient, which is an additional visit “as the need arises.” A full list is included in Table 5.
Table 5.
How HHC clinicians’ respond to risk factors
| Response theme | Interviews exemplars | No. of clinicians, N (%) |
|---|---|---|
| Communicate with prescribing providers | “The doctors don't like us to bother them if the patient is doing well. If the patient is deteriorating, that's one of the reasons why we will call the doctor. Anytime we change the plan of care we contact the doctor, but most of the time we call the doctor if we are trying to get more visits or add a discipline, like the patient fell and needs physical therapy. The more we contact the doctor the more it indicates the patient may need an intervention.” RN2 | 13 (86.7%) |
| Follow-up with patients or Caregivers | “I try to keep caregivers up to date with everything that's going on, especially for patients with dementia where the caregiver is very involved. I text them after each visit. To let them know ‘I saw your mother, we did this, she was able to do this, she's progressing well with X, Y and Z,’ and I’ll be back every day, so that everyone is in the loop as to what's going on.” PT3 | 13 (86.7%) |
| Communicate with team members | “Patients will call customer care and I will check my notes and get a phone call from customer care directly to follow up with the patient.” CTM2 | 5 (33.3%) |
| Increase visit frequency | “If I have a patient that is not hitting the mark, then I would have to do that reassessment every single visit and determine if the patient is not doing their part, or if we need more time and visits.” PT1 | 2 (13.3%) |
| PRN visit | “So, if let's say a patient calls me and says ‘my wound is leaking. Something's wrong and you're not coming back until next week.’ I would call my manager and say I need a PRN visit. Then he would put in a new PRN order.” RN2 | 2 (13.3%) |
| Educate the patient or caregivers | “I talk to the patient and determine what the barriers are, if it's something in terms of their behavior, then I would re-educate. If the patient feels that they're not getting enough support then it's education for the caregiver.” PT1 | 2 (13.3%) |
CTM: Clinical Team Manager; PRN: as needed (Pro Re Nata); PT: Physical Therapist; RN: Registered Nurse; HHC: home healthcare.
Documentation of risk factors
Clinicians described where risk factors were captured in EHR documentation. Clinicians stated that they document everything to provide a record of interventions and assessments performed and document to show the added value of HHC. Risk factors were commonly documented in clinical notes, care coordination (53.3%) or visit notes (46.7%) contained in the EHR but were also in email communications contained outside the EHR (33.3%). The complete list and supporting quotes are described in Table 6.
Table 6.
Where HHC clinicians document risk factors
| Codes | Excerpt | No. of clinicians, N (%) |
|---|---|---|
| Care Coordination Notes | “When the symptom is significant enough, we would write that down in the care coordination, now that we reached out to someone especially a doctor” RN6 | 8 (53.3%) |
| Visit Notes | “We included within the note of a particular patient, in order to capture everything, especially what's going on in the house at that moment, I would have to write a narrative note, so I don't take pictures, I don't take videos, it's really just written documentation.” PT1 | 7 (46.7%) |
| Email Communication | “And documentation is always there, as soon as something happens, I will communicate with the therapists and then I'll follow up with an email later as a reminder.” RN3 | 5 (33.3%) |
| EHR Documentation (Structured data) | “In structured data there are certain things that are already there and I simply plug in the answers. It's kind of cookie cutter in that way, which is nice.” PT2 | 2 (13.3%) |
| HHC Agency Hotline | “As a centralized call system, when the patient calls the hotline the call center will then send an email to the nurse, covering nurse, and the nurse’s manager. Then, the manager can triage it.” RN2 | 2 (13.3%) |
EHR: electronic health records; HHC: home healthcare; PT: Physical Therapist; RN: Registered Nurse.
DISCUSSION
This manuscript describes the perspectives of 15 multidisciplinary HHC clinicians on their identification, response, and documentation of risk factors for deterioration (eg, hospitalization and ED visit). We found that HHC clinicians identified risk factors for deterioration related to a patient’s physiological, environmental, psychological, and health-related behavior characteristics, which include all 4 domains of the Omaha System. HHC clinicians highlighted the physiological risk factors “Skin” and “Circulation.” These risk factors were often the result of acute exacerbations of a chronic illness (eg, wounds and heart failure) which was previously reported to be the cause of 80% of acute care hospitalizations.33
Despite the emphasis recognized by the study participants on the social risk factors (eg, environmental, psychological, and health-related behaviors) in HHC, there are a limited number of studies that examine social factors as precursors to deterioration in HHC. While limited “Social contact” is commonly known as a risk factor,26,28,32 limitations in “Healthcare supervision”, or inadequate treatment plans, are less frequently studied.34 Other risk factors included “Medical diagnoses”26,28,32 and “Patient–clinician interaction.” Studies found that high visit intensity35,36 and recent hospitalizations33 were associated with hospitalization; however, there is little literature that investigates how patient–clinician interaction across settings influence rehospitalization. With additional studies highlighting the feasibility of extracting social factors from the EHR,37–39 this study reinforces the importance of understanding social risk factors in assessing for hospitalization risk in HHC.
The advantage of mapping risk factors to standardized terminologies is that it enables a more comprehensive and potentially interoperable extraction of risk factors. These standardized terminologies can be integrated into early warning systems such as clinical decision support.15 While the Omaha system30 captures a majority of the risk factors described by the HHC clinicians, a multi-terminology approach is needed to generate a comprehensive information model of HHC risk. An advantage of the Omaha System is its focus on symptoms; however, a drawback is that it is less focused on medical diagnosis compared to other terminologies. Therefore, in addition to the Omaha System, a medical terminology such as the Systematized Nomenclature of Medicine-Clinical Terms (SNOMED-CT) might be used to represent medical diagnoses to provide greater context to symptoms. Since the Omaha System has also been mapped to other terminologies, it is feasible to use this multi-terminology approach in developing a HHC early warning system to identify patients at risk for a hospitalization.28
Our results yielded a list of actions HHC clinicians perform in response to risk factors. HHC clinicians noted that the most frequent response was to communicate with the prescribing provider and following-up with the family or caregiver. Care coordination in HHC relies upon effective clinician and provider communication40 and most coordination is conducted remotely. The importance of effective communication was shown by a study that found that poor communication increases hospitalization readmission risk in HHC.37 Our results highlight the need for monitoring communication between clinicians and follow-up activities in order to understand patient risk.
Clinicians discussed how most of the documentation of risk factors is contained in clinical notes (eg, visit notes and care coordination notes) rather than structured data (ie, assessment drop-down menus). Previous studies have supported the value of using clinical notes in the HHC setting to identify patients with high risk of hospitalization.9,41 Aligned with a report on advancing HHC informatics,42 these findings highlight that advanced data science methods such as natural language processing might be needed to identify risk factors from EHR documentation. The integration of data from structured and unstructured data has demonstrated improved prediction of early warning systems which suggest the value of including both data structures in early warning systems.43
With further research, our results can be used in the development of a robust early warning systems that would highlight risk information to HHC clinicians. Including clinicians early on in the development of early warning systems is vital to its success.22 Thus, this study adds to the literature by showing feasibility and providing a basis for the development of an early warning system that is grounded in HHC clinician expertize and mapped to a standard terminology to identify patients at risk for deterioration in HHC.
Study limitations
We recognize that patients transitioning from the hospital to home are a significant subgroup included in this population; however, this study did not specifically investigate care transitions but instead focused on general risk factors influencing a majority of patients admitted to HHC. While interviews were ended when saturation was achieved, a limitation of our study is its small sample size. Clinicians were also only representative of 1 HHC agency. Thus, generalizability is limited; however, this agency is one of the largest HHC agencies in the country. In addition, while this study did include interdisciplinary clinicians, there was not an equal distribution of HHC clinicians interviewed across disciplines.
CONCLUSION
This study uses HHC clinician expertize to identify risk factors for deterioration and understand HHC clinicians’ response and documentation of these risk factors. Our results reinforce the importance of including physiological, environmental, psychological, and health-related behavior data in risk identification. There is significant potential in studying clinical notes to understand patient risk for hospitalization, but to retrieve this information novel data analysis such as natural language processing are needed. Through further inclusion of risk factors and improved clinician communication, prompt recognition of deterioration over time may lead to timely intervention and decreased hospitalizations and ED visits. In conclusion, this study supports the development of a robust early warning system by providing a comprehensive list of risk factors grounded in HHC clinician expertize mapped to standardized terminologies.
FUNDING
This work was supported by the Agency for Healthcare Research and Quality (AHRQ) grant number R01 HS027742. The content is solely the responsibility of the authors and does not necessarily represent the official views of the AHRQ. Ms Hobensack is supported by the National Institute for Nursing Research training grant Reducing Health Disparities through Informatics (RheaDI) grant number T32NR007969 and the Jonas Scholarship. Ms Kennedy is supported by the National Institute of Nursing Research Ruth L. Kirschstein Predoctoral Individual National Research Service Award grant number F31NR019919.
AUTHOR CONTRIBUTIONS
MH, JS, MO, KB, MM, and MM developed the study design. MH, MO, SS, and MT participated in data collection. All authors supported the interpretation of the data. MH, MO, and MT drafted the manuscript and all authors reviewed, edited, and approved the final draft prior to submission.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Journal of the American Medical Informatics Association online.
CONFLICT OF INTEREST STATEMENT
None declared.
DATA AVAILABILITY
The data in this article cannot be shared publicly due to privacy of the individuals who participated. Data may be shared upon reasonable request to the corresponding author.
Supplementary Material
Contributor Information
Mollie Hobensack, Columbia University School of Nursing, New York City, New York, USA.
Marietta Ojo, Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA.
Yolanda Barrón, Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA.
Kathryn H Bowles, Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA; Department of Biobehavioral Health Sciences, University of Pennsylvania School of Nursing, Philadelphia, Pennsylvania, USA.
Kenrick Cato, Columbia University School of Nursing, New York City, New York, USA; Emergency Medicine, Columbia University Irving Medical Center, New York City, New York, USA.
Sena Chae, College of Nursing, University of Iowa, Iowa City, Iowa, USA.
Erin Kennedy, Department of Biobehavioral Health Sciences, University of Pennsylvania School of Nursing, Philadelphia, Pennsylvania, USA.
Margaret V McDonald, Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA.
Sarah Collins Rossetti, Columbia University School of Nursing, New York City, New York, USA; Department of Biomedical Informatics, Columbia University, New York City, New York, USA.
Jiyoun Song, Columbia University School of Nursing, New York City, New York, USA; Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA.
Sridevi Sridharan, Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA.
Maxim Topaz, Columbia University School of Nursing, New York City, New York, USA; Center for Home Care Policy & Research, Visiting Nurse Service of New York, New York City, New York, USA; Data Science Institute, Columbia University, New York City, New York, USA.
REFERENCES
- 1. Report to Congress: Medicare Payment Policy. Washington, DC: Medicare Payment Advisory Commission; 2021. [Google Scholar]
- 2. Ellenbecker CH, Samia L, Cushman MJ, et al. Patient safety and quality in home health care. In: Hughes RG, ed. Patient Safety and Quality: An Evidence-Based Handbook for Nurses. Rockville, MD: Agency for Healthcare Research and Quality (US); 2008. [PubMed] [Google Scholar]
- 3. Melby L, Obstfelder A, Hellesø R. “We tie up the loose ends”: homecare nursing in a changing health care landscape. Glob Qual Nurs Res 2018; 5: 233339361881678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gray E, Currey J, Considine J. Hospital in the home nurses’ recognition and response to clinical deterioration. J Clin Nurs 2018; 27(9–10): 2152–60. [DOI] [PubMed] [Google Scholar]
- 5. Jones CD, Falvey J, Hess E, et al. Predicting hospital readmissions from home healthcare in Medicare beneficiaries. J Am Geriatr Soc 2019; 67(12): 2505–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Agency for Healthcare Research & Quality. Potentially avoidable hospitalizations: Ambulatory care-sensitive conditions, 2015. https://www.ahrq.gov/research/findings/nhqrdr/chartbooks/carecoordination/measure3.html. Accessed April 19, 2021.
- 7.Centers for Medicare and Medicaid Services. Home Health Quality Measures, 2019. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HomeHealthQualityInits/Home-Health-Quality-Measures.html. Accessed April 19, 2021.
- 8. Wood C, Chaboyer W, Carr P. How do nurses use early warning scoring systems to detect and act on patient deterioration to ensure patient safety? A scoping review. Int J Nurs Stud 2019; 94: 166–78. [DOI] [PubMed] [Google Scholar]
- 9. Topaz M, Woo K, Ryvicker M, Zolnoori M, Cato K. Home healthcare clinical notes predict patient hospitalization and emergency department visits. Nurs Res 2020; 69(6): 448–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Demner-Fushman D, Chapman WW, McDonald CJ. What can natural language processing do for clinical decision support? J Biomed Inform 2009; 42(5): 760–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. De Vliegher K, Paquay L, Vernieuwe S, Van Gansbeke H. The experience of home nurses with an electronic nursing health record. Int Nurs Rev 2010; 57(4): 508–13. [DOI] [PubMed] [Google Scholar]
- 12. Sockolow PS, Bowles KH, Adelsberger MC, Chittams JL, Liao C. Challenges and facilitators to adoption of a point-of-care electronic health record in home care. Home Health Care Serv Q 2014; 33(1): 14–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Korach ZT, Yang J, Rossetti SC, et al. Mining clinical phrases from nursing notes to discover risk factors of patient deterioration. Int J Med Inform 2020; 135: 104053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ahmadian L, van Engen-Verheul M, Bakhshi-Raiez F, Peek N, Cornet R, De Keizer NF. The role of standardized data and terminological systems in computerized clinical decision support systems: literature review and survey. Int J Med Inform 2011; 80(2): 81–93. [DOI] [PubMed] [Google Scholar]
- 15. Dissanayake PI, Colicchio TK, Cimino JJ. Using clinical reasoning ontologies to make smarter clinical decision support systems: a systematic review and data synthesis. J Am Med Inform Assoc 2020; 27(1): 159–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. About Us. New York, NY: Visiting Nurse Services of New York. https://www.vnsny.org/who-we-are/about-us/. Accessed April 19, 2021.
- 17. Hobensack M, Ojo M, Bowles K, McDonald M, Song J, Topaz M. Home healthcare clinicians’ perspectives on electronic health records: a qualitative study. Stud Health Technol Inform 2021; 284: 426–30. [DOI] [PubMed] [Google Scholar]
- 18. Guven N. The development of nurses’ individualized care perceptions and practices: Benner’s novice to expert model perspective. Int J Caring Sci 2019; 12(2): 1279. [Google Scholar]
- 19. Bowles KH, Ratcliffe S, Potashnik S, et al. Using electronic case summaries to elicit multi-disciplinary expert knowledge about referrals to post-acute care. Appl Clin Inform 2016; 7(2): 368–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Reñosa MDC, Mwamba C, Meghani A, et al. Selfie consents, remote rapport, and Zoom debriefings: collecting qualitative data amid a pandemic in four resource-constrained settings. BMJ Glob Heal 2021; 6(1): e004193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dedoose Version 8.0.35. A Web Application forManaging, Analyzing, and PresentingQualitative and Mixed Method Research Data. LosAngeles, CA: SocioCultural ResearchConsultants, LLC; 2018. [Google Scholar]
- 22. Schaaf J, Prokosch HU, Boeker M, et al. Interviews with experts in rare diseases for the development of clinical decision support system software—a qualitative study. BMC Med Inform Decis Mak 2020; 20(1): 230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res 2005; 15(9): 1277–88. [DOI] [PubMed] [Google Scholar]
- 24. Topaz M, Ronquillo C, Peltonen LM, et al. Nurse informaticians report low satisfaction and multi-level concerns with electronic health records: results from an international survey. AMIA Annu Symp Proc 2017; 2016: 2016–25. [PMC free article] [PubMed] [Google Scholar]
- 25. Topaz M, Peltonen LM, Mitchell J, et al. How to improve information technology to support healthcare to address the COVID-19 pandemic: an international survey with health informatics experts. Yearb Med Inform 2021; 30(1): 61–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kiger ME, Varpio L. Thematic analysis of qualitative data: AMEE Guide No. 131. Med Teach 2020; 42(8): 846–54. [DOI] [PubMed] [Google Scholar]
- 27. Aldosari B, Alanazi A, Househ M. Pitfalls of ontology in medicine. Stud Health Technol Inform 2017; 238: 15–8. [PubMed] [Google Scholar]
- 28. Martin KS. The Omaha System: A Key to Practice Documentation, and Information Management. St Louis, MO: Elsevier Saunders; 2005: 484. [Google Scholar]
- 29. Garvin JH, Martin KS, Stassen DL, Bowles KH. Omaha System: coded data that describe patient care. J AHIMA 2008; 79(3): 44–52. [PubMed] [Google Scholar]
- 30. Topaz M, Golfenshtein N, Bowles KH. The Omaha System: a systematic review of the recent literature. J Am Med Inform Assoc 2014; 21(1): 163–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Monsen KA. The Omaha system as an ontology and meta-model for nursing and healthcare in an era of Big Data. Kontakt 2018; 20(2): e109–10 [Google Scholar]
- 32.The Omaha System. https://www.omahasystem.org/. Accessed July 2021.
- 33. O’Connor M. Hospitalization among Medicare-reimbursed skilled home health recipients. Home Health Care Manag Pract 2012; 24(1): 27–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Ma C, Shang J, Miner S, et al. The prevalence, reasons, and risk factors for hospital readmissions among home health care patients: a systematic review. Health Care Manag Pract 2018; 30(2): 83–92. [Google Scholar]
- 35. O'Connor M, Hanlon A, Naylor MD, Bowles KH. The impact of home health length of stay and number of skilled nursing visits on hospitalization among Medicare-reimbursed skilled home health beneficiaries. Res Nurs Health 2015; 38(4): 257–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Bick I, Dowding D. Hospitalization risk factors of older cohorts of home health care patients: a systematic review. Home Health Care Serv Q 2019; 38(3): 111–52. [DOI] [PubMed] [Google Scholar]
- 37. Pesko MF, Gerber LM, Peng TR, Press MJ. Home health care: nurse–physician communication, patient severity, and hospital readmission. Health Serv Res 2018; 53(2): 1008–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Monsen KA, Rudenick JM, Kapinos N, Warmbold K, McMahon SK, Schorr EN. Documentation of social determinants in electronic health records with and without standardized terminologies: a comparative study. Proc Singapore Healthc 2019; 28(1): 39–47. [Google Scholar]
- 39. Arons A, DeSilvey S, Fichtenberg C, Gottlieb L. Documenting social determinants of health-related clinical activities using standardized medical vocabularies. JAMIA Open 2019; 2(1): 81–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Lohman MC, Scherer EA, Whiteman KL, Greenberg RL, Bruce ML. Factors associated with accelerated hospitalization and re-hospitalization among Medicare home health patients. J Gerontol A Biol Sci Med Sci 2018; 73(9): 1280–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Topaz M, Koleck TA, Onorato N, Smaldone A, Bakken S. Nursing documentation of symptoms is associated with higher risk of emergency department visits and hospitalizations in homecare patients. Nurs Outlook 2021; 69(3): 435–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Sockolow PS, Bowles KH, Topaz M, et al. The time is now: informatics research opportunities in home health care invited editorial 100. Appl Clin Inform 2021; 12(1): 100–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Song J, Ojo M, Bowles KH, et al. Detecting language associated with home healthcare patient’s risk for hospitalization and emergency department visit. Nurs Res 2022; (4). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data in this article cannot be shared publicly due to privacy of the individuals who participated. Data may be shared upon reasonable request to the corresponding author.