- Email: [email protected]
Perceived Barriers to Communication Between Hospital and Nursing Home at Time of Patient Transfer
ORIGINAL STUDIES
Perceived Barriers to Communication Between Hospital and Nursing Home at Time of Patient Transfer Faraaz Shah, MD, Orah Burack, MA, and Kenneth S. Boockvar, MD, MS
Objectives: To identify perceived barriers to communication between hospital and nursing home at the time of patient transfer and examine associations between perceived barriers and hospital and nursing home characteristics. Design: Mailed survey. Setting: Medicare- or Medicaid-certified nursing homes in New York State. Participants: Nursing home administrators, with input from other nursing home staff. Measurements: Respondents rated the importance as a barrier to hospital-nursing home communication of (1) hospital providers’ attitude, time, effort, training, payment, and familiarity with nursing home patients; (2) unplanned and off-hours transfers; (3) HIPAA privacy regulations; and (4) lost or failed information transmission. Associations were determined between barriers and the following organizational characteristics: (1) hospital-nursing home affiliations, pharmacy or laboratory agreements, cross-site staff visits, and cross-site physician care; (2) hospital size, teaching status, and frequency of geriatrics specialty care; (3) nursing home size, location, type, staffing, and Medicare quality indicators; and (4) hospitalto-nursing home communication, consistency of
hospital care with health care goals, and communication quality improvement efforts. Results: Of 647 questionnaires sent, 229 were returned (35.4%). The most frequently reported perceived barriers to communication were sudden or unplanned transfers (44.4%), transfers that occur at night or on the weekend (41.4%), and hospital providers’ lack of effort (51.0%), lack of familiarity with patients (45.0%), and lack of time (43.5%). Increased hospital size, teaching hospitals, and urban nursing home location were associated with greater perceived importance of these barriers, and cross-site staff visits and hospital provision of laboratory and pharmacy services to the nursing home were associated with lower perceived importance of these barriers. Conclusions: Hospital and nursing home characteristics and interorganizational relationships were associated with nursing home administrators’ perceptions of barriers to hospital-nursing home communication. These findings may inform design and targeting of interventions to improve intersite communication processes. (J Am Med Dir Assoc 2010; 11: 239–245) Keywords: Communication barriers; nursing homes; hospitals; patient transfer
Nursing home patients are hospitalized at high rates, at significant cost.1–3 Although intended to accommodate patients’ shifting care needs, transfers between hospitals and nursing
homes are associated with errors in communication4,5; preventable adverse events6,7; and patient, caregiver, and provider dissatisfaction. Potential barriers to communication
University of Pittsburgh Medical Center, Pittsburgh, PA (F.S.); Jewish Home Lifecare, New York, NY(O.B., K.S.B.); J.J. Peters Veterans Affairs Medical Center, Bronx, NY(K.S.B.); Mount Sinai School of Medicine, New York, NY(K.S.B.)
Address correspondence to Kenneth S. Boockvar, MD, MS, James J. Peters VA Medical Center, 130 West Kingsbridge Road, Bronx, NY 10468. E-mail: [email protected]
Financial support was provided by a Mount Sinai School of Medicine Summer Research Fellowship (F.S.), a Pfizer/Foundation for Health in Aging Junior Faculty Scholarship for Research on Health Outcomes in Geriatrics (K.S.B.), and the New York State Department of Health. K.S.B. is currently a Research Career Development Awardee from the VA Health Services Research and Development Service.
Published by Elsevier Inc. on behalf of the American Medical Directors Association
ORIGINAL STUDIES
DOI:10.1016/j.jamda.2009.08.006
Shah, Burack, and Boockvar 239
between hospitals and nursing homes include provider-level factors, information transmission failure, and regulatory restrictions.8 In addition, characteristics of hospitals and nursing homes (eg, management structure, clinical information systems) and the relationships between them (eg, interorganizational affiliations and agreements) may affect communication and coordination9 and efforts to improve intersite transfers.10 Understanding these effects may help stakeholders predict whether interventions designed to improve intersite transfer processes are likely to work and identify potential targets for change. The objective of this study was to identify perceived barriers to communication between hospital and nursing home at the time of patient transfer and examine associations between barriers, hospital and nursing home characteristics, and hospital-nursing home interorganizational relationships. We hypothesized that nursing homes that had relationships with their referral hospitals such as corporate or academic affiliations, or whose patients were cared for by providers who visited patients at both sites would report decreased barriers to intersite communication. METHODS Survey Development We created a model of factors that might affect care at the time of patient transfer using a traditional scheme for modeling quality of health care11 and building on a previously published model of care transfer between organizations.12 A questionnaire based on this model was developed, as has been described previously.13 Briefly, it ascertained (1) hospital characteristics (number of beds, distance from nursing home, teaching status, frequency of inpatient geriatrics consultation or of admission to a dedicated geriatrics care unit); (2) hospital-nursing home relationships (affiliation with the same health system, same corporate owner, trainees from the same academic institution, pharmacy or laboratory agreements); (3) cross-site physician and nonphysician visits; (4) quality of care (hospital-to-nursing home communication, consistency of care with health care goals, and patient and family satisfaction with hospital care); (5) perceived barriers to communication between hospital and nursing home; and (6) communication improvement initiatives. The questionnaire was sent to nursing home administrators and it referred to the hospital to which patients in the nursing facility were transferred most often. Respondents were asked not to name the hospital so as to promote candid answers about hospital care. With regard to perceived barriers to communication between hospital and nursing home, respondents were asked to rate the importance of prespecified factors as a cause of their facility ‘‘failing to receive all the information required to take care of an individual transferred from the hospital.’’ These factors were patient privacy regulations (Health Insurance Portability and Accountability Act [HIPAA]); lost or misplaced information; failed computer link between the hospital and the nursing home; transfers occurring at night or on the weekend; sudden or unplanned transfers; and hospital 240 Shah, Burack, and Boockvar
providers who are not familiar with patient histories, who do not think the transfer process is important, who lack time to dedicate to transfer tasks, who give little effort when completing transfer tasks, who are untrained or unqualified, and who lack reimbursement for transfer tasks. Responses were rated on a Likert scale from 1 (not at all important) to 5 (extremely important). Respondents could also provide open-ended responses. Participants Addresses and telephone numbers of all 647 Medicare- or Medicaid-certified nursing homes in New York State were obtained from a public database14 and the name of each facility’s administrator verified by telephone. From May to June 2005, each administrator was sent an introductory letter, a questionnaire with a stamped return envelope, and a reminder letter, each 2 weeks apart, addressed to him or her by name. Of 647 questionnaires sent, 229 were returned (35.4%). To encourage obtaining information from front-line providers, respondents were instructed to consult with other staff at the facility if needed to answer items on the questionnaire. The large majority of respondents (88.6%) indicated that they did: 67% consulted nursing directors, 32.1% medical directors, 25.2% medical records, 58.4% other staff members, and 14.1% other sources. Whether or not a response was received, the following data for each nursing home were obtained from a public database14: number of beds, type of ownership, chain membership, prevalence of pressure sores, prevalence of physical restraint use, prevalence of bladder catheter use, and nurse staffing. The human subjects committees of the Mount Sinai School of Medicine and The Jewish Home and Hospital Lifecare System approved study procedures. Analysis Responses from hospital-based nursing homes (n 5 31, 13.5%) were excluded from the analysis because many questionnaire items regarding nursing home/hospital interorganizational relationships did not apply to those facilities. For the current study, the main outcome measure was perceived barriers to hospital-nursing home communication. We calculated mean ratings of importance for each perceived barrier, with higher ratings indicating greater perception of the factor as a barrier to communication. We examined associations between perceived barriers and predictor variables (nursing home and hospital characteristics, interorganizational relationships, and quality of care) by calculating differences in mean barrier ratings and by calculating correlation coefficients. A positive correlation coefficient indicated a variable was associated with greater perception of a factor as a barrier to communication, and a negative correlation coefficient indicated a variable was associated with decreased perception of the factor as a barrier. We conducted 2-sided significance tests and considered an association significant if P was less than or equal to .05. All analyses were performed using SPSS Version 15.0 (SPSS, Inc., Chicago, IL). JAMDA – May 2010
Table 1.
Characteristics of Responding and Nonresponding Nursing Homes, Their Primary Referral Hospitals, and Interorganizational Relationships
Nursing Home Characteristics
Characteristic
Respondents (N5198)
Nonrespondents (N5418)
Beds, mean (SD)
201.4 (138.0)
182.6 (127.7)
Proprietary status: For profit, n (%) Not-for-profit, n (%) Government, n (%) Location: Urban, n (%) Rural, n (%) Chain, n (%) Staffing minutes per patient per day: Registered nurse, mean (SD) Licensed practical nurse, mean (SD) Certified nursing assistant, mean (SD) Prevalence of: Pressure ulcers (high-risk), mean % (SD) Pressure ulcers (low-risk), mean % (SD) Restraints, mean % (SD) Bladder catheter, mean % (SD) Hospital Characteristics
Nursing homehospital relationships
Beds, mean (SD)
86 (43.4) 92 (46.5) 18 (9.1)
229 (54.8)* 156 (37.3)* 31 (7.4)*
189 (82.5) 39 (17) 20 (10.1)
369 (88.3) 19 (11.7) 59 (14.1)
36.2 (16.1) 46.2 (44.8) 135.9 (35.8)
35.9 (26.2) 44.9 (23.8) 139.3 (52.2)
14.6 (6.7) 2.9 (3.0) 4.8 (6.4) 4.7 (4.4)
16.0 (7.5)* 3.0 (2.8) 4.4 (2.9) 4.7 (3.6)
255.2 (173.8)
—
Teaching, n (%) Geriatrics specialty care (often or always), n (%)
96.0 (48.5) 23.0 (12.6)
—
Distance (travel time), minutes, mean (SD) Affiliated with same health system, n (%) Same owner, n (%) Trainees from same institution, n (%) Pharmacy agreement, n (%) Laboratory agreement, n (%) Cross-site physician care (often or always), n (%) Other cross-site staff visits (often or always), n (%)
11.0 (7.8) 30 (15.1) 13 (6.6) 30 (15.1) 19 (9.6) 87 (43.9) 75 (37.9) 67 (33.8)
— — — — — — — —
* Indicates P # .05 for comparison with respondents.
RESULTS Organization Characteristics Table 1 shows characteristics of responding and nonresponding nursing homes, characteristics of the nursing homes’ primary referral hospitals, and nursing home-hospital interorganizational relationships. Thirteen nursing homes (6.6%) had the same corporate owner as the hospital. Thirty nursing homes (15.1%) supported trainees from the same academic Table 2.
institution as the hospital, 19 (9.6%) received pharmacy services from the hospital, and 87 (43.9%) received laboratory services from the hospital. Seventy-nine (37.9%) respondents reported cross-site physician care in the nursing home and hospital and 67 (33.8%) reported other cross-site staff visits. Perceived Barriers to Communication Respondents’ perceptions of barriers to communication are shown in Table 2. The most important were lack of hospital
Perceived Barriers to Communication
How important is each of the following factors as a cause of your facility failing to receive all the information required to take care of an individual transferred from the hospital?
Barrier Importance*
Standard Deviation
% Rating Importance as 4 or 5*
Communication between hospital and nursing facility is hindered by patient privacy regulations (HIPAA) Information is lost or misplaced Computer link between hospital and nursing facility fails Transfer occurs at night or on the weekend Transfer is sudden or unplanned Hospital providers who perform transfer tasks. .are not familiar with residents’ medical histories .think transfer process is unimportant .do not have enough time for transfer tasks .do not put enough effort into transfer tasks .are not trained and/or qualified to perform transfer tasks .are not paid enough to perform transfer tasks
2.82
1.46
33.9
2.91 1.90 2.97 3.22
1.34 1.32 1.43 1.38
33.4 13.2 41.4 44.4
3.27 3.14 3.33 3.45 2.91 2.12
1.26 1.28 1.16 1.25 1.29 1.17
45.0 38.9 43.5 51.0 29.8 9.1
* Barrier importance rated from 1 to 5; 1 indicates ‘‘not at all important’’ and 5 indicates ‘‘extremely important.’’ ORIGINAL STUDIES
Shah, Burack, and Boockvar 241
Table 3.
Nursing Home and Hospital Characteristics and Perceived Barriers to Communication Perceived Barrier:
Nursing home characteristics: Nursing home location Nursing home type Nursing home size Cross-site physician care Cross-site staff visits Hospital characteristics: Hospital geriatrics care Hospital size Teaching hospital Hospital/NH system affiliation Hospital as NH pharmacy Hospital as NH laboratory
HIPAA
Low provider effort*
Provider lacks time*
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
Urban Rural Profit Nonprofit Govt Corr. Coeff. Corr. Coeff. Corr. Coeff.
2.81 2.23† 2.96 2.56 2.48 0.005 0.073 0.151†
3.59 2.76‡ 3.57 3.45 2.95 0.089 0.044 0.057
3.38 2.56‡ 3.36 3.26 3.05 0.092 0.047 0.159†
3.56 3.31‡ 3.33 3.12 3.00 0.131 0.113 0.197‡
3.22 2.79 3.23 3.12 3.10 0.045 0.087 0.160†
3.23 2.69† 3.15 3.12 3.14 0.046 0.008 0.206‡
Corr. Coeff. Corr. Coeff. No Yes No Yes No Yes No Yes
0.032 0.018 2.69 2.72 2.73 2.61 2.80 2.28† 3.05 2.38‡
0.064 0.188† 3.28 3.64† 3.45 3.45 3.47 3.39 3.61 3.30
0.015 0.101 3.21 3.36 3.27 3.34 3.30 3.20 3.41 3.16
0.025 0.177† 3.07 3.31 3.14 3.36 3.26 2.83† 3.34 3.04
0.024 0.087 3.02 3.31 3.17 3.07 3.17 3.10 3.21 3.10
0.042 0.057 3.11 3.18 3.10 3.27 3.11 3.29 3.23 3.05
All figures are mean barrier importance ratings, or correlation coefficients (Corr. Coeff.), as indicated. NH, nursing home. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ‡ P \.01.
provider effort, with 101 (51.0%) respondents rating it as important or extremely important, hospital provider’s unfamiliarity with patients (89; 45.0%), lack of hospital provider time (86; 43.5%), sudden or unplanned transfers (88; 44.4%), transfers that take place on nights or weekends (82; 41.4%), and hospital providers’ belief that transfer process is unimportant (77; 38.9%). One respondent remarked in the open-ended portion of the survey that ‘‘difficulty obtaining transportation causes admissions to often arrive on off shifts, creating risk of missed information/communication.’’ Another respondent indicated that ‘‘details like wandering, alcoholism . are omitted purposely because the hospital feels facilities will be less likely to admit these types of residents.’’ Nursing Home and Hospital Characteristics and Perceived Barriers Associations between nursing home and hospital characteristics and perceived barriers are shown in Table 3. Urban nursing homes reported increased importance of barriers to communication. Compared with rural nursing homes, urban nursing homes indicated greater importance of lack of hospital provider effort, hospital provider unfamiliarity with patients, and lack of hospital provider time dedicated to transfer tasks. Nursing homes receiving patients from larger hospitals reported significantly greater perceived barriers to communication for 2 of the 6 presented barriers (low effort by hospital providers, and unfamiliarity of hospital provider with patients). In contrast, when nursing homes used pharmacy or laboratory services from the hospital, perceived hindrance from HIPAA and from provider unfamiliarity with patients was lower. Cross-site visits by nursing home staff was also associated 242 Shah, Burack, and Boockvar
with lower perceived barriers to communication, correlating most strongly with decreased perceptions that hospital providers are unfamiliar with nursing home patients or that hospital providers believe the transfer process is ‘‘unimportant.’’ Quality of Care and Perceived Barriers Table 4 demonstrates the relationships between quality of care and perceived barriers to hospital and nursing home communication. Greater consistency of goals of care between hospital and nursing home was associated with decreased ratings of lack of hospital provider effort and time as barriers. Consistency of hospital care with patient end-of-life preferences also correlated strongly with lower perceived barriers for 5 of the 6 barriers examined. Nursing homes reporting that they consistently received from the hospital all the information needed to care for their residents reported lower ratings for all major potential barriers. Conversely, nursing homes reporting incidents of harm in the preceding 12 months from incomplete hospital-to–nursing home communication rated barriers related to hospital providers much higher than those that did not, especially lack of effort. In contrast, almost all Medicare measures of nursing home quality of care, including nurse staffing, were not associated with perceived barriers to communication from the hospital (Table 4). Communication Improvement Efforts and Perceived Barriers Associations between communication improvement efforts and perceived barriers were variable (Table 5). Nursing homes reporting in-facility meetings and new computerized JAMDA – May 2010
Table 4.
Quality of Care and Perceived Barriers to Communication Perceived Barrier:
Hospital care: Care consistent with goals Care adherent to preferences Hospital patient satisfaction
HIPAA
Corr. Coeff. Corr. Coeff. Corr. Coeff.
Low provider effort*
Provider lacks time*
0.155† 0.143† 0.128
0.149† 0.085 0.063
0.030 0.162† 0.095
0.039 0.168† 0.027
0.130 0.169† 0.075
0.301‡ 0.148† 0.232‡ 3.35 4.07‡
0.268‡ 0.083 0.126 3.20 3.77†
0.282‡ 0.061 0.108 3.10 3.70†
0.229‡ 0.033 0.035 3.13 3.33
0.251‡ 0.077 0.157† 3.04 3.77‡
0.121 0.075 0.100 0.110 0.077 0.040 0.050
0.078 0.108 0.109 0.106 0.038 0.137† 0.052
0.123 0.027 0.066 0.164† 0.000 0.032 0.076
0.101 0.058 0.048 0.056 0.041 0.002 0.026
0.139 0.026 0.086 0.089 0.072 0.015 0.086
0.005 0.19‡ 0.132
Hospital-to-nursing home transfer communication and care: Transfer information complete Corr. Coeff. 0.19‡ Clear purpose for meds Corr. Coeff. 0.002 Legible plan of care Corr. Coeff. 0.046 Incidents of transfer harm No 2.65 Yes 3.10 Nursing home care: Pressure sores (low risk) Corr. Coeff. 0.079 Pressure sores (high risk) Corr. Coeff. 0.061 Use of restraints Corr. Coeff. 0.037 Indwelling catheter use Corr. Coeff. 0.152† RN minutes per patient Corr. Coeff. 0.066 LPN minutes per patient Corr. Coeff. 0.129 CNA minutes per patient Corr. Coeff. 0.082
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
All figures are mean barrier importance ratings, or correlation coefficients (Corr. Coeff.), as indicated. CNA, certified nursing assistant; LPN, licensed practical nurse; RN, registered nurse. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ‡ P \.01.
links with the hospital did not report any significant differences in perceived barriers to communication compared with those that did not. Nursing homes that met with hospital staff in the preceding few months reported HIPAA as a less important barrier but reported night and weekend transfers as a more important one. Implementation of new forms to help with transitional care was associated with a decreased importance of barriers, especially sudden transfers. Implementation of a new telephone call system during patient transfers was associated with an increase in the perceived importance of lack of provider effort and sudden transfers as barriers to communication. Respondents providing openended responses indicated implementation of additional
Table 5.
interventions to improve transitional care such as case managers, cross-site visits, and case review for problematic cases. DISCUSSION Findings from this statewide survey of nursing home administrators suggest that factors perceived to be the most significant barriers to communication between hospital and nursing home may be amenable to improvement efforts, including lack of hospital provider effort and time, sudden or unplanned transfers, and hospital provider’s unfamiliarity with patients. Supporting the importance of the role of these barriers, nursing homes that reported lower communication barriers also reported fewer communication lapses with
Communication Improvement Interventions and Perceived Barriers to Communication
Improvement intervention: Within nursing home meetings Meetings that include hospital staff New transfer forms New telephone protocol New cross-site computer system
Perceived Barrier:
HIPAA
Low provider effort*
Provider lacks time*
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
No Yes No Yes No Yes No Yes No Yes
2.83 2.61 2.93 2.50† 2.83 2.43 2.70 2.78 2.75 2.50
3.35 3.53 3.40 3.50 3.47 3.44 3.36 3.78† 3.50 3.18
3.18 3.37 3.19 3.37 3.33 3.18 3.23 3.48 3.30 3.21
3.19 3.18 3.27 3.11 3.22 3.13 3.13 3.39 3.20 3.18
3.09 3.20 3.02 3.27 3.26 2.85† 3.02 3.55† 3.17 3.00
3.00 3.25 3.13 3.15 3.22 2.97 3.10 3.31 3.19 2.86
All figures are mean barrier importance ratings. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ORIGINAL STUDIES
Shah, Burack, and Boockvar 243
hospitals and fewer incidents of harm related to transfer. When we examined associations between barriers and hospital and nursing home characteristics, increased hospital size, teaching hospitals, and urban nursing home location were associated with greater perceived importance of communication barriers, and cross-site staff visits and hospital provision of laboratory and pharmacy services to the nursing home were associated with lower perceived importance of these barriers. Finally, associations between quality improvement efforts and perceived barriers were mixed, perhaps reflecting different reasons that communication improvement efforts were undertaken. Our results contribute to a previously unstudied perspective on communication during the hospital-to–nursing home transfer process, but are concordant with findings from previous studies. One study using qualitative focus groups identified communication as the most frequently reported barrier to effective transfers, with other significant factors being unexpected transfers, infrequent verbal reports, and illegible paperwork.8 A study of nursing home patient safety culture found that institutions reporting incidents of harm also reported worse patient safety culture in the domains of communication openness, feedback and communication about errors, nonpunitive response to error, organizational learning and continuous improvements, and teamwork.15,16 Finally, another prior study of 3 nursing homes and 4 home health agencies12 found that specialized geriatrics care in the hospital and nursing home affiliation with the referring hospital were associated with better intersite communication. This study also found that communication was best at hospitals with the largest bed occupancy but poorest for those with intermediate bed sizes. Our study’s strengths include (1) a quantitative examination of nursing home administrators’ and other employees’ perceptions of communication during the transfer process, (2) respondents from a representative sample of all New York State nursing homes, and (3) an examination of associations between nursing home perceptions of the transfer process and hospital and nursing home characteristics to identify associations that can help inform communication improvement efforts. This study has several limitations. First, our response rate was lower than some previously published response rates to mail surveys of nursing homes,17–19 and in this study response rates were lower from for-profit nursing homes. Second, our respondents were nursing home administrators who are not directly involved with clinical processes. To overcome this we instructed respondents to discuss the questionnaire items with front-line staff and medical and nursing directors, which almost all respondents did. Third, there may be other factors involved in the transfer process that we did not test for as barriers, such as purposeful omission of patient information by hospital providers in order to reduce the likelihood of rejection of admission by nursing homes, although this was captured in the open-ended response section. Fourth, the questionnaire only uncovered associations and could not prove causality. Our findings have implications that may be useful for communication and transfer improvement efforts. First, hospital 244 Shah, Burack, and Boockvar
provider effort, hospital provider attitudes, and sudden transfers are identified as the greatest perceived barriers to effective communication during patient transfer. That hospital provider factors are given such importance suggests that education and training of providers and improving work flow during patient handoffs to increase efficiency might have as big or greater impact than updating standardized transfer forms, which has been a suggested8 and widely used approach. This may need to include mechanisms to (1) allow flexibility in what technology providers use to communicate— according to provider preferences and available resources, (2) allow providers to prioritize information contained in transfer communications according to the needs of each patient and the circumstances of transfer, and (3) hold facilities and providers accountable when transfer communication breaks down. Given that sudden transfers are associated with greater perception of barriers to communication, nursing homes and hospitals may also institute protocols to reduce sudden, night, or weekend transfers, and to increase communication consistency when unplanned transfers do occur so that they are as effective as planned ones. The findings that increased hospital bed size, teaching hospitals, and urban settings are associated with worse perceptions of barriers to communication may inform strategies for Quality Improvement Organizations and other health system stakeholders (eg, Medicare, Institute for Healthcare Improvement) to target communication improvement efforts. Based on our findings, interventions such as meetings between hospital and nursing home staff and implementing standardized forms may have a positive effect on communication. In addition, increasing frequency of intersite staff visits, as has been tested with some success in Medicare demonstration projects and transition interventions,20,21 may improve communication and patient outcomes. Our study also suggests that provision of laboratory and pharmacy services by hospitals to the nursing home is associated with lower barriers to communication, perhaps because cross-site access to these data are automated. In fact, our study found that failed computer linkage was an important barrier to only a few respondents, likely because there are few nursing homes where computers are used for this purpose. This may change as nursing homes and hospitals become involved in Regional Health Information Organizations, which are organizations designed to build capacity for electronic health information exchange. Of note, affiliation of hospital and nursing home with the same health system was not associated with decreased perceived barriers to communication. This may be because a relationship that exists at a management level such as health system affiliation may not have a strong influence on intersite care or quality improvement efforts, which might be good or poor irrespective of the management-level relationship. CONCLUSION Our findings suggest that lower barriers to communication between hospitals and nursing homes are associated with better patient processes and outcomes, including better intersite communication, higher intersite consistency of patient JAMDA – May 2010
care, and lower incidence of transfer-related harm. Education and training of providers at both sites, more intersite provider visits, and automated data exchange (such as occurs when hospitals provide laboratory and pharmacy services for the nursing home) have the potential to lower barriers to communication. ACKNOWLEDGMENT We gratefully acknowledge Elizabeth Hamilton, DO (St. John Macomb Hospital, Warren, MI), for administering the survey. REFERENCES 1. Castle NG, Mor V. Hospitalization of nursing home residents: A review of the literature, 1980–1995. Med Care Res Rev 1996;53:123–148. 2. Grabowski DC, O’Malley AJ, Barhydt NR. The costs and potential savings associated with nursing home hospitalizations. Health Aff 2007;26: 1753–1761. 3. Boockvar KS, Gruber-Baldini AL, Stuart B, et al. Medicare expenditures for nursing home residents triaged to nursing home or hospital for acute infection. J Am Geriatr Soc 2008;56:1206–1212. 4. Jones JS, Dwyer PR, White LJ, Firman R. Patient transfer from nursing home to emergency department: Outcomes and policy implications. Acad Emerg Med 1997;4:908–915. 5. Lee V, Westley CJ, Fletcher K. If at first you don’t succeed: Efforts to improve collaboration between nursing homes and a health system. Topics in Advanced Nursing eJournal 2004;4. Available at: http://www. medscape.com/viewarticle/487323. 6. Libow LS. Another type of iatrogenic problem. Geriatrics 1978;33:92, 94, 99. 7. Boockvar K, Fishman E, Kyriacou CK, et al. Adverse events due to discontinuations in drug use and dose changes in patients transferred between acute and long-term care facilities. Arch Intern Med 2004;164: 545–550.
ORIGINAL STUDIES
8. Terrell KM, Miller DK. Challenges in transitional care between nursing homes and emergency departments. J Am Med Dir Assoc 2006;7: 499–505. 9. Wagner EH. Chronic disease management: What will it take to improve care for chronic illness? Eff Clin Pract 1998;1:2–4. 10. Coleman EA, Boult C. Improving the quality of transitional care for persons with complex care needs. J Am Geriatr Soc 2003;51:556–557. 11. Donabedian A. Evaluating the quality of medical care. Milbank Mem Fund Q 1966;44:166–206. 12. Anderson MA, Helms LB. Comparison of continuing care communication. Image J Nurs Sch 1998;30:255–260. 13. Boockvar KS, Burack OR. Organizational relationships between nursing homes and hospitals and quality of care during hospital-nursing home patient transfers. J Am Geriatr Soc 2007;55:1078–1084. 14. Medicare. Nursing Home Compare. Available at: www.medicare.gov/ nhcompare. Accessed June 18, 2009. 15. Bonner AF, Castle NG, Men A, Handler SM. Certified nursing assistants’ perceptions of nursing home patient safety culture: Is there a relationship to clinical outcomes? J Am Med Dir Assoc 2009;10:11–20. 16. Handler SM, Castle NG, Studenski SA, et al. Patient safety culture assessment in the nursing home. Qual Saf Health Care 2006;15:400–404. 17. Coburn AF, Fralich JT, McGuire C, Fortinsky R. Variations in outcomes of care in urban and rural nursing facilities in Maine. J Appl Gerontol 1996;15:101–123. 18. Kisor J. Nursing facility admission agreements: An analysis of selected content. J Appl Gerontol 1996;15:294–313. 19. Zinn JS, Weech RJ, Brannon D. Resource dependence and institutional elements in nursing home TQM adoption. Health Serv Res 1998;33: 261–273. 20. Coleman EA, Smith JD, Frank JC, et al. Preparing patients and caregivers to participate in care delivered across settings: The Care Transitions Intervention. J Am Geriatr Soc 2004;52:1817–1825. 21. Naylor MD, Brooten D, Campbell R, et al. Comprehensive discharge planning and home follow-up of hospitalized elders: A randomized clinical trial. JAMA 1999;281:613–620.
Shah, Burack, and Boockvar 245
Perceived Barriers to Communication Between Hospital and Nursing Home at Time of Patient Transfer Faraaz Shah, MD, Orah Burack, MA, and Kenneth S. Boockvar, MD, MS
Objectives: To identify perceived barriers to communication between hospital and nursing home at the time of patient transfer and examine associations between perceived barriers and hospital and nursing home characteristics. Design: Mailed survey. Setting: Medicare- or Medicaid-certified nursing homes in New York State. Participants: Nursing home administrators, with input from other nursing home staff. Measurements: Respondents rated the importance as a barrier to hospital-nursing home communication of (1) hospital providers’ attitude, time, effort, training, payment, and familiarity with nursing home patients; (2) unplanned and off-hours transfers; (3) HIPAA privacy regulations; and (4) lost or failed information transmission. Associations were determined between barriers and the following organizational characteristics: (1) hospital-nursing home affiliations, pharmacy or laboratory agreements, cross-site staff visits, and cross-site physician care; (2) hospital size, teaching status, and frequency of geriatrics specialty care; (3) nursing home size, location, type, staffing, and Medicare quality indicators; and (4) hospitalto-nursing home communication, consistency of
hospital care with health care goals, and communication quality improvement efforts. Results: Of 647 questionnaires sent, 229 were returned (35.4%). The most frequently reported perceived barriers to communication were sudden or unplanned transfers (44.4%), transfers that occur at night or on the weekend (41.4%), and hospital providers’ lack of effort (51.0%), lack of familiarity with patients (45.0%), and lack of time (43.5%). Increased hospital size, teaching hospitals, and urban nursing home location were associated with greater perceived importance of these barriers, and cross-site staff visits and hospital provision of laboratory and pharmacy services to the nursing home were associated with lower perceived importance of these barriers. Conclusions: Hospital and nursing home characteristics and interorganizational relationships were associated with nursing home administrators’ perceptions of barriers to hospital-nursing home communication. These findings may inform design and targeting of interventions to improve intersite communication processes. (J Am Med Dir Assoc 2010; 11: 239–245) Keywords: Communication barriers; nursing homes; hospitals; patient transfer
Nursing home patients are hospitalized at high rates, at significant cost.1–3 Although intended to accommodate patients’ shifting care needs, transfers between hospitals and nursing
homes are associated with errors in communication4,5; preventable adverse events6,7; and patient, caregiver, and provider dissatisfaction. Potential barriers to communication
University of Pittsburgh Medical Center, Pittsburgh, PA (F.S.); Jewish Home Lifecare, New York, NY(O.B., K.S.B.); J.J. Peters Veterans Affairs Medical Center, Bronx, NY(K.S.B.); Mount Sinai School of Medicine, New York, NY(K.S.B.)
Address correspondence to Kenneth S. Boockvar, MD, MS, James J. Peters VA Medical Center, 130 West Kingsbridge Road, Bronx, NY 10468. E-mail: [email protected]
Financial support was provided by a Mount Sinai School of Medicine Summer Research Fellowship (F.S.), a Pfizer/Foundation for Health in Aging Junior Faculty Scholarship for Research on Health Outcomes in Geriatrics (K.S.B.), and the New York State Department of Health. K.S.B. is currently a Research Career Development Awardee from the VA Health Services Research and Development Service.
Published by Elsevier Inc. on behalf of the American Medical Directors Association
ORIGINAL STUDIES
DOI:10.1016/j.jamda.2009.08.006
Shah, Burack, and Boockvar 239
between hospitals and nursing homes include provider-level factors, information transmission failure, and regulatory restrictions.8 In addition, characteristics of hospitals and nursing homes (eg, management structure, clinical information systems) and the relationships between them (eg, interorganizational affiliations and agreements) may affect communication and coordination9 and efforts to improve intersite transfers.10 Understanding these effects may help stakeholders predict whether interventions designed to improve intersite transfer processes are likely to work and identify potential targets for change. The objective of this study was to identify perceived barriers to communication between hospital and nursing home at the time of patient transfer and examine associations between barriers, hospital and nursing home characteristics, and hospital-nursing home interorganizational relationships. We hypothesized that nursing homes that had relationships with their referral hospitals such as corporate or academic affiliations, or whose patients were cared for by providers who visited patients at both sites would report decreased barriers to intersite communication. METHODS Survey Development We created a model of factors that might affect care at the time of patient transfer using a traditional scheme for modeling quality of health care11 and building on a previously published model of care transfer between organizations.12 A questionnaire based on this model was developed, as has been described previously.13 Briefly, it ascertained (1) hospital characteristics (number of beds, distance from nursing home, teaching status, frequency of inpatient geriatrics consultation or of admission to a dedicated geriatrics care unit); (2) hospital-nursing home relationships (affiliation with the same health system, same corporate owner, trainees from the same academic institution, pharmacy or laboratory agreements); (3) cross-site physician and nonphysician visits; (4) quality of care (hospital-to-nursing home communication, consistency of care with health care goals, and patient and family satisfaction with hospital care); (5) perceived barriers to communication between hospital and nursing home; and (6) communication improvement initiatives. The questionnaire was sent to nursing home administrators and it referred to the hospital to which patients in the nursing facility were transferred most often. Respondents were asked not to name the hospital so as to promote candid answers about hospital care. With regard to perceived barriers to communication between hospital and nursing home, respondents were asked to rate the importance of prespecified factors as a cause of their facility ‘‘failing to receive all the information required to take care of an individual transferred from the hospital.’’ These factors were patient privacy regulations (Health Insurance Portability and Accountability Act [HIPAA]); lost or misplaced information; failed computer link between the hospital and the nursing home; transfers occurring at night or on the weekend; sudden or unplanned transfers; and hospital 240 Shah, Burack, and Boockvar
providers who are not familiar with patient histories, who do not think the transfer process is important, who lack time to dedicate to transfer tasks, who give little effort when completing transfer tasks, who are untrained or unqualified, and who lack reimbursement for transfer tasks. Responses were rated on a Likert scale from 1 (not at all important) to 5 (extremely important). Respondents could also provide open-ended responses. Participants Addresses and telephone numbers of all 647 Medicare- or Medicaid-certified nursing homes in New York State were obtained from a public database14 and the name of each facility’s administrator verified by telephone. From May to June 2005, each administrator was sent an introductory letter, a questionnaire with a stamped return envelope, and a reminder letter, each 2 weeks apart, addressed to him or her by name. Of 647 questionnaires sent, 229 were returned (35.4%). To encourage obtaining information from front-line providers, respondents were instructed to consult with other staff at the facility if needed to answer items on the questionnaire. The large majority of respondents (88.6%) indicated that they did: 67% consulted nursing directors, 32.1% medical directors, 25.2% medical records, 58.4% other staff members, and 14.1% other sources. Whether or not a response was received, the following data for each nursing home were obtained from a public database14: number of beds, type of ownership, chain membership, prevalence of pressure sores, prevalence of physical restraint use, prevalence of bladder catheter use, and nurse staffing. The human subjects committees of the Mount Sinai School of Medicine and The Jewish Home and Hospital Lifecare System approved study procedures. Analysis Responses from hospital-based nursing homes (n 5 31, 13.5%) were excluded from the analysis because many questionnaire items regarding nursing home/hospital interorganizational relationships did not apply to those facilities. For the current study, the main outcome measure was perceived barriers to hospital-nursing home communication. We calculated mean ratings of importance for each perceived barrier, with higher ratings indicating greater perception of the factor as a barrier to communication. We examined associations between perceived barriers and predictor variables (nursing home and hospital characteristics, interorganizational relationships, and quality of care) by calculating differences in mean barrier ratings and by calculating correlation coefficients. A positive correlation coefficient indicated a variable was associated with greater perception of a factor as a barrier to communication, and a negative correlation coefficient indicated a variable was associated with decreased perception of the factor as a barrier. We conducted 2-sided significance tests and considered an association significant if P was less than or equal to .05. All analyses were performed using SPSS Version 15.0 (SPSS, Inc., Chicago, IL). JAMDA – May 2010
Table 1.
Characteristics of Responding and Nonresponding Nursing Homes, Their Primary Referral Hospitals, and Interorganizational Relationships
Nursing Home Characteristics
Characteristic
Respondents (N5198)
Nonrespondents (N5418)
Beds, mean (SD)
201.4 (138.0)
182.6 (127.7)
Proprietary status: For profit, n (%) Not-for-profit, n (%) Government, n (%) Location: Urban, n (%) Rural, n (%) Chain, n (%) Staffing minutes per patient per day: Registered nurse, mean (SD) Licensed practical nurse, mean (SD) Certified nursing assistant, mean (SD) Prevalence of: Pressure ulcers (high-risk), mean % (SD) Pressure ulcers (low-risk), mean % (SD) Restraints, mean % (SD) Bladder catheter, mean % (SD) Hospital Characteristics
Nursing homehospital relationships
Beds, mean (SD)
86 (43.4) 92 (46.5) 18 (9.1)
229 (54.8)* 156 (37.3)* 31 (7.4)*
189 (82.5) 39 (17) 20 (10.1)
369 (88.3) 19 (11.7) 59 (14.1)
36.2 (16.1) 46.2 (44.8) 135.9 (35.8)
35.9 (26.2) 44.9 (23.8) 139.3 (52.2)
14.6 (6.7) 2.9 (3.0) 4.8 (6.4) 4.7 (4.4)
16.0 (7.5)* 3.0 (2.8) 4.4 (2.9) 4.7 (3.6)
255.2 (173.8)
—
Teaching, n (%) Geriatrics specialty care (often or always), n (%)
96.0 (48.5) 23.0 (12.6)
—
Distance (travel time), minutes, mean (SD) Affiliated with same health system, n (%) Same owner, n (%) Trainees from same institution, n (%) Pharmacy agreement, n (%) Laboratory agreement, n (%) Cross-site physician care (often or always), n (%) Other cross-site staff visits (often or always), n (%)
11.0 (7.8) 30 (15.1) 13 (6.6) 30 (15.1) 19 (9.6) 87 (43.9) 75 (37.9) 67 (33.8)
— — — — — — — —
* Indicates P # .05 for comparison with respondents.
RESULTS Organization Characteristics Table 1 shows characteristics of responding and nonresponding nursing homes, characteristics of the nursing homes’ primary referral hospitals, and nursing home-hospital interorganizational relationships. Thirteen nursing homes (6.6%) had the same corporate owner as the hospital. Thirty nursing homes (15.1%) supported trainees from the same academic Table 2.
institution as the hospital, 19 (9.6%) received pharmacy services from the hospital, and 87 (43.9%) received laboratory services from the hospital. Seventy-nine (37.9%) respondents reported cross-site physician care in the nursing home and hospital and 67 (33.8%) reported other cross-site staff visits. Perceived Barriers to Communication Respondents’ perceptions of barriers to communication are shown in Table 2. The most important were lack of hospital
Perceived Barriers to Communication
How important is each of the following factors as a cause of your facility failing to receive all the information required to take care of an individual transferred from the hospital?
Barrier Importance*
Standard Deviation
% Rating Importance as 4 or 5*
Communication between hospital and nursing facility is hindered by patient privacy regulations (HIPAA) Information is lost or misplaced Computer link between hospital and nursing facility fails Transfer occurs at night or on the weekend Transfer is sudden or unplanned Hospital providers who perform transfer tasks. .are not familiar with residents’ medical histories .think transfer process is unimportant .do not have enough time for transfer tasks .do not put enough effort into transfer tasks .are not trained and/or qualified to perform transfer tasks .are not paid enough to perform transfer tasks
2.82
1.46
33.9
2.91 1.90 2.97 3.22
1.34 1.32 1.43 1.38
33.4 13.2 41.4 44.4
3.27 3.14 3.33 3.45 2.91 2.12
1.26 1.28 1.16 1.25 1.29 1.17
45.0 38.9 43.5 51.0 29.8 9.1
* Barrier importance rated from 1 to 5; 1 indicates ‘‘not at all important’’ and 5 indicates ‘‘extremely important.’’ ORIGINAL STUDIES
Shah, Burack, and Boockvar 241
Table 3.
Nursing Home and Hospital Characteristics and Perceived Barriers to Communication Perceived Barrier:
Nursing home characteristics: Nursing home location Nursing home type Nursing home size Cross-site physician care Cross-site staff visits Hospital characteristics: Hospital geriatrics care Hospital size Teaching hospital Hospital/NH system affiliation Hospital as NH pharmacy Hospital as NH laboratory
HIPAA
Low provider effort*
Provider lacks time*
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
Urban Rural Profit Nonprofit Govt Corr. Coeff. Corr. Coeff. Corr. Coeff.
2.81 2.23† 2.96 2.56 2.48 0.005 0.073 0.151†
3.59 2.76‡ 3.57 3.45 2.95 0.089 0.044 0.057
3.38 2.56‡ 3.36 3.26 3.05 0.092 0.047 0.159†
3.56 3.31‡ 3.33 3.12 3.00 0.131 0.113 0.197‡
3.22 2.79 3.23 3.12 3.10 0.045 0.087 0.160†
3.23 2.69† 3.15 3.12 3.14 0.046 0.008 0.206‡
Corr. Coeff. Corr. Coeff. No Yes No Yes No Yes No Yes
0.032 0.018 2.69 2.72 2.73 2.61 2.80 2.28† 3.05 2.38‡
0.064 0.188† 3.28 3.64† 3.45 3.45 3.47 3.39 3.61 3.30
0.015 0.101 3.21 3.36 3.27 3.34 3.30 3.20 3.41 3.16
0.025 0.177† 3.07 3.31 3.14 3.36 3.26 2.83† 3.34 3.04
0.024 0.087 3.02 3.31 3.17 3.07 3.17 3.10 3.21 3.10
0.042 0.057 3.11 3.18 3.10 3.27 3.11 3.29 3.23 3.05
All figures are mean barrier importance ratings, or correlation coefficients (Corr. Coeff.), as indicated. NH, nursing home. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ‡ P \.01.
provider effort, with 101 (51.0%) respondents rating it as important or extremely important, hospital provider’s unfamiliarity with patients (89; 45.0%), lack of hospital provider time (86; 43.5%), sudden or unplanned transfers (88; 44.4%), transfers that take place on nights or weekends (82; 41.4%), and hospital providers’ belief that transfer process is unimportant (77; 38.9%). One respondent remarked in the open-ended portion of the survey that ‘‘difficulty obtaining transportation causes admissions to often arrive on off shifts, creating risk of missed information/communication.’’ Another respondent indicated that ‘‘details like wandering, alcoholism . are omitted purposely because the hospital feels facilities will be less likely to admit these types of residents.’’ Nursing Home and Hospital Characteristics and Perceived Barriers Associations between nursing home and hospital characteristics and perceived barriers are shown in Table 3. Urban nursing homes reported increased importance of barriers to communication. Compared with rural nursing homes, urban nursing homes indicated greater importance of lack of hospital provider effort, hospital provider unfamiliarity with patients, and lack of hospital provider time dedicated to transfer tasks. Nursing homes receiving patients from larger hospitals reported significantly greater perceived barriers to communication for 2 of the 6 presented barriers (low effort by hospital providers, and unfamiliarity of hospital provider with patients). In contrast, when nursing homes used pharmacy or laboratory services from the hospital, perceived hindrance from HIPAA and from provider unfamiliarity with patients was lower. Cross-site visits by nursing home staff was also associated 242 Shah, Burack, and Boockvar
with lower perceived barriers to communication, correlating most strongly with decreased perceptions that hospital providers are unfamiliar with nursing home patients or that hospital providers believe the transfer process is ‘‘unimportant.’’ Quality of Care and Perceived Barriers Table 4 demonstrates the relationships between quality of care and perceived barriers to hospital and nursing home communication. Greater consistency of goals of care between hospital and nursing home was associated with decreased ratings of lack of hospital provider effort and time as barriers. Consistency of hospital care with patient end-of-life preferences also correlated strongly with lower perceived barriers for 5 of the 6 barriers examined. Nursing homes reporting that they consistently received from the hospital all the information needed to care for their residents reported lower ratings for all major potential barriers. Conversely, nursing homes reporting incidents of harm in the preceding 12 months from incomplete hospital-to–nursing home communication rated barriers related to hospital providers much higher than those that did not, especially lack of effort. In contrast, almost all Medicare measures of nursing home quality of care, including nurse staffing, were not associated with perceived barriers to communication from the hospital (Table 4). Communication Improvement Efforts and Perceived Barriers Associations between communication improvement efforts and perceived barriers were variable (Table 5). Nursing homes reporting in-facility meetings and new computerized JAMDA – May 2010
Table 4.
Quality of Care and Perceived Barriers to Communication Perceived Barrier:
Hospital care: Care consistent with goals Care adherent to preferences Hospital patient satisfaction
HIPAA
Corr. Coeff. Corr. Coeff. Corr. Coeff.
Low provider effort*
Provider lacks time*
0.155† 0.143† 0.128
0.149† 0.085 0.063
0.030 0.162† 0.095
0.039 0.168† 0.027
0.130 0.169† 0.075
0.301‡ 0.148† 0.232‡ 3.35 4.07‡
0.268‡ 0.083 0.126 3.20 3.77†
0.282‡ 0.061 0.108 3.10 3.70†
0.229‡ 0.033 0.035 3.13 3.33
0.251‡ 0.077 0.157† 3.04 3.77‡
0.121 0.075 0.100 0.110 0.077 0.040 0.050
0.078 0.108 0.109 0.106 0.038 0.137† 0.052
0.123 0.027 0.066 0.164† 0.000 0.032 0.076
0.101 0.058 0.048 0.056 0.041 0.002 0.026
0.139 0.026 0.086 0.089 0.072 0.015 0.086
0.005 0.19‡ 0.132
Hospital-to-nursing home transfer communication and care: Transfer information complete Corr. Coeff. 0.19‡ Clear purpose for meds Corr. Coeff. 0.002 Legible plan of care Corr. Coeff. 0.046 Incidents of transfer harm No 2.65 Yes 3.10 Nursing home care: Pressure sores (low risk) Corr. Coeff. 0.079 Pressure sores (high risk) Corr. Coeff. 0.061 Use of restraints Corr. Coeff. 0.037 Indwelling catheter use Corr. Coeff. 0.152† RN minutes per patient Corr. Coeff. 0.066 LPN minutes per patient Corr. Coeff. 0.129 CNA minutes per patient Corr. Coeff. 0.082
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
All figures are mean barrier importance ratings, or correlation coefficients (Corr. Coeff.), as indicated. CNA, certified nursing assistant; LPN, licensed practical nurse; RN, registered nurse. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ‡ P \.01.
links with the hospital did not report any significant differences in perceived barriers to communication compared with those that did not. Nursing homes that met with hospital staff in the preceding few months reported HIPAA as a less important barrier but reported night and weekend transfers as a more important one. Implementation of new forms to help with transitional care was associated with a decreased importance of barriers, especially sudden transfers. Implementation of a new telephone call system during patient transfers was associated with an increase in the perceived importance of lack of provider effort and sudden transfers as barriers to communication. Respondents providing openended responses indicated implementation of additional
Table 5.
interventions to improve transitional care such as case managers, cross-site visits, and case review for problematic cases. DISCUSSION Findings from this statewide survey of nursing home administrators suggest that factors perceived to be the most significant barriers to communication between hospital and nursing home may be amenable to improvement efforts, including lack of hospital provider effort and time, sudden or unplanned transfers, and hospital provider’s unfamiliarity with patients. Supporting the importance of the role of these barriers, nursing homes that reported lower communication barriers also reported fewer communication lapses with
Communication Improvement Interventions and Perceived Barriers to Communication
Improvement intervention: Within nursing home meetings Meetings that include hospital staff New transfer forms New telephone protocol New cross-site computer system
Perceived Barrier:
HIPAA
Low provider effort*
Provider lacks time*
Provider unfamiliar w/patient*
Sudden transfer
Process ‘‘unimportant’’*
No Yes No Yes No Yes No Yes No Yes
2.83 2.61 2.93 2.50† 2.83 2.43 2.70 2.78 2.75 2.50
3.35 3.53 3.40 3.50 3.47 3.44 3.36 3.78† 3.50 3.18
3.18 3.37 3.19 3.37 3.33 3.18 3.23 3.48 3.30 3.21
3.19 3.18 3.27 3.11 3.22 3.13 3.13 3.39 3.20 3.18
3.09 3.20 3.02 3.27 3.26 2.85† 3.02 3.55† 3.17 3.00
3.00 3.25 3.13 3.15 3.22 2.97 3.10 3.31 3.19 2.86
All figures are mean barrier importance ratings. * Refers to beliefs or characteristics of hospital providers who perform transfer tasks. † P \.05. ORIGINAL STUDIES
Shah, Burack, and Boockvar 243
hospitals and fewer incidents of harm related to transfer. When we examined associations between barriers and hospital and nursing home characteristics, increased hospital size, teaching hospitals, and urban nursing home location were associated with greater perceived importance of communication barriers, and cross-site staff visits and hospital provision of laboratory and pharmacy services to the nursing home were associated with lower perceived importance of these barriers. Finally, associations between quality improvement efforts and perceived barriers were mixed, perhaps reflecting different reasons that communication improvement efforts were undertaken. Our results contribute to a previously unstudied perspective on communication during the hospital-to–nursing home transfer process, but are concordant with findings from previous studies. One study using qualitative focus groups identified communication as the most frequently reported barrier to effective transfers, with other significant factors being unexpected transfers, infrequent verbal reports, and illegible paperwork.8 A study of nursing home patient safety culture found that institutions reporting incidents of harm also reported worse patient safety culture in the domains of communication openness, feedback and communication about errors, nonpunitive response to error, organizational learning and continuous improvements, and teamwork.15,16 Finally, another prior study of 3 nursing homes and 4 home health agencies12 found that specialized geriatrics care in the hospital and nursing home affiliation with the referring hospital were associated with better intersite communication. This study also found that communication was best at hospitals with the largest bed occupancy but poorest for those with intermediate bed sizes. Our study’s strengths include (1) a quantitative examination of nursing home administrators’ and other employees’ perceptions of communication during the transfer process, (2) respondents from a representative sample of all New York State nursing homes, and (3) an examination of associations between nursing home perceptions of the transfer process and hospital and nursing home characteristics to identify associations that can help inform communication improvement efforts. This study has several limitations. First, our response rate was lower than some previously published response rates to mail surveys of nursing homes,17–19 and in this study response rates were lower from for-profit nursing homes. Second, our respondents were nursing home administrators who are not directly involved with clinical processes. To overcome this we instructed respondents to discuss the questionnaire items with front-line staff and medical and nursing directors, which almost all respondents did. Third, there may be other factors involved in the transfer process that we did not test for as barriers, such as purposeful omission of patient information by hospital providers in order to reduce the likelihood of rejection of admission by nursing homes, although this was captured in the open-ended response section. Fourth, the questionnaire only uncovered associations and could not prove causality. Our findings have implications that may be useful for communication and transfer improvement efforts. First, hospital 244 Shah, Burack, and Boockvar
provider effort, hospital provider attitudes, and sudden transfers are identified as the greatest perceived barriers to effective communication during patient transfer. That hospital provider factors are given such importance suggests that education and training of providers and improving work flow during patient handoffs to increase efficiency might have as big or greater impact than updating standardized transfer forms, which has been a suggested8 and widely used approach. This may need to include mechanisms to (1) allow flexibility in what technology providers use to communicate— according to provider preferences and available resources, (2) allow providers to prioritize information contained in transfer communications according to the needs of each patient and the circumstances of transfer, and (3) hold facilities and providers accountable when transfer communication breaks down. Given that sudden transfers are associated with greater perception of barriers to communication, nursing homes and hospitals may also institute protocols to reduce sudden, night, or weekend transfers, and to increase communication consistency when unplanned transfers do occur so that they are as effective as planned ones. The findings that increased hospital bed size, teaching hospitals, and urban settings are associated with worse perceptions of barriers to communication may inform strategies for Quality Improvement Organizations and other health system stakeholders (eg, Medicare, Institute for Healthcare Improvement) to target communication improvement efforts. Based on our findings, interventions such as meetings between hospital and nursing home staff and implementing standardized forms may have a positive effect on communication. In addition, increasing frequency of intersite staff visits, as has been tested with some success in Medicare demonstration projects and transition interventions,20,21 may improve communication and patient outcomes. Our study also suggests that provision of laboratory and pharmacy services by hospitals to the nursing home is associated with lower barriers to communication, perhaps because cross-site access to these data are automated. In fact, our study found that failed computer linkage was an important barrier to only a few respondents, likely because there are few nursing homes where computers are used for this purpose. This may change as nursing homes and hospitals become involved in Regional Health Information Organizations, which are organizations designed to build capacity for electronic health information exchange. Of note, affiliation of hospital and nursing home with the same health system was not associated with decreased perceived barriers to communication. This may be because a relationship that exists at a management level such as health system affiliation may not have a strong influence on intersite care or quality improvement efforts, which might be good or poor irrespective of the management-level relationship. CONCLUSION Our findings suggest that lower barriers to communication between hospitals and nursing homes are associated with better patient processes and outcomes, including better intersite communication, higher intersite consistency of patient JAMDA – May 2010
care, and lower incidence of transfer-related harm. Education and training of providers at both sites, more intersite provider visits, and automated data exchange (such as occurs when hospitals provide laboratory and pharmacy services for the nursing home) have the potential to lower barriers to communication. ACKNOWLEDGMENT We gratefully acknowledge Elizabeth Hamilton, DO (St. John Macomb Hospital, Warren, MI), for administering the survey. REFERENCES 1. Castle NG, Mor V. Hospitalization of nursing home residents: A review of the literature, 1980–1995. Med Care Res Rev 1996;53:123–148. 2. Grabowski DC, O’Malley AJ, Barhydt NR. The costs and potential savings associated with nursing home hospitalizations. Health Aff 2007;26: 1753–1761. 3. Boockvar KS, Gruber-Baldini AL, Stuart B, et al. Medicare expenditures for nursing home residents triaged to nursing home or hospital for acute infection. J Am Geriatr Soc 2008;56:1206–1212. 4. Jones JS, Dwyer PR, White LJ, Firman R. Patient transfer from nursing home to emergency department: Outcomes and policy implications. Acad Emerg Med 1997;4:908–915. 5. Lee V, Westley CJ, Fletcher K. If at first you don’t succeed: Efforts to improve collaboration between nursing homes and a health system. Topics in Advanced Nursing eJournal 2004;4. Available at: http://www. medscape.com/viewarticle/487323. 6. Libow LS. Another type of iatrogenic problem. Geriatrics 1978;33:92, 94, 99. 7. Boockvar K, Fishman E, Kyriacou CK, et al. Adverse events due to discontinuations in drug use and dose changes in patients transferred between acute and long-term care facilities. Arch Intern Med 2004;164: 545–550.
ORIGINAL STUDIES
8. Terrell KM, Miller DK. Challenges in transitional care between nursing homes and emergency departments. J Am Med Dir Assoc 2006;7: 499–505. 9. Wagner EH. Chronic disease management: What will it take to improve care for chronic illness? Eff Clin Pract 1998;1:2–4. 10. Coleman EA, Boult C. Improving the quality of transitional care for persons with complex care needs. J Am Geriatr Soc 2003;51:556–557. 11. Donabedian A. Evaluating the quality of medical care. Milbank Mem Fund Q 1966;44:166–206. 12. Anderson MA, Helms LB. Comparison of continuing care communication. Image J Nurs Sch 1998;30:255–260. 13. Boockvar KS, Burack OR. Organizational relationships between nursing homes and hospitals and quality of care during hospital-nursing home patient transfers. J Am Geriatr Soc 2007;55:1078–1084. 14. Medicare. Nursing Home Compare. Available at: www.medicare.gov/ nhcompare. Accessed June 18, 2009. 15. Bonner AF, Castle NG, Men A, Handler SM. Certified nursing assistants’ perceptions of nursing home patient safety culture: Is there a relationship to clinical outcomes? J Am Med Dir Assoc 2009;10:11–20. 16. Handler SM, Castle NG, Studenski SA, et al. Patient safety culture assessment in the nursing home. Qual Saf Health Care 2006;15:400–404. 17. Coburn AF, Fralich JT, McGuire C, Fortinsky R. Variations in outcomes of care in urban and rural nursing facilities in Maine. J Appl Gerontol 1996;15:101–123. 18. Kisor J. Nursing facility admission agreements: An analysis of selected content. J Appl Gerontol 1996;15:294–313. 19. Zinn JS, Weech RJ, Brannon D. Resource dependence and institutional elements in nursing home TQM adoption. Health Serv Res 1998;33: 261–273. 20. Coleman EA, Smith JD, Frank JC, et al. Preparing patients and caregivers to participate in care delivered across settings: The Care Transitions Intervention. J Am Geriatr Soc 2004;52:1817–1825. 21. Naylor MD, Brooten D, Campbell R, et al. Comprehensive discharge planning and home follow-up of hospitalized elders: A randomized clinical trial. JAMA 1999;281:613–620.
Shah, Burack, and Boockvar 245