- Email: [email protected]
Standardizing Hospital Discharge Planning at the Mayo Clinic
The Joint Commission Journal on Quality and Patient Safety Continuity of Care
Standardizing Hospital Discharge Planning at the Mayo Clinic Diane E. Holland, Ph.D., R.N.; Michele A. Hemann, M.S.N., R.N.
I
mproving the quality of patient transitions across health care settings is a top-priority health care concern. Inadequate transitions are known to lead to suboptimal outcomes, the development of new or worsening symptoms, unplanned rehospitalizations, and medical errors and other adverse events, especially in older adults.1–5 Hospital discharge planning (DP) is the primary vehicle for managing care coordination from the hospital. The Centers for Medicare & Medicaid Services (CMS) Hospital Conditions of Participation in the Medicare Program require that hospitals have a DP process in effect that applies to all patients.6 Furthermore, hospitals must identify early in the hospital stay those patients who without further DP evaluation would suffer adverse consequences.7 The CMS condition also indicates that the additional DP evaluation must include an assessment of the patient’s capacity for independence, the possibility of the patient being cared for in the environment from which he or she entered the hospital, the likelihood of a patient needing posthospital services, and the availability of the services.6 Components linked to a successful or an effective DP process include early identification of at-risk patients, a systematic approach to the process, multidisciplinary teamwork, and well-working communication systems.8–12 CMS does not mandate any one specific DP structure, which allows for flexibility in the use of available resources to provide DP services, although this also results in a wide variation in DP service structures across institutions. Despite the large body of literature on DP, the impact on indicators such as readmission rates, hospital length of stay, and costs, as found in systematic reviews of DP, remains uncertain.13–15 Recent emphasis on patient-centered care and better coordination of care has led to a search for patient-oriented indicators, in recognition of the fact that patients are central to the DP process and, as such, are key stakeholders.16 In this article, we report how we evaluated a change in DP practice from the perspective of recently discharged patients. We were further interested in possible differences in patient per-
January 2011
Article-at-a-Glance Background: Improving the quality of patient coordina-
tion in the transition from hospital to home is a highpriority health care concern. The Centers for Medicare & Medicaid Services (CMS) Hospital Conditions of Participation in the Medicare Program require that hospitals have a discharge planning (DP) process in effect that applies to all patients. The impact of a practice change in DP practice on the quality of care coordination at discharge was evaluated from patients’ perspectives. Methods: A multifactor, evidence-based DP practice change, which included merging of DP specialist roles and use of an early screen for DP decision support tool, was initiated in a large, Midwestern academic medical center and evaluated in a nonequivalent comparison group design with separate pre- and postpractice change samples. The threeitem Care Transitions MeasureTM (CTM-3TM) was mailed to adults recently discharged from one medical and one surgical nursing unit before and after the practice change. Results: Response rates were 52.4% before (218/416) and 39.5% (153/387) after the practice change. There were no significant differences between characteristics of the preand postpractice change participants. The mean CTM-3 score of patients who received assistance from the nurse/ social worker DP team improved by 14 points (67.2 to 81.2), although the data were skewed with a ceiling effect, rendering the results inconclusive. Conclusions: Although the CTM-3 results were inconclusive, the practice change resulted in a clinically meaningful decrease in length of stay for a group of older patients at greater risk for complex discharge plans. The proactive approach to DP proved to be a valuable shift. The successes of the standardization of DP processes and improved multidisciplinary teamwork were important considerations for implementation throughout the organization.
Volume 37 Number 1
Copyright 2011 © The Joint Commission
29
The Joint Commission Journal on Quality and Patient Safety Table 1. Discharge Planning (DP) Practice Before and After Changes Before Practice Change ■ 3 roles related to discharge planning − Infusion therapy coordinator (ITC) − DP nurse − Social worker (SW)
After Practice Change ■ ITC and DP nurse combined into same role (DP nurse specialist [DPNS])
■ DP nurse, ITC, and SW in separate departments − Role overlap − Duplicated efforts − Lack of teamwork
■ DP nurse specialists and SW merged into one department − Role clarification − Separation of DP duties − DPNS/SW team approach
■ No standardized processes for DP assessment or documentation
■ Standardized DP process including the following: − Early identification of patients who need further DP evaluation using decision support tool − Electronic consult to DPNS/SW team − Daily multidisciplinary rounds on all units − Comprehensive assessment tool and documentation format
ceptions between patients who received assistance with DP from a nurse/social worker team (after the practice change) and patients who received services from a DP nurse or social worker (before DP team development), as part of the practice change.
Methods INSTITUTING A DISCHARGE PLANNING PRACTICE CHANGE In 2008, members of the nursing department, in collaboration with the medical social services division, proposed a multicomponent change in DP practice at our institution. Aligning DP Support. DP support was provided to direct care providers (the staff nurses and physicians) by persons in three different specialized positions, each based in a different division: ■ DP nurses, who were assigned to units and available by consult whenever the staff nurse or physician perceived the patient’s discharge plan was more complex than what they felt capable of handling themselves ■ Nurse infusion therapy coordinators (ITCs), who provided specialized DP services for patients with postdischarge intravenous (IV) therapy ■ Social workers, who were assigned to medical services and were also available by consult to assist with complex discharge plans It was not uncommon for all three DP specialists, each requested by a different direct care staff member, to consult on the same patient, resulting in duplicated effort and poor communication. Teamwork between the disciplines assigned to DP roles was further hampered by different administrative reporting struc30
January 2011
tures for the roles. Supervision of all three DP roles was merged in early 2008. This led to discussions regarding how to improve our practice to ensure that we were providing the best service to our key stakeholders. A multidisciplinary DP team of key stakeholders, including nurses, social workers, pharmacists, dietitians, and physical therapists planned the DP practice changes. DP nurses and leadership from two nursing units (one general medical and one general surgical unit) provided development and oversight expertise. The multifactor change in DP practice sought to move from a reactive to a more proactive standardized approach. As shown in Table 1 (above), the ITCs and DP nurses were combined into one role—renamed discharge planning nurse specialists. The nurses and social workers merged into one division with the same administrator. Further team development ensued, as well as role clarification for the direct care registered nurse (RN), DP nurse, and social worker to minimize duplication of effort. DP Decision Support Tool. In 2009, a DP team oversaw the development of the evidence-based Early Screen for Discharge Planning (ESDP) DP decision support tool which was embedded into the RN admission assessment in the electronic medical record. The ESDP tool, which was based on patient admission assessment data, had been developed and tested within the institution’s hospitals.17 The ESDP, designed to assist in the early engagement of DP resources for those patients with potentially complex DP needs, consists of four patient characteristics readily available on admission—age, self-reported walking limitation, prior living status, and overall level of disability. A cut-point score of 10 or more points indicates that a referral to the DP team is warranted. For example, patients with
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety the following characteristics would meet or exceed the threshold score: ■ A patient 80 years or older (8 points) with a walking limitation (3 points) or who lives alone (3 points) ■ A 70-year-old patient (6 points) who needs help daily with dressing or hygiene (9 points) ■ A 50-year-old (4 points) with a walking limitation (3 points) and who lives alone (3 points) After the ESDP was completed, a score was electronically computed. When the score met or exceeded the predesignated threshold, a consult was issued by the staff RN and sent electronically to the DP work list. This provided greater ease and timeliness for making consults to the DP team. DP Rounds. Multidisciplinary DP rounds were instituted daily on the two nursing units to facilitate communication and timely planning for continuing care needs. DP Assessment Format. An evidence-based, standardized comprehensive DP assessment format18 was introduced in 2009 for use with high-risk patients, such as those who would likely suffer adverse consequences without a further evaluation of their continuing care needs. Content domains consisted of cognitive/behavioral/emotional status, health status, functional status, finances, environmental factors in postdischarge care, anticipated skilled care requirements, and resources to meet continuing care needs. The DP nurse completes the assessment by interviewing the patient and caregiver(s). The caregiver has the opportunity to discuss whether he or she is able to offer support for continuing care needs or prefers that formal community service providers be engaged. The DP team members use the comprehensive assessment information to communicate the patient’s continuing care needs to hospital staff and community providers (home care agencies, nursing homes, outpatient centers, infusion companies, and equipment companies). The information is shared with continuing care providers via telephone and transferred into the hospital dismissal summary document before patient discharge. A copy of the dismissal summary is available at patient discharge and shared with continuing care providers. After they are engaged in DP for a patient, members of the DP team remain actively involved in the initial implementation of the discharge plan until the patient is discharged. Staff nurses continue to be responsible for formulating and implementing discharge plans of patients with relatively straightforward discharge plans. Consistent with the Medicare Conditions of Participation,19 the DP team can also be consulted at any point throughout the hospital stay.
January 2011
EVALUATING CHANGE
THE
DISCHARGE PLANNING PRACTICE
Design. The evaluation study, approved by the Institutional Review Board, was designed by the DP team [including D.E.H., M.A.H.] as a nonequivalent comparison group study with separate pre- and postintervention samples. The practice change was implemented during a period of four weeks (July 7–August 1, 2008). The comparison group sample consisted of patients hospitalized on two nursing units—general medical and general surgical—six weeks before the practice change. Patients in the intervention sample of patients were hospitalized on the same two units in the six-week period after the practice change implementation period was completed. (These units were chosen because the practice change was intended to affect all hospitalized adults.) Patients on these units represented adults admitted to general acute care and did not represent any one specific clinical specialty practice. Outcome Measure. We used the Care Transitions Measure (CTM-3; Figure 1, page 32),20,21 which consists of three questions derived from qualitative studies of recently hospitalized patients. The National Quality Forum (NQF) has endorsed it as standard measure for assessing patients’ perspectives of the quality of care coordination from the hospital to the home setting.22 The tool has been tested in both mailed and interview formats, with patients experiencing either medical or surgical problems,20 and with ethnically diverse populations.23 Results are scored by computing a mean score for each responder. Scores are transformed to a 0–100 scale for comparison and public reporting. Higher scores indicate higher perceived quality of care coordination at discharge.22 Sample. All adult patients hospitalized on the two nursing units who met inclusion criteria regardless of payor source were surveyed. Although the NQF guidelines recommend excluding proxy responses,22 we mailed the surveys, which neccessitated adding a question as to who responded to the survey (patient or other person). Patients younger than 18 years of age and patients on observation status (did not meet hospital admission criteria) were also excluded. Responders were further separated into two naturally occurring groups—those whose discharge plans were formulated and implemented by a staff nurse and those who received assistance with their discharge plans from the nurse/social worker DP team (Table 1). A sample size of 120 patients pre- and postpractice change provided sufficient power to detect a clinically meaningful (10point) difference in CTM-3 scores. On the basis of a reported standard deviation (SD) for the CTM-3 of 13.67,22 a minimum of 60 patients in each group provided 80% power to detect a
Volume 37 Number 1
Copyright 2011 © The Joint Commission
31
The Joint Commission Journal on Quality and Patient Safety The Three-Item Care Transitions Measure 1. The hospital staff took my preferences and those of my family or caregiver into account in deciding what my health care needs would be when I left the hospital. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
2. When I left the hospital, I had a good understanding of the things I was responsible for in managing my health. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
3. When I left the hospital, I clearly understood the purpose for taking each of my medications. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
Figure 1. The three-item Care Transitions MeasureTM (CTM-3TM) was developed by Eric A. Coleman, M.D., M.P.H., of the University of Colorado Denver Care Transitions Program and is available in the public domain from http://www.caretransitions.org (last accessed Nov. 23, 2010). The CTM-3 is endorsed by the National Quality Forum for hospital and nursing home performance measurement.
minimal clinically meaningful difference in mean CTM-3 scores of 10, alpha set at .05. Survey Administration. Consistent with the NQF guidelines, which indicate that survey administration can occur between 48 hours and six weeks after discharge and can be mailed or administered by telephone,22 the CTM-3 was mailed at approximately four–six days after patient discharge to the primary address listed in the medical record. This time period was chosen on the basis of DP literature indicating that the first few weeks after discharge are critical to patient recovery.11,24,25 In 2008, the CTM-3 was mailed by members of the study team to 416 eligible patients discharged from the two nursing units during six weeks before the practice change and to 387 patients discharged during the six weeks after the practice change was implemented. The patients were asked to complete and return the survey (in a stamped envelope provided) the same day they received it, if possible. Sociodemographic and health characteristics of the patients, documented assistance by the DP team, and referrals to formal community services were obtained by record review. Characteristics of nonresponders who had previously granted the general authorization for the use of their medical records for research purposes under the state law were also recorded.
Results RESPONSE RATES The response rates (including proxy responses) were 52.4% before (218/416) and 39.5% (153/387) after the practice change. Because almost one-third (29.4%; 10/34) of the questionnaires returned by patients who received assistance from DP nurses or social workers were completed by proxy responders, the decision was made to include proxy responses in the data analysis.
RESPONDERS AND NONRESPONDERS There were some differences between responders and nonresponders (Table 2, page 33). Before the practice change, more respondents were female and hospitalized on the surgical unit with longer lengths of stay than nonresponders. After the practice change, responders were older and, again, more likely to be hospitalized on the surgical unit than nonresponders. More nonresponders were discharged with formal services such as home health care or to a nursing facility than were responders. Responders were generally in their late 50s. There were no statistically significant differences in characteristics between the pre- and postpractice change responders. Few responders went home with a referral to formal community services.
CTM-3 SCORES DATA ANALYSIS Descriptive statistics were used to describe sample characteristics. The Mann-Whitney U test was used to measure differences between the pre- and postpractice change CTM-3 scores. Nonparametric tests were used because of the skewed nature of the data. 32
January 2011
The overall mean CTM-3 scores were relatively the same in the pre- (82.9 ±24.3) and postpractice change groups (80.1 ±23.0; Table 3, page 33). The distribution of both pre- and postpractice change scores was markedly skewed. The overall median scores remained the same (88.9), as did the interquartile ranges (IQR; 66.7–100.0). Of note, the 75th quartile score
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety Table 2. Sample Characteristics of Before and After Practice Change Responders and Nonresponders*
Age† Length of stay‡
Sex‡ Female Male Primary insurance Medicare Commercial Other Discharge disposition‡ Self care Home health care Nursing facility Nursing unit†‡ Medical Surgical
Before Practice Change Responders Nonresponders N = 218 N = 192
After Practice Change Responders Nonresponders N = 153 N = 222
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
Mean (+ SD)
58.5 (⫾14.9) 3.2 (⫾3.2)
58.5 (⫾18.6) 4.5 (⫾4.8)
59.9 (⫾17.1) 3.4 (⫾2.8)
54.9 (⫾18.2) 4.3 (⫾4.5)
N (%)
N (%)
N (%)
N (%)
218 133 (61.0) 85 (39.0) 218 87 (39.9) 104 (47.7) 27 (12.4) 218 207 (95.0) 2 (0.9) 9 (4.1) 218 68 (31.2) 150 (68.8)
192 114 (59.4) 78 (40.6) 192 80 (41.7) 87 (45.3) 25 (13.0) 188 159 (84.0) 7 (3.7) 22 (11.7) 192 103 (53.7) 89 (46.4)
153 99 (64.7) 54 (35.3) 153 63 (41.2) 82 (53.6) 8 (5.2) 152 143 (94.0) 3 (2.0) 6 (3.9) 153 43 (28.1) 110 (71.9)
222 138 (62.2) 84 (37.8) 218 75 (34.4) 122 (56.0) 21 (9.6) 220 193 (87.7) 10 (4.6) 17 (7.7) 219 106 (48.4) 113 (51.6)
* SD, standard deviation. † After practice change responders and nonresponders differed significantly (p < .05). ‡ Before practice change participants and nonresponders differed significantly (p < .05).
Table 3. CTM-3 Scores* Before Practice Change† Overall
N 218
M (⫾SD) 82.9 (⫾24.3)
Median (IQR) 88.9 (66.7–100.0)
After Practice Change N 153
M (⫾SD) 80.1 (⫾23.0)
Median (IQR) 88.9 (66.7–100.0)
DP staff involvement‡
21
67.2 (⫾31.1)
66.7 (66.7–83.3)
34
81.2 (⫾18.1)
66.7 (66.7–100.0)
No DP staff involvement
197
84.8 (⫾22.6)
100.0 (66.7–100.0)
120
79.9 (⫾24.4)
88.9 (66.7–100.0)
* SD, standard deviation; IQR, interquartile range; DP, discharge planning. † Before practice change mean CTM-3 scores were statistically significantly different between the “DP staff involvement” and “No DP staff involvement” groups
(p < .05). ‡ After practice change reflects DP nurse/social worker team
in both groups was 100, the maximum CTM-3 score. A large ceiling effect was observed overall in both the pre- (49.1%) and postpractice change (39.9%) CTM-3 scores.
PERCEIVED CARE COORDINATION AT DISCHARGE The difference in perceived care coordination at discharge between the subgroups of patients who received and those who did not receive assistance from DP staff was also of interest. Patients seen by DP staff members were older and had longer
January 2011
lengths of stay (Table 4, page 34). The DP staff interacted with more patients hospitalized on the medical units than on the surgical units. Before the nurse/social worker DP team was developed as part of the practice change, DP nurses, ITCs, or social workers were consulted for only 9.6% of responders (21/218). After the practice change, the DP team was involved with 22.2% of the responders (34/153). More patients seen by the DP team went home with formal community services than those not assisted by the DP team.
Volume 37 Number 1
Copyright 2011 © The Joint Commission
33
The Joint Commission Journal on Quality and Patient Safety Table 4. Characteristics of Participants by Subgroups (With and Without DP Staff Involvement)* Before Practice Change With DP, RN, or SW No DP, RN, or SW N = 21 N = 199 Age†‡ Length of Stay†‡ Length of Time Until Survey Returned§
Sex Female Male Nursing Unit|| Medical Surgical Primary Insurance‡ Medicare Commercial Other Discharge Disposition†‡ Self Care Home Health Care Nursing Facility Received Post Acute Assistance (includes family)‡§
After Practice Change With DP Team No DP Team N = 34 N = 119
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
66.1 (⫾14.0) 8.0 (⫾5.2)
57.5 (⫾14.8) 2.7 (⫾2.4)
70.4 (⫾17.2) 6.4 (⫾6.4)
55.6 (⫾15.8) 2.8 (⫾2.6)
12.2 (⫾5.4)
12.4 (⫾7.6)
15.9 (⫾7.3)
18.6 (⫾10.4)
N (%)
N (%)
N (%)
N (%)
9 (42.9) 12 (57.1)
123 (61.8) 76 (38.2)
22 (64.7) 12 (35.3)
77 (64.7) 42 (35.3)
10 (47.6) 11 (52.3)
58 (29.1) 141 (70.9)
14 (41.2) 20 (58.8)
28 (23.5) 91 (76.5)
13 (61.9) 5 (23.8) 3 (14.3)
74 (37.2) 101 (50.8) 24 (12.1)
22 (64.7) 12 (35.3) 0 (0)
39 (32.8) 71 (59.7) 8 (6.7)
15 (71.4) 1 (4.8) 5 (23.8)
194 (97.4) 1 (0.5) 4 (2.0)
25 (73.5) 3 (8.8) 6 (17.6)
119 (100.0) 0 (0) 0 (0)
14 (66.7)
102 (51.2)
26 (76.5)
47 (39.4)
* DP, discharge planning; RN, registered nurse; SW, social worker; SD, standard deviation. † Before practice change difference between “with DP, RN, or SW involvement” and “no DP, RN, or SW involvement” groups (p < .05). ‡ After practice change difference between “with DP team” and “no DP team” groups (p < .05). § Difference in “no DP, RN, or SW” and “no DP team” groups before to after practice change (p < .05). ||
Difference before to after practice change within both “with DP, RN, or SW” and “no DP team” groups (p < .05)
The mean CTM-3 scores for the both the pre- and postpractice change participants who received assistance from the DP staff were appreciably lower than those who did not receive assistance. The mean CTM-3 score of the patients who received assistance improved by 14 points from 67.2 before to 81.2 after the practice change, exceeding the predetermined difference deemed clinically meaningful. There was also less variability of mean CTM-3 scores after the practice change. The SD of the mean decreased from 31.1 to 18.1. The 75th percentile score increased from 83.3 before to 100.0 after the practice change, although the median scores did not change.
Discussion Evaluation of the changes in DP practice in terms of the patient’s perspective were inconclusive, which may reflect a number of factors. First, the response rate was low—but com34
January 2011
parable to those found by the developers of the CTM-3.23 A higher response rate might have been obtained if the measure was implemented along with the hospital’s routine postacute satisfaction survey tool. Second, only a small number of patients in this sample received DP team involvement, with fewer than expected patients discharged home with formal community services. The relatively low percentage of patients with Medicare coverage and their relatively low mean age also suggest that this particular sample would have been less likely to have shown a practice-change effect. Yet, although the CTM-3 was designed based on studies of older adults with a greater likelihood of requiring care after discharge,26,27 DP is mandated as an essential care process for all hospitalized patients.7,28 Similarly, the NQF guidelines require sampling all patients discharged from general acute care hospitals, not just patients with Medicare coverage.22 Using the CTM-3 to evalu-
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety ate DP interventions that target at-risk patients (for example, older patients with known multiple long-term health needs) may yield different results.29–31 Third, a few of the components of the practice change involved coordination of the DP nurse and social work roles to improve communication and decrease duplicated effort. Patients may not be aware of the care coordination, which involves using past findings, evaluations, and decisions in current management among a number of providers to facilitate the appropriate delivery of services,32–34 occurring on their behalf. Finally, there also was a very high ceiling effect with the CTM-3 scores in this sample. More than half of the patients rated the quality of their care transition extremely high (maximum score) before the practice change, making it difficult to capture the practice-change effect.
length of stay, more likely to be covered by Medicare than commercial insurance), who typically are at greater risk for complex discharge plans. More of these patients were referred to formal community services such as home health care or a nursing facility for postacute care. Readers should keep in mind that the data reported were from participants hospitalized on only two nursing units in one academic medical center. The DP practice components require further testing before implementation in other health care delivery systems, in other geographic locations, and with more ethnically diverse patient populations. Evaluating a practice change should also include perspectives of other key stakeholders, such as community service providers and hospital staff.
Postscript LESSONS LEARNED There are a number of lessons learned from attempting to capture the patient perspective of a DP practice change. The response rate after the practice change was low (39.5%), which is a concern, given that response rate is widely used as a measure of survey quality35 and that nonresponse bias can be a serious threat to the validity of the estimates derived from studies.36 In general, surveys in health care tend to under represent males, those with lower socioeconomic status, the unemployed, and those with lower education,37 which may have contributed to an underestimation of the practice-change effect. The CTM-3 is a self-report instrument. Excluding proxy responses, as recommended by NQF, as stated,22 limits the information that can be gathered from patients with cognitive impairment or those with more complex continuing care needs—the very patients who are usually targeted for specialized DP services—who require complex discharge plans. Capturing perceptions of the quality of care coordination from the patients with the greatest care coordination needs may be problematic. Response rates have generally been higher for the CTM-3 when the data were obtained via phone calls, with no significant floor or ceiling effects reported.26 The significant ceiling effects in this study may have reflected social desirability response bias. Telephone administration of the CTM-3 by an outside company and combining the items with other hospitalspecific questions may minimize this bias. Although the CTM-3 results are inconclusive, a clinically meaningful decrease occurred in the length of stay for the group of patients who received assistance from the DP team— from 8.0 to 6.4 days. The patients who received assistance from the DP team were a more disadvantaged group (older, longer
January 2011
The proactive approach to DP proved to be a valuable shift. The standardization of DP processes and improved multidisciplinary teamwork were important considerations for implementing the practice change throughout our institution. The ESDP and the comprehensive DP assessment format have been incorporated into hospitalized patients’ electronic medical records. The DP nurses and social workers now meet quarterly to discuss common DP issues. A multidisciplinary DP council now exists to oversee DP practice issues that span clinical specialties in both the inpatient and outpatient settings. J This study was funded in part by NINR T32 Grant Number NR009356, School of Nursing, University of Pennsylvania; Mayo Clinic Rochester Department of Nursing; and Grant Number 1 UL1 RR024150-01, National Center for Research Resources. The authors wish to thank the members of the Discharge Planning Oversight Group, Department of Nursing, Mayo Clinic, without whose help this study could not have been successful.
Diane E. Holland, Ph.D., R.N., is Clinical Nurse Researcher, Department of Nursing, Mayo Clinic, Rochester, Minnesota, and Assistant Professor of Nursing, College of Medicine, Mayo Clinic. Michele A. Hemann, M.S.N., R.N., is Nursing Education Specialist, Department of Nursing, Mayo Clinic. Please address correspondence to Diane E. Holland, [email protected].
References 1. Forster A.J., et al.: The incidence and severity of adverse events affecting patients after discharge from the hospital. Ann Intern Med 138:161–167, Feb. 4, 2003. 2. Greenwald J.L., et al.: The hospital discharge: A review of a high risk care transition with highlights of a reengineered discharge process. J Patient Saf 3:97–106, Jun. 2007. 3. Institute of Medicine: Priority Areas for Nation Action: Transforming Health Care Quality. Washington, DC: National Academy Press, 2003. 4. Kripalani S., et al.: Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient
Volume 37 Number 1
Copyright 2011 © The Joint Commission
35
The Joint Commission Journal on Quality and Patient Safety safety and continuity of care. JAMA 297:831–841, Feb. 28, 2007. 5. Weinberg D.B., et al.: Beyond our walls: Impact of patient and provider coordination across the continuum on outcomes for surgical patients. Health Serv Res 42(1 pt 1):7–24, 2007. 6. Centers for Medicare & Medicaid Services: State Operations Manual: Appendix A-Survey Protocol, Regulations and Interpretive Guidelines for Hospitals. Rev. 47, Jun. 5, 2009. http://www.cms.hhs.gov/manuals/ downloads/som107ap_a_hospitals.pdf (last accessed Nov. 23, 2006). 7. Chiplin A.J. Jr.: Breathing Life into Discharge Planning. Willimantic, CT: Center for Medicare Advocacy, Inc., 2005. 8. Efraimsson E., et al.: How to get one’s voice heard: The problems of the discharge planning conference. J Adv Nurs 53:646–655, Mar. 2006. 9. Wells D.L., et al.: An integrated model of discharge planning for acutely-ill elderly patients. Can J Nurs Leadersh 12:6–12, Sep.–Oct. 1999. 10. Bull M.J., Roberts J.: Components of a proper hospital discharge for elders. J Adv Nurs 35:571–581, Aug. 2001. 11. Naylor M.D., Bowles K.H., Brooten D.: Patient problems and advanced practice nurse interventions during transitional care. Public Health Nurs 17:94–102, Mar.–Apr. 2000. 12. Naylor M.D.: Transitional care of older adults. Annu Rev Nurs Res 20:127–147, 2002. 13. Shepperd S., et al.: Discharge planning from hospital to home. Cochrane Database Syst Rev 1:CD000313, Aug. 6, 2004. 14. Mistiaen P., Francke A.L., Poot E.: Interventions aimed at reducing problems in adult patients discharged from hospital to home: A systematic metareview. BMC Health Serv Res 7:47, Apr. 4, 2007. 15. Pearson P., et al.: The process of hospital discharge for medical patients: A
36
January 2011
model. J Adv Nurs 46:496–505, Jun. 2004. 16. Institute of Medicine: Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press, 2001. 17. Holland D.E., et al.: Development and validation of a screen for specialized discharge planning services. Nurs Res 55:62–71, Jan.–Feb. 2006. 18. Holland D.E., et al.: Continuity of care: A nursing needs assessment instrument. Geriatr Nurs 19:331–334, Nov.–Dec. 1998. 19. Medicare and Medicaid programs; Revisions to conditions of participation for hospitals—HCFA. Final rule. Fed Regist 59:64141–64153, Dec. 13, 1994. 20. Coleman E.A., Mahoney E., Parry C.: Assessing the quality of preparation for posthospital care from the patient’s perspective: The Care Transitions Measure. Med Care 43:246–255, Mar. 2005. 21. University of Colorado Health Sciences Center, Coleman E.A.: The Care Transitions Program. http://www.caretransitions.org (last accessed Nov. 23, 2010). 22. National Quality Forum: National Voluntary Consensus Standards for Hospital Care: Additional Priority Areas 2005–2006. Washington, DC: National Quality Forum, 2006. 23. Parry C., et al.: Assessing the quality of transitional care: Further applications of the Care Transitions Measure. Med Care 46:317–322, Mar. 2008. 24. Bowles K.H., Naylor M.D, Foust J.B.: Patient characteristics at hospital discharge and a comparison of home care referral decisions. J Am Geriatr Soc 50:336–342, Feb. 2002. 25. Naylor M.D., et al.: Transitional care of older adults hospitalized with heart failure: A randomized, controlled trial. J Am Geriatr Soc 52:675–684, May 2004. Erratum in: J Am Geriatr Soc 52:1228, Jul. 2004. 26. Coleman E.A., et al.: Development and testing of a measure designed to assess the quality of care transitions. Int J Integr Care 2:e02, 2002. Epub Jun. 1, 2002. 27. Coleman E.A., et al.: The central role of performance measurement in improving the quality of transitional care. Home Health Care Serv Q 26(4):93–104, 2007. 28. Chugh A., et al.: Better transitions: Improving comprehension of discharge instructions. Front Health Serv Manage 25:11–32, Spring 2009. 29. Dedhia P., et al.: A quality improvement intervention to facilitate the transition of older adults from three hospitals back to their homes. J Am Geriatr Soc 57:1540–1546, Sep. 2009. Epub Aug. 18, 2009. 30. Parrish M.M., et al.: Implementation of the Care Transitions Intervention: Sustainability and lessons learned. Prof Case Manage 14:282–293, Nov.–Dec. 2009. 31. Wallis M., et al.: Effectiveness of an advanced practice emergency nurse role in a minor injuries unit. Aust J Adv Nurs 27:21–29, Sep.–Nov. 2009. 32. Donabedian A.: The Definition of Quality and Approaches to Its Assessment. Ann Arbor, MI: Health Administration Press, 1980. 33. McGuiness C., Sibthorpe B.: Development and initial validation of a measure of coordination of health care. Int J Qual Health Care 15:309–318, Aug. 2003. 34. National Transitions of Care Coalition (NTOCC), Measures Work Group: Transitions of Care Measures. 2008. http://www.ntocc.org/Portals/0/ TransitionsOfCare_Measures.pdf (last accessed Nov. 23, 2010). 35. Atrostic B.K., et al.: Nonresponse in U.S. government household surveys: Consistent measures, recent trends, and new insights. J Off Stat 17(2):209–226, 2001. 36. Sackett D.L.: Bias in analytic research. J Chronic Dis 32(1–2):51–63, 1979. 37. Groves R.M.: Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quarterly 70(5):646–675, 2006. http://poq.oxfordjournals.org/content/70/5/646.full (last accessed Nov. 29, 2011).
Volume 37 Number 1
Copyright 2011 © The Joint Commission
Standardizing Hospital Discharge Planning at the Mayo Clinic Diane E. Holland, Ph.D., R.N.; Michele A. Hemann, M.S.N., R.N.
I
mproving the quality of patient transitions across health care settings is a top-priority health care concern. Inadequate transitions are known to lead to suboptimal outcomes, the development of new or worsening symptoms, unplanned rehospitalizations, and medical errors and other adverse events, especially in older adults.1–5 Hospital discharge planning (DP) is the primary vehicle for managing care coordination from the hospital. The Centers for Medicare & Medicaid Services (CMS) Hospital Conditions of Participation in the Medicare Program require that hospitals have a DP process in effect that applies to all patients.6 Furthermore, hospitals must identify early in the hospital stay those patients who without further DP evaluation would suffer adverse consequences.7 The CMS condition also indicates that the additional DP evaluation must include an assessment of the patient’s capacity for independence, the possibility of the patient being cared for in the environment from which he or she entered the hospital, the likelihood of a patient needing posthospital services, and the availability of the services.6 Components linked to a successful or an effective DP process include early identification of at-risk patients, a systematic approach to the process, multidisciplinary teamwork, and well-working communication systems.8–12 CMS does not mandate any one specific DP structure, which allows for flexibility in the use of available resources to provide DP services, although this also results in a wide variation in DP service structures across institutions. Despite the large body of literature on DP, the impact on indicators such as readmission rates, hospital length of stay, and costs, as found in systematic reviews of DP, remains uncertain.13–15 Recent emphasis on patient-centered care and better coordination of care has led to a search for patient-oriented indicators, in recognition of the fact that patients are central to the DP process and, as such, are key stakeholders.16 In this article, we report how we evaluated a change in DP practice from the perspective of recently discharged patients. We were further interested in possible differences in patient per-
January 2011
Article-at-a-Glance Background: Improving the quality of patient coordina-
tion in the transition from hospital to home is a highpriority health care concern. The Centers for Medicare & Medicaid Services (CMS) Hospital Conditions of Participation in the Medicare Program require that hospitals have a discharge planning (DP) process in effect that applies to all patients. The impact of a practice change in DP practice on the quality of care coordination at discharge was evaluated from patients’ perspectives. Methods: A multifactor, evidence-based DP practice change, which included merging of DP specialist roles and use of an early screen for DP decision support tool, was initiated in a large, Midwestern academic medical center and evaluated in a nonequivalent comparison group design with separate pre- and postpractice change samples. The threeitem Care Transitions MeasureTM (CTM-3TM) was mailed to adults recently discharged from one medical and one surgical nursing unit before and after the practice change. Results: Response rates were 52.4% before (218/416) and 39.5% (153/387) after the practice change. There were no significant differences between characteristics of the preand postpractice change participants. The mean CTM-3 score of patients who received assistance from the nurse/ social worker DP team improved by 14 points (67.2 to 81.2), although the data were skewed with a ceiling effect, rendering the results inconclusive. Conclusions: Although the CTM-3 results were inconclusive, the practice change resulted in a clinically meaningful decrease in length of stay for a group of older patients at greater risk for complex discharge plans. The proactive approach to DP proved to be a valuable shift. The successes of the standardization of DP processes and improved multidisciplinary teamwork were important considerations for implementation throughout the organization.
Volume 37 Number 1
Copyright 2011 © The Joint Commission
29
The Joint Commission Journal on Quality and Patient Safety Table 1. Discharge Planning (DP) Practice Before and After Changes Before Practice Change ■ 3 roles related to discharge planning − Infusion therapy coordinator (ITC) − DP nurse − Social worker (SW)
After Practice Change ■ ITC and DP nurse combined into same role (DP nurse specialist [DPNS])
■ DP nurse, ITC, and SW in separate departments − Role overlap − Duplicated efforts − Lack of teamwork
■ DP nurse specialists and SW merged into one department − Role clarification − Separation of DP duties − DPNS/SW team approach
■ No standardized processes for DP assessment or documentation
■ Standardized DP process including the following: − Early identification of patients who need further DP evaluation using decision support tool − Electronic consult to DPNS/SW team − Daily multidisciplinary rounds on all units − Comprehensive assessment tool and documentation format
ceptions between patients who received assistance with DP from a nurse/social worker team (after the practice change) and patients who received services from a DP nurse or social worker (before DP team development), as part of the practice change.
Methods INSTITUTING A DISCHARGE PLANNING PRACTICE CHANGE In 2008, members of the nursing department, in collaboration with the medical social services division, proposed a multicomponent change in DP practice at our institution. Aligning DP Support. DP support was provided to direct care providers (the staff nurses and physicians) by persons in three different specialized positions, each based in a different division: ■ DP nurses, who were assigned to units and available by consult whenever the staff nurse or physician perceived the patient’s discharge plan was more complex than what they felt capable of handling themselves ■ Nurse infusion therapy coordinators (ITCs), who provided specialized DP services for patients with postdischarge intravenous (IV) therapy ■ Social workers, who were assigned to medical services and were also available by consult to assist with complex discharge plans It was not uncommon for all three DP specialists, each requested by a different direct care staff member, to consult on the same patient, resulting in duplicated effort and poor communication. Teamwork between the disciplines assigned to DP roles was further hampered by different administrative reporting struc30
January 2011
tures for the roles. Supervision of all three DP roles was merged in early 2008. This led to discussions regarding how to improve our practice to ensure that we were providing the best service to our key stakeholders. A multidisciplinary DP team of key stakeholders, including nurses, social workers, pharmacists, dietitians, and physical therapists planned the DP practice changes. DP nurses and leadership from two nursing units (one general medical and one general surgical unit) provided development and oversight expertise. The multifactor change in DP practice sought to move from a reactive to a more proactive standardized approach. As shown in Table 1 (above), the ITCs and DP nurses were combined into one role—renamed discharge planning nurse specialists. The nurses and social workers merged into one division with the same administrator. Further team development ensued, as well as role clarification for the direct care registered nurse (RN), DP nurse, and social worker to minimize duplication of effort. DP Decision Support Tool. In 2009, a DP team oversaw the development of the evidence-based Early Screen for Discharge Planning (ESDP) DP decision support tool which was embedded into the RN admission assessment in the electronic medical record. The ESDP tool, which was based on patient admission assessment data, had been developed and tested within the institution’s hospitals.17 The ESDP, designed to assist in the early engagement of DP resources for those patients with potentially complex DP needs, consists of four patient characteristics readily available on admission—age, self-reported walking limitation, prior living status, and overall level of disability. A cut-point score of 10 or more points indicates that a referral to the DP team is warranted. For example, patients with
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety the following characteristics would meet or exceed the threshold score: ■ A patient 80 years or older (8 points) with a walking limitation (3 points) or who lives alone (3 points) ■ A 70-year-old patient (6 points) who needs help daily with dressing or hygiene (9 points) ■ A 50-year-old (4 points) with a walking limitation (3 points) and who lives alone (3 points) After the ESDP was completed, a score was electronically computed. When the score met or exceeded the predesignated threshold, a consult was issued by the staff RN and sent electronically to the DP work list. This provided greater ease and timeliness for making consults to the DP team. DP Rounds. Multidisciplinary DP rounds were instituted daily on the two nursing units to facilitate communication and timely planning for continuing care needs. DP Assessment Format. An evidence-based, standardized comprehensive DP assessment format18 was introduced in 2009 for use with high-risk patients, such as those who would likely suffer adverse consequences without a further evaluation of their continuing care needs. Content domains consisted of cognitive/behavioral/emotional status, health status, functional status, finances, environmental factors in postdischarge care, anticipated skilled care requirements, and resources to meet continuing care needs. The DP nurse completes the assessment by interviewing the patient and caregiver(s). The caregiver has the opportunity to discuss whether he or she is able to offer support for continuing care needs or prefers that formal community service providers be engaged. The DP team members use the comprehensive assessment information to communicate the patient’s continuing care needs to hospital staff and community providers (home care agencies, nursing homes, outpatient centers, infusion companies, and equipment companies). The information is shared with continuing care providers via telephone and transferred into the hospital dismissal summary document before patient discharge. A copy of the dismissal summary is available at patient discharge and shared with continuing care providers. After they are engaged in DP for a patient, members of the DP team remain actively involved in the initial implementation of the discharge plan until the patient is discharged. Staff nurses continue to be responsible for formulating and implementing discharge plans of patients with relatively straightforward discharge plans. Consistent with the Medicare Conditions of Participation,19 the DP team can also be consulted at any point throughout the hospital stay.
January 2011
EVALUATING CHANGE
THE
DISCHARGE PLANNING PRACTICE
Design. The evaluation study, approved by the Institutional Review Board, was designed by the DP team [including D.E.H., M.A.H.] as a nonequivalent comparison group study with separate pre- and postintervention samples. The practice change was implemented during a period of four weeks (July 7–August 1, 2008). The comparison group sample consisted of patients hospitalized on two nursing units—general medical and general surgical—six weeks before the practice change. Patients in the intervention sample of patients were hospitalized on the same two units in the six-week period after the practice change implementation period was completed. (These units were chosen because the practice change was intended to affect all hospitalized adults.) Patients on these units represented adults admitted to general acute care and did not represent any one specific clinical specialty practice. Outcome Measure. We used the Care Transitions Measure (CTM-3; Figure 1, page 32),20,21 which consists of three questions derived from qualitative studies of recently hospitalized patients. The National Quality Forum (NQF) has endorsed it as standard measure for assessing patients’ perspectives of the quality of care coordination from the hospital to the home setting.22 The tool has been tested in both mailed and interview formats, with patients experiencing either medical or surgical problems,20 and with ethnically diverse populations.23 Results are scored by computing a mean score for each responder. Scores are transformed to a 0–100 scale for comparison and public reporting. Higher scores indicate higher perceived quality of care coordination at discharge.22 Sample. All adult patients hospitalized on the two nursing units who met inclusion criteria regardless of payor source were surveyed. Although the NQF guidelines recommend excluding proxy responses,22 we mailed the surveys, which neccessitated adding a question as to who responded to the survey (patient or other person). Patients younger than 18 years of age and patients on observation status (did not meet hospital admission criteria) were also excluded. Responders were further separated into two naturally occurring groups—those whose discharge plans were formulated and implemented by a staff nurse and those who received assistance with their discharge plans from the nurse/social worker DP team (Table 1). A sample size of 120 patients pre- and postpractice change provided sufficient power to detect a clinically meaningful (10point) difference in CTM-3 scores. On the basis of a reported standard deviation (SD) for the CTM-3 of 13.67,22 a minimum of 60 patients in each group provided 80% power to detect a
Volume 37 Number 1
Copyright 2011 © The Joint Commission
31
The Joint Commission Journal on Quality and Patient Safety The Three-Item Care Transitions Measure 1. The hospital staff took my preferences and those of my family or caregiver into account in deciding what my health care needs would be when I left the hospital. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
2. When I left the hospital, I had a good understanding of the things I was responsible for in managing my health. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
3. When I left the hospital, I clearly understood the purpose for taking each of my medications. 1. ■ Strongly disagree
2. ■ Disagree
3. ■ Agree
4. ■ Strongly agree
5. ■ Don’t know/don’t remember/not applicable
Figure 1. The three-item Care Transitions MeasureTM (CTM-3TM) was developed by Eric A. Coleman, M.D., M.P.H., of the University of Colorado Denver Care Transitions Program and is available in the public domain from http://www.caretransitions.org (last accessed Nov. 23, 2010). The CTM-3 is endorsed by the National Quality Forum for hospital and nursing home performance measurement.
minimal clinically meaningful difference in mean CTM-3 scores of 10, alpha set at .05. Survey Administration. Consistent with the NQF guidelines, which indicate that survey administration can occur between 48 hours and six weeks after discharge and can be mailed or administered by telephone,22 the CTM-3 was mailed at approximately four–six days after patient discharge to the primary address listed in the medical record. This time period was chosen on the basis of DP literature indicating that the first few weeks after discharge are critical to patient recovery.11,24,25 In 2008, the CTM-3 was mailed by members of the study team to 416 eligible patients discharged from the two nursing units during six weeks before the practice change and to 387 patients discharged during the six weeks after the practice change was implemented. The patients were asked to complete and return the survey (in a stamped envelope provided) the same day they received it, if possible. Sociodemographic and health characteristics of the patients, documented assistance by the DP team, and referrals to formal community services were obtained by record review. Characteristics of nonresponders who had previously granted the general authorization for the use of their medical records for research purposes under the state law were also recorded.
Results RESPONSE RATES The response rates (including proxy responses) were 52.4% before (218/416) and 39.5% (153/387) after the practice change. Because almost one-third (29.4%; 10/34) of the questionnaires returned by patients who received assistance from DP nurses or social workers were completed by proxy responders, the decision was made to include proxy responses in the data analysis.
RESPONDERS AND NONRESPONDERS There were some differences between responders and nonresponders (Table 2, page 33). Before the practice change, more respondents were female and hospitalized on the surgical unit with longer lengths of stay than nonresponders. After the practice change, responders were older and, again, more likely to be hospitalized on the surgical unit than nonresponders. More nonresponders were discharged with formal services such as home health care or to a nursing facility than were responders. Responders were generally in their late 50s. There were no statistically significant differences in characteristics between the pre- and postpractice change responders. Few responders went home with a referral to formal community services.
CTM-3 SCORES DATA ANALYSIS Descriptive statistics were used to describe sample characteristics. The Mann-Whitney U test was used to measure differences between the pre- and postpractice change CTM-3 scores. Nonparametric tests were used because of the skewed nature of the data. 32
January 2011
The overall mean CTM-3 scores were relatively the same in the pre- (82.9 ±24.3) and postpractice change groups (80.1 ±23.0; Table 3, page 33). The distribution of both pre- and postpractice change scores was markedly skewed. The overall median scores remained the same (88.9), as did the interquartile ranges (IQR; 66.7–100.0). Of note, the 75th quartile score
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety Table 2. Sample Characteristics of Before and After Practice Change Responders and Nonresponders*
Age† Length of stay‡
Sex‡ Female Male Primary insurance Medicare Commercial Other Discharge disposition‡ Self care Home health care Nursing facility Nursing unit†‡ Medical Surgical
Before Practice Change Responders Nonresponders N = 218 N = 192
After Practice Change Responders Nonresponders N = 153 N = 222
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
Mean (+ SD)
58.5 (⫾14.9) 3.2 (⫾3.2)
58.5 (⫾18.6) 4.5 (⫾4.8)
59.9 (⫾17.1) 3.4 (⫾2.8)
54.9 (⫾18.2) 4.3 (⫾4.5)
N (%)
N (%)
N (%)
N (%)
218 133 (61.0) 85 (39.0) 218 87 (39.9) 104 (47.7) 27 (12.4) 218 207 (95.0) 2 (0.9) 9 (4.1) 218 68 (31.2) 150 (68.8)
192 114 (59.4) 78 (40.6) 192 80 (41.7) 87 (45.3) 25 (13.0) 188 159 (84.0) 7 (3.7) 22 (11.7) 192 103 (53.7) 89 (46.4)
153 99 (64.7) 54 (35.3) 153 63 (41.2) 82 (53.6) 8 (5.2) 152 143 (94.0) 3 (2.0) 6 (3.9) 153 43 (28.1) 110 (71.9)
222 138 (62.2) 84 (37.8) 218 75 (34.4) 122 (56.0) 21 (9.6) 220 193 (87.7) 10 (4.6) 17 (7.7) 219 106 (48.4) 113 (51.6)
* SD, standard deviation. † After practice change responders and nonresponders differed significantly (p < .05). ‡ Before practice change participants and nonresponders differed significantly (p < .05).
Table 3. CTM-3 Scores* Before Practice Change† Overall
N 218
M (⫾SD) 82.9 (⫾24.3)
Median (IQR) 88.9 (66.7–100.0)
After Practice Change N 153
M (⫾SD) 80.1 (⫾23.0)
Median (IQR) 88.9 (66.7–100.0)
DP staff involvement‡
21
67.2 (⫾31.1)
66.7 (66.7–83.3)
34
81.2 (⫾18.1)
66.7 (66.7–100.0)
No DP staff involvement
197
84.8 (⫾22.6)
100.0 (66.7–100.0)
120
79.9 (⫾24.4)
88.9 (66.7–100.0)
* SD, standard deviation; IQR, interquartile range; DP, discharge planning. † Before practice change mean CTM-3 scores were statistically significantly different between the “DP staff involvement” and “No DP staff involvement” groups
(p < .05). ‡ After practice change reflects DP nurse/social worker team
in both groups was 100, the maximum CTM-3 score. A large ceiling effect was observed overall in both the pre- (49.1%) and postpractice change (39.9%) CTM-3 scores.
PERCEIVED CARE COORDINATION AT DISCHARGE The difference in perceived care coordination at discharge between the subgroups of patients who received and those who did not receive assistance from DP staff was also of interest. Patients seen by DP staff members were older and had longer
January 2011
lengths of stay (Table 4, page 34). The DP staff interacted with more patients hospitalized on the medical units than on the surgical units. Before the nurse/social worker DP team was developed as part of the practice change, DP nurses, ITCs, or social workers were consulted for only 9.6% of responders (21/218). After the practice change, the DP team was involved with 22.2% of the responders (34/153). More patients seen by the DP team went home with formal community services than those not assisted by the DP team.
Volume 37 Number 1
Copyright 2011 © The Joint Commission
33
The Joint Commission Journal on Quality and Patient Safety Table 4. Characteristics of Participants by Subgroups (With and Without DP Staff Involvement)* Before Practice Change With DP, RN, or SW No DP, RN, or SW N = 21 N = 199 Age†‡ Length of Stay†‡ Length of Time Until Survey Returned§
Sex Female Male Nursing Unit|| Medical Surgical Primary Insurance‡ Medicare Commercial Other Discharge Disposition†‡ Self Care Home Health Care Nursing Facility Received Post Acute Assistance (includes family)‡§
After Practice Change With DP Team No DP Team N = 34 N = 119
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
Mean (⫾SD)
66.1 (⫾14.0) 8.0 (⫾5.2)
57.5 (⫾14.8) 2.7 (⫾2.4)
70.4 (⫾17.2) 6.4 (⫾6.4)
55.6 (⫾15.8) 2.8 (⫾2.6)
12.2 (⫾5.4)
12.4 (⫾7.6)
15.9 (⫾7.3)
18.6 (⫾10.4)
N (%)
N (%)
N (%)
N (%)
9 (42.9) 12 (57.1)
123 (61.8) 76 (38.2)
22 (64.7) 12 (35.3)
77 (64.7) 42 (35.3)
10 (47.6) 11 (52.3)
58 (29.1) 141 (70.9)
14 (41.2) 20 (58.8)
28 (23.5) 91 (76.5)
13 (61.9) 5 (23.8) 3 (14.3)
74 (37.2) 101 (50.8) 24 (12.1)
22 (64.7) 12 (35.3) 0 (0)
39 (32.8) 71 (59.7) 8 (6.7)
15 (71.4) 1 (4.8) 5 (23.8)
194 (97.4) 1 (0.5) 4 (2.0)
25 (73.5) 3 (8.8) 6 (17.6)
119 (100.0) 0 (0) 0 (0)
14 (66.7)
102 (51.2)
26 (76.5)
47 (39.4)
* DP, discharge planning; RN, registered nurse; SW, social worker; SD, standard deviation. † Before practice change difference between “with DP, RN, or SW involvement” and “no DP, RN, or SW involvement” groups (p < .05). ‡ After practice change difference between “with DP team” and “no DP team” groups (p < .05). § Difference in “no DP, RN, or SW” and “no DP team” groups before to after practice change (p < .05). ||
Difference before to after practice change within both “with DP, RN, or SW” and “no DP team” groups (p < .05)
The mean CTM-3 scores for the both the pre- and postpractice change participants who received assistance from the DP staff were appreciably lower than those who did not receive assistance. The mean CTM-3 score of the patients who received assistance improved by 14 points from 67.2 before to 81.2 after the practice change, exceeding the predetermined difference deemed clinically meaningful. There was also less variability of mean CTM-3 scores after the practice change. The SD of the mean decreased from 31.1 to 18.1. The 75th percentile score increased from 83.3 before to 100.0 after the practice change, although the median scores did not change.
Discussion Evaluation of the changes in DP practice in terms of the patient’s perspective were inconclusive, which may reflect a number of factors. First, the response rate was low—but com34
January 2011
parable to those found by the developers of the CTM-3.23 A higher response rate might have been obtained if the measure was implemented along with the hospital’s routine postacute satisfaction survey tool. Second, only a small number of patients in this sample received DP team involvement, with fewer than expected patients discharged home with formal community services. The relatively low percentage of patients with Medicare coverage and their relatively low mean age also suggest that this particular sample would have been less likely to have shown a practice-change effect. Yet, although the CTM-3 was designed based on studies of older adults with a greater likelihood of requiring care after discharge,26,27 DP is mandated as an essential care process for all hospitalized patients.7,28 Similarly, the NQF guidelines require sampling all patients discharged from general acute care hospitals, not just patients with Medicare coverage.22 Using the CTM-3 to evalu-
Volume 37 Number 1
Copyright 2011 © The Joint Commission
The Joint Commission Journal on Quality and Patient Safety ate DP interventions that target at-risk patients (for example, older patients with known multiple long-term health needs) may yield different results.29–31 Third, a few of the components of the practice change involved coordination of the DP nurse and social work roles to improve communication and decrease duplicated effort. Patients may not be aware of the care coordination, which involves using past findings, evaluations, and decisions in current management among a number of providers to facilitate the appropriate delivery of services,32–34 occurring on their behalf. Finally, there also was a very high ceiling effect with the CTM-3 scores in this sample. More than half of the patients rated the quality of their care transition extremely high (maximum score) before the practice change, making it difficult to capture the practice-change effect.
length of stay, more likely to be covered by Medicare than commercial insurance), who typically are at greater risk for complex discharge plans. More of these patients were referred to formal community services such as home health care or a nursing facility for postacute care. Readers should keep in mind that the data reported were from participants hospitalized on only two nursing units in one academic medical center. The DP practice components require further testing before implementation in other health care delivery systems, in other geographic locations, and with more ethnically diverse patient populations. Evaluating a practice change should also include perspectives of other key stakeholders, such as community service providers and hospital staff.
Postscript LESSONS LEARNED There are a number of lessons learned from attempting to capture the patient perspective of a DP practice change. The response rate after the practice change was low (39.5%), which is a concern, given that response rate is widely used as a measure of survey quality35 and that nonresponse bias can be a serious threat to the validity of the estimates derived from studies.36 In general, surveys in health care tend to under represent males, those with lower socioeconomic status, the unemployed, and those with lower education,37 which may have contributed to an underestimation of the practice-change effect. The CTM-3 is a self-report instrument. Excluding proxy responses, as recommended by NQF, as stated,22 limits the information that can be gathered from patients with cognitive impairment or those with more complex continuing care needs—the very patients who are usually targeted for specialized DP services—who require complex discharge plans. Capturing perceptions of the quality of care coordination from the patients with the greatest care coordination needs may be problematic. Response rates have generally been higher for the CTM-3 when the data were obtained via phone calls, with no significant floor or ceiling effects reported.26 The significant ceiling effects in this study may have reflected social desirability response bias. Telephone administration of the CTM-3 by an outside company and combining the items with other hospitalspecific questions may minimize this bias. Although the CTM-3 results are inconclusive, a clinically meaningful decrease occurred in the length of stay for the group of patients who received assistance from the DP team— from 8.0 to 6.4 days. The patients who received assistance from the DP team were a more disadvantaged group (older, longer
January 2011
The proactive approach to DP proved to be a valuable shift. The standardization of DP processes and improved multidisciplinary teamwork were important considerations for implementing the practice change throughout our institution. The ESDP and the comprehensive DP assessment format have been incorporated into hospitalized patients’ electronic medical records. The DP nurses and social workers now meet quarterly to discuss common DP issues. A multidisciplinary DP council now exists to oversee DP practice issues that span clinical specialties in both the inpatient and outpatient settings. J This study was funded in part by NINR T32 Grant Number NR009356, School of Nursing, University of Pennsylvania; Mayo Clinic Rochester Department of Nursing; and Grant Number 1 UL1 RR024150-01, National Center for Research Resources. The authors wish to thank the members of the Discharge Planning Oversight Group, Department of Nursing, Mayo Clinic, without whose help this study could not have been successful.
Diane E. Holland, Ph.D., R.N., is Clinical Nurse Researcher, Department of Nursing, Mayo Clinic, Rochester, Minnesota, and Assistant Professor of Nursing, College of Medicine, Mayo Clinic. Michele A. Hemann, M.S.N., R.N., is Nursing Education Specialist, Department of Nursing, Mayo Clinic. Please address correspondence to Diane E. Holland, [email protected].
References 1. Forster A.J., et al.: The incidence and severity of adverse events affecting patients after discharge from the hospital. Ann Intern Med 138:161–167, Feb. 4, 2003. 2. Greenwald J.L., et al.: The hospital discharge: A review of a high risk care transition with highlights of a reengineered discharge process. J Patient Saf 3:97–106, Jun. 2007. 3. Institute of Medicine: Priority Areas for Nation Action: Transforming Health Care Quality. Washington, DC: National Academy Press, 2003. 4. Kripalani S., et al.: Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient
Volume 37 Number 1
Copyright 2011 © The Joint Commission
35
The Joint Commission Journal on Quality and Patient Safety safety and continuity of care. JAMA 297:831–841, Feb. 28, 2007. 5. Weinberg D.B., et al.: Beyond our walls: Impact of patient and provider coordination across the continuum on outcomes for surgical patients. Health Serv Res 42(1 pt 1):7–24, 2007. 6. Centers for Medicare & Medicaid Services: State Operations Manual: Appendix A-Survey Protocol, Regulations and Interpretive Guidelines for Hospitals. Rev. 47, Jun. 5, 2009. http://www.cms.hhs.gov/manuals/ downloads/som107ap_a_hospitals.pdf (last accessed Nov. 23, 2006). 7. Chiplin A.J. Jr.: Breathing Life into Discharge Planning. Willimantic, CT: Center for Medicare Advocacy, Inc., 2005. 8. Efraimsson E., et al.: How to get one’s voice heard: The problems of the discharge planning conference. J Adv Nurs 53:646–655, Mar. 2006. 9. Wells D.L., et al.: An integrated model of discharge planning for acutely-ill elderly patients. Can J Nurs Leadersh 12:6–12, Sep.–Oct. 1999. 10. Bull M.J., Roberts J.: Components of a proper hospital discharge for elders. J Adv Nurs 35:571–581, Aug. 2001. 11. Naylor M.D., Bowles K.H., Brooten D.: Patient problems and advanced practice nurse interventions during transitional care. Public Health Nurs 17:94–102, Mar.–Apr. 2000. 12. Naylor M.D.: Transitional care of older adults. Annu Rev Nurs Res 20:127–147, 2002. 13. Shepperd S., et al.: Discharge planning from hospital to home. Cochrane Database Syst Rev 1:CD000313, Aug. 6, 2004. 14. Mistiaen P., Francke A.L., Poot E.: Interventions aimed at reducing problems in adult patients discharged from hospital to home: A systematic metareview. BMC Health Serv Res 7:47, Apr. 4, 2007. 15. Pearson P., et al.: The process of hospital discharge for medical patients: A
36
January 2011
model. J Adv Nurs 46:496–505, Jun. 2004. 16. Institute of Medicine: Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press, 2001. 17. Holland D.E., et al.: Development and validation of a screen for specialized discharge planning services. Nurs Res 55:62–71, Jan.–Feb. 2006. 18. Holland D.E., et al.: Continuity of care: A nursing needs assessment instrument. Geriatr Nurs 19:331–334, Nov.–Dec. 1998. 19. Medicare and Medicaid programs; Revisions to conditions of participation for hospitals—HCFA. Final rule. Fed Regist 59:64141–64153, Dec. 13, 1994. 20. Coleman E.A., Mahoney E., Parry C.: Assessing the quality of preparation for posthospital care from the patient’s perspective: The Care Transitions Measure. Med Care 43:246–255, Mar. 2005. 21. University of Colorado Health Sciences Center, Coleman E.A.: The Care Transitions Program. http://www.caretransitions.org (last accessed Nov. 23, 2010). 22. National Quality Forum: National Voluntary Consensus Standards for Hospital Care: Additional Priority Areas 2005–2006. Washington, DC: National Quality Forum, 2006. 23. Parry C., et al.: Assessing the quality of transitional care: Further applications of the Care Transitions Measure. Med Care 46:317–322, Mar. 2008. 24. Bowles K.H., Naylor M.D, Foust J.B.: Patient characteristics at hospital discharge and a comparison of home care referral decisions. J Am Geriatr Soc 50:336–342, Feb. 2002. 25. Naylor M.D., et al.: Transitional care of older adults hospitalized with heart failure: A randomized, controlled trial. J Am Geriatr Soc 52:675–684, May 2004. Erratum in: J Am Geriatr Soc 52:1228, Jul. 2004. 26. Coleman E.A., et al.: Development and testing of a measure designed to assess the quality of care transitions. Int J Integr Care 2:e02, 2002. Epub Jun. 1, 2002. 27. Coleman E.A., et al.: The central role of performance measurement in improving the quality of transitional care. Home Health Care Serv Q 26(4):93–104, 2007. 28. Chugh A., et al.: Better transitions: Improving comprehension of discharge instructions. Front Health Serv Manage 25:11–32, Spring 2009. 29. Dedhia P., et al.: A quality improvement intervention to facilitate the transition of older adults from three hospitals back to their homes. J Am Geriatr Soc 57:1540–1546, Sep. 2009. Epub Aug. 18, 2009. 30. Parrish M.M., et al.: Implementation of the Care Transitions Intervention: Sustainability and lessons learned. Prof Case Manage 14:282–293, Nov.–Dec. 2009. 31. Wallis M., et al.: Effectiveness of an advanced practice emergency nurse role in a minor injuries unit. Aust J Adv Nurs 27:21–29, Sep.–Nov. 2009. 32. Donabedian A.: The Definition of Quality and Approaches to Its Assessment. Ann Arbor, MI: Health Administration Press, 1980. 33. McGuiness C., Sibthorpe B.: Development and initial validation of a measure of coordination of health care. Int J Qual Health Care 15:309–318, Aug. 2003. 34. National Transitions of Care Coalition (NTOCC), Measures Work Group: Transitions of Care Measures. 2008. http://www.ntocc.org/Portals/0/ TransitionsOfCare_Measures.pdf (last accessed Nov. 23, 2010). 35. Atrostic B.K., et al.: Nonresponse in U.S. government household surveys: Consistent measures, recent trends, and new insights. J Off Stat 17(2):209–226, 2001. 36. Sackett D.L.: Bias in analytic research. J Chronic Dis 32(1–2):51–63, 1979. 37. Groves R.M.: Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quarterly 70(5):646–675, 2006. http://poq.oxfordjournals.org/content/70/5/646.full (last accessed Nov. 29, 2011).
Volume 37 Number 1
Copyright 2011 © The Joint Commission