Journal of Gerontological Nursing

DISCHARGE PLANNING FOR Elderly Patients

Mary Ann Rosswurm, EdD, RN, CS, FAAN; Debra M Lanham, MSN, RNC

Abstract

ABSTRACT

The complex chronic health problems and functional limitations common in the elderly population place them at risk for complicated hospitalizations and discharge planning. The purpose of this study was to investigate the effectiveness of a discharge planning protocol in identifying elderly patients' home care needs. The sample in this quasiexperimental study consisted of 507 hospitalized patients age 65 years or older. The control group received the usual hospital discharge planning protocol. In the experimental group, nurse/social worker teams coordinated the discharge planning process, using an adapted form of the Discharge Planning Questionnaire (DPQ) to identify the home care needs of elderly patients. Thirty days after hospital discharge, both patient groups participated in a telephone survey to obtain information about health care problems they experienced during home recovery and their use of health care resources.

The findings indicated that the majority of the elderly patients had functional dependencies, which required the help of another person to carry out daily household duties and provide assistance with basic needs, especially ambulation. These functionally dependent patients only received home care referrals about 50% of the time. These findings raise questions about current reimbursable services. Logistic regression analysis indicated that patients with increased functional dependency and patient problems during home recovery had a greater likelihood of rehospitalization and emergency department usage. This information about the home care of elderly patients after hospitalization supports the need for comprehensive functional assessment as part of discharge planning. This study also suggests that the nurse/social worker team can provide effective screening and discharge planning coordination of home care. Physician involvement and effective communication networks must be in place.

Abstract

ABSTRACT

The complex chronic health problems and functional limitations common in the elderly population place them at risk for complicated hospitalizations and discharge planning. The purpose of this study was to investigate the effectiveness of a discharge planning protocol in identifying elderly patients' home care needs. The sample in this quasiexperimental study consisted of 507 hospitalized patients age 65 years or older. The control group received the usual hospital discharge planning protocol. In the experimental group, nurse/social worker teams coordinated the discharge planning process, using an adapted form of the Discharge Planning Questionnaire (DPQ) to identify the home care needs of elderly patients. Thirty days after hospital discharge, both patient groups participated in a telephone survey to obtain information about health care problems they experienced during home recovery and their use of health care resources.

The findings indicated that the majority of the elderly patients had functional dependencies, which required the help of another person to carry out daily household duties and provide assistance with basic needs, especially ambulation. These functionally dependent patients only received home care referrals about 50% of the time. These findings raise questions about current reimbursable services. Logistic regression analysis indicated that patients with increased functional dependency and patient problems during home recovery had a greater likelihood of rehospitalization and emergency department usage. This information about the home care of elderly patients after hospitalization supports the need for comprehensive functional assessment as part of discharge planning. This study also suggests that the nurse/social worker team can provide effective screening and discharge planning coordination of home care. Physician involvement and effective communication networks must be in place.

People older than age 64 occupy at least 40% of the hospital beds in the United States for an average hospital stay of 6.3 days (National Center for Health Statistics, 1991). Hospitals must become better prepared to serve the increasing number of elderly patients, particularly the frail elderly older than age 85. During hospitalization, between 34% and 50% of elderly patients experience functional decline unrelated to the primary diagnosis (Hirsch, Sommers, Olsen, Mullen, & Winograd, 1990). Health care professionals are challenged to respond not only to the acute care problems but also to the chronic health problems, functional limitations, and special discharge needs of elderly patients. Because of shortened hospital stays, sicker and more functionally impaired elderly patients are going home sooner, often to elderly family caregivers. This situation increases the likelihood of post-discharge problems and costly rehospitalization (Evans & Hendricks, 1993; Kelly, Huber, Johnson, McCloskey, & Maas, 1994) and nursing home admission (Rudberg, Sager, & Zhang, 1996). Re-admissions of elderly patients account for at least 25% of all hospital admissions and significantly contribute to increasing Medicare expenditures (Fethke, Smith, & Johnson, 1986).

Although functional dependency and increased age have been identified as factors associated with frequent and prolonged use of health care services (Evans & Hendricks, 1993; Parfrey et al., 1994), hospital discharge planning for elderly patients often is inadequate in the assessment of functional abilities and the home environment (Henderson, 1992). Studies suggest that more comprehensive assessment of elderly patients' health-recovery needs may decrease unplanned rehospitalizations (Jackson, 1994; Johnson & Fethke, 1985; Naylor, 1990).

Other critical factors for effective discharge planning are involvement of family caregivers and consideration of caregiver burden (Bull, 1990, 1992). The Discharge Planning Questionnaire (DPQ) (Bull, 1994) was developed to facilitate communication between health professionals and the patients/families about patients' post-hospital needs. The DPQ elicits information from patients and family members about their perceptions of the patient's functioning, social support, and selected physical features of the home environment. The relationship of the DPQ to post-discharge outcomes needs to be studied.

The findings of Anderson and Helms' (1993) study of dischargeplanning outcomes indicated that defined lines of responsibility for discharge planning and clear communication of the plan were essential. Other studies (Bull, 1992; Haddock, 1991) have described the importance of early development of a formalized discharge plan and of effective intraorganizational and interorganizational communication. Haddock (1994) explored the outcomes of a structured discharge planning program implemented collaboratively by a gerontological nurse specialist and a social worker on a hospital unit. The findings indicated that the nurse/social worker discharge planning model resulted in patients having shorter hospital stays, fewer re-admissions, and higher rates of needed post-discharge services.

PURPOSE AND RESEARCH QUESTIONS

The purpose of this study was to investigate the effectiveness of a discharge planning protocol in which nurse/social worker teams used an adapted form of the DPQ (Bull, 1994) to identify post-hospital needs of elderly patients. This study examined the relationships between posthospital use of health care services and selected patient characteristics and functioning. The specific research questions were:

1. What are the common types of functional dependencies of patients in the experimental group who are referred for home care services?

2. What are the common patient problems occurring in the experimental and control groups at home within 30 days after discharge?

3. Are there significant differences between the experimental and control groups' patient problems at home and use of health care services within 30 days after discharge?

4. What relationships exist among the use of health care services and functional abilities, severity of illness, social support, and health problems at home?

METHODS Design

The setting for this quasiexperimental, non-equivalent control group study was a 920-bed ternary care center. A convenience sample was selected from five nursing units (surgical cardiovascular, oncology, renal, neuroscience, and medicalsurgical). The sample was limited to patients 65 years of age or older who were admitted from home and were returning home. Because the researchers were primarily interested in the problems encountered during the recovery period at home and the use of home care services, patients being discharged to nursing homes were excluded from this study. The independent variable was the type of discharge planning protocol. Discharge planning for patients in the control group was completed by the social worker without a formalized assessment tool. In the experimental group, nurse/social worker teams screened patients using the adapted DPQ, a tool which facilitated communication of health professionals and patients and their family caregivers. Dependent variables included the use of acute and home health care services and problems at home within 30 days after discharge. Demographic variables included: age, gender, marital status, race, severity of illness, diagnosis, and length of hospital stay. Information for these variables were obtained from the hospital's computerized database. The hospital database included the Pittsburgh Research Institute's (1992) patient severity index, which had scores ranging from 1 (least severe) to 7 (most severe).

Figure 1 . Functional Abilities Assessment.

Figure 1 . Functional Abilities Assessment.

Instrument

The DPQ (Bull, 1994) is a screening tool which measures patients' and family caregivers' perspectives and expectations about post-hospital care needs. The tool consists of 51 items including questions related to activities of daily living (ADL), instrumental activities of daily living (IADL), social support, and environment. Questions on ADL and IADL are rated on a scale from 0 (complete independence) to 4 (complete dependence). Responses to social support and environment questions are either yes or no. All uncertain answers are scored with a decimal figure (1.1) intended to alert health professionals to monitor those needs more closely. Concurrent validity of the ADL questions was established with the criterion measure, the Performance ADL scale (Kane, Reigler, Bell, Potter, & Koshland, 1983). The DPQ significantly correlated with scores on the Performance ADL scale (r = .66). Bull (1994) also reported Cronbach alphas for the subscales ranging from .60 to .87.

In this study, the content of the DPQ items on functional abilities, social support, and home environment remained essentially the same. The format was changed to simplify patients' reading of the items. One item was added to the ADL (i.e., ablility to control bladder and bowels). The DPQ item about taking medications was moved from the ADL to the IADL section. Two items were added to the IADL (i.e., items relating to seeking medical attention and coping with stress). Figure 1 contains the adapted DPQ Functional Abilities Assessment items. The 15 ADL and IADL items were scored on a scale of 0 to 2 (0 = functional independence; 1 = assistance needed; 1.1 = don't know; 2 = total dependence). Individual item scores on the ADL and IADL sections of the questionnaire were summed for a total functional score, ranging from 0 to 30. Scoring for the social support and home environment sections included 0 = yes; 1.1 = don't know; 2 = no. The 1.1 items were further monitored by the nurse.

Procedure

Approval to conduct the study was obtained from the hospital's Institutional Review Board. The control group consisted of patients age 65 and older, who were admitted to the study's hospital units during the 2 months prior to the implementation of the new discharge screening and referral protocol. The control group received the hospital's usual discharge planning, which did not include a formal, comprehensive assessment tool. The researchers collected background data about the control group from their medicalrecord databases. Thirty days after hospital discharge the control group participated in a telephone survey, which included questions about their home care problems and use of health care resources.

Implementation of the new discharge planning protocol for the experimental group occurred during the 2-month period following the hospital discharge of the control group. The researchers held an inservice for nurses and social workers to explain the new discharge protocol and use of the DPQ. The proposed protocol was also discussed with physicians on the selected units. In the new protocol, nurses asked patients and/or caregivers to complete the DPQ within 24 to 48 hours of hospital admission. The nurse reviewed the DPQ and notified the social worker about any patients with functional limitations, which would require more assistance and resources than the family caregiver could provide after hospital discharge. A discharge planning conference was scheduled with input from the patient (when able), the family caregiver, the nurse, and social worker. The nurse and social worker summarized the DPQ data and their referral recommendations for home care. This information was written in the progress notes for physicians to review and discuss as needed, prior to physicians writing orders for home care referrals. As was done for the control group, the researchers collected background data from the medical-record databases of the experimental group and obtained the information about home care problems and health care use during a 30day follow-up phone survey.

FINDINGS Patient Characteristics

The total sample consisted of 575 hospitalized elderly subjects, with 373 in the control group and 202 in the experimental group. The groups were similar in most background characteristics (Table 1). The mean age for the total group was 74 years (range 65 to 100). Ninety-five percent of the patients were Caucasian, which was representative of the geographic area. The majority of patients were married (control 64%; experimental 55%). The most common diagnostic categories included: cardiovascular, neurologic, and respiratory illnesses. The control group had a larger percentage of male patients (54%) and patients with cardiovascular disease (47%). The experimental group had 40% male patients and 28% with cardiovascular disease. The mean severity of illness score for both groups clustered around the midpoint of the scale (i.e., 3.8 for the control group and 3.4 for the experimental group). Both groups had a mean hospital length of stay of 8.2 days.

Experimental Croup's Functional Dependencies and Home Care (Research Question No. 1)

Because data about functional abilities were not recorded in the control group's discharge plan, the findings about functional dependencies refer only to the experimental group. The mean total DPQ functional score for the experimental group was 8.0 (SD 8.4; range 0 to 30). The scores varied significantly according to diagnosis, with surgical cardiovascular patients having the least functional dependency or lowest total functional score (mean = 5.35) and renal patients having the greatest functional dependency or highest score (mean = 13.27; F = 4.57; ? = 0.002). The DPQ functional assessment indicated that 113 (56%) patients were totally dependent in at least one ADL. The types of functional dependencies, the nurse/social worker recommendations for home care follow up, and the number of patients who actually had a home visit are reported in Table 2. The largest percentages of patients (31% to 45.7%) were totally dependent on another person for assistance with the IADLs of cleaning house, laundry, shopping, and/or cooking.

Table

TABLE 1BACKGROUND VARIABLES OF PATIENTS

TABLE 1

BACKGROUND VARIABLES OF PATIENTS

Both the nurse/social worker recommendations for home care assistance and the actual home care assistance received for these IADLs occurred in approximately 50% of the cases. Total dependencies in ADLs were greatest for walking, bathing, and dressing. Although nearly 14% were totally dependent in these basic activities, only 7.4% to 8.4% received a nurse/social worker recommendation for home care follow up and less than half actually received a home visit.

Although the majority of patients reported that a family caregiver or friend would be available to assist them with ADLs and IADLs as needed, most family caregivers were elderly spouses. Accessibility to resources was limited. Approximately 50% of the patients lived at least a 15-minute drive from their physicians* offices. Drug and grocery store delivery services were available for only 18% of patients. The majority of homes were not equipped with assistive and safety devices. Thirty-eight percent had safety bars for the tub, and only 26% had safety bars for the toilet.

Table

TABLE 2NUMBER OF PATIENTS IN EXPERIMENTAL GROUP WITH A FUNCTIONAL DEPENDENCY, HOME CARE RECOMMENDATION, AND ACTUAL HOME CARE VISIT

TABLE 2

NUMBER OF PATIENTS IN EXPERIMENTAL GROUP WITH A FUNCTIONAL DEPENDENCY, HOME CARE RECOMMENDATION, AND ACTUAL HOME CARE VISIT

Figure 2. Percentage of health care services used post-hospitalization.

Figure 2. Percentage of health care services used post-hospitalization.

Experimental and Control Groups' Problems at Home (Research Questions No. 2 and No. 3)

Pain (40%) and activity/exercise problems (33%) were the two major problems of patients experienced at home during the 30 days after discharge. Other problems experienced by 15% to 24% of the sample were with medications, diet, anxiety/stress, depression, and other medical/health conditions. There was no difference between groups in the mean number of patient problems at home post-discharge (mean = 1.6; SD 1.3; range = 0 to 6). The severity of these problems was not measured.

Table

TABLE 3SIGNIFICANT RELATIONSHIPS BETWEEN HEALTH OUTCOMES AND SELECTED VARIABLES USING LOGISTIC REGRESSION

TABLE 3

SIGNIFICANT RELATIONSHIPS BETWEEN HEALTH OUTCOMES AND SELECTED VARIABLES USING LOGISTIC REGRESSION

Experimental and Control Groups' Use of Health Care Resources (Research Question No. 3)

There were no significant differences between groups in the percentage of patients who were rehospitalized or who made unplanned visits to the emergency room or physician within 30 days of the hospital discharge (Figure 2). Patients in the experimental group reported significantly more visits by health care professional services such as speech therapy, physical therapy, and social services (χp 2 = 7.7; p < .005); community rehabilitation services (χp 2 = 7.5; p < .005); and home care equipment (χp 2 = 6.0, p < .01). No differences were noted in relation to the use of other home care services or community services such as meals-onwheels, transportation, or hospice.

Variables Within the Experimental and Control Groups Related to Health Care Resources Use (Research Question No. 4)

Logistic regression was used to examine relationships between the use of health care services within 30 days after discharge and the total scores for functional abilities, social support, severity of illness scores, and the number of patient problems. Significant relationships are shown in Table 3. As functional dependency and the number of patient problems increased within the experimental group, so did rehospitalization and emergency room visits. For patients with a total functional score of 5 (needs little assistance), the estimated likelihood of rehospitalization was 18%, whereas patients with a score of 15 had a 31% likelihood of rehospitalization. Patients with a score of 30 (totally functionally dependent) had a 58% estimated likelihood of rehospitalization.

Patients in the total sample who experienced four health management problems after discharge had a 37% estimated likelihood of rehospitalization. They had a 54% estimated likelihood with six problems. A higher number of patient health management problems was also significantly related to more use of home care services. For example, the likelihood of using home care services was 41% with three problems and 64% with six problems.

DISCUSSION

In this study, the patients' total functional score was the best predictor of rehospitalization. The hospital severity of illness score was not a predictor. The more dependent the patient was on others for ADLs, the greater the risk of their being rehospitalized within 30 days. This finding is consistent with reports of earlier studies (Evans & Hendricks, 1993; Parfrey et al., 1994) and strongly supports the use of tools such as the DPQ for assessing elderly patients' post-discharge care needs.

The control group in this study had a greater percentage of males with cardiovascular disease. Because the control group was discharged without the use of a functional assessment tool, data about their functional dependencies are not available. However, the assessment of functional dependencies in the experimental group indicated that patients with cardiovascular disease were significantly less functionally dependent than other diagnostic groups. With more males in the control group, one can hypothesize that more wives were also more available as family caregivers. Thus, the rehospitalization and resource use data in the control group may have been decreased by the diagnostic and gender differences of the control and experimental groups.

In the experimental group of 202 patients, 56% were totally dependent and 57% were partially dependent on others for assistance with at least one basic ADL. A larger percentage were dependent on others for IADLs, especially household duties. This information further verifies the care demands placed on family caregivers of elderly patients and supports the need to involve family caregivers in the planning.

Post-discharge problems of patients were also significantly related to use of health care services. Patients identified pain and impaired activity levels as the most common problems. Findings about post-discharge problems were limited to number of problems without measurement of the severity of the problem and how it may have impacted the patient and family caregiver.

In this study, the nurse/social worker teams recommended home care services based on their assessments of patients' functional, social, and environmental needs. However, their recommendations and the implementation of their recommendations were limited by their experience about what would be reimbursed and by the lack of control over physician follow up on their recommendations. The number of recommended referrals by nurses/social workers was small considering the degree of functional dependencies of patients in the experimental group. This finding may have been affected by the strong cultural value of family caregivtng rather than formal caregiving. The number of home care services actually received was less than the number recommended by nurses and social workers, suggesting a need for further study of more efficient and effective ways to involve and communicate with the physician. The difference in the number of recommended and actual home visits also raises questions about the types of services currently reimbursable, as well as nurses' and social workers' lack of authority to order reimbursable home care services.

The lack of significant differences between the experimental and control group use of acute health care resources may have been related to the limited referral follow up for the experimental group, as well as the gender and diagnostic differences already mentioned. Also, the 30-day follow up may have been too short of a time period for measuring rehospitalization rates associated with complications from functional disabilities.

IMPLICATIONS FOR PRACTICE

Assessment of functional performance is an essential component of the comprehensive assessment needed for the effective discharge planning of elderly patients. With the increasing number of elderly family caregivers, it also may be important to evaluate the caregiver's functional performance. Based on the findings of the types of patient problems experienced in the post-discharge period, health care professionals need to implement measures to prevent further deterioration of elderly patients' functional abilities during hospitalization, particularly the ability to ambulate. Specific instructions about activity, exercise, and safety in the home should be a routine part of the discharge teaching program. The findings of this study support the need for more effective teaching about managing pain at home. In general, it appears that health care professionals could benefit from additional information about geriatric care and approaches to prevent deterioration.

While all health professionals have a responsibility in discharge planning, assigning specific nurse/social worker teams to coordinate the planning may be an effective and cost-efficient approach. Establishing specific processes for communicating discharge recommendations to physicians and home care agencies may enhance implementation of the discharge plan.

Functionally impaired patients tend to fall between the cracks of the reimbursement system. They may not qualify for the reimbursable skilled home care services, but they may not need to be admitted to a nursing home. Alternative health care services need to be explored. Such things as day hospitals, day rehabilitation centers, and hospitalbased extended care facilities may provide alternative sites for the needed services for elderly patients. Such services would ease the burden of the family caregivers and may decrease the risk of rehospitalization.

The findings of this study support the participation of the patient and family caregiver in discharge planning and decision making. The goal must be to maintain elderly patients' maximum level of functioning and comfort as they move from one level of care to another. Achieving this goal not only will improve the quality of care but also will decrease the inappropriate use of costly acute health care services.

REFERENCES

  • Anderson, M.A., & Helms, L. (1993). An assessment of discharge planning models: Communication in referrals for home care. Orthopedic Nursing, 12(4), 41-49.
  • Bull, MJ. (1990). Factors influencing family caregiver burden and health. Western Journal of Nursing Research, 12(h), 758-776.
  • Bull, MJ. (1992). Managing the transition from hospital to home. Qualitative Health Research, 2(1), 27 -41.
  • Bull, MJ. (1994). A discharge planning questionnaire for clinical practice. Applied Nursing Research, 7(4), 193-207.
  • Evans, R.L., & Hendricks, R.D. (1993). Evaluating hospital discharge planning: A randomized clinical trial. Medical Care, 31(4), 358-370.
  • Fethke, CC, Smith, I.M., & Johnson, N. (1986). Risk factors affecting readmission of the elderly into the health care system. Medical Care, 24(5), 429-437.
  • Haddock, K.S. (1991). Characteristics of effective discharge planning programs for the frail elderly. Journal of Gerontological Nursing, 17(7), 10-14.
  • Haddock, K. (1994). Collaborative discharge planning: Nursing and social services. Clinical Nursing Specialist, 8(5), 248-252.
  • Henderson, M.C. (1992). Families in transition: Caring for the rural elderly. Family Community Health, 14(4), 61-70.
  • Hirsch, C.H., Sommers, L., Olsen, ?., Mullen, L., & Winograd, CH. (1990). The natural history of functional morbidity in hospitalized older patients. Journal of the American Geriatrics Society, 38, 1296-1303.
  • Jackson, M.F. (1994). Discharge planning: Issues and challenges for gerontological nursing: A critique of the literature. Journal of Advanced Nursing, 19, 492-502.
  • Johnson, N., & Fethke, C. (1985). Postdischarge outcomes and care planning for the hospitalized elderly. In C. McClelland, K. Kelly, & K. Buckwalter (Eds.), Continuity of care: Advancing the concept of discharge planning (pp. 229-240). Orlando, FL: Grune & Stratton.
  • Kane, R.L., Riegler, S., Bell, R., Potter, R., & Koshland, G. (1983). Predicting the course of nursing home patients: A progress report (Report No. N-1786-NCHSR). Santa Monica, CA: Rand.
  • Kelly, CK., Huber, D.G., Johnson, M., McCloskey, J.C, & Maas, M. (1994). The medical outcomes study: A nursing perspective. Journal of Professional Nursing, 10, 209216.
  • National Center for Health Statistics. (1991). Prevention profiles, health, United States 1990. Hyattsville, MD: Public Health Service.
  • Naylor, M.D. (1990). Comprehensive discharge planning for hospitalized elderly: A pilot study. Nursing Research, 39(3), 156-161.
  • Parfrey, P.S., Gardner, E., Vavasour, H., Harnett, J.D., McManamon, C, McDonald, J., & Dawe, J. (1994). The feasibility and efficacy of early discharge planning initiated by the admitting department in two acute care hospitals. Clinical Investigations in Medicine, 17, 88-96.
  • The Pittsburgh Research Institute. (1992). Patient management categories. Pittsburgh, PA: Author.
  • Rudberg, M.A., Sager, M.A., & Zhang, J. (1996). Risk factors for nursing home use after hospitalization for medical illness. The Journals of Gerontology, 15(5), Ml 89-Ml 94.

TABLE 1

BACKGROUND VARIABLES OF PATIENTS

TABLE 2

NUMBER OF PATIENTS IN EXPERIMENTAL GROUP WITH A FUNCTIONAL DEPENDENCY, HOME CARE RECOMMENDATION, AND ACTUAL HOME CARE VISIT

TABLE 3

SIGNIFICANT RELATIONSHIPS BETWEEN HEALTH OUTCOMES AND SELECTED VARIABLES USING LOGISTIC REGRESSION

10.3928/0098-9134-19980501-08

Sign up to receive

Journal E-contents