| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
a Office of Geriatric Medicine, University of Cincinnati Medical Center, Ohio.
b Scripps Gerontology Center, Miami University, Oxford, Ohio.
Shahla Mehdizadeh, Scripps Gerontology Center, Miami University, Oxford, OH 45056 E-mail: mehdizk{at}muohio.edu.
Decision Editor: William B. Ershler, MD
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Using both Medicare and Medicaid records from a sample of dually eligible elderly people in Ohio, we identified patients receiving antibiotic prescriptions in the nursing home and measured the frequency of nursing home physician visits and the hospital transfer rate.
Results. Among the study sample (N = 1306), two thirds experienced a total of 3685 episodes of infections. Just under 5% of the sample were hospitalized as a result of the infection. In one third of the episodes, physicians saw the resident in person within 5 days (before or after) of the initiation of the medication. The hospital transfer rate was slightly higher (7% vs 3.5%) for those patients directly evaluated by a physician before receiving the prescription.
Conclusions. A majority of prescriptions were written without direct physician examination, raising key questions about practice patterns and the effect on patient care and costs.
EACH year more than 25% of patients living in nursing homes are transferred to the hospital or emergency room for evaluation and treatment (1)(2). In addition, recurrent transfers occur in some patients, leading to the "ping-pong" pattern between nursing homes and hospitals (2). The most frequent reasons for transfer to the hospital are infections, specifically of the urinary and respiratory tracts (3)(4)(5)(6)(7). With over 1.5 million people living in nearly 17,000 nursing homes in the United States, these frequent transfers may have an adverse impact on the quality and cost of care these patients receive.
The risk for iatrogenic problems resulting from these frequent transfers is high (1)(8)(9). Communication between the hospital and the nursing home is often poor. A different team of physicians and nurses usually directs the individual's care at each site. Important information on symptoms, baseline functioning, ongoing or new treatments, and advance directives does not always come to the hospital with the patient. Many of these patients experience confusion aggravated by the transfer, thus further complicating assessment and treatment in the hospital. Overtreatment, undertreatment, and misdiagnosis are often the result of these problems.
Several authors have addressed the economic, policy, and social influences on the nursing home-hospital interface (2)(10)(11)(12)(13). The patient's attending physician in the nursing home is responsible for the decision to transfer a patient to the hospital. Financial and time constraints are frequently given as explanations for infrequent and short visits by physicians to nursing homes (9)(14)(15). Limited physician involvement in nursing home care is a particular concern when evaluating the quality of assessment and management of acute medical problems. Clinical experience and preliminary research suggest that many decisions to transfer occur without a physician first examining the patient (1)(11)(16)(17)(18)(19).
This study used a longitudinal design to assess infections in Ohio nursing homes, physician involvement in care, and the transfer rate to hospitals. The study utilized an existing sample of elderly persons that were Medicare and Medicaid eligible at the time they were evaluated by the Medicaid preadmission review program. Our intent was to learn more about the occurrence and management of infections in Ohio nursing homes as an important measure of quality and cost of long-term care.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
To examine infection patterns in nursing homes, we utilized data for each individual between January 1994 and December 1996. Depending on the date of preadmission review, in some cases, data included up to 12 months prior to preadmission review. Our focus for this work was on those sample members that experienced infections while they were residents of a nursing facility. The unit of analysis was either the individual or an episode of infection.
The data for this study came from several sources. Demographic and functional status information came from an in-person and/or paper review of clients at the time of preadmission review. Hospital and physician visits were captured from Medicare claim files. Prescription medication and nursing home use were recorded from Medicaid claims data, and the information surrounding the clients' death (date and place) came from the Ohio Department of Health, Bureau of Vital Statistics. Nursing home residents who were experiencing an infection were identified by the use of an antibiotic prescription. The Ohio Medicaid formulary was reviewed and systemic antibiotics that could be delivered by oral, intramuscular or intravenous means were noted. Topical and ophthalmologic preparations were excluded.
Next we also examined how each episode of infection was managed. Using Medicare data on physician visits, the date of the first prescription in each episode was compared with the dates of physician visits that occurred in the nursing home or physician's office. If a visit was within 5 days (before or after) of the prescription's date it was considered to be related to that episode of infection. All clients that manifested an infection were then examined against Medicare hospitalization claims. If a hospitalization had occurred within 15 days of the first prescription in each episode, it was assumed to be related to the infection. All clients were examined against mortality data from the vital statistics records. If the death occurred within 45 days of the first prescription date, it was presumed to be linked to the episode of infection.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Of the 1306 sample members that spent some time in a nursing home, 67% (874 patients) had received an antibiotic prescription in a nursing home, and 3685 episodes of infection were identified (see Fig. 1). Table 1 reviews the frequency of infection episodes among the 874 patients. Seventy-five percent of the 874 patients had five or fewer infections during the study period, and about four of ten patients had either one or two episodes. Less than 5% of the sample had 12 or more episodes. Nine of ten patients received one or two antibiotics during each of the 3685 episodes of infections. Less than 1% of the patients received six or more antibiotics per episode (Table 2 ). Although it is not unusual for physicians to adjust antibiotics during the management of infections, such as pneumonia or urinary tract infections, four out of five patients received only one prescription per episode. Six persons had more than 40 antibiotic prescriptions during the 3 years of study. These six patients were excluded from the study. For the 20% of the sample using multiple medications, several factors could account for this outcome: (i) an antibiotic might be started empirically, and when definitive culture and sensitivity results are obtained, an adjustment is made; (ii) if a patient fails to respond to initial therapy, a change to an alternative drug is not unusual; or (iii) a patient may be treated initially with intramuscular or intravenous therapy and then changed to an oral medication.
|
|
|
In just under 5% (170) of the episodes of infection, nursing home residents were transferred to the hospital. Of the episodes of infection associated with a physician visit, about 7% (83) resulted in a hospital transfer. When no physician visit occurred, the hospital transfer rate was 3.5%. This difference may be the result of the following: (i) physicians chose to see the sicker patients; (ii) when physicians actually examined the patients, they became more aware of the severity of the illness; or (iii) the physicians had more direct contact with the nursing home staff or the patient's family, who encouraged the transfer.
During the study period, 43% of those residents presumed to have an infection died. However, only half of those deaths were attributed to the infections. The mean time between the beginning of an episode of infection and death was 15.7 days. Of the 140 patients that were transferred to the hospital, about 40% died within 45 days. Half of the patients hospitalized and seen by a physician prior to the transfer died, compared to just under one third of those patients hospitalized and not seen by a physician.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The availability of broad-spectrum oral and intramuscular antibiotics has greatly facilitated the treatment of serious infections in the nursing home. Supportive therapies and procedures (e.g., oxygen, intravenous of subcutaneous fluids, laboratory support) also are widely available. In this study, we observed that the vast majority of patients received three or fewer antibiotic prescriptions per episode. However, 3% of patients received four or more prescriptions per episode, and four patients received ten or more prescriptions during the 21-day infection period. These few cases with high numbers of prescriptions cannot be easily explained and represent an opportunity for quality review.
The nursing home and home setting are increasingly sites for the delivery of acute and subacute care, but unlike the office and hospital settings, physicians have not reorganized their practices to arrange scheduled time for the evaluation of patients in these settings. Patients in the hospital are seen daily, and it is unusual for a physician to prescribe antibiotics over the telephone for serious infections before the patient is examined in the office. It is likely that the organization of physician practice within the nursing home shapes optimal physician performance.
The calculated hospitalization rate in the 1991 Cincinnati study was 16.0% annually (20). Comparative transfer rates in other studies for nursing home patients for all acute medical problems (e.g., infections, cardiovascular illness, fractures) ranged from 21.0% to 55.0%, with infection being the most frequent acute medical problem (1). In these studies, infections accounted for approximately one half of nursing home-to-hospital transfers. The lower transfer rate in this study may have resulted from the following: (i) we may not have captured all hospitalizations through our definition, which required an antibiotic prescription within 15 days of a hospitalization; (ii) our subjects may not have been as ill as in other samples; (iii) practice patterns are changing and physicians are becoming more comfortable managing sick patients in the nursing home; (iv) the study sample was not residing in a nursing home for the entire study period; and (v) the dual eligible status of our patients resulted in fewer hospital transfers. A significant barrier to nursing home management of sick patients is the cost of delivering expensive care in the nursing home. For patients with Medicare coverage only, without a prior 3-day hospital stay, the costs of expensive medications and treatments are the patient's responsibility. However, for dual eligible patients, the Medicaid program assumes these costs, allowing for acute and subacute care to be delivered to this population in the nursing home setting without a hospital transfer.
In summary, we have documented that physician management of infections in nursing homes is largely directed over the telephone. We cannot assess the impact of lack of direct physician examination of the patient on patient outcomes, but this type of patient management would not be considered appropriate in the hospital or outpatient setting. The evaluation of an acutely ill nursing home patient should take place in the nursing home, in a timely manner, by a physician or midlevel practitioner familiar with the patient's clinical condition. These data could encourage third-party payers to create reimbursement incentives for treatment of acute illness in long-term care facilities. In fact, the increasing numbers of older adults in managed Medicare has helped to shift financial incentives for some patients by eliminating the need for the 72-hour hospital stay before Medicare can provide reimbursement for subacute services in the nursing home. With financial incentives, nursing facilities and physicians could provide quality, cost-effective, acute care to appropriate patients.
|
Acknowledgments |
|---|
Received November 11, 1999
Accepted March 29, 2000
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. A. Caplan, A. Meller, B. Squires, S. Chan, and W. Willett Advance care planning and hospital in the nursing home Age Ageing, November 1, 2006; 35(6): 581 - 585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Gasquet, S. Dehe, P. Gaudebout, and B. Falissard Regular Visitors Are Not Good Substitutes for Assessment of Elderly Patient Satisfaction With Nursing Home Care and Services J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2003; 58(11): M1036 - 1041. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Morley Editorial: Hot Topics in Geriatrics J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2003; 58(1): M30 - 36. [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Mehdizadeh Health and Long-Term Care Use Trajectories of Older Disabled Women Gerontologist, June 1, 2002; 42(3): 304 - 314. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
