| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
1 Research Institute on Aging of the Hebrew Home of Greater Washington, Rockville, Maryland.
2 George Washington University Medical Center, Washington, D.C.
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionAppendix 1References |
|---|
Methods. Six male physicians and 3 female nurse practitioners completed questionnaires that described the medical decision-making process for 70 residents of a large nonprofit nursing home.
Results. Hospitalization was the most frequently cited treatment considered and chosen; family members were involved in 39% of decisions, and nurses were involved in 34%. The most important considerations in making a decision were reported to be the resident's quality of life, the relative effectiveness of the treatment options, and the family's wishes. The levels of importance ascribed to the considerations were related to the physician's identity, specific resident characteristics (such as estimated life expectancy), and communication between the physician and resident (such as sharing knowledge of family wishes).
Conclusions. The decision at the time of a status change event involves multiple conditions, multiple considerations, and multiple treatment options, and tends to result in either an active route, such as hospitalization, or a passive one, such as comfort care. The impact of the individual physician and the physician–resident relationship on this process deserves further investigation.
THIS paper provides a first step in the investigation of medical decisions in the nursing home. Such decisions frequently involve interplay between medical/nursing personnel, family, and advance directives (Figure 1). When the resident is unable to represent himself/herself, advance directives provide a focal representation, and as such influence the family and the formal caregivers, usually a nurse and a physician. The real interaction between formal and informal caregivers, together with their implicit interaction with the advance directives, dictates the actual decision. However, the relative role of each of these players in the decision-making is as yet unclear.
|
|
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionAppendix 1References |
|---|
An interview ascertained the physician's personal opinion regarding a range of questions relating to the status change event and the decision-making process. Three of the six physicians completed nine questionnaires each (12.9%), and the three other physicians completed 1 (1.4%), 13 (18.6%), and 24 (34.3%) questionnaires, respectively. Two of the nurse practitioners completed one questionnaire each (1.4%), and the third completed three questionnaires (4.3%). The questionnaire contained both open- and close-ended questions in five general areas (Table 1 and Appendix 1). It included both quantitative and qualitative portions (2). A research staff member usually contacted the physician the day after the status change event, when the event appeared on internal nursing home reports. The actual time of the interview varied from that same day to several days later. Based on the questionnaire results, we calculated the number of passive treatments (e.g., comfort care) considered and chosen as well as the number of active treatments (e.g., medication, hospitalization) considered and chosen. This study was approved by our institutional review board.
|
The Role of Family and Resident Wishes
The Impact of Short Life Expectancy
Determinants of Physician Satisfaction
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionAppendix 1References |
|---|
Sixty-eight percent of the residents had an estimated life expectancy of greater than 3 months, whereas 10% had a life expectancy of less than 3 months. The physicians did not know the life expectancy for 22% of the residents.
Nature of status change event.-- The most common incidents during status change events were trouble breathing (29%), aspiration/pneumonia (11%), fracture (11%), and hypotension (10%) (Table 2). Many of the residents experienced multiple incidents during a status change event. The residents averaged 1.5 incidents per person, with 65.2% having 1 incident, 26.1% having 2 incidents, 7.2% having 3 incidents, and 1.4% having 4 incidents. However, multiple incidents could be related. For example, chest pain and trouble breathing could be rated as two separate incidents yet have the same underlying condition. In 63% of the cases, residents had a history of medical problems related to these events.
|
Treatments considered and chosen.-- Hospitalization was the most frequently cited treatment considered and chosen (Table 3). It was considered for 76% of the cases and chosen for 39%. Medication was considered for 45% of cases and chosen for 36%. Diagnostic testing, although considered for 33% of the cases, was chosen for only 18%. Comfort care was considered in 28% of the cases but was chosen for only 13% of the cases. Observation was considered for 27% of cases and chosen for 18% of cases. On average, there were 2.7 treatments considered (range 1–7 treatments) and 1.6 treatments chosen (range 0–4 treatments).
|
Persons involved in making the care decision.-- When asked who else was involved in making the medical decision, physicians indicated that family members (son, daughter, wife, niece, nephew, sister, and son-in-law/daughter-in-law) were most frequently involved (42%), followed by nurses (34%), other physicians (9%), social workers (2%), and others, for example, a resident's friend or rabbi (5%); for 29%, no one else was involved. The values total more than 100% because for some cases, more than one other person was involved (N = 66; four cases were missing data due to an unanswered section on the questionnaire).
Considerations in making treatment decisions.-- The most important considerations in making a decision were reported to be the resident's quality of life, the relative effectiveness of the treatment options, and family wishes (Table 4). Cost of alternate treatments and the resident's wishes, which were frequently unavailable, were least important. However, the physician reported being at least somewhat familiar with the wishes of only 21 residents. For these, there was a near-significant correlation between familiarity and importance attributed to residents' wishes (r =.45, p =.05).
|
Evaluation of the decisions.-- The physicians evaluated their feelings about the decision as very positive for 18% of the residents, positive for 45%, comfortable for 27%, and indifferent for 3%. Ratings were somewhat negative and very negative for 2% and 5%, respectively. For the negatively rated decisions, physicians would have wanted less treatment (three decisions) or more treatment (one decision). When asked to evaluate their perceptions of the family caregiver's feelings about the decision, the physicians assessed it as very positive in 15% of the events, positive in 32%, comfortable in 19%, and somewhat negative in 3%. They could not evaluate the family's feelings in 31% of the events (N = 68; two cases were missing data due to an unanswered section on the questionnaire).
In evaluating the medical impact of the treatment, physicians expected two-thirds (66%) of the residents who were alive to improve (relative to their condition immediately after the event), expected treatment to limit deterioration for 9%, and expected treatment would not make a difference for 14%. The physicians were unsure about the impact of treatment for the other residents (11%) (N = 56; 14 cases were missing data due to an unanswered section on the questionnaire).
When trying to evaluate the patient's perspective, the physicians were asked, "If the patient could tell you, what do you think he/she would want as his/her treatment?" For over half of the residents (63%), the physicians thought that the resident would have chosen the same action as was taken. Physicians could not evaluate the resident's point of view in another 22% of the cases. For others, the physicians thought that the resident would prefer less treatment (5%) or pain relief (5%), as well as specific treatments, such as tube feeding or stitches (5%).
When asked what treatment options they would have wanted to attempt if they were the patient in question (in his/her condition and age), for the majority of cases (72%) physicians would have wanted the treatment provided. In more than one-fifth of the cases (21%), they would have wanted less treatment (e.g., no treatment, to be left alone, comfort care, treatment in the nursing home), and in 3% of cases, they would have wanted more treatment (e.g., surgery and amputation). Other treatments, opted for in only one event each (4%), were tube feeding and temporary feeding. In one case, the physician said it was too personal a question.
Relationship Between Process and Outcome Variables
We evaluated the following hypotheses.
The role of family's and resident's wishes.--
The impact of short life expectancy.--
Determinants of physician satisfaction.--
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionAppendix 1References |
|---|
In rating the importance of considerations for making the decision, physicians rated quality of life as the highest, and clearly higher than prolongation of life. Families' wishes were also rated as important, whereas residents' wishes were given lower importance, probably because they were most often not available. For the three considerations that tended to be rated as highest (i.e., quality of life, effectiveness of treatment options, and family's wishes), these considerations tended to be rated as either very important or not important rather than given intermediary ratings. In contrast, less important considerations received low ratings more often. Nevertheless, all considerations received ratings of "very important" for at least some of the residents. These findings show that considerations related to multiple factors (Figure 2) were viewed as important, including personal, medical, and autonomy factors, without any single factor dominating the physicians' considerations.
Considerations were related to a resident's status and to the physician–resident relationship. When relatively imminent death was anticipated, prolongation of life was rated as less important. Similarly, residents' cognitive impairment resulted in higher importance attributed to relatives' wishes. Physicians' level of familiarity with wishes also was associated with the importance attributed to those wishes.
We found significant differences among individual physicians in the ratings of the factors that impact their decisions and in the number of treatments considered and chosen, highlighting the importance of individual physician style of practice and of reporting this practice. This finding is consistent with the study of Eisemann and colleagues (3) of vignette-based decision-making for incompetent elderly persons by physicians of three European countries. They found significant differences among the countries in treatment decisions and in the level of importance ascribed to considerations, such as patient's or family's wishes, in making a decision. Similarly, Kellogg and Ramos (4) found that Do Not Resuscitate decisions were more likely to be made by specific physician and social worker teams than by others.
The physicians were satisfied with the decision-making process in the vast majority of the events and thought that the family members and the nursing staff were pleased with it. Physicians expressed satisfaction with the decision even when they would have opted for a different treatment for themselves. However, they tended to feel even more satisfied when the decision matched their preferences for treatment for themselves had they been in the resident's condition. Their perceptions of the concordance between the hypothesized resident's preferences and the treatment provided were not significantly related to their own satisfaction with the decision. Conversely, there were a few cases in which they were displeased with the decision when the decision was different from how they would have wanted to be treated. Perceived family satisfaction was related to the importance of family wishes in making the decision.
Future research needs to examine several issues concerning the considerations, including:
|
Appendix 1 |
|---|
|
TopAbstractMethodsResultsDiscussionAppendix 1References |
|---|
Background
Physician name:
Dr. A Dr. B Dr. C Dr. D Dr. E Dr. F
NP A NP B NP C
Physician's Characteristics
Gender: Male Female
Years of experience in geriatrics:
Years at this nursing home:
Research Assistant name:
Resident name: ID#:
Unit: Date:
1. Is this your resident?
Yes No
2. If yes, then how long have you been treating the resident?
Please rate how familiar you are with each of the following items.
3. How familiar are you with the family's wishes concerning the resident's care?
Not familiar Somewhat familiar Familiar Very familiar
4. How familiar are you with this resident's wishes concerning care?
Not familiar Somewhat familiar Familiar Very familiar
Nature and severity of status change event
5. What occurred in the recent event?
Infection CVA Aspiration/pneumonia Trouble swallowing Fracture
Trouble breathing Not eating/weight loss Passed out Chest pain Other (specify in 5a)
5a.
6. Before the acute event, would you have estimated the resident's life expectancy to be less than 3 months?
Yes No Don't Know
7. Is/Was the change acutely life threatening? **note: this question was deleted from the questionnaire pertaining to residents who died by the time of the interview
Yes No Don't Know
8. Does the resident have a history of medical problems related to this event/change?
Yes No Don't Know
Decision-making process
9. What options did you consider for treating this situation?
Medication Comfort Care Rehabilitation Observation Diagnostic tests Hospitalization Surgery IV Amputation Dialysis Blood transfusion
Radiation Other (specify in 9a.)
9a.
10. What options did you choose for treating this situation?
Medication Comfort care Rehabilitation Observation Diagnostic tests Hospitalization Surgery IV Amputation Dialysis Blood transfusion
Radiation Other (specify in 10a.)
10a.
11. Who else was involved in making the care decision regarding this resident?
Nurse 11a. specify who
Social worker 11b. specify who
Other physician 11c. specify who
Family member 11d. specify who
Other 11e. specify who
No one
Considerations in making treatment decisions
12. Why did you choose the above treatments?
Please rate how important the following considerations were in your treatment decisions, using the following scale.
Relative effectiveness/futility of treatment options
Not important at all Somewhat important Important Very important
Potential liability issues
Not important at all Somewhat important Important Very important
General treatment practice for this condition
Not important at all Somewhat important Important Very important
Cost of alternative treatments
Not important at all Somewhat important Important Very important
Resident's quality of life
Not important at all Somewhat important Important Very important
Family's wishes
Not important at all Somewhat important Important Very important
Resident's wishes
Not important at all Somewhat important Important Very important
Prolongation of resident's life
Not important at all Somewhat important Important Very important
Resident's advance directive **note: this question was added to the questionnaire late, and thus is not addressed in this paper
Not important at all Somewhat important Important Very important
Your own preference if you were in the resident's place
Not important at all Somewhat important Important Very important
Other (specify in 12a.)
Not important at all Somewhat important Important Very important
12a.
Evaluation of the decision
13. What is the likelihood that this treatment will improve the patient's condition as compared to the patient's condition immediately after the status change event?
Likely to improve Likely to limit deterioration Unsure Not likely to make any difference
Other (specify in 13a.)
13a.
**note: for the questionnaire pertaining to residents who died, the previous question was changed to:
13. Was the patient's death caused by the:
Status change event Treatment Both Neither Chronic illness Lack of treatment Don't know
14. How does the family feel about the decision?
Very positive Positive Comfortable Indifferent Somewhat negative Very negative Don't know
15. How do you feel about the decision?
Very positive Positive Comfortable Indifferent Somewhat negative Very negative Don't know
16. If the patient could tell you, what do you think he/she would want as his/her treatment?
(This was coded as: less treatment, same, more treatment, other treatment)
17. If you were this patient (in his/her condition and age), what treatment option would you most want attempted?
(This was coded as: less treatment, same, more treatment, other treatment)
18. Additional comments
|
Acknowledgments |
|---|
Address correspondence to Jiska Cohen-Mansfield, PhD, ABPP, Director, Research Institute on Aging, Hebrew Home of Greater Washington, 6121 Montrose Rd. Rockville, MD 20852. E-mail: cohen-mansfield{at}hebrew-home.org
Received April 15, 2002
Accepted October 7, 2002
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionAppendix 1References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. S. Allen, L. D. Burgio, S. E. Fisher, J. Michael Hardin, and J. L. Shuster Jr., Behavioral Characteristics of Agitated Nursing Home Residents With Dementia at the End of Life Gerontologist, October 1, 2005; 45(5): 661 - 666. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
