Research Proposal Running head: RESEARCH PROPOSAL

1

Research Proposal Kathryn Keiper Duke University School of Nursing Research Methods N307-01 Dr. Carla Gene Rapp Apr 20, 2005

Research Proposal Table of Contents

2

Introduction……………………………………………………………………………….3 Review of the Literature…………………………………………………………………..4 Purpose……………………………………………………………………………………8 Methods……………………………………………………………………………………9 Design and research questions…………………………………………………….9 Sample……………………………………………………………………………..9 Survey instrument……………………………………………………..…………10 Research procedures……………………………………………………………..10 Informed consent and IRB approval……………………………………………..11 Strengths and Weaknesses……………………………………………………………….11 Timeline and Cost Considerations……………………………………………………….12 Conclusion……………………………………………………………………………….13 Concept Model………………………………………………………………..Appendix A Sample Questions……………………………………………………………..Appendix B References……………………………………………………………………………….18

Research Proposal Research Proposal Introduction Intensive care units (ICUs) were designed to provide highly skilled, lifesaving nursing care to viable patients with acute illnesses or injuries. Patients with chronic and/or terminal illness were not expected to be admitted to these units, with the possible exception of acute exacerbations of reversible complications. Patients whose care needs changed from curative to palliative were intended to be transferred out of critical care to patient care environments more suited to end-of-life care. However, as more patients become “chronically critically ill”, critical care nurses are being asked more often to provide care to patients on their deathbeds (Puntillo et al., 2001). Deciding which ICU patients are actually dying remains an extremely inexact science, and the transition to palliative care is not one easily made. ICU mortality rates are as high as 69% (Puntillo et al., 2001); almost 20% of Americans die in intensive care units (Hodde, Engelberg, Treece, Steinberg, & Curtis, 2004). The majority of patients who die in ICU have had DNR orders written, many of them within a day or two of their deaths. There are a number of reasons these patients are not transferred out: it may be too disruptive to the patient and/or family; there may be no appropriate bed available; or the level of care may still be such that a med/surg or hospice-type floor is not equipped or staffed to handle it (Puntillo et al.). Evidence shows that end-of-life care in ICUs is highly inconsistent, indicating that caregivers are not in agreement on how best to care for this patient population. There is evidence that dying patients experience inadequate relief of pain and other undesirable symptoms, and also that their wishes concerning end-of-life care are not always taken into account (Hodde et al.)

3

Research Proposal Most ICU nurses did not enter their specialty because they enjoyed caring for patients that have no hope of recovery, and now they find themselves being called on to provide end-oflife care in what is often the worst possible setting. It has been this author’s experience that

4

many ICU nurses feel that they have not provided their patients with the best possible death for a variety of reasons. Many feel unprepared to give end-of-life care if it does not involve resuscitation, and feel unsupported in their attempt to provide quality palliative care in the hightech environment of the ICU. This author is proposing a research study that would investigate the experiences and attitudes of critical care nurses about providing end-of-life care in the ICU environment. Review of the Literature In her article Nursing Knowledge for the 21st Century: Opportunities and Challenges (2000), Ada Sue Hinshaw identifies end-of-life research as one of the major fields of clinical study in which nurses need to be heavily involved. This research needs to include not only approaches to palliative care but also end-of-life decision making for patients and families, identification of transition phases, and what is important to patients and families during their end-of-life experience. In 2004, Bryce et al. conducted a study to determine how much healthy living time people would trade for quality end-of-life care in four areas: pain and discomfort, daily surroundings, treatment decisions and family support. Three-quarters of the subjects were willing to trade an average of 7.5 months of healthy living for a better end-of-life experience, indicating that quality end-of-life care is highly valued. However, while it is clear that patients and families want to have quality end-of-life care, most don’t want it quite yet – thus the transition from curative to palliative care is usually fraught with tension and uncertainty on the part of patient, family, and caregivers.

Research Proposal

5

In 1999, Singer, Martin and Kelner conducted a qualitative study to identify and describe end-of-life issues from the patient’s perspective. They conducted face-to-face interviews with 126 patients with potentially life-ending conditions: dialysis patients, patients with HIV infection, and residents of a long-term care facility. The major issues of concern identified by these patients were adequate pain and symptom relief, inappropriate prolongation of the dying process, having a sense of control, not being a burden to their loved ones, and being able to reinforce relationships before death. It is important to keep these issues in mind when providing end-of-life care, regardless of setting or circumstance. The Robert Wood Johnson Foundation Critical Care End-of-Life Peer Workgroup and fifteen nurse-physician teams from a variety of ICUs in the U.S. and Canada conducted an indepth review of the literature followed by an eighteen-month iterative consensus process in order to identify key end-of-life care domains, quality indicators, and possible interventions/behaviors that could be implemented in intensive care units (Clarke et al., 2003). Based on this process, the following domains were identified: patient- and family-centered decision-making, communication, continuity of care, emotional and logistical support for patients and families, pain and symptom relief, spiritual care, and emotional and institutional support for caregivers. This author plans to use these domains as guidelines within which nurses can be surveyed regarding their attitudes and experiences. There is evidence that suggests that many patients dying in ICUs, and their families, are not satisfied with the quality of end-of-life care that they received. Many report suffering from pain and other unpleasant symptoms, families complain that communication is poor, and conflict between caregivers and family members is all too common (Clarke et al., 2003). A questionnaire regarding satisfaction with patient comfort, communication, and decision-making at the end of

Research Proposal life was administered to family members/surrogate respondents for 767 seriously ill adults who died in the hospital; results showed room for improvement in all areas (Baker et al., 2000). Studies done over the last decade substantiate findings that too many ICU patients die in pain while receiving interventions not desired by themselves or their loved ones. Advance directives have had little effect on the end-of-life care patients receive, and critical care nurses are often

6

frustrated and demoralized by the manner in which their patients are dying (Rubenfeld & Curtis, 2002). It is clear that standards of care must be established for patients who die in ICUs; in order to do that, we must understand current practice patterns, what nurses perceive as barriers to providing quality end-of-life care, and how they think that care can be improved (Prendergast, Claessens, & Luce, 1998). In 2003, a convenience sample of Australian cardiorespiratory nurses participated in focus groups in order to discuss the delivery and environment of end-of-life care (Davidson et al., 2003). Most of these nurses felt uncertain about how to incorporate effective end-of-life care into their daily acute care practice. Four main areas of concern were identified: finding structure and meaning in death from chronic illness, lack of a treatment plan, uneasiness in dealing with death and dying, and lack of knowledge about palliative care methods and resources. Although this sample consisted of a small group of acute care nurses in Australia rather than ICU nurses from the United States, it did identify some broad themes and supports the need for the integration of palliative care into environments that have a curative biomedical focus (Davidson et al.). In late 1999, White, Coyne and Patel surveyed practicing oncology nurses to determine end-of-life core competencies and perceived educational needs, and to compare responses according to respondents’ age, education, practice role and setting, and employment status

Research Proposal (2001). While this survey did not include critical care nurses, it did identify core competencies that need to be included in basic and continuing education curricula and what nurse characteristics need to be considered when planning these programs (White, Coyne, & Patel). Kirchhoff et al. conducted a qualitative study on critical care nurses’ perceptions of endof-life care in the ICU (2000). Four focus groups were held with nurses from four ICUs in two hospitals. Some common themes emerged as to what constituted “good” end-of-life care: adequate control of pain and discomfort, maintenance of dignity, family involvement, a clear prognosis and treatment plan, and continuity of care. Barriers to good care include conflict between and among family members and caregivers, uncertain prognosis, and lack of communication. Some things that would improve end-of-life care include changes in physical

7

layout of the unit, more education about end-of-life care, and better communication. Limitations of this study include its small sample size and the fact that the nurses that volunteered may have been those that actually felt more comfortable with the topic (Kirchhoff et al.). In 2001 a survey was sent out to 3000 members of the American Association of Critical Care Nurses to investigate their knowledge, beliefs and ethical concerns regarding caring for dying patients in the ICU (Puntillo et al., 2001). The survey contained scenarios dealing with pain management, life support issues and euthanasia. The portion of the survey of interest to this author dealt specifically with pain management and nurses’ ability to provide comfort care to patients dying in ICU. Although most respondents felt that they (and their colleagues) are skilled at pain management, both pharmacological and non-pharmacological, 78% still felt that patients did not always receive adequate pain medication. Most felt that communication with physicians regarding end-of-life care is effective, but 17% felt that that it’s highly ineffective or conflicted. The majority of nurses felt that support services for dying patients, their families, and

Research Proposal caregivers is unavailable or inadequate. Limitations of this study included a low response rate (30%), possibly due to the sensitive nature and/or extreme length of the survey, and the fact that only members of a professional organization were questioned. Their responses may not be representative of the majority of critical care nurses (Puntillo et al.).

8

Based on a review of these studies, the author has identified the need for further research into end-of-life care in the ICU. This research needs to be based on issues that have been identified as important to patients and families, concentrating on those areas that most need improvement, such as pain/symptom management, communication, conflict, and decisionmaking. Current practice patterns need to be described, and critical care nurses must identify barriers that prevent them from delivering quality end-of-life care. They also need to define what interventions or changes, individually and organizationally, could help them to overcome these barriers. The study will build on the findings of Kirchhoff et al. (2000) and responses will be correlated with demographic information such as age, educational level, experience, gender, and type/size of unit. Purpose The overarching purpose of the proposed study will be to identify current practice in delivery of end-of-life care in ICU and begin to look at how that care can be improved (see Concept Model – Appendix A). Issues that have been found to be important to patients and families will be stressed, especially those that cause the most dissatisfaction. Topics to be examined will include current end-of-life care delivery methods, barriers to quality care, and suggestions on how to improve end-of-life care.

Research Proposal

9

Methods Design and research questions An exploratory survey design will be used to investigate the previously identified issues. Questions will be divided into the following sections: 1. What are your current practice patterns in providing palliative care to dying patients on your unit, and how effective are they? 2. What are the barriers that keep you from providing quality end-of-life care to patients and their families? 3. What are some things that could be done to improve the quality of end-of-life care provided to patients dying in your unit? Sample Recognizing that various parts of the country may have cultural differences regarding death and dying, and due to practical limitations, the survey will include only nurses in Pennsylvania, New York and New Jersey. A list of all hospitals with ICUs in these three states will be obtained from the American Hospital Association (American Hospital Association [AHA], 2005) and divided according to bed number: less than 200 beds, 200-349 beds, 350-499 beds, and 500 beds or more. A proportional, stratified random sample will be selected from the list; these hospitals will be contacted and asked to provide names and addresses of all nurses providing direct patient care at least half-time in their adult critical care units. Three thousand nurses will be randomly selected from those lists to receive surveys. This sampling method was chosen to ensure proportional representation from hospitals of all sizes. It is economical and not overly time-consuming; it has a low risk of bias. Chance of

Research Proposal

10

sampling error is increased, however, and data analysis will be more complex (LoBiondo-Wood & Haber, 2002). Survey instrument The survey instrument will be a questionnaire based on the literature review – qualitative findings (Clarke et al., 2003; Davidson et al. 2003; Kirchhoff et al., 2000), patient/family issues (Baker et al., 2000; Singer, Martin, & Kelner, 1999) and relevant surveys already done (Puntillo et al., 2001;White et al., 2001). Content validity will be evaluated by experts in the fields of critical care, palliative care and psychometrics. A pilot study will be done using a convenience sample of nurses working in ICUs at four hospitals of varying sizes in the Scranton/WilkesBarre, Pocono, and Lehigh Valley regions of Pennsylvania (approximately fifty nurses). This will be done to determine reliability and validity as well as variance in item response (LoBiondoWood & Haber, 2002). The pilot survey will also contain some open-ended questions for respondents to suggest topics that should be added to the survey. Revisions will be made as necessary after the pilot survey and will be re-evaluated by the consultants. Demographic information will include age, gender, years of experience, educational level, practice setting, and size of hospital. Likert-type responses and other close-ended questions will be used. See Appendix B for sample questions. Research procedures Questionnaires will be mailed to subjects with a stamped, self-addressed return envelope and a cover letter explaining the purpose of the study (to study and improve end-of-life care in ICUs) and guaranteeing anonymity; a reminder card will be sent two weeks later. Completed surveys will be scanned into SPSS (Statistical Package for Social Sciences), which is a software

Research Proposal package for analyzing data (Polit & Hungler, 1999). Responses will also be correlated with demographic data. Informed consent and IRB approval Informed consent means that subjects have received complete information about the

11

research, are able to comprehend it, and are capable of agreeing to participation. When the study involves minimal risk (such as filling out an anonymous questionnaire), informed consent is optional and implied by the fact that the participant filled out the survey (Polit & Hungler, 1999). Studies that involve human subjects must be approved by an Institutional Review Board (IRB) in order to make sure every precaution has been taken to protect the participants. Certain types of research in which there are no apparent risks to human subjects are totally exempt from IRB review; this includes research involving survey procedures such as the proposed study. Exceptions to this exemption would be if human subjects could be identified, if responses could put the subjects at risk for legal action or financial loss, or if the research deals with sensitive aspects of the participant’s behavior (such as sexual behavior or alcohol abuse) (Polit & Hungler, 1999). The proposed study meets the criteria for exemption from IRB review. Strengths and Weaknesses Surveys can provide the researcher with a lot of surprisingly accurate information at relatively low cost (LoBiondo-Wood & Haber, 2002). Stratified random sampling of hospitals ensures that hospitals and adult ICUs of all sizes are represented. This topic is of great interest to many critical care nurses and is becoming something they have to deal with on a regular basis; hopefully this will inspire them to complete the survey and express their opinions on the subject. Limitations include the possibility of sampling error (Polit & Hungler, 1999) due to the sampling method utilized, and the fact that information provided via surveys tends to be a bit

Research Proposal superficial. It is also common to have a low rate of return (30%) on surveys mailed out to

12

subjects – hopefully this will not happen! Threats to internal validity include history – imagine if the Terry Schiavo case was going on at the time the survey was being conducted! That might have some effect on responses. There is also the threat of selection – respondents may be those that are most concerned about end-of-life care, whereas those who do not bother with the survey may not be a bit concerned about the quality of end-of-life care they are providing. Generalizability will be limited to ICU nurses in Pennsylvania, New York and New Jersey, although some of the findings will surely be applicable to ICUs all over the country. It was also be generalizable only to adult ICUs, as no pediatric or neonatal ICUs will be included in the survey. Timeline and Cost Considerations This author would like to receive at least partial funding through a grant from the American Association of Critical Care Nurses; six months will be allotted for application and disbursement of funds. During this time development of the instrument can be taking place. Pilot testing and revision of the survey will take another three months. Selection of the participants will take another three months. Surveys will be mailed out; reminder cards sent two weeks later; collection of the surveys should be complete ninety days after initial mailing. One month will be allotted for data analysis. The study should be complete one year and three months after its inception. Costs that will be incurred during this study include paying the consultants and doing the pilot study for instrument development, the cost of sampling, the costs of printing and mailing the surveys (with SASE) and reminder cards, and the cost of data analysis.

Research Proposal Conclusion Most deaths in the ICU occur under less than ideal circumstances – patients may be

13

intubated, unable to speak or unconscious. Many are in isolation. Visitation may be limited and privacy may be difficult to provide. Family members worry that the patient is in pain or uncomfortable; they fear that the treatments they have consented to are doing more harm than good. They view the ICU environment as high-tech, cold, and impersonal, and often feel that staff members do not treat their loved one as a cherished human being (Kirchhoff et al., 2000). Evidence suggests that many patients dying in the ICU still suffer from pain and other unpleasant symptoms; some studies have shown that patients are less satisfied with pain management in ICU than in other care settings. Families feel that communication is poor and conflict often occurs (Clarke et al., 2003). It is clear that end-of-life care in the ICU setting is in need of some improvement. Few studies have been done on critical care nurses’ reports of caring for dying patients (Kirchhoff et al., 2000). Qualitative studies have shown that acute care nurses are frustrated when they feel they are providing futile, often painful, high-tech care to patients that are clearly dying; they also feel that lack of a palliative plan of care means that patients’ deaths are not as comfortable and “good” as nurses think they should be. In a study done by Kincaid and Powers, more than 50% of nurses described their end-of-life skills as insufficient (Davidson et al., 2003). The evidence suggests that many nurses working in critical care feel uncomfortable and unprepared providing care to dying patients and their families; if care is to improve, some changes need to occur. Current research is leading us towards a more integrated model of critical care, in which palliative care is not postponed until death is imminent. However, the focus on technology,

Research Proposal economic restraints, and lack of data concerning end-of-life care in the ICU are preventing implementation of this new model (Nelson & Danis, 2001). There is no reason palliative and curative care cannot co-exist in critical care units; society’s view of ICUs as places for heroic life-saving technology must be broadened to include viewing them as places where patients sometimes die. With this new perspective on critical care more hospice and palliative care philosophies can be integrated into care provided in ICUs (Puntillo et al., 2001). Even in the high-tech environment of critical care, “rituals of meeting the other, leave-

14

taking, being with, according respect, and providing comfort for the dying and their loved ones open a social, sentient space for meeting particular other human beings. When the person is met and death is acknowledged, death can be faced as a human passage rather than merely a failed attempt to save a life” (Benner, 2001, p. 355). Attentive, compassionate care needs to be provided for dying patients regardless of the setting; critical care nurses are aware of this but face unique challenges in being able to provide this kind of care. The first step in improving the situation is to examine how end-of-life care is being practiced in critical care units and begin thinking about how we can make it better; this research proposal hopes to do exactly that.

Research Proposal APPENDIX A Concept Model

15

Pain control
Physical layout of unit
Lack of privacy
Staffing issues
Physician issues

Symptom relief
Dignity

Barriers

Patient

Control
Closure
Relationships
Emotional support
Spiritual care

Education
Support
Decision making involvement Non-pharm comfort measures
Comp/alter therapies
Continuity of care
Palliative philosophy

Critical Care Nurse
4/24/2005 - v3

Communication

Potential for improvement

Absence of conflict

Family

Emotional support
Practical help
Visitation
Decision making

Research Proposal APPENDIX B Sample Questions Current practice patterns and their effectiveness 1.

16

How often do you think your dying patients’ pain is well controlled? (never, rarely, sometimes, usually, always)

2.

How often do your dying patients suffer from other unpleasant symptoms that are not well controlled, such as air hunger or seizures? (never, rarely, sometimes, often, always)

3.

How effective is communication regarding end-of-life issues between the dying patient’s family and the nursing staff? (totally ineffective, not too effective, moderately effective, quite effective, very effective)

4.

How often does conflict occur between family members and nursing staff? (never, rarely, sometimes, often, always)

5.

How involved does the nursing staff usually get in end-of-life decisionmaking? (not at all, a little, somewhat, quite, very)

Barriers to quality end-of-life care 1. Is the physical layout of your unit conducive to providing the kind of family visitation that should be occurring during the dying process? (not at all, barely, somewhat, quite, very) 2. Do staffing levels make it difficult to provide the level of care necessary at the end of life? (never, rarely, sometimes, often, always) 3. Are physicians supportive in your efforts to provide palliative care to your dying patients? (never, rarely, sometimes, often, always)

Research Proposal 4. Is an attempt made to provide continuity of care as far as nursing staff assignments throughout the patient’s ICU stay? (never, rarely, sometimes, often, always) 5. Are you comfortable giving end-of-life care with family members present at the bedside? (never, rarely, sometimes, often, always) Possible ways to improve end-of-life care 1.

17

Do you feel that you have received enough education regarding how to provide quality end-of-life care? (not at all, somewhat, absolutely)

2.

Does your hospital provide support services for families of dying patients (to provide help with emotional and practical matters)? (not at all, barely, somewhat, often, always)

3.

Does your hospital provide support for caregivers of dying patients? (not at all, barely, somewhat, often, always)

4.

Are you familiar with non-pharmacologic comfort measures and their uses with dying patients? (not at all, barely, somewhat, quite, very)

5.

Are you familiar with complementary/alternative therapies available for dying patients (not at all, a little, somewhat, quite, very) and is their use encouraged at your hospital? (not at all, a little, somewhat, often, always)

Research Proposal References American Hospital Association (2005). Retrieved April 23, 2005, from http://www.aha.org. Baker, R., Wu, A. W., Teno, J. M., Kreling, B., Damiano, A. M., & Rubin, H. R. et al. (2000, May). Family satisfaction with end-of-life care in seriously ill hospitalized adults. Journal of the American Geriatrics Society, 48(5), S61-S69.

18

Benner, P. (2001, September). Death as a human passage: Compassionate care for persons dying in critical care units. American Journal of Critical Care, 10(5), 355-359. Bryce, C. L., Loewenstein, G., Arnold, R. M., Schooler, J., Wax, R. S., & Angus, D. C. (2004, May). Quality of death: Assessing the importance placed on end-of-life treatment in the intensive-care unit. Medical Care, 42(5), 423-431. Clarke, E. B., Curtis, J. R., Luce, J. M., Levy, M., Danis, M., & Nelson, J. et al. (2003, September). Quality indicators for end-of-life care in the intensive care unit. Critical Care Medicine, 31(9), 2255-2262. Davidson, P., Introna, K., Daly, J., Paull, G., Jarvis, R., & Angus, J. et al. (2003, January). Cardiorespiratory nurses' perceptions of palliative care in nonmalignant disease: Data for the development of clinical practice. American Journal of Critical Care, 12(1), 47-53. Hinshaw, A. S. (2000). Nursing knowledge for the 21st century: Opportunities and challenges. Journal of Nursing Scholarship, 32(2), 117-123. Hodde, N. M., Engelberg, R. A., Treece, P. D., Steinberg, K. P., & Curtis, J. R. (2004, August). Factors associated with nurse assessment of the quality of dying and death in the intensive care unit. Critical Care Medicine, 32(8), 1648-1653.

Research Proposal Kirchhoff, K. T., Spuhler, V., Walker, L., Hutton, A., Cole, B. V., & Clemmer, T. (2000, January). Intensive care nurses' experiences with end-of-life care. American Journal of Critical Care, 9(1), 36-42. LoBiondo-Wood, G., & Haber, J. (2002). Nursing research: Methods, critical appraisal, and utilization (5th ed.). St. Louis, Missouri: Mosby. Nelson, J. E., & Danis, M. (2001, February). End-of-life care in the intensive care unit: Where are we now? Critical Care Medicine, 29(2), N2-N9. Polit, D. F., & Hungler, B. P. (1999). Nursing research: Principles and methods (6th ed.). Philadelphia: Lippincott.

19

Prendergast, T. J., Claessens, M. T., & Luce, J. M. (1998). A national survey of end-of-life care for critically ill patients. American Journal of Respiratory Critical Care Medicine, 158, 1163-1167. Puntillo, K. A., Benner, P., Drought, T., Drew, B., Stotts, N., & Stannard, D. et al. (2001, July). End-of-life issues in intensive care units: A national random survey of nurses' knowledge and beliefs. American Journal of Critical Care, 10(4), 216-229. Rubenfeld, G. D., & Curtis, J. R. (2002, December 18). Beyond ethical dilemmas: Improving the quality of end-of-life care in the intensive care unit. Critical Care, 7, 11-12. Retrieved January 24, 2005, from http://ccforum.com/content/7/1/11. Singer, P. A., Martin, D. K., & Kelner, M. (1999, January 13). Quality end-of-life care: Patients' perspectives. Journal of the American Medical Association, 281(2), 163-168. White, K. R., Coyne, P. J., & Patel, U. B. (2001). Are nurses adequately prepared for end-of-life care? Journal of Nursing Scholarship, 33(2), 147-151.