The “PSOST”: Providers' Signout for Scope of Treatment

Introduction

The Physicians Orders for Life Sustaining Treatment (POLST) ¹ document is becoming widely known as a tool that allows patients to specify which life-sustaining treatments they do and do not desire. The POLST has been shown to be most useful in long-term care facilities, where providers may not have personal relationships with each patient for whom they are responsible. The POLST improves care by communicating the patients' desires regarding escalation of care in the form of actionable orders.

On our inpatient Palliative Care unit, we are daily faced with POLST-worthy discussions and decisions. Our 11-bed unit and hospital consult team serve a very ill population, with approximately half of our admissions coming from intensive care units. Our unit also provides symptom management for patients who are not at the end of life and, in fact, are expected to be cured of their conditions and/or are predicted to have extended survival.

Our residents, fellows and attending physicians cover the unit during the day and are always available, but at night the medical residents from the inpatient hematology/oncology service provide first-call service in the case of an emergency. Nurses on our unit know their patients and families well, but cannot provide a consistent presence for all decision-making as they also rely on handoffs. The nurses and physicians who care for our patients at night are not always involved in the intense daytime conversations that take place between providers and patients and/or families regarding goals of care. Therefore, they are in a similar situation to those physicians who cover calls from nursing homes when it comes to their knowledge of patients' specific wishes regarding escalation of care. Without accurate communication, they may be unsure of what level of care they should provide to palliative care patients. We have always been concerned that some patients might have their care escalated to a level of unwanted care (resuscitation, vasopressors, transfer to intensive care, etc.) or inappropriately not aggressive enough (failure to provide resuscitation in someone who requests it for a correctable condition, decision not to give an intravenous bolus of fluids, diagnose and treat pulmonary embolism or pneumonia, etc.)

In the hospital it is routine to document “code status”, referring to whether or not a patient desires cardiopulmonary resuscitation. It is evident in our institution, just like those outside of the hospital, that the black and white delineation of “full code” versus “do not resuscitate” is not descriptive enough to guide care. This realization is the basis of what led to the POLST paradigm and is also what spawned the development of this internal tool.

Tool Development and Use

The importance of the patient “handoff” has gained considerable attention in the medical literature, especially as continuity has been reduced by system changes like limited resident duty hours and hospitalist shift-work. While there is no published proof, it stands to reason that improved communication between providers improves continuity and thus outcomes. ² The “signout” form is a method of communication that is widely used to facilitate accurate handoffs between providers. ³ In order to make nighttime decisions as simple and appropriate as possible, we evolved a comprehensive but concise tool to facilitate signout. It was based on the format used by the medical residents for signout regarding the care of inpatients. This format is used frequently by house staff across institutions and disciplines. The goal is to communicate a snapshot of each patient, enough to allow the covering house staff to appropriately respond to an emergency or incidental phone call without needing to perform a complete chart review. This type of signout routinely includes a brief History of Present Illness (HPI), Past Medical History (PMH), code status, significant laboratory or diagnostic test results, and a “to do” list of studies, labs or other things that need to be checked on during their shift. We included these elements but added a column, labeled “Care Plan.” We discussed among the fellows and contributing medicine residents what topics seem to arise during off hours. Those that seemed to cause the most angst among the residents are decisions regarding intravenous fluids, antibiotics, response to fever and escalation of care (including transfer to a higher acuity floor). We also felt it would be useful to provide information that may be needed if a patient dies, including cause of death, whether the patient needs to be evaluated by the medical examiner or not, and who (and who not) to call with medical information. This information is deliberately omitted on patients who are not expected to die so as to elicit a further investigation into the situation.

Table 1 is an example of the signout tool. The following helps to describe the patients on the list and how the signout helps to guide their care. The first patient on our sample signout, Patient A, was admitted to the palliative care unit for symptom control during treatment of her stage I colon cancer. She and her providers expect that she will tolerate her treatment and be cured of her malignancy. History and physical revealed a good functional status. Discussion with the patient reveals that she is hopeful to return to her career as a teacher and to do so she is willing to undergo any and all medical interventions necessary. Therefore, her Plan of Care would include intravenous fluids, work-up, and treatment of a fever and escalation of care if needed in addition to full cardiopulmonary resuscitation as indicated by her code status.

In contrast, Patient B is a 93-year-old patient who has little to no chance of meaningful recovery. Her family indicated that she had frequently told them that she had “lived a full life” and did not want to be maintained by artificial means. The goal of her family and her providers is to keep her comfortable and avoid any unnecessary medical interventions. Therefore, her plan of care would not include intravenous fluids even if she were hypotensive. She would be treated with an antipyretic if she had a fever but would not have cultures drawn or be started on an antibiotic. She would remain in her quiet, private room and would not be moved to a monitored bed, even if her condition declined further. Patient C is a man with extensive metastatic colon cancer. He understands that he will likely die from his cancer and therefore has decided to have the order, “do not resuscitate entered,” because resuscitation likely would not help to meet his goal of maintaining comfort during the last few months that he has to live. He has an Eastern Cooperative Oncology Group (ECOG) performance score of 2, however, and would like to maintain the function that he currently has and return home with his family. Therefore, if he were to become acutely febrile and hypotensive, interventions such as intravenous fluids, antibiotics, and an investigation into possible infectious sources would be consistent with his goal. All aggressive management would be done on the palliative care unit without transfer to higher level of care, for example, to the intensive care unit for vasopressors.

The signout is updated in a Word document daily on one of the password- protected computers and saved on a shared drive that is accessible throughout the hospital by medical house staff and attending physicians. It is reviewed by the fellow with the covering team, and a printed copy is left available to the Palliative Care unit nursing staff in their workroom. The form also includes the pager numbers of the responsible fellow and attending. Like other signout tools used by providers, it is not part of the medical record nor is a permanent record maintained. Printed copies are shredded on a daily basis and replaced by updated sheets.

Like the POLST and its other iterations such as the Physician Orders for Scope of Treatment (POST), ⁴ or Medical Orders for Life Sustaining Treatment (MOLST), ⁵ our tool addresses a limited set of interventions. We included the interventions that seemed most relevant for the providers who do not know the patients in order to address the majority of issues that arise at nighttime. Also like the POLST, the PSOST is meant to serve only as a starting point, a framework to guide the provider toward an appropriate response in the absence of the ability to communicate with the patient in a timely and thorough manner. It can only be completed if the daytime provider has communicated well with the patient and/or family regarding their goals.

Discussion

Open, honest communication and medically appropriate goal setting are imperative elements of effective palliative medicine that cannot be achieved without due diligence. ⁶ Decisions regarding appropriate levels of care are difficult for patients and families. Even when discussed and documented in an advance directive, such forms may not be applicable to the current situation and often do not accompany the patient. Patients' conditions and care preferences can sometimes change on a daily basis. The key to helping patients meet their goals will always rest in effective reassessment and communication of those goals. This is difficult to achieve with any stagnant form such as a living will or other advance directive, which may be part of the reason why advance directive use has not increased in the past 20 years, and why POLST forms are required to be regularly reviewed and updated. This tool offers a unique way for a medical team to ensure that patients' changing goals are accurately communicated to others. It should be noted, however, that this tool is reliant on effective communication with the patient. The utility of this tool is lost if it is not based on the true wishes of the patient as gleaned through open, honest discussion.

This is a simple tool that documents resuscitation status, goals of care, desired and undesired interventions, cause of death, medical examiner case or not, who (and who not) to call with medical information in addition to standard signout information. The benefits of this form are that it is personalized, dynamic and applicable to multiple patient populations. It is updated on a daily basis by those who know the patients best. There are two significant limitations of this form. The first is that it is not a part of the medical record and therefore does not travel with the patient to other healthcare settings. It therefore may not facilitate the handoff of the patient outside of the healthcare setting in which it is utilized. The second limitation is the reliance on providers to regularly update the signout. This is a time intensive process that may allow for human error. However, the concept of this type of form could be adapted for use in multiple settings. These could include, but are not limited to, use with hospitalized patients who are not on the palliative care services, patients in hospice care at home or in the inpatient setting, or for patients residing in long-term care facilities. The concept of including the “Plan of Care” could be adapted to fit into already established signout tools, including verbal signouts, between nurses, or informal notes written between other care providers. We believe inclusion of this topic in signout is imperative in order to best apply medical treatments that are appropriate for the patient. It also provides the ability to quickly look at a patient's plan of care without having to look through a chart and possibly delay the next appropriate step. It is similar to other lists such as the Top Ten List, which does not include any notation about escalation of care or the goals of treatment. ⁷ The emphasis on a written plan, inclusion of all the care givers, and direct person to person communication fits with current theory of how to reduce errors ⁸ and may prevent missed opportunities. ⁹

Our tool has not been formally studied for its effectiveness but since we started using it, residents covering for our service have consistently reported a greater level of comfort with the care of our patients as a result of having access to the new tool. The fellows caring for patients also report that they notice less frequent interventions that seem out of alignment with their patients' goals. Our nursing staff has also reported satisfaction, feeling that they have a guide at their fingertips to assist in their own decision-making. Possible ways to validate the tool could include a formal survey to assess satisfaction amongst the residents, fellows, attending physicians and nursing staff.

We would encourage others in the palliative care field to share their experience with PSOSTs and tools like them in order to evolve the best tools for coordinated patient care.