A Collaborative ICU Staffing Model: A Model Comprised of Nurse Practitioners and Physician

The aging population and advanced medical therapies has contributed to an increased need for more onsite medical coverage in Intensive Care Units (ICU) across the country. However, there has been a shortage of adequately trained critical care provider to staff these ICUs. This has contributed to current critical care providers being overworked, leading to provider burnout and provider turnover. Inadequate coverage in the ICU has contributed to inadequate medical care, compromise patient safety, loss of revenue for the hospital, and finally a decrease in provider, nursing staff, and patient satisfaction. The purpose of the project was to assess current provider coverage in a local ICU, determine future needs based on current admission rates and patient acuities, and propose a collaborative staffing model comprised of both physicians and nurse practitioners. The collaborative staffing model was formulated after a review of current literature and assessment of other area ICU staffing models. Finally, the proposed staffing model was presented to the key stakeholders for review and comments. This project had four objectives: 1) Collect and assess current local intensive care unit staffing levels and models. 2) Compare local findings with current research and identify potential barriers. 3) Propose a collaborative provider staffing model incorporating current research and current ICU needs. 4) Present and disseminate the proposed staffing model to stakeholders and submit an abstract to a local conference's poster presentation. Based on the literature review, there is a growing need for provider coverage in most ICUs across the country. Most hospitals are experiencing an increased in patient acuity and ICU admission rates. There is currently a nationwide shortage of critical care providers due to stagnant levels of new critical care providers and increases in both turnover and burnout rates. Critical care provider staffing models that include nurse practitioners have been successful in many ICUs nationwide. Billing for ICUs using a nurse practitioner/physician model has not been disrupted or found to decrease potential revenue. Also the physician/nurse practitioner model has shown to improve medical care in the ICU by providing consistency in the medical care provided, and increased in both patient and medical staff satisfaction levels. Results from the project indicate that units were potentially understaffed based on the highest potential patient to provider ratio. However, the assessment of one particular unit was identified as facing a critical provider shortage. These results supported the need for a change in provider coverage in order to meet the current trend of increasing patient acuity and increasing admissions rates. Based on these findings a collaborative staffing model comprised of both physician and nurse practitioner providers was proposed to the administrative team of this facility for their consideration.

Running head: COLLABORATIVE ICU STAFFING MODEL 1 A Collaborative ICU Staffing Model: A Model Comprised of Nurse Practitioners and Physician Intensivists Jorge A. Delgadillo University of Utah In partial fulfillment of the requirements for the Doctorate of Nursing Practice COLLABORATIVE ICU STAFFING MODEL 2 Executive Summary The aging population and advanced medical therapies has contributed to an increased need for more onsite medical coverage in Intensive Care Units (ICU) across the country. However, there has been a shortage of adequately trained critical care provider to staff these ICUs. This has contributed to current critical care providers being overworked, leading to provider burnout and provider turnover. Inadequate coverage in the ICU has contributed to inadequate medical care, compromise patient safety, loss of revenue for the hospital, and finally a decrease in provider, nursing staff, and patient satisfaction. The purpose of the project was to assess current provider coverage in a local ICU, determine future needs based on current admission rates and patient acuities, and propose a collaborative staffing model comprised of both physicians and nurse practitioners. The collaborative staffing model was formulated after a review of current literature and assessment of other area ICU staffing models. Finally, the proposed staffing model was presented to the key stakeholders for review and comments. This project had four objectives: 1) Collect and assess current local intensive care unit staffing levels and models. 2) Compare local findings with current research and identify potential barriers. 3) Propose a collaborative provider staffing model incorporating current research and current ICU needs. 4) Present and disseminate the proposed staffing model to stakeholders and submit an abstract to a local conference's poster presentation. Based on the literature review, there is a growing need for provider coverage in most ICUs across the country. Most hospitals are experiencing an increased in patient acuity and ICU admission rates. There is currently a nationwide shortage of critical care providers due to stagnant levels of new critical care providers and increases in both turnover and burnout rates. Critical care provider staffing models that include nurse practitioners have been successful in many ICUs nationwide. Billing for ICUs using a nurse practitioner/physician model has not been disrupted or found to decrease potential revenue. Also the physician/nurse practitioner model has shown to improve medical care in the ICU by providing consistency in the medical care provided, and increased in both patient and medical staff satisfaction levels. Results from the project indicate that units were potentially understaffed based on the highest potential patient to provider ratio. However, the assessment of one particular unit was identified as facing a critical provider shortage. These results supported the need for a change in provider coverage in order to meet the current trend of increasing patient acuity and increasing admissions rates. Based on these findings a collaborative staffing model comprised of both physician and nurse practitioner providers was proposed to the administrative team of this facility for their consideration. The committee includes Clint Child, DNP, MBA, RN, Denise E. Ward, DNP, ACNP-BC, FNP-BC, & Assistant Dean Dr. Pamela K. Hardin, PhD, RN, CNE. The content expert for this project is Omar Mendez, MD COLLABORATIVE ICU STAFFING MODEL 3 Acknowledgements I would like to first thank my mom and my dad for the encouragement they have provided me throughout my undergraduate and graduate education. I also want to thank my siblings for their encouragement and a special thanks to my brother for expressing on multiple occasions how proud he is of my educational accomplishments. Very special thanks go out to my stepsons for their patience with me during the many long hours I was absent from many family activities and the several family vacations that were cut short due to general school responsibilities and work on this project. Lastly, my everlasting thanks, gratitude, and love goes out to my wife who put up with more than I know and more than she will ever tell me during the course of my general graduate education and the work on this project. I thank and love you all. COLLABORATIVE ICU STAFFING MODEL 4 Table of Contents Executive Summary ........................................................................................................................ 2 Acknowledgements ......................................................................................................................... 3 Problem Statement .......................................................................................................................... 5 Clinical Significance ....................................................................................................................... 5 Purpose............................................................................................................................................ 8 Objectives ....................................................................................................................................... 8 Literature Review............................................................................................................................ 8 Intensivist Shortage .................................................................................................................................. 9 Intensivist/Advanced Practitioner Staffing Models ................................................................................ 11 Theoretical Framework ................................................................................................................. 12 Implementation & Evaluation ....................................................................................................... 14 Results ........................................................................................................................................... 18 Recommendations ......................................................................................................................... 20 DNP Essentials.............................................................................................................................. 21 Conclusion .................................................................................................................................... 23 References ..................................................................................................................................... 24 Appendices .................................................................................................................................... 27 Appendix A - Lewin's Model of Change............................................................................................... 28 Appendix B - IRB Non-Human Research Determination...................................................................... 30 Appendix C - Critical Care Staffing Questionnaire ............................................................................... 32 Appendix D - APACHE II Score ........................................................................................................... 35 Appendix E - PowerPoint Presentation to Project Site Stakeholders ..................................................... 37 Appendix F - Abstract Submission Confirmation.................................................................................. 43 Appendix G - Local ICU Comparison ................................................................................................... 45 Appendix H - 2016 ICU Admits Per Day of the Week.......................................................................... 47 Appendix I - 2016 ICU Bi-hourly Admit Breakdown ........................................................................... 49 Appendix J - Proposal Defense PowerPoint ........................................................................................... 51 Appendix K - Poster............................................................................................................................... 59 COLLABORATIVE ICU STAFFING MODEL 5 Problem Statement The demand for critical care (CC) services continues to rise as the availability of CC providers falls short. The shortfall in CC providers we are seeing today was identified by specialty societies and the federal government a decade ago, yet we still do not have a solution to meet this growing demand for this specialty healthcare service (Halpren, Pastores, Oropello, & Kvetan, 2013). Halpern, Pastores, Oropello, and Kvetan point out that despite the many years of publicized warnings from different groups and even the federal government, the issue has not been adequately addressed (2013). They secondly also point out the difficulty of accurately assessing the current nationwide shortfall of CC service providers (Halpren, Pastores, Oropello, & Kvetan, 2013). A wide variety of CC staffing models are being used across the nation in an attempt to close the CC provider shortage. Having an effective and adequate provider staffing model is key to preventing the negative effects and consequences that can develop as a result of an understaffed CC service. An effective and efficient staffing model is especially important for ICUs with limited resources, such as the site for this project and similar facilities. Clinical Significance The shortage in CC providers is leading to staffing models that overstretch and overwork available providers. Staffing models comprised of one full time CC provider covering all ICU beds for extended periods of time are currently being utilized. In many cases these providers also have a variety of other non-clinical/administrative responsibilities. These additional responsibilities can greatly increase their overall workload. It is easy to see how these models may not be sustainable long term and may compromise the care these providers deliver. Critical care physicians who are being overworked and stressed are at risk for developing burnout syndrome. Burnout syndrome can lead to increased provider turnover, reductions in COLLABORATIVE ICU STAFFING MODEL 6 patient satisfaction, and also decreases in the quality and consistency of the care being delivered by burned out providers. Critical care providers are especially susceptible to burnout due to the high stresses associated with caring for this specific patient population (Moss, Good, Gozal, Kleinpell, & Sessler, 2016). This issue is compounded by the increasing numbers of patients requiring CC services across the nation. A study by Merlani, Verdon, Businger, Domenighetti, Pargger, Ricou, & the STRESI1 Group found that burnout syndrome can affect up to 45% of nurses and physicians providing care in the intensive care setting (2011). This is an alarmingly high rate of burnout. This specific study was conducted in Switzerland and cultural / personality differences may affect the application and generalization of these results to ICUs in the United States, but stressors such as staffing shortages, patient acuity, and patient deaths are universally experienced by all intensive care providers, regardless of the culture or country in which they practice. If shortages in provider coverage are not addressed and resolved by the ICUs facing them, the likelihood of losing not just intensivist providers but also nursing and ancillary staff increases greatly. Provider shortages will also affect an ICUs ability to admit and safely care for ICU level patients. This can lead to more out-of-facility transfers, which can have a direct impact on the revenue stream of these ICUs. A negative impact to an ICUs revenue stream may also occur as a result of decreases in billable CC services due to a shortage of providers available to perform these services. Losses in revenue can occur when an ICU has to hire and train CC providers in order to replace intensivist who turnover, burnout, or retire. There are many estimates about the actual cost of physician turnover. One study found that the cost of simply recruiting a physician can be in the tens of thousands of dollars (Misra-Herbert, Kay, & Stoller, 2004). Misra-Herbert, Kay, COLLABORATIVE ICU STAFFING MODEL 7 and Stoller found that the loss of revenue from physician turn over can be over $500,000 for an inpatient service (2004). The actual amounts vary based on the specialty or service. The $500,000 estimate is for inpatient primary care services, there were no specific figures in this study for losses of intensive care revenue from CC physician turnover (Misra-Herbert, Kay, & Stoller, 2004). The assumption can be made that the losses and the costs incurred for replacing a CC provider could be more than $500,000 due to the longer training period required to become a board certified CC provider. The shortage of CC providers affects each facility differently. The shortfall is an issue that in some facilities has been alleviated by staffing models that include Advanced Practice Providers (APP) such as physician's assistants and nurse practitioners (NP). This has been done on a facility by facility basis, with various different staffing models being utilized nationwide. A few of those models have been studied. An example of this type of staffing model has been employed at Vanderbilt Medical Center since 2005 (Squiers, King, Wagner, Ashby, & Parmley, 2013). These models can serve as a means of closing provider gaps, preventing MD intensivist burnout, and maintaining consistent high quality evidence based care. The ICU patient population and the staff providing CC services are easily identified as stakeholders. It is important to remember that the stakeholders affected by the intensivist shortage extends beyond the CC providers and ICU staff; they include the hospital administration (CEO, CNO, and CFO), emergency room department, surgical services, medical services, and any other service or provider that relies on the availability and quality of CC services. It is also important to recognize the surrounding communities and patient populations, being served by any ICU facing provider shortfalls as stakeholders. COLLABORATIVE ICU STAFFING MODEL 8 Purpose The purpose of this project was to assess current CC provider coverage in relation to patient acuity and admission rates at a local ICU. Next, propose a critical care provider staffing model that incorporates current trends supported in the literature and the review of staffing models used at similar sized ICUs in the area. Objectives The first objective of this project was to determine local ICU provider staffing models. The second objective of this project was to compare current research to staffing models used in similar ICUs within the local and identify potential barriers. The third objective of this project was to develop a needs based collaborative staffing model for the ICU of the site for this project. The fourth objective of this project was to propose the staffing model to the identified stakeholders on the administrative team and disseminate findings by submitting an abstract to a local CME conference's poster presentation. Literature Review The CC services being provided by our healthcare system are paramount and an important piece of the overall services we have available for the acutely ill individual. In the United States there are "approximately 6,500 ICUs with 94,000 ICU beds in nearly 5,000 U.S. acute care hospitals" (Halpern, Pastores, Oropello, & Kvetan, 2013, p.2755). Occupancy of ICU beds ranges from 65% to 72%, this amounts to between 4 million and 6 million ICU admission annually across the United States (Halpern, Pastores, Oropello, & Kvetan, 2013). We are currently facing a shortage of CC providers across the Country. This shortage is leading to new, and in many cases innovative staffing models, in order to meet the increasing need for CC services. Between 2000 and 2005, the number of ICU beds grew by 6.5%, while the average COLLABORATIVE ICU STAFFING MODEL 9 length of ICU stays increased by 10.6% (Mullins, Goyal, & Pines, 2013). Between 4 to 6 million patients are admitted to the ICU yearly. In 2011, there were 11,806 physicians with CC medicine board certification (Halpren, Pastores, Oropello, & Kvetan, 2013). A study conducted by Mullins, Goyal, and Pines provides a glimpse of the increase of intensive care admissions from the emergency department (2013). They found an almost 50% increase in ICU admissions from the ER between 2002 and 2009 (by Mullins, Goyal, & Pines, 2013). The combination of increasing ICU admissions, decreasing ICU beds, and decreasing numbers of CC providers may be putting at risk patient safety, patient satisfaction, and provider satisfaction. Intensivist Shortage The shortage of CC medicine providers, also called intensivist, was identified as early as 2004 by Ewart, Marcus, Gaba, Bradner, Medina, and Chandler in their article entitled "The Critical Care Medicine Crisis: A Call for Federal Action," published in the medical journal CHEST. In their article the authors discussed the upcoming shortage of CC medicine providers that our healthcare system would face. They predicted the shortage would become "severe" by 2007 and continue to worsen through the year 2030 (Ewart, Marcus, Gaba, Bradner, Medina, & Chandler, 2004, p.1518). They attributed the shortage of qualified CC medicine providers to the current delivery system's inability to meet the current and future demand for this specialty and the increasing numbers of critically ill patients nationwide (Ewart, Marcus, Gaba, Bradner, Medina, & Chandler, 2004). At this time the authors of this article laid out their recommendations of what the Federal government could do to help alleviate this shortfall of CC service providers. Their main suggestions revolved around incentivizing the education process for this specialty and simplifying the reimbursement of CC services (Ewart, Marcus, Gaba, Bradner, Medina, & Chandler, 2004). COLLABORATIVE ICU STAFFING MODEL 10 Ewart, Marcus, Gaba, Bradner, Medina, and Chandler point out that additional stressors are affecting the amount of new providers and the retention of currently practicing CC service providers (2004). Stressors such as an increasing critically ill patient population, flat reimbursement rates for services and care provided, and the inherent toll that working with the very sick have all been identified as contributing to worsening CC provider recruitment rates. These stressors are also adding to growing provider retention issues being faced by this medical specialty (Halpern, Pastores, Oropello, & Kvetan, 2013). Staffing shortfalls and a mix of stressors are leading to an increasing number of CC service providers experiencing "burnout syndrome" (Moss, Good, Gozal, Kleinpell, &Sessler, 2016). Burnout syndrome can develop in all types of healthcare service providers, but is becoming more and more common in those that work with the sickest patient populations. The development of burnout syndrome can not only affect the individual healthcare worker, but those that work along side them, and their place of employment as well. The effects faced by the healthcare worker's place of employment can be varying combinations of the following: increased healthcare costs, lower staff morale, reduced quality of care, and reduction in overall productivity of individuals or groups of individuals facing burnout (Moss, Good, Gozal, Kleinpell, &Sessler, 2016). Studies have shown that as the number of patients being care for by a CC provider increases, the risk for negative impact on patient care, staff well-being, and workforce stability also increases (Ward & Howell, 2015). In 2012, the Society of Critical Care Medicine convened a task force and they produced a consensus statement where they presented their recommendations on provider to patient ratios. This taskforce found that the literature dose not support specific patient to provider ratio recommendations (Ward et al., 2013). The process of COLLABORATIVE ICU STAFFING MODEL 11 determining appropriate CC provider to patient ratios is complex and there are no one size fits all recommendations. They recommend that each facility use a "common-sense approach" to evaluating their specific needs and demands based on patient census, patient acuity, and other non clinical factors (Ward et al., 2013, p. 643). They do provide a guide that can be used to evaluate the demand on CC providers. This calculation takes into account both clinical and non clinical responsibilities being carried out by individual CC providers (Ward et al., 2013). Intensivist/Advanced Practitioner Staffing Models One of the solutions to the CC provider shortage that some organizations are using is the implementation of staffing models that utilize advanced practice providers. These models can be made up of nurse practitioners or physician's assistants, working together with experienced physician intensivist to provider CC services. One example of such a staffing model is the one developed and implemented at the Vanderbilt University Medical Center in Nashville, Tennessee. The staffing model being used by this organization specifically uses acute care nurse practitioners as the advanced practice providers working alongside physician intensivist (Squiers, King, Wagner, Ashby, & Parmley, 2013). This organization has found many advantages to the staffing model that they implemented. Some of the advantages have been: increased ability to bill for critical services provided, versus the old model of using resident physicians exclusively; improved consistency of care over time; improved staff and patient satisfaction levels; and more cost-effective care (Squiers, King, Wagner, Ashby, & Parmley, 2013). As part of their staffing model, Vanderbilt University Medical Center also included an education curriculum that they implemented alongside their staffing model. This curriculum is specific to the care and interventions provided to critically ill patients in the ICU setting (Squiers, King, Wagner, Ashby, & Parmley, 2013). COLLABORATIVE ICU STAFFING MODEL 12 Several studies have found that the overall patient outcomes of the care provided by advanced level practitioners are comparable to those of physician providers. One such study was conducted by Gershengorn, Wunsch, Wahab, Leaf, Brodie, Li, and Factor (2011). They found that mortality rates did not vary considerably between NP CC providers and their MD counterparts, specifically MD residents providing CC services (Gershengorn, Wunsch, Wahab, Leaf, Brodie, Li, and Factor, 2011). Past and current literature has identified the shortage of CC providers, the negative consequences and effects that can occur when these shortages are present, and that innovative solutions have proven effective in resolving provider shortages and preventing negative consequences and effects (Squiers, King, Wagner, Ashby, & Parmley, 2013). Theoretical Framework A change in CC provider staffing model can present an organization or system with various hurdles. This type of change can push and challenge an organization's established practices, cultures, and norms. This can lead to the creation of barriers that can have an impact on the successful development, dissemination, implementation, and adoption of new ideas and processes. Using a theoretical framework such as Lewin's Three Stage Theory of Planned Change can guide the development, implementation, and establishment of new ideas, processes, and models and can assist in easing the process of change. An illustration of this model can be found in appendix A. Using a formal framework can also lead to an increase in acceptance and buy in by all stakeholders and improving the likelihood that the proposed changes will not only be initially implemented but lead to long lasting changes and improvements. The landscape of healthcare is ever changing; developing and implementing new ideas is a crucial step in avoiding practice failure (Gesme & Wiseman, 2010). COLLABORATIVE ICU STAFFING MODEL 13 A general definition of an organizational and systems change model is: a process were "individuals, organizations, policies, and regulations come together to create a new way of doing things that is both feasible and sustainable. It involves getting individual people and individual organizations to change in a coordinated way that involves policies, financing, and services motivated toward a specific change or specific sets of changes" (Isett, n.d., para 1). Multiple organizational and systems change models have been developed, studied, and implemented over the years. The Three Stage Theory of Planned change was developed by Kurt Lewin in 1951 and has served as the prototype and model for many organizational and systems change theories and models (Mitchell, 2013). The goals of Lewin's theory are to assist in avoiding common pitfalls that can thwart change and offer a framework to guide the desired change (Shirey, 2013). Lewin's Three Staged Theory provides a simple yet very effective model and guide for the development and implementation of a new ICU staffing model that can result in financial and patient care benefits in the intensive care setting. The three stages of Lewin's Theory of Planned change are Unfreeze, Change, and Refreeze (Shirey, 2013). Each one of these stages can be further broken down in to more detailed and specific ideas, steps, and processes. For this project, assessing the dynamics of the staffing models currently being used, establishing the shared goals of CC services and providers, identifying the barriers and challenges that may impede implementation and staffing model change, and finally developing a plan that involves the many different stake holders makes up the "Unfreeze" stage of the process. The ultimate goal of this stage is to examine the status quo and to increase the driving forces for change (Mitchell, 2013). Disseminating the findings of the Unfreeze stage and implementing the developed model make up the "Change" stage of this process. This stage can be the toughest of all the stages and the one where much of the COLLABORATIVE ICU STAFFING MODEL 14 challenges will be faced, as ideas become actions and the individuals that make up the organization and the groups being affected by the changes react to the shifts taking place. Finally, thoroughly evaluating the benefits, consequences, and ongoing challenges are the key steps to making a change that is long lasting and allows the process to move forward to Lewin's final stage of "Refreezing" at a new and improved state of function and/or production. The ultimate goal of this final stage is to have a change that is long-lasting, yet dynamic. Implementation & Evaluation The process of implementation and evaluation for this project required multiple steps and data was collected in various forms (questionnaire, census data, and de-identified medical record data). The required steps included applying for Non-human research status from the IRB at the University of Utah and the site for this project (see Appendix B for University of Utah IRB determination). This project was a quality improvement project and determined to be non-human research by both IRBs listed above. Following is a breakdown of the individual steps taken in the implementation and evaluation of each individual objective. The first objective of this project was to determine local ICU provider staffing models. To implement objective 1, the researcher determined which tertiary facilities have ICUs within the selected local area and contacted one of the following individuals within each identified facility: ICU unit director, CC provider, or administrator. The identified individual at each site was provided with an overview of the project and consent was obtained by the researcher in order to administer the ICU Staffing questionnaire to this individual (see Appendix C for questionnaire used). The questionnaire was administered and the responses obtained were compiled. COLLABORATIVE ICU STAFFING MODEL 15 The first objective was considered completed and successful when all ICUs in the local area were identified; successful contact was made with an individual willing and qualified to complete the ICU staffing assessment questionnaire; and all the questionnaire responses were compiled, data analyzed, and staffing model practices established for each facility. The second objective of this project was to compare current research to staffing models used in similar ICUs within the local and identify potential barriers. To implement objective 2, the researcher used the findings from the data gathered by way of the questionnaire and compared the various staffing models. Comparison was made between each facility's staffing model and current research recommendations. The literature review focused on CC provider shortage, consequences associated with CC provider shortage, patient to CC provider ratio recommendations, and ICU staffing models that incorporate nurse practitioners. Literature searches were conducted in CINAHL, PubMed, and the Society of Critical Care Medicine's website. The individual ICUs were compared to similar units in the studied area. Units were determined to be similar based on the total number of beds of each facility. Each of the similar units were compared in the following areas: overall hospital bed size, total ICU beds in facility, level of trauma care provided, current intensive care provider staffing model, and presence of specialty services such as neurosurgery, trauma surgery, and cardiothoracic surgery. The second objective was considered completed after completion and summary of the literature review process, comparison of the current local ICU staffing practices with those found in the research, and assessment of current local CC provider staffing models. The third objective of this project was to develop a needs based collaborative staffing model for the ICU of the site for this project. To implement objective 3, the researcher assessed and considered the findings in objectives 1 and 2. This information served as a guide and COLLABORATIVE ICU STAFFING MODEL 16 reference that was used in the development of the intensive care provider staffing model that utilizes both physician providers and advanced level practitioner providers. In order to develop a model that closely met the demand for CC services of the project site, more detailed data was gathered in the form of overall census data which included: total number of ICU admits, date and times of admits, origin of admits (ER, OR, other hospital floor, direct admit, and outside facility transfer), admitting diagnosis, discharge date and time, and discharge disposition. The census data collection was limited to the dates between January 1, 2016 to December 31, 2016. This data was used to determine the overall need for CC services in this facility and to help propose a staffing model that covers the days of the week and time frames of the day where CC services are in highest demand. A second, separate data set was obtained with the purpose of calculating acuity scores using the APACHE II scoring tool for a 15% representative sample of the total number of ICU patients admitted between January 1, 2016 to December 31, 2016 (see Appendix D for APACHE II Score parameters). These scores were used to further support the need for improved coverage of CC services. With all above steps completed a staffing model proposal was developed that fits the needs of the selected mid-sized facility. The model took into consideration the facilities needs as determined by the data and analysis that took place in objectives 1 through 3. Both data sets collected were input into the REDcap database for storage and analysis. The third objective was considered completed after the data from the specified facility was gathered, input into the REDcap database, analyzed, and a staffing model proposal developed based on the data collected in this objective and objectives 1 and 2. The fourth objective of this project was to propose the developed staffing model to key stakeholders on the administrative team of the site for this project and to submit a project abstract COLLABORATIVE ICU STAFFING MODEL 17 for a poster presentation at a local CME conference. To implement objective 4, a PowerPoint presentation of the research findings and proposed staffing model was presented to the identified stakeholders on April 25, 2017 (see Appendix E for PowerPoint presentation). Finally, broader dissemination of the project findings and recommendations was accomplished by submitting a study abstract for consideration to a local CME conference's poster presentation session. The fourth and final objective was completed after the project findings were presented to the identified stakeholders and project abstract submitted to a local CME conference (see Appendix F for abstract submission receipt email). Objective 1: Determine local provider staffing levels / models Objective 2: Compare local findings with findings from the research Implementation: 1. Identify local tertiary facilities with ICUs. 2. Get in contact with each facility and obtain information concerning current provider staffing model. 3. Compile the information gathered and separate data based on overall facility bed size. Evaluation: 1. The ability to identify all ICU's in the local area; 2. Getting in contact with an individual knowledgeable enough and willing to discuss their intensive care provider staffing model; and 3. Successfully compiling the gathered data and determining staffing models used. Implementation: 1. Compare Evaluation: The success of the compiled data gathered in this objective will come from the first objective with data a proper and through gathered from latest research literature search, and a 2. Based on comparison through of the data compiled findings the researcher will from objectives one and two. assess local ICU staffing Further evaluation of the standards and need for results from this objective additional will come by way of a face to CC providers. 3. The goal of face meeting with the content this objective is to determine expert identified for this if there is an increasing need project. for critical care providers, a lack of intensive care provider staffing standard, the consequences that any current shortage is having. 18 COLLABORATIVE ICU STAFFING MODEL Objective 3: Propose a staffing model incorporating research and analysis of the collected data. Implementation: 1. Review the findings from objective 2. 2. Use gathered info from above step as a guide for staffing model development. 3. Gather detailed data from project site to be used as guide in the development of staffing model proposal. 4. Develop staffing model proposal that fits the needs of the project site. Objective 4: Disseminate / present recommendations Implementation: 1. Review final project and PowerPoint presentation draft with content expert. 2. Schedule presentation to project site administrative team stakeholders. 3. Present final project, and make time at end of presentation for feedback and questions. 4. Submit abstract to local CME conference for inclusion consideration to their poster presentation session. Evaluation: The evaluation will come in the form of reviews of drafts and final staffing model proposal by project chair, content expert, and identified stakeholders. The purpose of the evaluation of this objective is to obtain feedback that will be use to refine the model into something that is feasible, sustainable, and implementable. Evaluation: The evaluation for this objective will come in the form of presenting final project draft and PowerPoint draft to content expert, presenting findings and proposed staffing model to primary stake holders, and submitting a project abstract to a local CME conference. Results A total of 5 ICUs from 3 different healthcare systems were identified in the studied area. The identified ICUs ranged from no trauma designation to a level 2 trauma center. There was great variability in the overall size of each facility, as assessed by total number of beds. The facilities ranged from 44 to 395 total beds. Three of the facilities were between 90 to 124 beds, making them the most alike. The site for this project is in this group of three. There was also great variability in the number of total CC beds in each ICU, ranging from 4 to 49 beds (see Appendix G for local ICU comparison). The range of services also varied greatly between each facility. All facilities have an emergency room and general surgical services, 4 of the facilities COLLABORATIVE ICU STAFFING MODEL 19 have Cath labs, 2 of the facilities have cardiothoracic and vascular surgical services, 3 have trauma surgical services, and one has neurosurgical services. Four of the facilities have one ICU that admits all ICU level patients, with one facility having two ICUs divided into neuro/shock/trauma/medical and cardiothoracic ICUs. A Provider Staffing Questionnaire was completed for each of the identified ICUs in the selected local area. Analysis of the responses from these questionnaires showed that there is no standard CC provider staffing model being utilized in this area. The staffing models greatly ranged in number of providers on staff and in provider primary specialty. One ICU in this area was identified as having nurse practitioners incorporated into their CC team. Questionnaire response analysis showed that some facilities extend ICU admitting privileges to hospitalists services, in order to meet the demand for CC services. Analysis of the highest potential patient to provider ratio showed that all ICUs in this area are at risk for experiencing a shortfall in CC providers if all their available ICU beds were to be filled and their units staffed by their current provider staffing models (see Appendix G for local ICU comparison). Only a couple of the ICUs in this area have the ability to enhance their standard provider staffing models with additional providers during times of high or full census. The inability of the other ICUs to enhance their staffing with additional providers during times of high or full census puts them at a great risk of experiencing the negative effects associated with overworked provider/s and provider burnout. As part of this analysis the site for this project was identified as facing a CC provider shortage. This was determined by an analysis of the highest potential patient to provider ratio in the ICU, he lack of ability to enhance provider coverage with additional providers in times of full or high census, and finally the fact that the provider for this ICU is single handedly overseeing CC services for two of the ICUs in the local area. This provider oversees services at these two COLLABORATIVE ICU STAFFING MODEL 20 facilities 24 hours a day 7 days a week for stretches between 2 - 4 months at a time. Currently, the sister facility to the site of this project does employ NPs as part of the ICU provider staffing model. This does provide some relief to the overall workload for this provider. An assessment of the project site's ICU 2016 patient census was conducted. Acuity scores were calculated for a sample of patients from the 2016 census using APACHE II scores. A sample of 59 randomly selected patients was used to determine average patient acuity in 2016. This sample size was based on a population size of 382 with a confidence interval of 90% and a margin of error of 10%. It was determined that the above stated confidence interval and margin of error were appropriate for the purposes of this project. Results showed that this unit admitted 382 ICU status patients in 2016 with an average APACHE II score of 14.2. This corresponds to an average initial 24-hour mortality rate of 23.5%. The average length of stay for this unit in 2016 was 3.3 days. These results are in line with national ICU averages for patient acuity and ICU length of stay (Society of Critical Care Medicine, n.d.). These results establish a need for adequate CC provider coverage at this facility. Recommendations Census analysis showed that the highest demand for CC provider coverage for this unit was Monday thru Friday between the hours of 8:00 AM to 4:00 PM (see Appendices H and I for census analysis results). Based on these findings a collaborative staffing model was proposed to the administrative team of this facility for their consideration during their April 25, 2017 Administration team meeting (see Appendix E for PowerPoint presentation). The following project site administrative team member were in attendance: CEO, COO, CFO, CNO, Director of Quality and Risk, Director of Marketing, Director of Human Resources, and Associate Chief Nursing Officer. The proposed staffing model is based on a 7 days on 7 days off rotation for the COLLABORATIVE ICU STAFFING MODEL 21 NP providers. The purpose of this rotation is to assist in maintaining consistency of patient care. The proposed collaborative staffing model would provide onsite CC coverage by a full time nurse practitioner Monday thru Friday between 8:00 AM to 4:00 PM with an MD intensivist available at all time for consultation, either in person or by electronic means. Night time coverage would be provided by NP provider on an on call basis with MD intensivist available for consultation. Nurse Practitioner and MD providers will conduct daily multidisciplinary rounds (MDR). Weekends (Saturday and Sunday) will consist of MDRs and direct on call coverage for the remainder of the day by NP providers with MD intensivist available for consultation by electronics means. The optimal provider coverage for this proposed collaborative staffing model would be 2 MD intensivist providers and 2-4 nurse practitioner providers. This model can be implemented with a minimum of 2 NPs and 1 MD provider, but would still leave the challenge of finding MD coverage when primary MD intensivist takes time off. Currently the project site ICU is using locum tenants to fill this void. The advantage of the proposed model over current practice is the NPs on the CC team will provide and maintain the consistency in patient care, even when locum tenants are used. DNP Essentials The Doctor of Nursing Practice (DNP) essentials that are applicable to this project are essentials II, V and VI. These essentials have served as a guide during all of the steps of the scholarly project. These DNP essentials have played an important role in determining appropriate overall goals and have guided the various organization and analysis structures used throughout the entire study, particularly the development of the collaborative staffing model. The following descriptions provide details on the utilization of the above listed essentials in this scholarly project. COLLABORATIVE ICU STAFFING MODEL 22 Essential II: Organization and Systems Leadership for Quality Improvement and Systems Thinking provided the overall guiding principles for the goals and development of the collaborative staffing model. The overall goal of the staffing model is to improve the quality of CC services and prevent the adverse effects associated with CC provider shortages and provider burnout. CC provider staffing models such as this have the potential to "eliminate health disparities and to promote patient safety and excellence in practice" in facilities facing staffing shortages and challenges (AACN, 2006). Essential V: Healthcare Policy for Advocacy in Healthcare applies to this scholarly project as the aim of the developed staffing model is to change the organizational standards and create a new framework to guide CC provider staffing. Organizational policy and framework changes are both key elements in the process of advocating for improvements to meet the needs of the patient population served by this facility. This model also aims at advocating for improvements in consistency and quality of care (AACN, 2006). Essential VI: Interprofessional Collaboration for Improving Patient and Population Health Outcomes directly applies to this scholarly project. The staffing model developed for this project is comprised of both physician critical providers and nurse practitioners. The goal of the collaboration between these two types of providers is to deliver care that is of the highest quality, consistency, and on par with the latest and best evidence based practices. More often than not, in today's healthcare environment, the care being provided is delivered by multiple individuals from differing specialties and areas, making interprofessional collaboration with the shared goals of improving health outcome of the utmost importance (AACN, 2006). COLLABORATIVE ICU STAFFING MODEL 23 Conclusion In conclusion, the use of collaborative staffing models comprised of NPs and MD intensivists providers are on the rise. This places NPs in a position to advance NP practice in the CC setting and become part of the solution that bridges the CC provider gap. This type of ICU provider staffing models may be especially useful to smaller to midsized ICUs, who face provider shortages. Unfortunately, there is no one size fits all CC provider staffing model. Each facility will have to perform an assessment of their specific CC needs in order to determine which staffing model will meet the needs of their particular ICU. Meeting the growing demand for CC services continues to increase in importance as our population ages and lives longer. Nurse Practitioner CC providers, particularly those educated under the adult acute care specialty, are poised to meet this growing demand. The overall goal of the proposed staffing model is to improve the quality of the care and services provided in the ICU and hopefully prevent the negative effects that can arise from CC provider turnover and burnout. COLLABORATIVE ICU STAFFING MODEL 24 References American Association of Colleges of Nursing. (2006). The Essentials of Doctoral Education for Advanced Nursing. Retrieved March 8, 2017, from http://www.aacn.nche.edu/publications/position/DNPEssentials.pdf Ewart, G. W., Marcus, L., Gaba, M. M., Bradner, R. H., Medina, J. L., & Chandler, E. B. (2004, April). The Critical Care Medicine Crisis: A Call for Federal Action. Chest, 125(4), 1518-1521. doi:10.1378/chest.125.4.1518 Gershengorn, H. B., Wunsch, H., Wahab, R., Leaf, D., Brodie, D., Li, G., & Factor, P. (2011, June). Impact of Nonphysician Staffing on Outcomes in a Medical ICU. Chest, 139(6), 1347-1353. doi:10.1378/chest.10-2648 Gesme, D., & Wiseman, M. (2010). How to implement change in practice. Journal of Oncology Practice, 6(5), 257-259. Halpern, N. A., Pastores, S. M., Oropello, J. M., & Kvetan, V. (2013, December). Critical Care Medicine in the United States: Addressing the Intensivist Shortage and Image of the Specialty. Critical Care Medicine, 41(12), 2754-2761. doi:10.1097/ccm.0b013e318298a6fb Isett, K. (n.d.). Analyzing systems change. Retrieved September 5, 2016, from http://www.modelsforchange.net/about/research/isett.html Knaus, W. A., Draper, E. A., Wagner, D. P., & Zimmerman, J. E. (1985). APACHE II: a severity of disease classification system. Critical Care Medicine, 13(10), 818-829. doi:10.1097/00003246-198510000-00009 COLLABORATIVE ICU STAFFING MODEL 25 Merlani, Paolo, Mélanie Verdon, Adrian Businger, Guido Domenighetti, Hans Pargger, and Bara Ricou. "Burnout in ICU Caregivers." American Journal of Respiratory and Critical Care Medicine 184.10 (2011): 1140-146. Web. Misra-Herbert, A. D., Kay, R., & Stoller, J. K. (2004, March/April). A Review of Physician Turnover: Rates, Causes, and Consequences. American Journal of Medical Quality, 19(2), 56-66. Mitchell, G. (2013). Selecting the best theory to implement planned change. Nursing Management - UK, 20(1), 32-37. Moss, M., Good, V. S., Gozal, D., Kleinpell, R., & Sessler, C. N. (2016, July). An Official Critical Care Societies Collaborative Statement. Critical Care Medicine, 44(7), 14141421. doi:10.1097/ccm.0000000000001885 Mullins, P. M., Goyal, M., & Pines, J. M. (2013, May). National Growth in Intensive Care Unit Admissions From Emergency Departments in the United States from 2002 to 2009. Academic Emergency Medicine, 20(5), 479-486. doi:10.1111/acem.12134 Shirey, M. R. (2013). Strategic Leadership for Organizational Change. Lewin's Theory of Planned Change as a Strategic Resource. Journal Of Nursing Administration, 43(2), 6972. doi:10.1097/NNA.0b013e31827f20a9 Society of Critical Care Medicine. (n.d.). Critical Care Statistics. Retrieved February 1, 2017, from http://www.sccm.org/Communications/Pages/CriticalCareStats.aspx Squiers, J., King, J., Wagner, C., Ashby, N., & Parmley, C. L. (2012, February 28). ACNP intensivist: A new ICU care delivery model and its supporting educational programs. Journal of the American Academy of Nurse Practitioners, 25(3), 119-125. doi:10.1111/j.1745-7599.2012.00789.x COLLABORATIVE ICU STAFFING MODEL 26 Ward, N. S., & Howell, M. D. (2015). Intensivist-to-patient ratios in ICUs. Current Opinion in Anaesthesiology, 28(2), 172-179. doi:10.1097/aco.0000000000000170 Ward, N. S., Afessa, B., Kleinpell, R., Tisherman, S., Ries, M., Howell, M., . . . Kahn, J. (2013). Intensivist/patient ratios in closed ICUs: A statement from the Society of Critical Care Medicine taskforce on ICU staffing. Critical Care Medicine, 41(2), 638-645. doi:10.1097/CCM.0b013e3182741478 COLLABORATIVE ICU STAFFING MODEL Appendices 27 COLLABORATIVE ICU STAFFING MODEL Appendix A - Lewin's Model of Change 28 COLLABORATIVE ICU STAFFING MODEL 29 COLLABORATIVE ICU STAFFING MODEL Appendix B - IRB Non-Human Research Determination 30 COLLABORATIVE ICU STAFFING MODEL 31 IRB: PI: Title: IRB_00097174 Jorge Delgadillo A Collaborative ICU Staffing Model: A Model Comprised of Nurse Practitioners and Physician Intensivists Date: 2/23/2017 Thank you for submitting your request for approval of this project. The IRB has administratively reviewed your application and has determined on 2/23/2017 that your project does NOT meet the definitions of Human Subjects Research according to Federal regulations. Therefore, IRB oversight is not required or necessary for your project. DETERMINATION JUSTIFICATION: The investigator will review de-identified data and administer a provider questionnaire in order to develop a staffing model as a quality improvement initiative. This project is not a systematic investigation designed to develop or contribute to generalizable knowledge. Rather, this project is a quality improvement initiative. Please note: any publication regarding this projects should be described as a quality improvement initiative, and not as "research". This determination of non-human subjects research only applies to the project as submitted to the IRB. Since this determination is not an approval, it does not expire or need renewal. Remember that all research involving human subjects must be approved or exempted by the IRB before the research is conducted. If you have questions about this, please contact our office at 581-3655 and we will be happy to assist you. Thank you again for submitting your proposal. SUPPORTING DOCUMENTS Surveys, etc. Critical care staffing questionnaire Literature Cited/References Literature Cited/References Click IRB_00097174 to view the application. Please take a moment to complete our customer service survey. We appreciate your opinions and feedback. COLLABORATIVE ICU STAFFING MODEL Appendix C - Critical Care Staffing Questionnaire 32 COLLABORATIVE ICU STAFFING MODEL 33 Critical Care Service Questionnaire Questionnaire Number: ___________ Date: __________________________ Participant (Title/Position): ___________________________________________________ Facility (Name/location): _____________________________________________________ Facility trauma designation (if any): _____________________________________________ Total number of critical care beds covered by your group: ___________________________ Total number of critical care providers in your group: _______________________________ Is your group currently staffed by only MDs: ______________________________________ Is your critical care service strictly a consulting service or are you also an admitting service? ___________________ On average how many day shifts do the providers in your group work per week? ____________________________ On average how long are day shifts? _____________________________ On average how many on call and/or overnight shifts do the providers in your group work per week? ___________ On average how long are on call and/or overnight shifts? ______________________ On average how many patients are the providers in your group primarily responsible for, this does not include patients being cared for during on call and/or overnight shifts? __________________________________________ On average how many patients are the providers in your group responsible for during on call and/or overnight shifts? _____________________________________________ Does or has your critical care provider group ever used physician extenders (i.e. NPs or PAs)? _________________ If so, how many and what role did they have in providing critical care services? _____________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________ COLLABORATIVE ICU STAFFING MODEL 34 Do you feel that the current staffing model employed by your group or facility provides adequate critical care services to the facility your group provides coverage for? _______________________________________________ If not, why? ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________ Do you feel that the current staffing model employed by your group or facility helps to prevent provider burnout? ______________________________________________________________________________ _______________ If so, why? ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________ If not, why? ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ ______________________________ COLLABORATIVE ICU STAFFING MODEL Appendix D - APACHE II Score 35 COLLABORATIVE ICU STAFFING MODEL 36 COLLABORATIVE ICU STAFFING MODEL Appendix E - PowerPoint Presentation to Project Site Stakeholders 37 COLLABORATIVE ICU STAFFING MODEL 38 COLLABORATIVE ICU STAFFING MODEL 39 COLLABORATIVE ICU STAFFING MODEL 40 COLLABORATIVE ICU STAFFING MODEL 41 COLLABORATIVE ICU STAFFING MODEL 42 COLLABORATIVE ICU STAFFING MODEL Appendix F - Abstract Submission Confirmation 43 COLLABORATIVE ICU STAFFING MODEL 44 COLLABORATIVE ICU STAFFING MODEL Appendix G - Local ICU Comparison 45 46 COLLABORATIVE ICU STAFFING MODEL Total ICU Bed / Provider Comparison 60 50 40 30 20 10 0 ICU A ICU B Number of ICU Beds ICU C ICU D Providers On / 24 hr. period ICU E COLLABORATIVE ICU STAFFING MODEL Appendix H - 2016 ICU Admits Per Day of the Week 47 48 COLLABORATIVE ICU STAFFING MODEL 2016 ICU Admits per Day of the Week Saturday 12% Sunday Sunday 8% Monday Monday 18% Friday 14% Tuesday 11% Thursday 17% Wednesday 20% Tuesday Wednesda y Thursday Friday Saturday COLLABORATIVE ICU STAFFING MODEL Appendix I - 2016 ICU Bi-hourly Admit Breakdown 49 50 COLLABORATIVE ICU STAFFING MODEL 2016 ICU Admit bi-hourly breakdown Number of Admits 75 70 74 54 46 39 26 29 29 15 00:00 01:59 39 02:00 03:59 04:00 05:59 21 06:00 07:59 08:00 09:59 10:00 11:59 12:00 13:59 14:00 15:59 16:00 17:59 Bi-hourly block of time 18:00 19:59 20:00 21:59 22:00 23:59 COLLABORATIVE ICU STAFFING MODEL Appendix J - Proposal Defense PowerPoint 51 COLLABORATIVE ICU STAFFING MODEL 52 COLLABORATIVE ICU STAFFING MODEL 53 COLLABORATIVE ICU STAFFING MODEL 54 COLLABORATIVE ICU STAFFING MODEL 55 COLLABORATIVE ICU STAFFING MODEL 56 COLLABORATIVE ICU STAFFING MODEL 57 COLLABORATIVE ICU STAFFING MODEL 58 COLLABORATIVE ICU STAFFING MODEL Appendix K - Poster 59 COLLABORATIVE ICU STAFFING MODEL 60