A staff planning system with a monitoring and control function for the effective use of staff resources at nursing wards

(1)

A staff planning system with a monitoring and control

function for the effective use of staff resources at nursing

wards

Master Thesis

by Sander Roelfsema

University of Groningen

Faculty of Economics and Business

Master Technology and Operations Management

28 January 2019

Supervisor University:

dr. ir. D.J. van der Zee

Co-assessor University:

dr. M.J. Land

Supervisor field of study:

(2)

(3)

(4)

Abstract

Context: Many nursing wards are facing the challenge that the demand for nursing care is increasing and highly variable in the number of patients and work content while staff resources to serve the demand are limited. Simply increasing the amount of resources to deal with the increasing demand and highly variable environment is often not possible due to resource constraints, such as the limited supply of nursing staff. Hence, the effective use of staff resources is paramount. This puts pressure on staff planning in balancing supply, i.e. the number of staff, their skills, and efficiencies, and demand, i.e. work content related to patients’ care needs. Unfortunately, current staff planning models often build on historical nurse/pat ie nt ratios, that deny the observed need for balancing of resource use. To improve the effective use of nursing staff a more refined tuning of supply and demand in terms of nurse/patient ratio’s is required, as well as a monitoring and control function, allowing for a regular update of ratios found, to account for changes in composition of the workforce, and patient population.

Objective: The objective of this research is to develop a staff planning system with a monitoring and control function to safeguard the effective use of the limited nursing staff at nursing wards, i.e. to prevent over or underutilization of nursing staff, and guarantee a certain quality of care.

Method: A design science oriented research has been conducted based on the four-phase regulative cycle from Wieringa (2009). The first phase consists of a literature study and a case study at the MDL nursing ward from the UMCG to address the current knowledge gap and investigate the problem faced by the nursing ward. The staff planning system is developed during the second phase and the third phase concerns its validation.

Results: A staff planning system with a monitoring and control function has been proposed, which can be used by healthcare practitioners to optimally plan their nursing staff considering the workload to prevent over or underutilization of the nursing staff. This staff planning system includes: indicators to determine the nurse/patient ratio considering the workload, threshold values and a monitoring and control function for updating nurse/patient ratios.

Limitations and future research: The generalizability of the staff planning system is limited to the Netherlands, since the indicators to determine the nurse/patient ratio are based on a study performed at 24 hospitals within the Netherlands. Therefore, future research could focus on applying the staff planning system in other regions to improve its validity.

(5)

Preface

Dear reader of the master thesis “A staff planning system with a monitoring and control functio n for the effective use of staff resources at nursing wards” I hope you will enjoy reading my thesis as much as I enjoyed writing the master thesis. This master thesis has been written as a final step towards the completion of the master program Technology and Operations Management at the University of Groningen. Finishing this master thesis has greatly contributed to my understanding of healthcare organizations, especially the complexity of nursing wards and its current staffing challenges, furthermore it provided me with new insights for future career opportunities.

I would like to thank several persons who contributed to the completion of my master thesis. First, I would like to thank dr. ir. D.J. van der Zee for his feedback and guidance through my master thesis project. His enthusiasm for the healthcare sector and expertise was very pleasant and useful, furthermore his critical view on my master thesis contributed to continuo us ly improving my work. Second, I would like to thank ir. T.J. Hoogstins for his feedback and assistance in my master thesis project. T.J. Hoogstins helped me to gather data within the University Medical Center Groningen (UMCG) and introduced me to the MDL nursing ward. Furthermore, his experience within the healthcare sector and interest in the master thesis project contributed to finishing my work.

(6)

1. Introduction

Pressure on the healthcare sector is increasing due to the ageing population and growing healthcare costs (Hallam, 2018). The healthcare sector is facing the challenge of providing safe, reliable and affordable services with limited resources (Brackett, Comer & Whichello, 2013; Deblois & Lepanto, 2016). The provision of health care services requires several important resource inputs such as: human resources, physical capital and consumables (World Health Organization, 2000). Therefore, to cope with these challenges, it is necessary for healthcare organizations to monitor and control their limited resources to safeguard their effective use (Gomes et al., 2010).

This research is motivated by the Maag-Darm-Lever (MDL) nursing ward from the Univers it y Medical Center Groningen (UMCG). The UMCG primarily focusses on the diagnosis and treatment of complex patients. A great diversity of patients is being treated at the MDL nursing ward which suffer from diseases like functional bowel disorders, liver disorders, Crohn’s disease and patients which had an organ transplantatio n. Key resources at the MDL nursing ward include; beds, nursing staff and medical equipment. The problem faced by the MDL nursing ward is that they are structurally not able to serve the incoming flow of patients due to the limited amount of nursing staff.

There are possibilities to extend the capacity of the nursing ward by employing more nursing staff, however the acquisition of nursing staff is difficult due to a nationwide shortage in nursing staff (Oulton, 2006; Marć et al., 2018). Furthermore, before hiring new nursing staff it should be justified that this nursing staff is required to serve the demand. Since the staff resources of the nursing ward are limited the effective use of staff resources is paramount. This puts pressure on the staff planning in balancing supply, i.e. the number of nursing staff, their skills, and efficiencies, and demand, i.e. work content related to patients’ care needs.

(9)

2

control function could be used to determine the optimal nurse/patient ratio to be used in the staff planning based on historical data from the patient population, furthermore a monitor ing and control function includes thresholds values and control measures (van Amelsvoort, 1992) to adjust the nurse/patient ratio if required.

The main objective of this research is “develop a staff planning system with a monitoring and control function to safeguard the effective use of the limited nursing staff at nursing wards, i.e. to prevent over or underutilization of nursing staff, and guarantee a certain quality of care”. A design science approach will be adopted as a research methodology. To retain and effective ly apply the limited nursing staff of nursing wards a staff planning system will be designed, which should assist healthcare practitioners in optimizing the planning of the limited nursing staff at nursing wards considering the workload of nursing staff. Furthermore, the usefulness of the system will be evaluated by applying it to the MDL nursing ward of the UMCG. Therefore, the research contribution of this study will be a staff planning system with a four-step monitor ing and control function, which should assist healthcare practitioners in optimally planning their nursing staff considering the workload to prevent any related losses at the nursing ward.

(10)

3

2. Problem description and research design

This chapter will first provide a problem description in section 2.1. In section 2.2 the research objective will be presented. Next, in section 2.3 the conceptual model will be presented, which shows the system that will be investigated. Then, in section 2.4 the research- and sub questions will be presented. Within section 2.5 the research design will be presented and the differe nt phases of the research.

2.1 Problem background

The main challenge faced by many nursing wards is that the demand for nursing care is increasing and highly variable in the number of patients and work content, while the number of nursing staff required to serve the demand are limited. The shortage in nursing staff together with the increasing demand for nursing care is a major problem jeopardizing the healthcare system. Within hospitals a high degree of variability is present since acute patients can arrive at any moment (Litvak & Long 2000; Olsson & Aronsson, 2015). Simply increasing the amount of resources to deal with the increasing demand and highly variable environment is often not possible due to resource constraints such as the limited supply of nursing staff (Oulton, 2006; Marć et al., 2018). In addition, the nursing staff in the Netherlands is relatively scarce and this is not expected to change for the time being (Van den Oetelaar et al., 2016).

According to the World Health Organization (2013; 2015) the global nursing shortage will reach a deficit of 12.9 million nurses in 2035. A factor influencing declining supply of nursing is an unfavorable work environment due to high workloads (Oulton, 2006). The limited supply of nurses with an increased demand for care could trigger a chain of events such as: insuffic ie nt staffing, increased stress on the workforce and higher turnover rates, which will reduce the quality of patient care (Fasoli et al., 2011). Therefore, retaining experienced nursing staff at the nursing ward is of utmost importance (Aiken et al, 2002).

(11)

4

would like to give, which is one of the many reasons why nurses leave their jobs (Oulton, 2006). Maintaining appropriate staffing levels is essential in providing high quality care (Kortbeek et al., 2015). Therefore, predicting workloads and planning nursing staff accordingly are of utmost important to guarantee high quality care to patients (Broyles et al., 2010; Vericourt & Jennings, 2011).

2.2 Research objective

Considering the challenge introduced in section 2.1, the main objective of this study is “develop a staff planning system with a monitoring and control function to safeguard the effective use of the limited nursing staff at nursing wards, i.e. to prevent over or underutilization of nursing staff, and guarantee a certain quality of care”. The staff planning system will support healthcare practitioners in optimizing the staff planning by balancing supply, i.e. the number of staff, their skills, and efficiencies, and demand, i.e. work content related to patients’ care needs.

2.3 Conceptual model

Figure 1 visualizes the conceptual model of the research. The planning system includes planning activities related to the planning of the nursing staff. Furthermore, the planning system includes a monitoring and control function, which is required to optimize the effective use of the limited staff resources by providing support to the nursing staff planning with a weighted nurse/patient ratio. To determine the optimal nurse/patient ratio to be used in the staff planning the nurse/patient ratio should be monitored and reviewed periodically by using indicators which reflect the workload for the nursing staff. The control measures should provide decision makers with solutions on which actions to take (control measures) once the determined threshold values have been exceeded.

(12)

5

2.4 Research Questions

To provide a solution to the problem faced by nursing wards the following research question should be answered:

How to effectively apply the limited nursing staff of a nursing ward, i.e. prevent the over and underutilization, thereby accounting for changes in composition of the workforce, and patient population?

Sub-questions that should be answered to answer the main research question are as follows :

 Which indicators could be used to reflect a possible mismatch between the supply of nursing staff and the demand of patients?

 How often should these indicators be measured?

 Which threshold values should be applied to these indicators before undertaking action?  Which control measures could be taken if threshold values are reached?

2.5 Research Design

A design science approach will be adopted as the research methodology for this study. This type of research is appropriate since it is primarily aimed at practical problem solving by developing solution designs or “artifacts” for issues that have not been investigated before (Holmström, Ketokivi, Hameri, 2009). The design science method that will be used in this research is based on the four-phase regulative cycle from Wieringa (2009), figure 2. This method will be applied since it will help to solve a practical problem. The first three phases of the regulative cycle (problem investigation, solution design and design validation) will be conducted and the final phase (implementation) is considered to be beyond the scope of this research.

(13)

6 Figure 2: Regulative cycle by Wieringa (2009)

2.5.1 Case description

This research is motivated by the MDL department from the UMCG. The department consists of an endoscopic center, an outpatient department and a nursing ward. A great diversity of patients is being treated at this nursing ward which suffer from diseases like functional bowel disorders (FBD), liver disorder, Crohn’s disease and patients which had an organ transplantation. This nursing ward has a capacity of 30 beds and employs 30 nurses. Every year, roughly 700 patients are being treated at the nursing ward. However, this nursing ward is facing the problem of not being able to serve the incoming flow of patients due to the limited amount of nursing staff. Since the key resources of the nursing ward are limited it is important that the available resources are being used as effective as possible, i.e. to prevent over or underutilization.

2.5.2 Phase 1: Gathering information

During the first phase the problem and the system (MDL nursing ward) will be investigated to get an understanding of the occurring problem and the system. The goal of this phase is to describe the problem and possibly give predictions on what could happen if no intervent io ns would be applied (Wieringa, 2009). The information about the system will be gathered by interviewing the nursing staff, management and planners of the MDL nursing ward of the UMCG. These employees will be interviewed to get an understanding of the admission of patients, the treatment of patients, discharging of patients and the planning process of the nursing ward. Furthermore, a literature research, which serves as a knowledge base, will be conducted to gather an in-depth understanding of the problem.

2.5.3 Phase 2: Designing a planning system with a monitoring and control function

(14)

7

the nurse/patient ratio at the nursing ward. To monitor the nurse/patient ratio, it is important to identify indicators for measurement and determine how often these indicators should be measured. These indicators, threshold values, and control measures will be determined by performing a literature research and conducting interviews with employees from the UMCG.

2.5.4 Phase 3: Design validation

(15)

8

3. Theoretical background

Within this chapter literature will be reviewed concerning existing systems for planning nursing staff. Section 3.1 sketches challenges faced in nurse staff planning. Next, section 3.2 discusses staff planning systems and estimators for nurse work load. Section 3.3 elaborates on the concept of monitoring and control. The findings are summarized in 3.4

3.1 Challenges faced by nursing wards

A nursing ward can be modelled as a three-stage system, figure 3 (Pannick, Beveridge, Wachter and Sevdalis, 2014). The first stage is the admission stage where a patient can arrive at the nursing ward from different sources such as other hospital departments or nursing homes. The second stage is the care process where the patient receives the care that is needed. If the patient has been treated and is sufficiently recovered from a medical point of view the patient enters the final (third) stage which means that the patient will be discharged from the nursing ward

The main challenge faced by many nursing wards is that the demand for nursing care and the complexity is increasing and highly variable while the nursing staff required to serve the demand is limited (Litvak & Long 2000; Olsson & Aronsson, 2015). Since the staff resources of the nursing ward are limited the effective use of staff resources is paramount. This puts pressure on the staff planning in balancing supply, i.e. the number of nursing staff, their skills, and efficiencies, and demand, i.e. work content related to patients’ care needs (Brandeau et al., 2004; Harper et al., 2010; Kortbeek et al., 2015).

(16)

9

3.2 Planning nursing staff

Healthcare planning systems are often lacking because healthcare managers are dedicated to provide the best possible care to patients, however they often lack the knowledge on how to effectively use their resources (Carter, 2002). The function of a planning system is to coordinate resources (staff, equipment, and materials) in such a way that the objectives of the organiza t io n are realized (Robert, 1965). This research will focus on the “offline ” and “online” operational levels of the resource capacity planning, figure 4.

Hans, van Houdenhoven, & Hulshof (2012) proposed a four-by-four generic framework for healthcare planning, which includes four hierarchical control levels and four managerial areas, figure 4. The framework facilitates a dialogue between clinical staff and the management to design planning and control mechanisms, which are required to translate the organizatio na l objectives into effective and efficient healthcare processes (Delesie, 1998).

Figure 4: Framework for healthcare planning and control (Adapted from Hans et al., 2012)

(17)

10

Within the framework (figure 4) a distinction has been made between “offline” and “online” on the operational level of the planning. This distinction reflects the difference between in advance , i.e. “offline” planning, decision making and reactive, i.e. “online” planning. decision making. An example of “offline” operational planning is nursing staff rostering and an example of “online” operational planning is the add-on scheduling in case of emergencies. Due to the high variability in the system (Litvak & Long 2000; Olsson & Aronsson, 2015) reactive decision making is required that deals with monitoring the process and reacting to any unforeseen events (Hans et al., 2012).

A common issue in healthcare organizations is that managers would like to solve problems on the operational level by adding more resource capacity (on the strategic level), however they often overlook the opportunity to use their resources more effective (Hans et al, 2012). To accurately manage the resource capacity, i.e. like the number of nurses during each shift, operational planning issues should be handled together with tactical planning issues (Elkhuize n et al., 2007). Current staff planning models often build on historical nurse/patient ratios, that deny the observed need for balancing of resource use. Currently, there is a gap in literature on how to effectively apply nursing staff considering the fluctuations in supply and demand as mentioned before, which is where a monitoring and control function would have added value to regularly update the nurse/patient ratio (Kortbeek et al., 2015).

3.2.1 Staff planning - workload estimation

By implementing a staff planning system with a monitoring and control function, which supports healthcare practitioners in managing the workload for the nursing staff, the employees can be kept healthier, because a high workload is a predictor for burnouts (Ohue et al., 2011; Laschinger et al., 2012) and absenteeism (Mudaly & Nknosi, 2015) of nursing staff. Furthermore, Lacey et al. (2007) and Leone et al. (2015) have shown that a high workload influences nursing staff to leave their jobs, which makes the nursing shortage even worse.

(18)

11

There is a direct relationship between the workload of nursing staff, outcomes for patients and nurse-reported quality of care (Hinno et al., 2012; Liu et al., 2012; Aiken et al., 2014). Fasoli & Haddock (2010) identified several indicators reflecting the nursing workload. Table 1 shows the variables as identified by Fasoli & Haddock (2010) to consider in a staffing model for nurses.

Table 1: Nurse staffing indicators as identified by Fasoli & Haddock (2010)

In order to determine the nurse-to patient ratio Fasoli et al. (2011) identified several indicators, which could be used to determine the required number of nurses during a certain shift. The following indicators were identified by Fasoli et al. (2011):

1. Average length of stay: as a surrogate marker for patient severity of illness.

2. Average number of medication doses administered per day: multiple medications are another surrogate marker used as an indicator of complexity, severity, and staff time required for administration.

3. Percentage of patients older than 70 years, this could indicate that a patient requires additional assistance from the nursing staff in the completion of daily activities.

4. Percentage of patients with a body mass index (BMI) of 25 or higher, this could indicate that a patient requires additional assistance from the nursing staff with daily activities. 5. Top 3 diagnostic categories on the unit, this could indicate the complexity/scope of

nursing care required by the patients.

6. Average daily census, this is a unit- or organizational-level measure of patient volume and nursing staff workload.

(19)

12

3.3 Monitoring and control

To monitor and control an organization van Amelsvoort (1992) proposed a monitoring and control system, which can be used by an organization to adapt itself to changing external/internal circumstances. The system is characterized by a four-step approach, figure 5. This system supports organizations to take decisions based on measured information and established threshold values. The system consists of four functions which are the follow i ng: observing, assessing, deciding, controlling.

These functions will always be performed in the same order, from which the first phase is observing. During the observing phase the current state of the system under investigation is being observed to create a solid base for any further decision making. The next phase is the assessment phase were the performance of the current state of the system is being compared against the thresholds. The third phase is the decision-making phase where is decided whether intervention is required to adjust the performance of the system. If intervention is required control measures will be implemented during the controlling phase. However, it should be noted that not every situation requires intervention and thus the control measures might be skipped in some situations.

(20)

13 Figure 5: Monitoring and control system adapted from van Amelsvoort p.44 (1992)

3.4 Summary of findings

The main challenge faced by many nursing wards is that the demand for nursing care is increasing and highly variable while the nursing staff required to serve the demand is limited. Since the nursing staff in the Netherlands is relatively scarce and is not expected to change for the time being the nursing staff should be retained and effectively used, i.e. balancing supply and demand in terms of nurse/patient ratios.

(21)

14

4. System description – Nursing ward

In this chapter the case study system will be described by introducing the MDL nursing ward in terms of patients served, the care process, (staff) resources and their planning (Sections 4.1-4.6). Next, the planning system will be analyzed for its contribution to the effective use of nursing staff.

4.1 Overview MDL nursing ward

The MDL department consists of an endoscopic center, an outpatient department and a nursing ward (care process), figure 6. The MDL nursing ward is responsible for the treatment of patients and operates 24 hours per day and 7 days a week.

Figure 6: Flow diagram MDL nursing ward

4.2 Inflow of patients

Related to the variability in arrivals of patients there are two types of patients which could arrive at the MDL nursing ward; acute patients (58%) and elective patients (42%), figure 7. The acute patient group is causing a lot of variation within the hospital as stated by Olson & Aronsson (2015). Acute patients could arrive at the emergency department from the hospital at any moment from where they might need a diagnosis or treatment from the MDL department (Litvak & Long 2000). Data from January 1st_{2018 until November 13}th_{2018 related to the}

(22)

15 Figure 7: Division elective/acute patients at the MDL nursing ward

Elective patients do not require immediate care; therefore, the admission of these patients can be planned, i.e. in the “offline” planning, in contrast to acute patients, i.e. in the “online ” planning. Patients could arrive at the nursing ward coming from different sources as an elective patient or as an acute patient, figure 6. Since the UMCG is specialized in treating patients with very complex diseases patients sometimes come over from a community hospital. Another possibility is that patients arrive at the MDL nursing ward from other departments within the UMCG due to several reasons. Sometimes patients from other nursing wards are admitted at the MDL nursing ward due to capacity constraints at other nursing wards.

Furthermore, data from January 1st_{2018 until November 13}th_{2018 related to the length of stay}

of patients at the MDL nursing ward has been analyzed, figure 8. During this period the average length of stay was 6,3 days from the 1049 patients that were admitted at the MDL nursing ward. The length of stay refers to the actual stay of the patient at the MDL nursing ward. Sometimes patients are being transferred to another nursing ward during their stay and come back to the MDL nursing ward at a later moment. However, within figure 8 only the time is being included that the patient is staying at the MDL nursing ward.

Elective (Acute, within 24 hours); 99; 9% Elective; 347; 33% Acute; 603; 58%

(23)

16 Figure 8: Length of stay of patients at the MDL nursing ward

4.3 Care process

As can be seen from figure 6 the care process consists of several consecutive stages which are; the admission, diagnosis, treatment and discharging. During the first stage the acute or elective patient is being admitted at the nursing ward. The nursing staff is responsible to guide the patient trough the admission process and transfer the patient to an available bed at the nursing ward. In principle, all patients will be admitted if sufficient resources are available. Some patients require further diagnosis (second stage) before their treatment elsewhere. The nursing staff will transfer those patients. During diagnosis, the patient will remain at the nursing ward, requiring care services from the nursing staff.

After patients have been diagnosed they might receive treatment (third stage) at the nursing ward. The treatment can vary a lot between patients, which makes the treatment process variable for the nursing staff. However, since the UMCG is more focused on the treatment of complex patients the MDL nursing ward experiences that most of their beds are occupied by complex patients. This has consequences for the nursing staff, because they are responsible for the treatment of complex patients the nursing staff is required to perform more complex tasks.

(24)

17

4.4 Resources

Key ward resources include; beds (30 beds), nursing staff (30 nurses), and medical equipment. The MDL nursing ward is working in 3 shifts where each shift requires a different number of nursing staff to treat the patients. At the MDL nursing ward, several functions are present. The tasks and responsibilities for each function are explained in table 2.

Table 2: Functions with tasks and responsibilities

Function Tasks and responsibilities

Directing nurse Responsible for the whole care process of the MDL nursing ward including the admission and discharging of patients, also performing nursing activities.

Nursing staff Responsible for the treatment of patients and organizing the admissio ns and discharges of patients.

Planning nurse Responsible for making the planning related to the nursing staff and patients at the nursing ward, also performing nursing activities.

Senior nurse Responsible for the treatment of patients and actively participate in organizational activities that support the whole care process

Care assistants Responsible for supporting nursing staff in daily care tasks.

4.5 Current planning method MDL nursing ward

The MDL nursing ward is responsible for their own planning of nursing staff and patients. Whereas other nursing wards get their planning done by a central planning department, the MDL nursing ward has its own planning nurses who perform planning activities besides their nursing activities. A couple of weeks in advance, i.e. on the “offline” operational planning level (figure 9), the available nursing staff is known to the planning nurses. The nursing staff is divided over the patients considering the complexity of care these patients require.

(25)

18 Figure 9: Planning MDL nursing ward

If most of the patients at the nursing ward would be patients which require less complex treatment, then the nurse to patient ratio would be as follows according to the directing nurse of the MDL department:

 During the “day” shift, which is from 7:30AM until 4:00PM, six nurses should be sufficient to perform the treatment of patients at the nursing ward. Because the basic care of patients is being performed during this shift.

 During the “late” shift, which is from 3:00PM until 11:30PM, four nurses should be sufficient to perform the treatment of patients at the nursing ward.



During the “night” shift, which is from 11:15PM until 7:45AM, two nurses should be sufficient to perform the treatment of patients at the nursing ward.

4.6 Current workload evaluation MDL nursing ward

The MDL nursing ward currently has no structured monitoring and control function to determine whether the nurse/patient ratio being used in the “offline” staff planning is sustainable and effective. Currently, the nurse/patient ratio being used in the “offline” staff planning at the MDL nursing wards is a result of historical development since there is no tool available to base the current nurse/patient ratio on information about supply and demand . Therefore, control measures are implemented relatively last-minute and are reactive due to the very short term view while reviewing the effectiveness of the nurse/patient ratio, i.e. in the “online” staff planning, (figure 9).

(26)

19

results in the blocking of beds. The nursing staff planning is being made a couple weeks in advance, i.e. “offline” staff planning (figure 9), so that for each shift it is known how many nurses are present to treat the patients at the nursing ward. A monitoring a control functio n included in the planning system makes it possible to review the supply of nursing staff and demand of patients several weeks up to a few days in advance, which allows decision makers to implement control measures more in advance in the “offline” staff planning (figure 9) to adjust the nurse/patient ratio (preventive).

4.7 System analysis

To determine the effectiveness of the current nurse/patient ratio being used in the “offline” staff planning a description will be provided on how the current evaluation of the workload is being performed. Currently, the nurse/patient ratio being used in the “offline” staff planning at nursing wards is a result of historical development since healthcare practitioners lack the tools to base the current nurse/patient ratio on information about supply and demand (Brandeau et al., 2004; Harper et al., 2010; Kortbeek et al., 2015). The nurse/patient ratio that is being used by the MDL nursing ward during each shift is based on a pre-determined nurse/patient ratio by the management of the nursing ward. However, this nurse/patient ratio has been determined a long time ago and might not be appropriate anymore due to changes in the patient population and workforce.

Due to the ineffective use of limited nursing staff beds at nursing wards needs to be closed by the management to guarantee a certain quality of care. Therefore, the reasons of blocking beds at the MDL nursing ward have been analyzed. As can be seen from shown figure 10 beds are frequently unavailable due to nursing staff shortages. Data from January 1st_{2018 until}

November 6st_{2018 related to reasons for blocking beds at the MDL nursing ward have been}

analyzed. Within this period 1302 beds were blocked with an average blocking time of 46 hours and 30 minutes.

(27)

20

The complexity of patients at the nursing ward is increasing, which makes the treatment of patients more complex for the nursing staff. Furthermore, if nurses are responsible for the treatment of complex patients the physical and mental work pressure increases, which could lead to illness or burnout of the nursing staff. In addition, most of the nursing staff at the MDL nursing staff is relatively new to the nursing profession. Therefore, these nurses are not able to perform all complex treatment activities that are required, which results in a lower nurse/pat ie nt ratio. Thus, more nurses are required to treat the present patients at the nursing ward or beds should be closed to lower the number of patients at the nursing ward.

Currently, there is no structured approach to balance the supply of nursing staff with the demand of patients at the MDL nursing ward, i.e. determining and adjusting the nurse/patient ratio. This has several consequences, such as absenteeism and burnout of nursing staff due to ineffec t ive nurse/patient ratios, which results in an even worse working environment for the remaining nursing staff. If there are no sufficient nurses available, no patients can be admitted at the nursing ward since beds will be closed due to a shortage in nursing staff. However, by working in this manner in most cases the control measures will be implemented to late since the current workload evaluation is more reactive, i.e. “online” staff planning”, instead of preventive, i.e. “offline” staff planning”, figure 9.

Different; 72; 5%

Bed for general urgency; 25; 2% Bed for own urgency; 140; 11% No regular nursing bed; 4; 0% Reserved for internal takeover; 85; 7% Reserved for admission; 638; 49% Isolation; 65; 5% Staff availability; 255; 20% Rooming-in; 8; 1%

Reasons for blocking beds

(28)

21

5. System design – Planning system

This chapter discusses the development of the planning system aimed at safeguarding the effective use, i.e. preventing over and underutilization, of the nursing staff at the nursing ward. The approach of the development of the planning system and the system itself will be discussed in section 5.1. Section 5.2 will discuss the indicators to determine the nurse/patient ratio. Section 5.3 will discuss the thresholds values, and section 5.4 will discuss the control measures. Finally, section 5.5 will demonstrate the use of the monitoring and control function.

5.1 Approach

The designed staff planning system with a monitoring and control function will focus on optimizing the “offline” and “online” planning of nursing staff, figure 11. The system analysis in section 4.7 has shown that the lack of an adequate staff planning system is the main bottleneck in nursing care system, causing several losses such as the blocking of beds. The optimization of the staff planning will be done by a monitoring and control function allowing for regularly updating the nurse/patient ratio using indicators reflecting the workload, to account for changes in composition of the workforce, and patient population.

Several employees of the UMCG have been interviewed to gather relevant information for the design of the system. Information has been gathered related to workload indicators, which could be used to reflect a possible mismatch between the supply of nursing staff and the demand of patients. Furthermore, information has been gathered related to thresholds to determine when to undertake action if the performance of the system is declining. Finally, information has been gathered related to possible control measures, which could be implemented in the “offline” and “online” operational staff planning to adjust the nurse/patient ratio to improve the effective use of nursing staff. The different interviewees and the purpose of these interviews can be found in appendix A.

5.1.1 Planning system with a monitoring and control function

(29)

22 Figure 11: Planning system with a Monitoring and control function

The optimal nurse/patient ratio to be used in the “offline” operational planning, i.e. a couple of weeks in advance, will be provided by the monitoring and control function using indicators reflecting the workload. This ratio will be updated/reinitialized by the monitoring and control function periodically to account for changes in composition of the workforce, and patient population. However, since the nursing staff planning is made a couple of weeks in advance, i.e. the “offline” staff planning, it should be continuously reviewed on the short term due to the high variability in the system (Litvak & Long 2000; Olsson & Aronsson, 2015). On the short term, i.e. last-minute, control measures might be implemented in the “online” operational planning to adjust the nurse/patient ratio once the thresholds have been exceeded.

5.1.2 Methods

(30)

23

decision making in the “offline” and “online” operational planning of nursing staff. Thresholds are being used in the assessment phase were the nurse/patient ratio is compared against the thresholds. Control measures should be in place to intervene in the “offline” and “online ” operational planning if the nurse/patient ratio is unsatisfactory.

5.2 Indicators to determine the workload

To determine whether the current nurse/patient ratio being used in the “offline” staff planning is effective the optimal nurse/patient ratio should be determined beforehand, which is where a monitoring and control function is required. Several indicators have been identified to determine the nurse/patient ratio during a certain shift at the nursing ward to be used in the “offline” staff planning, table 3. The realized nurse/patient ratio should be monitored and compared against the proposed nurse/patient ratio. As the function of a monitoring and control system is to compare the planned performance against the actual performance and take corrective action if required.

The indicators to determine the optimal nurse/patient ratio are the following according to the interviewees of the UMCG; Average length of stay, Daily patient turnover, Probability of IC, BMI, Age, Diagnosis, Medication, Origin of the patient and the Psychosocial need. The indicators, which are being used by the UMCG to determine the nurse/patient ratio are based on the indicators proposed by Fasoli & Haddock (2010) and Fasoli et al. (2011) section However, two additional indicators have been proposed by interviewees of the UMCG, which are the origin of the patient and the psychosocial need. A description of these indicators and their relevance have been given in table 3. According to the interviewees of the UMCG a ll indicators are of equal importance in determining the nurse/patient ratio for the patient population at the nursing ward.

Table 3: Indicators to determine the nurse/patient ratio

Indicator Description Relevance

Average length of stay

The average total length of stay of patients which were located at the nursing ward during their admission.

Represents the severity of illness of the patient, which influences the care required by the patient.

Daily patient turnover

Admissions, transfers and discharges of patients.

Represents the workload and direct care activities related to the patient care transition processes.

Probability of IC

Probability that a patient needs intensive care (IC).

(31)

24

BMI Percentage of patients with a body mass index (BMI) of 30 or higher.

This could indicate that a patient requires additional assistance from the nursing staff with daily activities. Age Percentage of patients older than 70

years.

This could indicate that a patient requires additional assistance from the nursing staff in the completion of daily activities.

Diagnosis Number of different diagnoses of 80% percent of the patients at the nursing ward.

This indicates the complexity/scope of nursing care required by the patients.

Medication Average number of medication doses administered per day.

This can be used as an indicator of complexity, severity, and staff time required for administration.

Origin of the patient

Origin and destination of the patient. From or to another hospital.

Represents the severity of illness of the patient, which influences the care required by the patient.

Psychosocial need

The extent to which the psychosocia l needs of the patient populatio n contribute to the workload in the department based on an “online ” questionnaire completed by the nursing staff.

This could indicate that the patient requires additional assistance from the nursing staff due to psychologica l issues.

5.3 Determining threshold values

While monitoring the nurse/patient ratio thresholds should be set on the nurse/patient ratio to safeguard the effective use of nursing staff, i.e. a sustainable workload, and to guarantee a certain quality of care. These thresholds should be set by the decision makers and used as boundaries, once the monitored nurse/patient ratio exceeds these thresholds the nurse/pat ie nt ratio should be reconsidered. If the monitored nurse/patient ratio will exceed a certain threshold, a signal should indicate that the head nurse or directing nurse should take corrective actions preferably in advance, i.e. in the “offline” operational planning. Since the indicators explained in the previous section provide decision makers with a certain nurse/patient ratio, threshold values can be set accordingly.

(32)

25

The second threshold is related to the effective use of the nursing staff. To guarantee that the nursing staff is being employed in a sustainable working environment during a certain shift, a certain percentage of all shifts the nurse/patient ratio should be equal to or above the nurse/patient ratio proposed by the indicators from table 3. This percentage can be determined by the manager of the nursing ward, who should keep in mind that this will be a consideratio n between extra staffing costs or closing beds at the nursing ward due to insufficient nursing staff.

5.4 Determining control measures

When the threshold values are exceeded control measures could be applied to adjust the nurse/patient ratio in the “offline” or “online” operational staff planning. There are two main control measures to adjust the nurse/patient ratio: (re)deploying nursing staff, and closing beds. The decision to implement control measures will be taken by the head nurse of directing nurse. The decision maker might also decide not to intervene in the nursing care process. Since the focus of the proposed system is to safeguard the effective use of the nursing staff the control measures are related to adjusting the workload of the nursing staff, while keeping in mind that quality care should be delivered.

Control measures should preferably be implemented preventive, i.e. in the “offline” operational planning (figure 11) since most of the times the only control measure that is possible last-minute, i.e. in the “online” operational planning, is closing beds at the nursing ward. This is due to the fact that there is a shortage in nursing staff (Oulton, 2006; Marć et al., 2018), which makes it difficult to gather extra nursing staff last-minute, i.e. in the “online” operational planning (figure 11).

(33)

26

However, it could also occur that more nursing staff is being planned in the “offline” operational planning for a certain shift than required. If this is the case, the nursing staff might have no tasks to perform during some time of their shift, which results in the ineffective use of nursing staff. A control measure which could be applied is to adjust the number of nursing staff during the shift, i.e. in the “online” operational planning, by sending some nurses back home or letting them perform other tasks.

Another control measure to deal with the unsustainable nurse/patient ratio would be to close beds at the nursing ward. If insufficient nursing staff can be planned for a certain shift beds could be blocked in advance, i.e. in the “offline” planning. By blocking beds the total number of patients being admitted at the nursing ward will be reduced which makes the nurse/pat ie nt ratio more sustainable if there is a shortage in nursing staff. However, this control measure is less favorable, since for each absent nurse on average 4 or 5 beds should be blocked.

5.5 How to use the monitoring and control function

The monitoring and control function which has been added to the planning system in figure 11 has been developed based on the four-step approach from van Amelsvoort (1992), where the goal is to support healthcare practioners, i.e. the planning staff and the management of the nursing ward, to take decisions based on measured information and established threshold values. Next, these four steps will be explained to demonstrate how to use the monitoring and control function to optimize the “offline” and “online” nursing staff planning, figure 11.

Phase 1: Observing

A first step in regulating the staff resources would be to observe the current situation. During this phase the current nurse/patient ratio for each shift, is being observed to create a solid base for any further decision making. Furthermore, other aspects could be observed to assess the current nurse/patient ratio, which are the job satisfaction of the nursing staff or the absenteeism of nursing staff, figure 11.

Phase 2: Assessing

(34)

27

on how many hours of direct care are required during a shift, and thus how many nurses are required during a shift.

By calculating the total hours of direct care required by the patient population at the nursing ward the nurse/patient ratio can be determined and the nursing staff can be planned accordingly. This “optimal” nurse/patient ratio will be fixed for a certain period (e.g. 3 months as proposed by the interviewees from the UMCG) and reviewed periodically to account for changes in the composition of the workforce and patient population. To perform these workload calculat io ns the UMCG will be using a software tool from HotFlo, which will be implemented in 2019. Due to confidentiality reasons the calculations which are being used by Hotflo cannot be demonstrated.

While determining the nurse/patient ratio for a certain shift it should be kept in mind that the experience of the nursing staff also plays an important role in delivering quality care as stated by Fasoli & Haddock (2010). If all nursing staff would be considered as equally employable in the planning it could reduce the quality of care since insufficient staffing will reduce the quality of patient care (Fasoli et al., 2011). Insufficient staffing might occur if all nurses are considered to be equal while making the “offline” operational nursing staff planning. To include the different levels of experience of the nursing staff in the planning a certain correction factor should be considered related to the experience of the nursing staff to consider their employability at the nursing ward, table 4.

Table 4: Correction factor for employability of nursing staff

Nursing staff work experience Correction factor

Less than 1 year 0,6

Between 1 and 2 years 0,8

More than 2 years 1

Phase 3 & 4: Deciding & Controlling

(35)

28

6. Case study validation

This chapter will report on the results obtained from the application of the proposed staff planning system with a monitoring and control function at the MDL nursing ward from the UMCG. The first section will explain how the planning system with a monitoring and control function has been validated. Within the second section the application of the indicators to determine the nurse/patient ratio will be demonstrated for the MDL nursing ward.

6.1 Approach

The monitoring and control function to determine the nurse/patient ratio to be used in the “offline” staff planning has been put to the test to illustrate the use since there is currently no such approach to optimize the “offline” staff planning. Historical data related to the patient population from the MDL nursing ward has been gathered from the EHR of the UMCG. Based on this data the nurse/patient ratio has been determined by using a software tool from Hotflo. Then, based on the outcomes of the software tool a situation at the MDL nursing ward has been simulated to illustrate the use of the threshold values and control measures of the proposed monitoring and control function, figure 11. Furthermore, to validate the staff planning system with a monitoring and control function the system design has been discussed with the manager of the MDL nursing ward and the advisor care logistics of the UMCG.

6.2 Demonstrating the use of the monitoring and control function

A monitoring and control function is required for a more refined tuning of supply and demand in terms of nurse/patient ratios to improve the effective use of staff. Therefore, this section will illustrate how the monitoring and control function could improve the “offline” and “online ” operational staff planning by applying it to the MDL nursing ward.

(36)

29

Based on the outcomes in table 5 the number of direct care hours have been calculated by the software tool from Hotflo, which can be used to determine the nurse/patient ratio. Based on the outcomes in table 5 the number of direct care hours per patient would be 4 hours per 24-hours. Where the workload is divided over the shifts as follows; 50% during the day-shift, 30% during the late-shift and 20% during the night-shift. This division of the workload is being used since most of the nursing tasks are performed during the day-shift (e.g. the admission and transfers). These results have been used to determine the nurse/patient ratio at the MDL nursing ward, which are shown in table 6 for each shift. As can be seen from table 6 the proposed nurse/pat ie nt ratio by the software tool from HotFlo differs for each shift, since during the day shift the percentage of direct care hours required by patients is more than during late and night shifts.

To provide an example of the staff planning system with a monitoring and control function the following situation will be used; Monday 14 January 24 patients will be staying at the MDL nursing ward. Related to the thresholds the following percentages are established; 50% of the nurses in a shift should be a senior nurse and for simplification purposes all shifts the nurse/patient ratio should be above the ratio provided by the indicators.

Based on monitoring the data from the patient population the proposed nurse patient ratio has been established as shown in table 6. This would imply that during the day shift 6 (24/4 = 6) nurses would be required to perform the care activities. For the late shift 3,63 (24 / 6,6 = 3,33) nurses and for the night shift 2,42 (24 / 9,9 = 2,42) nurses. Considering the employability of the different types of nursing staff as discussed in section 5.5 these shifts could be performed by a different combination of nurses.

(37)

30 Table 5: Validation of the workload indicators at the MDL nursing ward

Table 6: Nurse/patient ratio MDL nursing ward

Indicator Description Score

The average length of stay of patients at the nursing ward 8,65

The number of admissions, transfers and admissions per day. 0,3 The number of transfers from and to the operation room per day. 0,03 The number of transfers to the radiology department per day. 0,13 The number of transfers to the function department per day. 0,11 The share of admissions with a part of their stay at the IC and the nursing ward. 9,23

The share of patients at the MDL nursing ward with a BMI of 25 or higher. 38,78 The share of patients at the MDL nursing ward with a BMI of 30 or higher. 11,81 The share of patients at the MDL nursing ward with a BMI of 35 or higher. 3,58 The average BMI of patients beinig admitted at the nursing ward. _23,34 Age The share of patients beining admitted at the MDL nursing older than 70 years. 15,82 The number of different diagnosis for 80% of the admissions at the nursing ward. 26

The number of different ATC-codes from all patients at the nursing ward. 10,81 The average score on the questionnaire filled in by the nursing staff varying

between 0 and 1. 0,53

Diagnosis

The number of different specialisms and diagnosis combinations for 80% of the

admissions at the nursing ward. 27

The number of different specialisms and diagnosis combinations for 70% of the

admissions at the nursing ward. 19

The number of different ATC-codes from the patients at the nursing ward with

at least one ATC-code. 11,34

Medication Psychosocial need 9,08 6,63 BMI Daily patient turnover

The share of admissions with a part of their stay at the IC (for longer than 4 hours) and the nursing ward.

The share of admissions with a part of their stay at the IC (for longer than 24 hours) and the nursing ward.

Probability of IC

Share of patients being admitted at the nursing ward coming from another

hospital and going to another hospital after their admission. 1,55

Share of patients being admitted at the nursing ward coming from another

hospital and not going to another hospital after their admission. 6,33 Length of stay

Share of patients being admitted at the nursing ward going to another hospital

(38)

31

Another option would be to try and gather more nursing staff from the internal pool in the hospital, however it should be kept in mind that these nurses have a lower employabilit y (correction factor 0,6), table 4. Which would imply that 5 nurses from the internal pool would be required to fill up the gap of 3 experienced nurses. If, however it would be impossible to gather more nursing staff a final control measure would be to close beds at the nursing ward for that specific day. Since only 3 nurses will be available all the beds which they cannot serve should be blocked by the decision maker.

6.3 Evaluation

(39)

32

7. Discussion

Within this chapter a discussion will be provided related to the staff planning system with a monitoring and control function, the research contributions, the research limitations and possibilities for future research.

7.1 Research contributions

The main objective of this study was to “develop a staff planning system with a monitoring and control function to safeguard the effective use of the limited nursing staff at nursing wards, i.e. to prevent over or underutilization of nursing staff, and guarantee a certain quality of care”.

Current staff planning models often build on historical nurse/patient ratios, that deny the observed need for balancing of resource use (Brandeau et al., 2004; Harper et al., 2010; Kortbeek et al., 2015). Case study results indicate that the lack of an adequate staff planning system is the main bottleneck in the nursing care process, causing several losses such as the blocking of beds. To improve the effective use of nursing staff a more refined tuning of supp ly and demand in terms of nurse/patient ratio’s is required, as well as a monitoring and control function, allowing for a regular update of ratios found, to account for changes in compositio n of the workforce, and patient population.

Quite often beds are blocked by the head- or directing nurse due to staffing shortages, which decreases the capacity of the nursing ward. These staffing shortages occur mainly because the workload for the nursing staff is not sufficiently balanced, which causes the nursing staff to become ill or burnout. By implementing a staff planning system with a monitoring and control function to smooth the workload the nursing staff can be kept healthier, because an unsustainable workload is a predictor for burnouts (Laschinger et al., 2012; Ohue et al., 2011) and absenteeism (Mudaly & Nknosi, 2015) of nursing staff. Furthermore, Lacey et al. (2007) and Leone et al. (2015) have shown that unsustainable workloads influences nursing staff to leave their jobs, which makes the nursing shortage even worse.

(40)

33

By implementing a planning system with a monitoring and control function the workload for nurses can be smoothed by balancing supply and demand, which could improve the effective use of nursing staff and the quality of care. In addition, the monitoring and control functio n improves the robustness of the planning system since the nurse/patient ratio being used in the planning is periodically reconsidered to account for changes in supply and demand instead of just using a historical ratio. Furthermore, this could prevent resource losses at the nursing ward since the nursing staff can be retained by regulating their workload, which is of utmost importance to prevent high turnover rates and guarantee a certain quality of care (Aiken et al, 2002).

7.2 Research limitations and future research

The proposed planning system with a monitoring and control function is an initial step towards safeguarding the effective use of nursing staff at the nursing ward, however the staff planning system could be further improved. Several research limitations should be mentioned.

The indicators to determine the optimal nurse/patient ratio are based on a study conducted within the Netherlands including 24 hospitals. Therefore, for the Netherlands it might reflect the workload of nursing staff in a decent way. However, by performing the study in another country different indicators might be found which reflect the workload of nursing staff in that specific region. In addition, the proposed threshold values and control measures might also differ for other nursing wards. Another remark that should be made is related to the correction factor. The correction factor that has been included related to the employability of the differe nt types of nursing staff has been established in consultation with the manager of the MDL department based on assumptions from the manager. Therefore, more research could be conducted related to the employability of different types of nursing staff at the nursing ward.

(41)

34

8. Conclusion

This research addressed the challenge faced by nursing wards of not being able to serve the demand of patients due to a limited supply of nursing staff. These staffing shortages occur mainly because the workload for the nursing staff is not sufficiently balanced, which causes the nursing staff to become ill or burnout. Since there exists a nationwide shortage in nursing staff hiring more nursing staff is a difficult task, therefore it is important that the available staff resources are being retained and used as effective as possible. Currently, the workload for nursing staff is not being considered in the operational planning while determining how many nurses are required during a certain shift, which could result in the ineffective use of nursing staff.

(42)

35

References

Aiken, L., Clarke, S., Sloane, D., Sochalski, J., & Silber, J. (2002). Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. The Journal of the American Medical

Association, 288 (16), 1987.

Aiken, L., Sloane, D., Bruyneel, L., Van, D., Griffiths, P., Busse, R. (2014). Nurse staffing and education and hospital mortality in nine european countries: A retrospective observationa l study. Lancet (london, England), 383 (9931), 1824-30.

Amelsvoort, P. J. L. M. (1992). Het vergroten van de bestuurbaarheid van produktie -organisaties. Eindhoven: Technische Universiteit Eindhoven.

Brackett, T., Comer, L. and Whichello, R. (2013). ‘Do Lean Practices Lead to More Time at the Bedside?’, Journal for Healthcare Quality, 35 (2), pp. 7-14.

Brandeau, M. L., Sainfort, F., & Pierskalla, W. P. (Eds.). (2004). Operations research and health care: a handbook of methods and applications (Vol. 70).

Broyles, J. R., Cochran, J. K., & Montgomery, D. C. (2010). A statistical Markov chain approximation of transient hospital inpatient inventory. European Journal of Operational

Research, 207 (3), 1645-1657.

Carter, M. (2002). Diagnosis: mismanagement of resources. OR MS TODAY, 29 (2), 26-33. Deblois, S. and Lepanto, L. (2016). ‘Lean and Six Sigma in Acute Care: A Systematic Review of Reviews’, International Journal of Health Care Quality Assurance, 29 (2), pp. 192-208. Delesie, L. (1998). Bridging the gap between clinicians and health managers. European Journal

of Operational Research, 105 (2), 248-256.

Elkhuizen, S. G., Bor, G., Smeenk, M., Klazinga, N. S., & Bakker, P. J. (2007). Capacity management of nursing staff as a vehicle for organizational improvement. BMC health services

research, 7 (1), 196.

Fasoli, D. R., & Haddock, K. S. (2010). Results of an Integrative Review of Patient Classi cation Systems. Annual review of nursing research, 28 (1), 295-316.

Fasoli, D., Fincke, B., & Haddock, K. (2011). Going beyond patient classification systems to create an evidence-based staffing methodology. The Journal of Nursing Administration, 41 (10), 434-439.

Gomes, C., Yasin, M., & Yasin, Y. (2010). Assessing operational effectiveness in healthcare organizations: A systematic approach. International Journal of Health Care Quality

Assurance, 23 (2), 127-40

Gupta, D., & Denton, B. (2008). Appointment scheduling in health care: Challenges and opportunities. IIE transactions, 40 (9), 800-819.

Hallam, C., Contreras, C., & Hurst, K. (2018). Lean healthcare: Scale, scope and sustainability. International Journal of Health Care Quality Assurance, 6 (6).

A staff planning system with a monitoring and control function for the effective use of staff resources at nursing wards