This article explores nurses shift scheduling in Kenya: what options can be explored, from the well known morning off, buffer, half day shifts to the proposition of self scheduling shift. The scholarly approach borrows from decision support point of view and evidence based practice.
Problem classification, decomposition, and ownership
The Kenya Health Service Delivery Indicator (KHSDI) Report of 2018 released in August 2019 revealed the state of absenteeism in majority of Kenyan hospitals. Imagine going to a hospital and waiting for long hours on end, only to realize that there are no nurses around to attend to you.
The findings show absenteeism of doctors at a staggering 60.7% followed closely by nurses at 54.5% and clinical officers at 49.5% countrywide. Hospitals had the highest levels of absenteeism at 60.4% whereas dispensaries and clinics had the lowest absenteeism rate at 44.5%. This showed that the Kenyan healthcare system could be getting worse by the day.
A similar study 5 years earlier, the Hivos study (2013) further observed apart from absenteeism even when present some providers spent less than half of their time in patient care. The survey found that over 29 percent of health providers in public facilities were absent from work,. Dr Martin Gayle, the Service Delivery Indicators (SDI) program leader emphasized that this absentee rate amongst clinicians was one of the primary causes of inefficiency in development efforts, yet despite its widespread occurrence, there were few measures of how often it occurred and the impact it had. Furthermore, it was alleged that 80 percent of this was sanctioned absence e.g. calls in to report they were sick
In concurrence to these study findings, a common observation among health sector employees in Kenya is that a good number of them use their employers’ time and sometimes resources to do their own business whether at their place of work or away.
Some findings from a study commissioned by Family Care International (2003) covering Homa Bay and Migori districts, several community leaders observed that facility-based staffs were increasingly splitting their time between government facilities and their own private clinics.
As a result, providers were not at the hospital, even when they were supposed to be on duty. Such absences only exacerbated the already-severe staffing shortages at health facilities. “You go to hospital, and you find the nurse is not there, but in their private clinics, and it is working hours…”It was hard to justify absenteeism on one hand and complain of a low nurse to patient ratio on the other hand.
Our public health system has had to cope with several shocks: At every opportunity it became normal in some settings to operate at bare minimum, if we could get a valid reason to do so. These included the union-sanctioned strikes and go-slows which had become all too common since 2016. During such times vital deliverables have to wait and scheduled appointments cancelled. Then we have to start all over when we come back.
The health sector cannot change some of the unfavorable working conditions that characterize it, such as night work and work during weekends and holidays, it would have to provide other incentives that will encourage young people to consider entering and remaining in the work force (Wiskow et al., 2010). Furthermore, most health care environments required rest periods before and after each shift type.
When looking at some particular units one teaching hospital in Kenya, it was shown that the turnover rate annually was 50%. The cost of training for one nurse was approximately Kshs 1.5 Million therefore losing around 40 nurses annually was a financial hemorrhage for the hospital. Morale was low and in turn patient care was suffering. Therefore this was a human resource issue with far reaching implications on care delivery and patient management outcomes. The chief nurse and human resource engaged in decision making process (Sharda et al., 2011).
Statement of the Problem
There was a need to create an attractive, supportive and effective work environment. When viewing turnover rates and staff satisfaction in the units, upon exit interviews it was discovered that scheduling was a concern for staff when considering work-life balance.
It was not likely that staff had somewhere else to go and that was why they left. Not unless the competition was luring them with niceties. What was it that we were not aware of? From interviews, it was discovered that one of the top staff’s concerns was that scheduling was non-flexible, staff felt they had no control over their work-life balance. They were being rotated between days and nights shifts often and were unable to develop any type of routine. Data was needed to determine what would increase staff satisfaction, decrease sick calls, increase productivity and offer a better work life balance for the nurses (see figure 1 below). The goal of the personnel scheduling according to Smet et al., (2012) in ‘modeling and evaluation issues in nurse rostering’ is to find a schedule, i.e. a set of assignments (see figure 2 below).
Curtin (2003) did a meta-analysis of 50 conveniently sampled research studies based in the US on nurse staffing as an dependent variable and how it was influenced by a multitude of circumstances including schedules, patient ratios, administration support to name a few. The author concluded that nurse staffing related to nurse satisfaction had a definite and measurable impact on patient outcomes, medical error, impact of organizational characteristics of work ethic and nurse staffing patterns, patient outcomes and the cost of patient care. Impact on nurses’ experience related to overall satisfaction with their jobs, nurse turnover and in the end patient mortality had all been connected in this study. This system of scheduling contrasted with the traditional system of where the person in charge of the ward roster performs this task. The independent variable was: Shifts (days and times) – its influence could be manipulated to change other variables (dependent or effects) e.g. call-ins, staff turnover, patient outcomes, containment of costs. The Shift (the cause) was central to this (see figure 2 below).
Germany had implemented what was referred to as ‘Elements of family-friendly policies in German hospitals’ which implement among others; coordination of duty rosters of couples across different departments; establishment of time accounts (plus and minus hours); and flexible working hours (“flexitime”), with family-oriented core times (Wiskow et al., 2010). In Smet (2012) family members requested working different shifts to enable at least one of them to care for their children. In a study by Harton et al., (2012), staff members were allowed to choose the days and times they wanted to work following a predetermined criteria that assured the appropriate amount of staff per unit. Allowing team members to choose their own schedules promoted responsibility, accountability, job satisfaction and personal growth (Harton et al., 2012).
Determine the impact of self-scheduling on the unit operating from a point of uncertainty since we did not know of precedent or significant outcomes from a Kenyan perspective. Most of the data available was web-based gleaned from US and Europe. We made several assumptions such as: Inflation, relaxed immigration rules, nurses shortages. What ifs – did nursing staff who used self-scheduling, have decreased call-ins, decreased turnover, enhance patient outcomes and contain costs? Weighing the alternatives for (a)either shift: Buffer, Half day, Morning off, and 12 hour (Straight) shift with self-scheduling or (b) maintaining status quo i.e. same shifts out self-scheduling (see figure 2 below).
One representative from each unit was trained on self-scheduling, they were to monitor the process for appropriateness and effectiveness on each unit and directly report. We needed to have it in phases then roll out to the rest of the hospital at 4 months and at 8 months. Schedules were created accommodating personal scheduling issues e.g. enough time with family, have control over schedule, Self scheduling model that was tailor-made. Balancing user friendliness with unit and organizational policies.
How the decision was evaluated
Job Satisfaction– was rated for a participant to their hospital or department. This rating will ranged from 1 to 10. Staff Turnover– This was the number of employees that left their job on a monthly/ annually basis. Self-Scheduling – Whether a participant was satisfied self-scheduling model as a transitional model as measured by Yes or No. Life balance – Evaluating the amount of time the participant was able to spend with their family, socially as indicative of a life balance. Measured the hours spent per week with family or socially. Productivity – evaluated shifts productivity and patient safety, nurse: patient ratios for a unit, department the hospital. Criteria for self-scheduling – Evaluate up to how much and with what authority could an individual staff balance out to meet their own simultaneously with unit demands (Sharda et al., 2011).
Figure 1: Model for determining staff turnover
Figure 2: Model for alternatives
1.Curtin, L. (2003). An integrated analysis of nurse staffing and related variables: effects on patient outcomes, Online Journal of Issues in Nursing. 8(3); 118-129.
2.Family Care International (2003): Care-Seeking During Pregnancy, Delivery, and the Postpartum Period: A Study in Homabay and Migori Districts, Kenya. Available:
3.Harton, B., Marshburn, D., Kuykendall, J., Poston, C., Mears. D. (2012). Self-scheduling help or hindrance. Nursing Management
4.Keystone partner Feedback report 2013, Where is your doctor? Hivos SDI program.
5.Kenya Health Service Delivery Indicator (KHSDI) Report (2018). International Bank for Reconstruction and Development, the World Bank, GOK.
6. Smet, P., Bilgin, B., De Causmaecker, P., Vanden-Berghe, G, (2012). Modeling and evaluation issues in nurse rostering. Annals of Operations Research.
7. Sharda, R., Delen, D., Turban, E. (2011). Decision support and business intelligence systems (9th ed.). Prentice Hall.
8. Wiskow, C., Albreht, T., de Pietro, C. (2010).World Health Organization and World Health Organization, on behalf of the European Observatory on Health Systems and Policies.