Modelling fatigue and use of fatigue risk management system in health care systems

Effective performance is required to successfully and safely maintain 24/7 operations. However, such operations typically involve extended work hours, night shift work, and early starts and fatigue associated with sleep loss, long duty hours, night work. In addition, fatigue can induce sleepiness and drowsiness, decrease the ability of workers to operate safely, and, thereby, increase the risk of fatalities and injuries. In traditional fatigue preventing method, regulations restrict the number of hours one can work per duty period and over a series of duty periods and this method was especially used for transportation operations. Then studies obtained that reducing the hours of work is not adequate for effectively managing fatigue because they do not take into account the circadian rhythm and effects of working multiple duty cycles, and they universally restrict shift durations without regard to commute times and other non duty related activities. Fatigue risk management systems attempt to move away from the traditional risk management strategy of prescriptive hours of service regulations and aims to integrate science with operational realities in order to promote effective performance and safety. A fatigue risk manage¬ment system provides an alternative, scientifically based means of managing the risks associated with fatigue and can enable companies to safely conduct operations beyond existing prescriptive regulatory limits. The common definiton for fatigue risk management system is "A data-driven means of continuously monitoring and managing fatigue-related safety risks, based upon scientific principles and knowledge and operational experience that aims to ensure relevant personnel are performing at adequate levels of alertness."(ICAO-IATA-IFALPA, 2011) Fatigue risk management interventions aim to optimally maintain alert performance and minimize the risk of errors, incidents, and accidents, while preserving the integrity and productivity of the operation. Basic contributions of this study are as follows: • Information about FRMS applications, fatigue factors and the methodology followed in the field of aviation were given. • The applicability of the FRMS in the health care services could be applied as in the aviation industry was investigated and different aspects were identified. • The methodology which is also a guide for implementing FRMS in health care services successfully was determined. • Fatigue factors which are special to health care were investigated. • An ANP based method which aims measuring fatigue and considers the relationships between fatigue factors and factor weights was also developed. • This quantification model is proposed to integrate dimensions of fatigue in to a single index. Also it helps decision makers to prioritize fatigue dimensions. First of all, fatigue, fatigue risk factors, fatigue symptoms, safety management system components, fatigue risk management system components, advantages and disadvantages of fatigue risk management system according to previous experiences were explained. Next chapter described fatigue risk management practices in aviation industry which is pioneer industry in this area and methodology arranged by ICAO, IATA and IFALPA. In the third part of the report, studies about fatigue in health care industry, arrangements for reducing hours of work in the world, guide which belongs to AMA (Australian Medical Association) named as "National Code of Practice - Hours of Work, Shiftwork and Rostering for Hospital Doctors" were explained. According to the guides of ICAO, IATA, IFALPA and AMA, it was investigated that whether fatigue risk management system can be adapted from aviation industry or not. That is why a fatigue risk assessment control list was developed by extending previous control lists which belong to AMA and Safe Work Australia. The checklists created by AMA and Safe Work Australia assess fatigue on the basis of individual risk factors and focus heavily on factor in the high-risk group. However, this is not possible in measurement of cumulative fatigue. Furthermore, according to the assessments if more than one factor are identified in high-risk group, it is not known what factors should be given priority. It is necessary to measure fatigue in order to manage fatigue. Also in this measurement factors should not weigted equally and relationships between the factors should not be forgotten. This shortcoming had been identified in the literature study and therefore other factors which should be used for assessing fatigue in health care industry were determined. Then the relationship between those fatigue factors were considered and factors were weighted according to their priorities. With this method measuring fatigue among health care workers in the most accurate way was intended. Many problems in real life modeling not be done with hierarchical structure. Because the analytic hierarchy process (AHP) assumes that all factors are independent. In contrast, the analytic network process (ANP) problem considers the interactions and feedbacks between the factors are modeled. Because there are relationships between the fatigue factors determined for the health care industry and it is possible to work with the maximum number of seven factors simultaneously, ANP method was preferred to weight the factors. Fatigue modeling programs which are called biomathematical models are efficient methods for determining the work-related fatigue. However, the biggest disadvantage of this fatigue scores on the programs is determined solely on the basis of working hours. Workload, other factors such as working environment is not taken into consideration. Some of these models include the light amount and time of sleep.However, most inputs are based on working time only. Such models also do not consider personal factors such as age, gender, drug use. That is why these models can calculate the average fatigue but they are inadequte for calculating personal fatigue. Also all models have been developed for different purposes also used to predict the fatigue. For example, some measures have been developed purely military purposes, some models are suitable for industrial use. That is why developing a new method for health care industry was intended. On the other hand, one of the weaknesses of the model is the potential to create difficulties in practice. Based on different inputs such as legal requirements, fatigue and sleepiness reports, adverse event reports, the model should be updated periodically. That is why the establishment of relations between factors must be made by individuals who know ANP techniques better. Furthermore, the experts consulted during each update may cause prolongation of the process. In addition practical guidance on how to eliminate or minimise risks arising from the fatigue hazards in health care industry are provided in the study. Current state of health care industry and its workers were carried out by gap analysis method. Then a road map was developed for health centers in order to apply fatigue risk management system. In this road map steps which should be taken by organizastions were explained by using process flow and key roles were determined in this flow. When studies about determining rest allowances were analyzed, it was observed that they did not take into account the fatigue factor. However, due to factors that cause fatigue and tiredness may be considered in the determination of rest periods and thus rest periods can be planned more effectively. Using the values obtained from the ANP based fatigue model is proposed for calculating the rest allowances in future studies. Limitation of working hours to prevent fatigue can be considered. But this will allow doctors to transfer the patient to another doctor frequently and in this case the patient follow-up will be more complicated and it can cause orther errors. The solution of such problems is transferring of the patient to another doctor within a certain systematic approach. In addition, nurse-specific fatigue factors can be investigated in detail and differences from the factors in this study can be identified. So ANP based fatigue models can be developed for them too.