Journal Logo

COVID-19 Applications and Perspectives

Healthcare Failure Mode and Effect Analysis (HFMEA) as an Effective Mechanism in Preventing Infection Caused by Accompanying Caregivers During COVID-19—Experience of a City Medical Center in Taiwan

Tiao, Chi-Hui MS, RN; Tsai, Ling-Chin LPN; Chen, Li-Chin MS, RN; Liao, Yu-Mei MS, RN; Sun, Li-Chen MS, RN

Author Information
Quality Management in Health Care: January/March 2021 - Volume 30 - Issue 1 - p 61-68
doi: 10.1097/QMH.0000000000000295
  • Free


A “severe infectious atypical pneumonia” emerged in Wuhan, Hubei Province, China, in December 2019, and the infection subsequently spread worldwide through global travel movements. The index case in Taiwan was a passenger who returned from Wuhan on January 20, 2020. On February 11, 2020, the World Health Organization (WHO) renamed the “severe infectious atypical pneumonia” to COVID-19, later identified as SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2), a new coronavirus. Taiwan Centres for Disease Control (CDC) classified COVID-19 as a type 5 infectious disease by statutory authorities.1 As of September 7, according to WHO statistics, the cumulative number of diagnoses in the world was 27 093 454 and the number of deaths was at least 883 231. Meanwhile, Taiwan had 494 positive cases and 7 deaths.2 The WHO also declared COVID-19 as a highly transmissible disease and stated the need to assume a high global infection rate. Furthermore, the WHO has yet to establish the exact source and viral host.3 As the COVID-19 epidemic continues to spread, the 32nd confirmed case in Taiwan was a patient caregiver from Indonesia. This raised alarms and drew attention to the management of accompanying caregivers and third-party caregivers in various hospitals, as well as the mitigation practices observed by these personnel. Accompanying caregivers and third-party caregivers can lack good hygiene habits and possess minimal knowledge of infection prevention and epidemic prevention; these may be the factors contributing to the increase of infection among patients. Moreover, the patient caregiver could also bring the virus home to other family members. A majority of patient caregivers are contractors from nonclinical service providers. Therefore, the patient caregivers often work across multiple hospitals and can be assigned to different patients, making them a high-risk cohort for person-to-person viral transmissions.4,5

On March 3, 2020, the Ministry of Health and Welfare of Taiwan ordered all medical institutions in Taiwan to respond to COVID-19 and strengthen the control measures for the accompanying patient caregivers and third-party caregivers to prevent shortcomings in the epidemic prevention protocol. The infection prevention and control section of the Joint Commission International Accreditation stipulates the overall quality management of the hospital includes patients, employees, visitors accompanying patients, and caregivers in the hospital. Moreover, hospitals should establish an infection control system and preventive measures, as well as continuous monitoring and improvements.6

Failure Mode and Effect Analysis (FMEA) is a preventive analysis method that performs preventive risk assessment and management of high-risk medical procedures to mitigate occurrence. FMEA is also applied to health care and medical institutions to evaluate and improve the safety of patient care activities, which is called Healthcare Failure Mode and Effect Analysis (HFMEA).7 Research was conducted in 2017 to solve the infection problem of the peripherally inserted central catheter in the intensive care unit (ICU), which reduced the catheter-related bloodstream infection average rate from 5.19% to 1.45%.8 In 2016, Sorrentino9 also used this method to improve the handoffs among nurses, thereby solving the problem of emergency patient congestion. It can be seen that the HFMEA method can indeed effectively prevent and improve clinical problems and improve patient safety.

This article uses the HFMEA method to examine the current process of the accompanying caregivers entering and leaving the hospital and contacting patients, to identify potential problems, and to provide appropriate personal protective equipment to maintain the quality of patient care and to ensure the safety of personal health and medical work environment to avoid infection.


The project was implemented in a city medical center in Taiwan, which had 1720 beds inclusive of 1250 beds requiring accompanying caregivers. The HFMEA working group, including the director and the vice director of the nursing department, the supervisor, the head of nursing station, and the team leader who manages the accompanying caregivers, as well as the infectious diseases specialist and the infectious control nurse, was established in February 2020 at the beginning of the pandemic. After the group was formed, weekly meetings were held for discussion. The meeting first evaluated the current process of the accompanying caregivers entering and leaving the hospital, establishing the process of entering the hospital during the epidemic prevention period, and explaining the HFMEA method. The task force continued to conduct failure mode discussions, evaluations, and chart records based on the HFMEA method. Patients and their families entered the hospital to review the flow of entering and exiting the hospital, implementation status, and problem analysis. After the evaluation and full discussion, the team members identified the failure mode and potential failure mode and determined the causal relationship of the process through the guidance of the infectious control expert and the team's brainstorming discussions in accordance with the regulations of the Taiwan CDC. At the same time, specific improvement measures were formulated, colleagues in the hospital were instructed to cooperate in the implementation process and then the improvement evaluation and statistics were carried out. The research period of this project was from February 2020 to May 2020, a total of 4 months.

The project followed 5 major steps of HFMEA to establish epidemic prevention and control of hospital accompanying caregivers, which are as follows: (1) define HFMEA theme; (2) organize a team; (3) draw flowchart; (4) perform hazard analysis; and (5) perform corrective measures and effectiveness measurement. The process of implementation steps is drawn from the beginning of the patient and family members entering the hospital to the end of the process of confirming the accompanying needs. The main process is divided into 4 items: (A) entering hospital, (B) hospital admission, (C) arriving at the ward, and (D) identification of companionship needs (Figure 1). The team developed the possible failure modes and potential causes of the 9 subprocesses based on the 4 main processes. Moreover, to quantify the risk index, team members conducted a hazard matrix analysis with scores based on the severity and incidence. Severity is graded on the basis of the degree of harm inflicted on the patients: 1 point being the least serious, and 4 points being the most severe, and the incidence is determined by the frequency of occurrence to give the following scale: 1 point represents rare and 4 points represents frequent. The highest score is 16 and the lowest score is 1. The higher the score, the more immediate action is needed (Table). Based on decision tree analysis, the process then puts forward preventive and improvement measures, discusses the implementation plan, and tests decision outcomes via real-world implementations.

Figure 1.
Figure 1.:
Workflow of caregivers entering the hospital.
Table. - Hazard Matrix
Process Failure Mode Hazard Matrix Decision Tree Analysis
Main Sub Potential Failure Mode Potential Failure Cause Severity Incidence Index Existing control Detection Whether to Correct
A. Entering hospital A1. Enter to the hospital by different entrance A1.1. Do not know which entrance the visitors enter the hospital A1.1.1. There are 12 entrance over the hospital 3 4 12 N N Y
A1.2. Do not know how many people enter the hospital A1.2.1. There are no limit of entrance people 3 3 9 N N Y
A1.3. Visitors can enter the hospital during 06:00-22:00 A1.2.1. There are no limit of entrance people 3 4 12 N N Y
A1.3.2. Night entry access control only 3 3 9 N N Y
A2. Implement (22:00-06:00) night entry control, leaving only one entrance and exit, and those who enter and exit need to show their companion card A2.1. Enter the hospital without companion card A2.1.1. Failure to ask for the companion card 3 3 9 N N Y
A2.1.2. Release without companion card 3 3 9 N N Y
B. Hospital admission B1. Apply for at most 2 companion cards B1.1. Overgiving companion card B1.1.1. Failure to implement the certificate control of companion card 3 2 6 Y Y
B1.1.2. Overapplication 2 2 4 Y Y
B2. Visiting time is 06:00-22:00 B2.1. Do not know the personal information and background of the personnel entering the hospital B2.1.1. Nonstandardized 2 3 6 Y Y
B2.1.2. There is no standard to register the background or information of personnel entering 4 3 12 N N Y
B2.2. No protective measures for personnel entering the hospital B2.2.1. Nonstandardized 4 4 16 N N Y
C. Arriving at the ward C1. Introduction to the environment: provide a manual and instructions for hospitalization, including instructions for night entry control C1.1 Do not explain one by one C1.1.1. Unfulfilled implementation instructions due to huge loading 3 2 6 Y Y
C1.1.2. No hospitalization instruction manual provided 3 1 3 Y Y
C2. Patient care evaluation: basic information, disease history, physical evaluation C2.1. The history of contact and grouping of patients not assessed C2.1.1. Unstandardized “patients to be evaluated” contact and group history 4 4 16 N N Y
C2.1.2. The patient did not answer honestly 3 1 3
C3. Disease-related nursing guidance and health education C3.1. No provision of health education programs for medical needs C3.1.1. Too many health education items, and unregulated essential health education items 3 2 6 Y Y
C3.2. Do not follow-up the implementation status after the hand hygiene instructions C3.2.1. Failure to evaluate the effectiveness of health education due to huge loading 3 2 6 Y Y
C3.2.2. Trust them that they will do 3 2 6 - -
C3.3. No health education or epidemic prevention policy C3.3.1. Unstandardized publicity and prevention key points 4 4 16 N N Y
C3.3.2. Unstandardized provision of anti-epidemic materials 4 4 16 N N y
D. Identify companionship needs D1. Accompanied by family members D1.1. Do not ask about the personal background and information of family members D1.1.1. Not regulated 4 4 16 N N y
D1.2. Do not ask family members for symptoms of physical discomfort, such as fever D1.2.1. Not regulated 4 4 16 - -
D1.3. No fixed family member to accompany with the patient and may be replaced at any time D.1.3.1. Unstandardized accompany persons to be fixed 3 3 9 N N y
D1.4. Unimplemented hand hygiene D1.4.1. Poor personal hygiene habits 2 3 6 - -
D1.4.2. Have not educated relevant training for health education guidance 4 4 16 N N Y
D1.4.3. The sensory control monitoring is not targeted at all people in the hospital, only part of the trial is selected 3 2 6 Y Y
D2. Accompany by employed caregiver D2.1. Not clear about the status of pandemic policy D2.1.1. Unstandardized publicity and prevention key points 4 4 16 N N Y
D2.1.2. No epidemic prevention materials 4 4 16 N N Y
D2.2. The contract does not meet the current situation D2.2.1. The content of the contract does not contain COVID-19 epidemic specifications 4 4 16 N N Y
D2.3. No record for sign-in and temperature data D2.3.1. No compliance record 4 3 12 N N Y
D2.3.2. Do not measure body temperature 3 3 9 N N Y
D2.4. No uniform and identification card D2.4.1. Do not cooperate with identification 4 3 12 N N Y
D2.5. Do not ask the caregiver if they have symptoms of physical discomfort, such as fever and diarrhea D2.5.1. Not regulated 3 3 9 N N Y
D2.6. Do not hand in the roster every day D2.6.1. Not regulated 4 4 16 N N Y
D2.7. Take care of different patients in morning and evening D2.7.1. Not regulated to take care of at most 2 patients at the same time 3 3 9 N N Y
D2.8. Enter ward of another patient D2.8.1. Failure to comply with work specifications 3 3 9 N N Y
D2.9. Unimplemented hand hygiene D2.9.1. Failure to comply with work specifications 4 4 16 N N Y
D2.9.2. Poor personal hygiene habits 3 2 6 - -
D2.9.3. Not included in personal work assessment 2 3 6 Y Y
D2.10. Not clear about the epidemic situation and hospital policies and regulations D2.10.1. No channels to obtain new knowledge of the epidemic 4 4 16 N N Y
D2.10.2. No epidemic prevention courses related to the epidemic have been held 4 4 16 N N Y
D2.10.3. No epidemic prevention materials 4 4 16 N N Y


The hazard matrix is used to perform a thorough study on the scores of the 4 main processes and 9 subprocesses of patient caregivers entering the hospital (Figure 2). A total of 26 potential failure modes and 42 potential failure causes were identified. Team members used the HFMEA model to improve the 29 factors, with a hazard index greater than 8 points for the caregivers entering the hospital. Factors with a hazard index of 16 points included: people entering the hospital have no protective measures, such as washing hands and wearing masks, not knowing the contact history, physical condition, or fever of the caregiver of the patient; not clear about the regulations of Taiwan's epidemic policy; and the contract of employment management company does not meet the COVID-19 epidemic regulations, etc. The hazard index is 12 points for: people without a companion card can enter and leave the hospital. Through implementing 24-hour access control, asking accompanying persons to wear masks and disinfect their hands when entering the hospital, registering contact history, measuring body temperature, organizing education, and training on hand hygiene and epidemic regulations of Taiwan government, the potential causes of failure decrease from 42 to 13 (Table).

Figure 2.
Figure 2.:
Improved policies for hospital entries by person(s) accompanying patients. TOCC indicates Travel history, Occupation, Contact history, and Cluster.

Management support and assistance were sought in the establishment of epidemic prevention control and COVID-19 policy for accompanying persons, and a visitation management plan was developed, including caregiver management standards (Figure 2). The standards include the following: implement access control management to accompany visitors and visitors; posters of “Guidelines for Access Control and Accompanying Caregivers During Epidemic Prevention in Hospitals” must be posted at the entrances and exits of the hospital and at various nursing stations; visiting hours are strictly controlled; visitors are not allowed during nonvisiting periods; the limit of visits in general wards is reduced to 2 periods, with a maximum of 2 persons per bed; the limit of visits in the ICU is reduced to 1 period, with a limit of 1 person per bed; visitors are not allowed on Saturday afternoons and on holidays throughout the day. Quarantine checkpoints must be set up at the gates of the hospital. It is necessary to wear a surgical mask, measure body temperature, and wash hands when entering the hospital. To enter the ward, all personnel must cooperate with the nursing station to request real-name registration, including name, relationship with patient, contact number, ID number, and place of residence, and answer TOCC (Travel history, Occupation, Contact history, Cluster) and whether there are respiratory symptoms. Multiple daily broadcasts to publicize visit management regulations during the epidemic and honestly fill in TOCC and other important notices must be complied with. During the introduction to the admission environment, instructions for key health education related to epidemic prevention are provided. The nursing station provides masks for accompanying patients every day to ensure that surgical masks are worn throughout the entire process, and body temperature, hand hygiene, and cough etiquette are undertaken. If the accompanying person has fever or respiratory symptoms, staff immediately assist in the outpatient or emergency medical treatment; daily health education and guidance for epidemic prevention instructions, such as the correct wearing of masks and hand hygiene, should be undertaken. Staff must monitor the behavior of accompanying patients to control infection, avoid nosocomial infections, and achieve the best epidemic prevention effect.


Following the SARS outbreak and the ensuing chaos experienced by Taiwan hospitals, several hospitals have implemented infection control training and management programs and conduct regularly organized epidemic prevention in preparation for the next epidemic/pandemic. In the face of the COVID-19 assault, this hospital deployed an early and comprehensive response plan. Response included the twice-daily (morning and evening) epidemic prevention meetings and immediate announcement of the epidemic prevention policies and updated specifications. Moreover, all medical and administrative staff members of the hospital participated in the implementation of epidemic prevention control actions. The HFMEA method is used to identify and assess the risks of caregivers after entering the hospital and provides an improved structure for the 4 main processes:

  1. Entering hospital: A quarantine checkpoint must be set up at the gate of the hospital, and a health insurance card or companion card must be used for checking in at the hospital. When entering the hospital, it is necessary to wear a surgical mask, measure body temperature, wash hands, and register with TOCC.
  2. Hospital admission: Only 1 person per bed is allowed to accompany, and a companion card will be issued. Other visitors can only visit the hospital at the specified time (1 hour in each time slot), and there can only be 2 visitors at a time.
  3. Arriving at the ward: While entering and leaving the ward, the accompanying person must cooperate with the nursing station to request registration, including legal name, relationship with patient, contact number, ID number, and place of residence. One mask per day will be provided by the hospital during the period, and wearing a mask and performing hand hygiene in accordance with the hospital's control standard will be necessary.
  4. Identify companionship needs: Provide education and training, including the method of virus transmission, the correct way to wear a mask, hand hygiene, and the way to prevent oneself from infection during the period. Employed caregivers must receive education and training relevant to COVID-19 before they are employed. They are able to ensure the safety of patients and themselves through daily in-hospital broadcasts of reminding self-protection measures.

After implementing improvement measures and adopting epidemic control for caregivers, the potential causes of failure decreased from 42 to 13 items, the pass rate of body temperature cutoff increased from 53.1% to 90.8%, the compliance rate of hand washing increased from 89.5% to 100%, and the accuracy rate of hand washing increased from 52.2% to 76%. Although the accuracy of hand washing is still poor, the hand hygiene compliance rate has reached 100% after improvement and there has been significant progress. It is still necessary to continue to strengthen health education and publicity and then continue to audit its implementation effectiveness.

In the process of improvement, the teams found that if the accompanying caregivers are family members of the patient, they may possibly switch over with another member to accompany the patient, which is not easy to control. Applying the HFMEA model can indeed and effectively target high-risk medical procedures by performing preventive risk assessment and management and identifying weaknesses in hospital control measures. Uncontrolled personnel flow into the hospital during the epidemic and may contribute to nosocomial infections. Positive epidemic prevention management strategies, including hygiene and health education of patients and their caregiver(s), monitor the behavior of the caregivers on infection control, provide education and training on anti-epidemic mechanisms, and can enhance societal awareness of epidemic prevention. Furthermore, people will cooperate with the implementation, which can effectively prevent hospitalizations caused by caregivers. Taken together, these are the most effective epidemic prevention management policies. The experiences of this hospital will be shared with other medical institutions for their consideration.


1. Tseng CY. Are new coronaviruses attacking, are you ready for protection? On how to prevent infection. Consumer Rep Taiwan. 2020;467:8–9.
2. Taiwan Centers for Disease Control. Press release. Published 2020. Accessed March 2020.
3. Wang QL. Wuhan pneumonia. WHO announces 10 typical symptoms of diagnosed patients! Fever, dry cough and burnout are the top three, 2020. Accessed March 2020.
4. Lin FY, Liao HJ. Implement the core work of nosocomial infection prevention ∼ National University Hospital “Hand Guardian Movement” promotes experience sharing. J Healthc Qual. 2007;1:50–54.
5. Lin CF, Yang CY, Lu MS, Kao CC. Effectiveness of a nosocomial infection control training in improving knowledge in patient-hired attendants and outsourced workers in Taiwan. J Nurs Res. 2008;16(3):187–194.
6. Hong CY, Chiu WT. JCI certification and recertification: experience of international hospital evaluation. Taiwan Med J. 2010;53:49–53.
7. Chen SH, Chen LC, Tu WL, Liu TT. Using Health Failure Mode and Effect Analysis to improve the workflow of measuring weights by hemodialysis patients. Taipei City Med J. 2012;9:61–77.
8. Li X, He M, Wang H. Application of failure mode and effect analysis in managing catheter-related blood stream infection in intensive care unit. Medicine (Baltimore). 2017;96(51):e9339.
9. Sorrentino P. Use of failure mode and effects analysis to improve emergency department handoff processes. Clin Nurse Spec. 2016;30(1):28–37.

COVID-19; caregiver; health education; hospital visitor; infection control; pandemic policy; preventing infection

© 2020 Wolters Kluwer Health, Inc. All rights reserved.