Modeling of adult patient falls and the repercussions to Nursing as a second victim

Abstract Objective: analyze the falls of adult hospitalized patients and their repercussions on the Nursing worker as the second victim. Method: an exploratory, descriptive and qualitative study conducted in two stages - identification of falls with moderate to severe damage and the modeling of falls using the software Functional Resonance Analysis Method; and analysis of the repercussions on the worker as the second victim by means of semi-structured interviews submitted to Content Analysis, with 21 Nursing workers. Results: a total of 447 falls of adult patients were identified, 12 of which with moderate to severe damage, occurred in the absence of the companion, while using sleep inducing, hypotensive or muscle strength altering medications. The modeling identified 22 functions related to the Standard Operating Procedure, to the fall prevention protocol and to the post-event actions. Of these, eight presented variability in their execution. In the second step, the categories “The complexity of care to prevent falls” and “Feelings of the second victim” emerged. Conclusion: falls are complex phenomena and prevention requires involvement of the patient, family and multi-professional team. The professionals involved are second victims and experience feelings of guilt, fear, anguish and helplessness. This study can contribute to the multi-professional spirit and to the early approach to second victims.

International Studies (12,14,16) have already signaled the repercussions of AE on second victims; however, the Brazilian context needs to be explored regarding this issue. This study is anchored on the relevance of falls in the global context of patient safety, the gap in knowledge about the perspective of the worker facing this scenario, using process modeling, the latter two being the differential of the studies published so far. The study aimed to analyze the falls of adult hospitalized patients and their repercussions on the Nursing worker as the second victim.

Method Type of study
This is an exploratory, descriptive study with a qualitative approach, supported by Content Analysis. The study was conducted in two sequential stages in order to achieve the proposed objective and its presentation followed the guidelines of the Consolidated Criteria for Reporting Qualitative Research (COREQ).

Location
The study was developed in a general university hospital, reference for high complexity, with capacity for 843 beds, in the South region of Brazil.

Selection criteria
Adult patients admitted to clinical and surgical inpatient units who had a fall inside the institution and had an injury grade ranging from two to four (moderate to death). Regarding the injury severity classification (8,(17)(18) , the institution of the study classifies as: (zero) no harm; (one) mild harm, one that involves minimal or moderate repercussion, but with rapid duration, requiring few interventions; (two) moderate harm, with minimal need for intervention, increased length of stay, harm or loss of function in a long term; (three) severe harm, one that has Quadros DV, Magalhães AMM, Wachs P, Severo IM, Tavares JP, Dal Pai D.
severe symptoms, with need for life support intervention or major clinical/surgical intervention, permanent or longterm loss of function and (four) death.
Exclusion criteria were considered to be patients with falls without damage or with mild damage (as these are considered to have less repercussion for the worker).

Sample
The sample consisted of 12 falls, which consisted of the total number of falls with moderate to severe damage recorded between July 2018 and July 2019.

Data analysis
The data was entered into the FRAM Model Visualiser Software * and their analysis occurred in two stages: initially, the data referring to the prescribed process

(SOP and Protocol for Prevention of Falls in Adults)
and, afterwards, the data referring to the real work (management information and medical records related to falls). The software used modeled (drew) each process (prescribed and real) presenting, as a result, a graphic illustration.

Period
The data collection was carried out in the period from March to May 2020.

Population
The population consisted of nurses and nursing technicians working in clinical and surgical admission units, consisting of a total of 265 professionals.

Selection criteria
Nurses and Nursing technicians allocated in one of the inpatient units where falls with moderate severity of injury to death occurred were included (according to the results of the first stage of the study). Exclusion criteria were: professionals who worked for less than a year in the institution, professionals on vacation or leave of any kind, besides those on temporary contract during the data collection period.

Sample
The sample was defined by simple random drawing.
One nurse and two Nursing technicians were chosen for each shift in the units where the falls occurred. Twenty-one professionals were selected, following the data saturation criterion, considering the proportional distribution to the five inpatient units selected in the first stage of the study and the number of professionals in each category (nurses and nursing technicians). Among those chosen, one professional declined the invitation and another, due to problems in the unit, was unable to participate.

Data collection
Semi-structured interviews were conducted with a script based on the study objective and the results The interviews were conducted by the master's student, the proponent of the study, who is a nurse and has experience in conducting semi-structured interviews and periodically attended meetings with the research team.
All interviews were conducted face-to-face, in a shift and time of preference of the participant. The interviews took place inside the professional's work unit, in a reserved place.
The professionals who agreed to participate in the interviews were provided with information about the institutional data and the theme of the study ** . * Functional Resonance Analysis Method (FRAM), software that makes it possible to operationalize the fallout modeling in four steps characteristic of FRAM: (i) identify and describe the main functions (and aspects) of the system; (ii) characterize the potential variability of each function; (iii) aggregate the variability, considering it in a real way in a given scenario; (iv) propose recommendations to monitor and influence the variability (4) . The description of these variables and functions allowed the analysis of the items that showed the greatest variability for the occurrence of the event, this being considered an instantiation. With the modeled process, it is possible to evidence the functional resonance through the coupling between the functions. ** All falls that comprised the sample of the first stage were modeled. Fall 10 was chosen to subsidize the interview, because this was the fall with the greatest repercussion for the patient. During the interview, the results of the first stage of the study were presented to the participant.

Data analysis
The Content Analysis technique was used according to Bardin (19) , following the three-stage systematics: (i) pre-analysis, (ii) exploration of the material, and (iii) treatment of results, inference and interpretation. The

Process Modeling in Falls of Hospitalized Adult Patients
The functions (represented by a hexagon) were modeled in the moments recommended for the application of the scale according to the Protocol for Prevention of Falls in Adults, in which the risk is identified by means of a stratified score. The patient identified as being at risk of falling receives a yellow bracelet, which provides visual signaling to the professionals. Next, the functions that concern the record in medical records and the guidelines that are given to the Nursing team were modeled, more specifically, to the Nursing technician responsible for implementing the care to the patient (and his/her caregiver or companion) in that shift.
The modeling sequence also allowed the identification of the care that is oriented and the adherence to all these orientations, both by the Nursing team and by the patient, his/her family member or caregiver, aiming at the non-occurrence of the fall. The functions listed for the prescribed work are presented in Figure 1.  The modeled fall 10 was used to support the interview in the second stage of the study (Figure 3).  The feelings that the patient awakens in the team reflect in a new configuration that must be made, in haste, to accommodate the demands, repercussions in a greater workload and even in the relationships of trust previously established between the teams, but that need to meet the demands of some family members.
Quadros DV, Magalhães AMM, Wachs P, Severo IM, Tavares JP, Dal Pai D. There are so many other attributions that they will not want to give psychological support to us because of a fall, I don't see The feeling of devaluation was also noted when the importance of support measures for the worker who was involved in a fall event was mentioned, which can be considered an addition to the condition of second victim.

Discussion
The analysis of the modeled falls allowed the identification of the variability of the functions represented (5,20)  The implementation of collaborative practices qualifies health services to the extent that there is understanding of the dimension of the work involved, resulting in better outcomes, patient safety, and promoting relationships of trust among professionals. In this context, this The feelings of the second victim provide negative reactions. Among the feelings reported, guilt, stress due to the concern for the patient, fear of judgment and the outcome for the patient elucidate what is also brought up in the international literature (11,(13)(14)(15) . These are common feelings in the face of error, as it demonstrates human fallibility and the consequent emotional impact, with increased absenteeism and intention to quit (11) . Spending days thinking about what happened, crying for the patient and suffering with the family are part of the professionals' daily practice. These repercussions are directly related to the degree of the patient's unfavorable outcome (15) and that, in a certain way, are corroborated by the chief's judgment when he changes a professional's scale at the request of a family member or when he doesn't receive support from his colleagues, given the volume of work that continues to be demanded, reiterating the need for a support infrastructure for the second victims (10)(11)13,16) .
The support to second victims of events with damage to patients is valuable for coping, and it ranges from the safety climate and staff satisfaction (30) , through early worker support.
Since this is a small-scale study, the results are sensitive to the context employed, which is a limitation.
However, this methodology can be replicated, allowing