Previous work

In 2018 our working group evaluated subjective views of multiprofessional care givers about how they experience patients’ and other health care workers’ reactions on isolation precautions.

In this study [8] we used facilitated discussion groups consisting of post-graduate health care workers from 11 different professions in an educational setting: Participants in this study were given two cards with a medical profession (other than their own) written down on it. They were expected to write down their impressions and experiences concerning how the professionals on the card react to MDRO. These results were pinned on a cardboard and discussed in the group. The cards were inductively coded by our working group consisting of three coders. After showing a relevant association with emotional aspects in the first coding round, we conducted a second deductive round using the set of the main emotions, described by Ekman et al. [9]: anger, disgust, anxiety, enjoyment and sadness. Considering this framework, the following themes were evaluated, showing health care providers experiences with MDRO management:

• Patients are admitted to hospital for high quality treatment of a severe and threatening illness or condition like coronary heart disease, cancer, or neurological disease.
• If confronted with MDRO, patients and relatives need further medical information.
• Too little knowledge and time or inability to communicate knowledge to patients lead to the lack of, and/or inconsistent and contradictory information.
• A lack of knowledge on the part of health care providers may lead to over- and underestimating risks.
• For patients and their relatives, contradictions in MDRO management provoke uncertainty, anxiety and anger.
• Combined with a severe underlying medical condition this aggravates fear and stress due to the need for time-consuming consultations, taxing valuable critical resources in the daily routine.

Scientific issues

Keeping our previous study in mind the follow-up described in this manuscript focuses on emotions applying a similar study design, but addressing specifically the emotional effects on patients, caregivers, and hospitals in conjunction with MDROs.

Our main hypothesis a priori was that inductive findings of the first study could be proven again in health care workers and may lead to a deeper insight on emotional impact than detectable in the first study. This article provides readers with our inductive and deductive analysis of a set of facilitated discussion groups made up of post-graduate health care providers of different professions und from several hospital-departments of anesthesiology, emergency medicine, critical care, pharmacy, and medical education. We report on the educational, psychological, and ethical perceptions of these three groups about MDRO-effects on patients, families and staff.

Study design

In this qualitative study, we engaged six facilitated discussion groups each containing 5 to 10 post-graduate medical professionals in service recruited from six hospitals (three primary, two secondary, one rehabilitation unit), and 11 medical educators from the Master of Medical Education course forming the seventh group for external validation. Hospital discussion groups were embedded within a six-hour interprofessional training session for postgraduate health care providers focusing on antimicrobial stewardship (AMS) in the German language in 2018. All participating trainees were enrolled in this study. Except for underage, there were no other inclusion or exclusion criteria for participants representing in-service medical professionals from these hospitals to participate in a learning program. None of the participants prohibited the use of their qualitative data derived from the lessons.

According to regulations from the ethics board and workers’ council further quantitative data aside professional groups was not obtained to guarantee anonymity. We aimed to enhance our findings’ generalizability by taking interprofessional and qualitative approach. We thus recruited participants from six hospitals and an external group of professionals to cross-validate our data. To guarantee anonymity, informed consent was obtained orally at the beginning and end of each course. Participants were informed about the use of anonymous data for scientific research during the registration process and the possibility to refuse or withdraw participation without incurring any disadvantage.

The learning format here, differing from our original study’s [8] was developed applying active learning methods [31] and curriculum development tools [32] and was open to physicians, critical care nurses, and pharmacists. It was divided into three lessons, each consisting of a passive part (short report lasting 20 minutes), an active part of different format (about 90 minutes) and a reflective part (card survey, about 10 minutes). These study data were collected in the first active part. The complete course is illustrated and described in detail in Fig 1. In the active part, 35 participants were exposed to five statements and were expected to write down their experiences, impressions, and thoughts on cards (Fig 2). The five statements were “MDRO leads to …” “…Anger”, “…Fear”, “…Enjoyment”, “…Disgust” and “…Sadness”. Participants had up to 15 minutes to write down their impressions and experiences on cards, followed by discussion. The moderator (facilitating the process, guiding the participants and when asked, answering other questions) took an active part as in a socio-constructivist (e.g., asking about details, clarifying answers) approach. The 11 participants in an external medical- education course fulfilled a validation purpose for us to seek for further tags and codes to ensure data saturation. For these, we applied the same technique of the facilitated discussion alone without additional training elements.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

https://doi.org/10.1371/journal.pone.0246820.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2.

https://doi.org/10.1371/journal.pone.0246820.g002

Approval from the ethics board of the Stuttgart Physicians’ Association was obtained prior to this investigation.

Research team and reflexivity

The main investigator of this study and facilitator is a 41-year-old male senior consultant anesthesiologist, critical care, and emergency physician, hospital hygiene & AMS expert and medical educator (MSc in medical education) with international certification as a medical risk manager. His main responsibilities at the time of this project were infection prevention and multiprofessional medical education (under- and postgraduate physicians, nurses, and anesthesiology assistants). The study supervisors were a male consultant (PhD) in neuropediatrics experienced in interprofessional medical education and qualitative research, and a male hospital hygiene expert (PhD, associate professor, head of department of the hospital-hygiene department) with practical experience in environmental and healthcare infection prevention. Coders were a male anesthesiologist with a master’s degree in design, a female social education worker specialized in systemic family therapy, a male resident in pediatrics and a female trauma surgeon all with experience in interprofessional education Two coders had worked in our preceding study. Coders were instructed in the coding procedure using individual- and team- reflections and supervision by the investigators.

About 40% of the study participants were known to the main investigator before the course. Ten percent had been in regular contact with him.

Analysis.

The main investigator as a single researcher [33] analyzed the data using a iterative inductive approach with 1) assignment of simplified “tags” to recorded items, 2) development of specific “codes” from these tags (recontextualized results and meanings) and 3) deriving themes (recontextualization and comprehension of results) [34]. In contrast to the first study, he did not participate in code allocation.

All coders participated in coding the inductive findings. For these, a code was accepted if three coders allocated a code to a tag. After acknowledgment of many “ethical” and “psychological” codes, two coders concentrated on the second round of deductive coding applying the four principles of biomedical ethics according to Beauchamps and Childress [30] on the one and the Calgary Intervention Model [24] on the other hand. The other two (both involved in the preceding study) coded additionally in line with our prior study’s findings [8]. In both deductive approaches, two coders had to agree to assign a code. Additionally, we took a semi-quantitative approach to illustrate our results and the frequency of code allocation (see Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Codes and main themes according to inductive codings.

https://doi.org/10.1371/journal.pone.0246820.t001

Codings were analyzed using MS Microsoft Excel (Microsoft Corporation, Redmond, USA). Illustrations were done using MindManager 2019 (Corel Corporation, Ottawa, Canada).

Participants’ characteristics

All 35 participants (27 ED and ICU physicians, 3 pharmacists, 5 ED and ICU nurses) of the hospital groups were recruited from six hospitals in Lake of Constance Region. Recruitment was done via the academy of health care professionals in Singen am Hohentwiel. Six courses were held from April 2018 to November 2018 at the Hegau-Bodensee-Klinikum Radolfzell. The external validation group (11 physicians all from different hospitals in Germany) was assembled in June 2018 and was recruited from a medical education event at the University Hospital Bonn.

Coding process

A total of 773 items were decontextualized (214 for anger (27.8%), 208 for fear (26.9%), 136 for sadness (17.6%) and 127 for disgust (16.4%), 88 for enjoyment (11.4%)), tags (comprehension and of items) allocated, and codes (meanings) and themes and subthemes formed.

We assigned 10 inductive codes for anxiety 10, 11 for anger, 7 for joy, 5 for sadness and 8 for disgust (see Table 1) forming each category’s sub-themes.

We identified three main-themes and 14 sub-themes.

Main-theme 1: Verifying the emotional impact of MDROs

Main theme 2: MDROs are linked to ethical considerations

In the first deductive part, all items were coded using the four principles of medical ethics (see Fig 3). These showed that benevolence and non-maleficence were coded to a large proportion of items linked to less care and giving up patients: “last on the operation schedule”, “no information to patients”, “information provokes fear”, “wrong consequences”, “the docs are no longer seeing the patient”, “these patients are uninformed”, “stigmatization”, “patients think they are going to die”, “nobody is visiting them” and “neglect of patients” were some of the phrases noted. In contrast, autonomy was seldom coded (“longer stay in hospital”, “I feel compassionate towards the patients, because they are not allowed to leave their room”, “The patients are locked into their room”) and mostly related to patients’ restricted freedom, due to being isolated. Justice codings was seldom and mainly focused on the subjectively perceived unfairness of the workload’s distribution (“more work for us–and there is no help”) and economic aspects (“economic loss”, “MDRO is expensive”, “beds have to be blocked”).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3.

https://doi.org/10.1371/journal.pone.0246820.g003

Main theme 3: MDROs impaired family functioning

Six deductive codes (3 structural, 2 functional and 1 developmental) according to the Calgary Family Intervention Modell by Wright and Leahey were assigned to all items (see Fig 4):

Structural—Contexts (ethnicity, race, social class, religion, spiritual) were coded only seven-times and were associated to knowledge and backgrounds of families. In contrast, 127 items were coded for the Structural-External (larger family composition, relatives, friends, and other networks). Most codes for emotions were linked to being socially isolated from friends and to shame and the fear of transmitting MDRO. Structural-Internal codes (family composition, gender, hierarchical rank, subsystems) yielded similar findings, like problems with knowledge (“Relatives don’t understand MDRO management”, “No knowledge leads to panic”) leading to “irrationality”, fear and despair (“O my God–what will happen to my children?”, “Will I ever get to go home?”). Comparable, functional codes revealed severe impairment of functional-instrumental (daily activities) and functional-expressive (family communication, problem-solving, beliefs, power) aspects witnessed in patients and families with MDRO: “They do not know, how to manage MDRO”, “helplessness”, “stigmatization”, “social isolation”, “loneliness” were mentioned in conjunction with anxiety and sadness. “being blamed” of different types and intensity (“malpractice”, “guilt”, “being prosecuted”, “getting even”) were coded for anger several times. Development items (tasks, plans) were coded only 25 times. Those codes were mainly associated with to fear and anxiety concerning MDRO management and uncertainty.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4.

https://doi.org/10.1371/journal.pone.0246820.g004

How themes from the last study were deductively coded

Finally, all inductively derived codes in our first study was coded deductively in our findings to confirm our primary study’s results. All codes, namely lack of knowledge on the part of specialists and lay persons, insufficient information transfer, lack of resources, impairment of prognosis, over-rating the MDRO leading to excessive isolation and panic, under-rating the MDRO leading to bacterial spread, poor communication techniques and inconsistent information within medical staff and globally, were coded several times.

1. MDRO cause an imbalance of information aggravated by and resulting in emotional reactions

According to the perceptions of health care workers our results show, that a MDRO diagnosis can overstretch the medical knowledge of health care providers, who complain about staff shortages and time pressure [36]. Caregivers’ perceptions of not having been adequately trained in MDRO management [8, 37] are also associated with their emotions (anger about a heavier workload due to barrier precautions, anxiety about contracting the MDRO or endangering their family, feeling sadness as compassion for the MDRO patient, disgust among younger staff), over- or underrating the MDRO risk, and the inability (due to lacking knowledge-transfer know-how [38, 39] or time pressure) to provide patients with valuable information. The discussion groups’ most frequently expressed emotions were especially anger and anxiety, and even the “enjoyment” items yielded items suspicious of triggering anger and revulsion toward external groups (i.e., scientists, reporters, hygiene experts), whom the caregivers participants thought might profit from MDROs.

Participants reported that patients and their relatives need information especially if there is a threatening underlying medical condition [14]. Medical staff having limited information and resources and with their own emotional reactions to MDROs, fail to transfer information, so that patients and their families feel helpless and unable to control the situation. Caregivers feel anger, sadness and despair as a consequence [40] and having to handle increased workload to compensate for these emotions to secure or reestablish the relationship between care givers and patients is experienced [18].

2. MDRO impairs family functioning

The deductive codes we derived applying the Calgary Family Intervention Model [24] show that health care providers perceive that MDRO can affect nearly all aspects of the model. In contrast to “classical” patient-focused psychological and intervention models, systemic psychology focuses on the patient’s environment and networks, especially the family. The Calgary model subdivides these aspects into three main themes: structural, functional, and developmental. Our findings suggest, that an MDRO diagnose can have a significant impact on the entire familiar system: An MDRO diagnose paired with insufficient information from medical professionals and the media (“We had the Killer-Klebsiella”) suggests “danger” somewhat comparable to pandemic threats like Ebola, leading them to being cut off from their family systems. This can occur because of the patient’s own wishes (“I do not want to endanger my family”), or the wishes of the family members to protect the family or both. Furthermore, our codes revealed interference with family functioning, like daily activities (“What do we do with the kids at home?”) and interaction with wider systems like neighbors (“Everybody in the town knows about my situation”). Isolation therefore seems to be a problem for both for the patient and their family. Considering that our discussion groups were not intended to evaluate whether general stigmatization, social tension [41] and prejudice is transferable from individuals to social groups (families) to organizations (hospitals, corporations) and to socio-political constructs like cities and even nations: epidemic outbreaks in the past have demonstrated that patients get separated from their families, families from their villages, villages from their countries and countries from the rest of the world: SARS, COVID-19 and Ebola [42] are merely recent examples.

3. Medical ethics

Taking the bio-ethical approach of four principles [30], especially the items “benevolence” and “non-maleficence”, we observed a major interference with MDRO management. Patients with MDRO are usually (following national guidelines) segregated from others to keep other from being MDRO-colonized and infected [43]. Especially in persons (lay and staff) with limited knowledge (as participants mentioned several times here and in our preceding study) managing MDRO is equated with highly contagious diseases. Taking a utilitarian approach [44], an MDRO-infected patient’s segregation is justified to protect the majority from the infected minority—putting MDRO-patients at risk for worsening social, economic and health-care related vulnerability. People perceive MDROs as a kind of pestilence. However, our findings only reveal bio-ethics and there is need for more investigation and work in this topic.

Limitations

Our findings may be limited by the following factors:

Our evaluation concentrated on the subjective perceptions and experiences of health care providers working in intensive care settings. We did not assess the perceptions of patients and family members. Thus, it is unclear, whether our findings reflect the perceptions of patients and their families or staff from other specialties. On the other hand, our findings may be useful for exactly that subgroup, as they concern patients and relatives already burdened with a critical medical condition. Our findings may also be biased by individual exceptional experiences. As emotion cause humans to be highly attentive, leading them to recall memories better [11]. Health care providers’ reports may just be “the tip of the iceberg” concerning their memories about MDROs. Therefore, the generalizability of their memories should be evaluated in other settings, e.g., in direct observation studies. Most of our codes concern socially significant isolation measures primarily and not the individual management of MDROs, like antibiotic treatments or surgery. Further studies should differentiate between MDROs themselves and MDRO management.

Over 50% of the coders in our first study also worked as coders in this follow up study. On the one hand, this confirms the grounded theory approach and on the other hand, fixation errors may have occurred. To avoid this, we decided to take a four-coder approach in contrast to the classical one or two-coder approaches. Additionally, we used a set of different frameworks (Atlas of Emotions, Beauchamps model, Calgary Model) for secondary deductive coding and not an inductively developed theme composition. We decided to use these well-established frameworks to enable better comparison and precise planning of ongoing research projects. We expect that where these frameworks interact (e.g., how emotions can affect ethical decision making with regard to MDRO) may uncover especially important information concerning MDRO management and the medical education of health care workers. Selection bias may be present but is limited by this evaluation by a group of 11 professionals from different hospitals. Nevertheless, a broad approach, e.g., involving international online-surveys and different populations (medical staff, visitors, unaffected persons, affected persons and their families) should be taken to expand upon our results.

Furthermore, we did not examine whether patients and their relatives by experiencing MDRO loose trust hospitals or even whole health care systems in [45, 46] due to their emotional reactions. Thus, future research could concentrate whether MDRO management in hospitals causes patients to loose trust in medical systems and if this distrust causes people to avoid hospitals or health care providers altogether.

Last, future investigations could employ other methodologies, i.e., audio or videorecording to obtain deeper insight in participants’ para- and nonverbal expressions, thereby revealing more information. At this stage of our investigation, we did not attempt that because of ethical considerations, but we will do so in future projects.

Additionally, we did not ask health care workers how long they spent talking to patients talking about MDRO or how they compensated for inadequate information or emotional crises. Nor did we determine whether care givers were trained in communication methods to help with strong emotions [47, 48]. We also did not assess whether time shortage and medical education were objectively or only subjectively perceived.

Next, the extent to which patients poorly informed about MDROs seek more information in the media and internet remains unclear to even more confusion, conflicts and impaired of the patient-physician-relationship as reported for dermatology [49].

All these limitations reveal this topics’ broad spectrum and the great need for qualitative and quantitative future investigations in different settings (intensive care, rehabilitation, family medicine, emergency medicine and home care) and populations (health care workers, social workers, educators, lay persons, family members with and without MDRO experience).

Generalizing our findings (with this study affirming our first one’s results), profound and ongoing multiprofessional medical education in MDRO management, perception and identification of emotional reactions known as “empathic accuracy” [17], communication techniques how to break bad news [50] and how to react to emotional reactions of oneself and of others [13] is needed to manage social systems in health care dealing with MDRO.