שנת סיום: 2021

כותרת עבודת הדוקטורט: 

Development of a Theoretical-Applied Model Based on the Positive Deviance Approach (PD), for Behavioral Change to Adhere with Infection Prevention and Control (IPC) Guidelines and Reduce Healthcare Associated Infections (HAIs) Rates in Israeli Hospitals

מנחים: פרופ' ענת גסר-אדלסבורג ופרופ' אלון מוזס

Abstract

Research Framework: Healthcare associated infections (HAIs) are considered one of the most critical and investigated issues in public health worldwide. Despite the accumulated knowledge and the implementation of varied strategies in the field, adherence with infection prevention and control (IPC) guidelines remains low, infection rates high, and there remains a gap between recommendations and implementation in practice. Paradoxically, health organizations guidelines are in many cases incompatible with what occurs on the ground. Furthermore, they do not always cover every situation on the care continuum. It is therefore clear that it is not enough to combine several strategies in order to achieve the goals, but rather there is a need to take into account factors that motivate building effective intervention programs. To address this issue, we chose a unique bottom-up approach— “The positive deviance (PD)” approach. PD is based on the premise that in every community there are certain individuals or groups whose uncommon behaviors and strategies enable them to find better solutions to problems than their peers, while facing worse challenges and having access to the same resources. They are "deviant" in that their behavior is not within the norms and "positive" because they found effective and successful solutions to address problems, without additional resources.

Objectives: The current PhD dissertation seeks to investigate the following: (1) To locate barriers that prevent healthcare professionals (HPs) from adhering to IPC guidelines in three hospitals; (2) To identify and classify PD practices in maintaining hygiene, by field and by sector, and validate the PD practices by their respective Infection Control Units; (3) To build social network maps to help spread and implement the PD practices (Preliminary study, Chapter 3); (4) To present the implementation and diffusion process of the PD practices, through deconstructing central line (CL) insertion processes guidelines for two intensive care units (ICUs) at Haddasha hospital (Manuscript 1, Chapter 4); (5) To characterize the socio cognitive profile of HPs who exhibit PD behaviours (Manuscript 2); (6) To examine the effectiveness of PD intervention in behavioral change in maintaining IPC guidelines. (Manuscript 2, Chapter 4); (7) To examine the effectiveness of PD intervention on HAIs rates from resistant bacteria and CL-associated bloodstream infections (CLABSI) (Chapter 5); and (8) To construct applied methodological tools, based on the PD approach, that will provide an infrastructure to develop more effective interventions to mitigate HPs barriers. Research methodology: A mixed-methods quasi-experiment study design that was conducted over 35 months (January 2017 to December 2019) at three Israeli governmental hospitals: Hadassah Medical Center (Hadassah), Rambam Health Care Campus (Rambam), and Bnai Zion Medical Center (Bnai Zion); (Figure 2). The study included three main phases: (Phase 1) Pre-Intervention - a retrospective baseline period (9 months); (Phase 2) PD Intervention - divided into two periods: Intervention 1 (9 months) - identifying HPs barriers to adhering with IPC guidelines, identifying and classifying positive deviant HPs, validating PD practices, and building social networks maps; Intervention 2 (12 months) - diffusion and dissemination of the PD practices (Manuscript 1); and (Phase 3) Post Intervention Follow-Up (5 months) - characterization of the socio-cognitive profile of PDs, evaluation of PD intervention on behavioral change (Manuscript 2), and HAIs rates (Chapter 5).

Main findings: Of 250 HPs who took part in the study from all sectors (nurses, physicians, support staff, and cleaning staff), 185 performed semi-structured interviews and 69 focused observations, while 23 participated in focus groups and 53 in simulations recorded on video (Table 1). In the context of barriers that adhere with IPC guidelines, in addition to barriers mentioned in the literature, we identified a new and significant barrier that has not yet been discussed, which we coined "gray area". This barrier encompasses the variety of situations on the care continuum that are not addressed in the accepted guidelines, where staff members interpreted or understood differently, and thus were unsure how to proceed. In 65% of interviews, participants spoke about "gray area" situations. These conditions evoked feelings of confusion, uncertainty, and the perception of the procedures as incoherent. Five key themes emerged from the interviews, dealing with the gray area, included: (1) Lack of uniformity in infection control procedures; (2) Vagueness in the guidelines concerning the extraction and sending of tests; (3) Uncertainty as to the definitions of "clean” or “contaminated” spaces, and lack of guidelines concerning mobile equipment and responsibility for performance; (4) Challenges in the transition from "clean" to "dirty" areas in the course of treatment and back; and (5) Hand hygiene (HH) training and reminders.
During the study, a total of 38 HPs were identified as PD individuals, responsible for 70 PD practices that were classified and validated by their respective Infection Control Units and did not exist in the official IPC guidelines. The practices were divided into 16 topics on the care continuum, among them: Removal and replacement of a dressing on a surgical cut; Removal of protective clothing when leaving an isolation room, and performing hand hygiene; Procedure of taking and sending blood samples; Procedure of CL insertion; Performing suction for a respirated patient; Sterilizing a stethoscope; Mixing IV meds and carrying them to the patient, etc. (Preliminary study)
The diffusion and implementation process of the CL insertion practices in two intensive care units (ICUs), was done through multiple platforms: (1) Tree hands-on learning simulations were performed by four PD physicians, in order to present, step-by-step, the process of CL insertion with a reasoned explanation ;(2) Two Discovery & Action Dialogue (DAD) meetings were held, during which the recorded videos were played, a discussion followed on the viability of disseminating these extraordinary practices among other physicians; (3) All the PD practices associated with the insertion of the CL identified videos were combined into one edited video and disseminated to all the ICU physicians via the WhatsApp application; (4) The edited video of CL insertion was presented in a unique professional conference dedicated to the PD approach and its implementation in Israel; (5) We opened a Facebook page entitled "Positive Deviance Israel" with the goal of creating a community and a scientific social network of public HPs interested in the approach, and to serve as a platform for raising awareness, dissemination and exchanging of information. (Manuscript 1)
Regarding to characterization the socio cognitive profiles of the HPs who exhibit PD behaviours, findings indicate that socio cognitive variables such as external locus of control, perceived threat, and social learning, were significant predictors of a person exhibiting PD behaviours. Almost 70% of HPs reported behavioural change and creating social networks as a result of the PD intervention in maintaining IPC guidelines, 16.9% of them are a "PD boosters" (a new group of HPs who adopted the positive practices of PDs that were originally identified, and added additional practices of their own). Moreover findings show that social networks can contribute to internalizing and raising personal accountability even among non-PD professionals, by creating a mind map that makes each person believe they are an important node in the network, regardless of their status and role. (Manuscript 2)
Regarding to the effectiveness of the PD intervention on HAIs rate, nine departments participated in the research, of which six were intervention departments and three control ones. Findings show that for Alert Bacteria at Hadassah hospital, after comparison of the three intervention phases together, showed a significant advantage (lower IRR) in the intervention department (MICU-EK) as compared to the control (MICU-MS) [ratio=0.27, 95% CI = (0.13,0.56), 2 (1) =12.66, P=0.0004]. Regarding BSI - in the intervention department (MICU-EK) a significant IR reduction of 47% was detected in all intervention phases together – relative to the baseline (IRR=0.53, 95% CI = 0.32, 0.87), although a comparison between the two departments showed no significant difference. Regarding CLABSI- due to the small number of events, an IRR comparison between the two departments showed no significant difference. For all intervention phases together, the ratio was 2.18 with 95% CI = 0.26,18.21 (2 (1) =0.52,
P=0.4706). Regarding Alert Bacteria in Bnei Zion- IRR comparison between intervention and control departments (Inter Med B vs. Inter Med A) showed a significant advantage (lower IRR) in the intervention department on post-intervention phase, relative to the baseline [ratio=0.10, 95% CI = (0.01,0.82), 2 (1) =4.62, P=0.0315]. Reduction in IR detected in the Orthopedic Department contributed to a 58% reduction on all intervention phases together – relative to the baseline (IRR=0.42, 95% CI = 0.1662, 1.0662). The annual average of IR from Alert Bacteria at Rambam, dropped from 4.7 in 2017 (Baseline phase) in the Intervention department (Inter Med H) to 3.9 after PD intervention at the end of 2019.

Research contribution:
Methodological contribution: This is pioneering study. To the best of our knowledge, it is the first to develop a detailed methodology that describes the use and application of the PD intervention step by step (from the data collection phase to the implementation and evaluation phases of the intervention). To date, most of the intervention programs in the area of IPC that used this approach focused on its effectiveness in decreasing HAIs and increasing HH compliance rates and failed to explain the methods and tools they used to find and implement them, as Baxter et al. (2016) reinforced this claim in their review. They concluded that despite the success of the PD approach, understanding of how the approach works is still limited. In this research, we expanded the PD approach methodology by detailing further guidance to help identify and classify positive deviants, selecting the methods used at each stage, and engaging front-line staff in the process of effectively disseminate findings (Baxter, R., Taylor, N., Kellar, I., & Lawton, 2016).

Theoretical contribution: The PD approach is an innovative iterative approach. In this study we expanded and added a theoretical layer to the existing literature, during the first two periods of the study. In phase 1, we coined the "gray area" barrier, which expanded the already known "circle of barriers" and provided significant insights into HPs behavioral difficulties related to adhering to IPC guidelines. In Phase 2, we examined if professionals who exhibit PD behaviours have a distinctive socio cognitive profile (locus of control, risk behavior, fatalism, thinking style, and social learning), compared to other HPs. Unlike most studies dealing with intervention programs that examine the effect of various variables on HPs behavior, the uniqueness of this study is that it works the other way around. We began by identifying the PD behaviors, and then sought to examine if there were differences in the characteristic's profiles. This socio-cognitive profile allowed fora better understanding of HPs, and their skills and capabilities (Figure 1).

Practical implications for future research: The PD intervention made an impact on HPs behavioral change in maintaining IPC guidlines, along with a significant reduction in HAIs rate among the intervention departments compared to the control departments. Moreover, the findings indicate that the use of various applied methodological tools to implement the PD approach, leads to the construction of a social networks. Social networks can create and generate a social and mind map that makes each person an important node in the network regardless of their status and role. Thus, if a person does not prevent the transmission of infections by their actions, the collective social and professional system is not disbanded. HPs are empowered when the social network is prominent and visible in the hospital system, so that each staff member perceives their contribution is recognized and appreciated by their colleagues and management. Finally, translating theory-based Positive Deviance approach into an applied tool, one that details action plans in a step-by-step manner, can help practitioners and researchers adopt and implement the same within intervention programs, thereby mitigating HPs barriers and reducing HAIs.

Conclusions: Traditional approaches in the field of IPC are mostly based on "Evidence Based Practice". Contemporary systematic reviews indicate that in order to overcome barriers, behavioral components must be focused on. The PD approach encourages HPs to find creative solutions that address the "circle of barriers" pointed out in this study. This approach is based on the strengths and resources of the community members and motivates the cooperation of all HPs. The finding of this study gives an in-depth view of the barriers of HPs, along with raising creative solutions that lead to behavioral change and a decrease in HAI rate.
Based on findings, this doctoral dissertation presents an advanced applied theoretical model, proposing integrative solutions based on the PD approach. The model lists four key parts, each part will detail the question that underlies challenges raised in the literature: How can HPs maintain a hygienic behavior routine in any situation over time? How to create HPs engagement when responsibility is transferred from the top to the community itself? How to turn PD practices into positive norms? How to translate the PD approach into an applied methodological tool? Following these challenges, we have refined the problems arising from those challenges, and present our solutions. This model may help professionals and researchers adopt and integrate it into community-based intervention programs, tailored to the hospital's specific profiles, thus reducing the HPs barriers and HAIs rates.