Evaluation of a new animal assisted intervention service for an adult intensive care unit

Introduction

In modern critical care medicine, the promotion of recovery, over mere survival, for both physical and non-physical domains, is the main rehabilitation objective (Wilson et al. 2019). Research has shown that a significant number of patients who survive critical illness have post-intensive care syndrome (PICS) and require a personalised rehabilitation prescription to meet the individual physical and non-physical needs of each patient (NICE 2018; Puthuchery et al. 2021). Wilson et al. (2019) highlighted that dehumanisation of patients can negatively impact on patient engagement and interest in their own well-being during the rehabilitative phase. An animal assisted intervention (AAI) is defined as an interaction between an animal and a patient during a medical, nursing or therapeutic intervention (Hosey et al. 2018). It is an umbrella term to describe different types of animal interventions. These range from goal driven interactions to informal introductions; both provide therapeutic benefits and humanise care. AAIs contribute to a holistic approach to treatment by registered volunteers visiting with their behaviourally assessed animals. Hosey et al. (2018) identify how AAIs may help reduce suffering through humanisation, aiming to improve mood and patient engagement thus facilitating achievement of rehabilitation goals. This in turn may reduce loneliness, physiological burden and the need for medication resulting in improved cost-effectiveness. There is an established evidence base for evaluation of AAI in healthcare settings such as nursing homes, community settings and psychiatric units (Lasa et al. 2011; Lundqvist et al. 2017; Uglow 2019). However, there is little research in AAI services in intensive care unit (ICU) settings, despite positive experiences being shared widely on social media platforms (ICS 2019). Uglow (2019) studied the use of AAIs on paediatric wards at a large NHS university teaching hospital in the United Kingdom (UK). Over 200 responses to a survey contributed to an overwhelmingly positive response resulting in the recommendation that similar services should be available across the UK. A similar AAI service had been implemented at a general ICU for adult patients at a large NHS teaching hospital in the East Midlands of the UK. The primary objective of this evaluation was to ascertain if the service was feasible and safe. Measuring the impact of the service for patients, visitors and staff would also inform future service development.

Method

Setting

The AAI service to a 20 bedded general ICU was implemented in September 2019. The registered charity pets as therapy (PAT) is a UK based charity which regulates and supports animal visits in various settings. A registered PAT volunteer and their dog visited the ICU on a weekly or fortnightly basis. The volunteer was also registered with the volunteer service at the NHS Trust and trained to the recommended level, as per the Intensive Care Society (ICS) guidance for AAI in a critical care setting (ICS 2020). The volunteer made contact with a senior member of nursing staff by telephone on the morning of the planned visit. Communication with the multi-professional team took place prior to the visit to identify any potential risks. Exclusions were identified as allergy, fear or an individual’s indication to decline the service. AAIs occurred during visiting hours.

The volunteer determined the length of the interaction dependent on the behaviour of the dog, person(s) involved and perceived benefit for all. Most interactions were less than 20 minutes long. The interaction involved verbal communication and touching the dog during periods of rest, delivery of care or treatment. The maximum total length of time the dog visited was 120 minutes on any one day, with regular welfare breaks for the dog (Figure 1).

Figure 1: The PAT dog on a visit in ICU.

Subjects

All participants were asked to give verbal consent to the AAI. If verbal consent could not be gained from the patient or their next of kin the AAI did not take place. The names of patients who interacted with the dog was compiled for the date of the interaction and stored according to trust and ICS policies. The service ran on a voluntary basis with no monetary cost implications.

Design

A questionnaire was designed by the author using the Survey Monkey application to collect opinions on the interaction between the individual concerned and the dog. The author purposefully chose similar questions to those used in a previous study by Uglow (2019) to allow for comparison. It comprised of ten questions, of mixed methods design, using 10-point Likert scales and free text responses. The questionnaire collected basic demographic information such as age, gender and for staff, their job role, using multiple-choice questions.

There were 2 questionnaire designs; 1 for patients and visitors and another 1 for staff (Appendix 1). Following an interaction, the person involved was asked if they would complete an electronic questionnaire using a convenience method of sampling. The questionnaire was pre-loaded on iPads or could be accessed by scanning a QR code with their personal electronic device. When the participant self-completed the questionnaire, implied consent was assumed.

The study was confirmed by the trust as a service evaluation that did not require ethical approval.

Data analysis

The responses were collected, stored and collated by the Survey Monkey application. Quantitative data was then further analysed by the author using simple descriptive statistics, whilst qualitative data was analysed using thematic analysis (Robson 1993).

Results

47 questionnaire responses were collected between September 2019 and March 2020. Responses comprised of 24 patient and visitor responses (70.83% male; demographics shown in Figure 2) and 23 staff member responses (91.30% nursing staff-various grades, 4.35% doctors and 4.35% physiotherapists).

Figure 2: To show the demographic data for patients, relatives and visitors who completed the questionnaire.

There were no concerns from staff, patients or relatives in relation to the dog’s cleanliness, presentation, welfare or handling. There were no reports from staff that AAIs were disruptive to patient care. 100% of staff and 95.8% of patients and visitors would recommend the service to other wards and hospitals. An overwhelming majority of responses highlighted a positive perceived benefit and high level of enjoyment for patients, visitors and staff with maximum scores of 10 being given by at least 70% of responders (range 70–95%).

Quantitative data: staff responses

44% of the interactions were observed by the member of staff without interacting with the patient or dog. However, 33% of patient interactions involved the patient and the staff interacting with the dog at the same time (Figures 3 and 4).

Only 4.35% (n = 1) of staff identified the interactions as an AAI.

Figure 3: To show the type of interaction experienced by the staff member (n = 23).

Figure 4: To show the staff members perceived level of enjoyment for patients, visitors and staff during the AAI (n = 23).

Quantitative data: Patient/visitor responses

The mode, median and mean length of ICU stay at the time of the AAI was 3, 4 and 7.7, respectively. A majority of the patients were resting or inactive, including sitting in a chair when the AAI took place (82.3%). Only 13% of the AAIs were during therapy (physiotherapy, occupational therapy or speech and language therapy).

83% percent of patients and visitors rated the interaction with the dog as a maximum score of 10, extremely enjoyable. 75% percent of patients and visitors rated the interaction as extremely beneficial (Figures 5 and 6).

Figure 5: To show the patients/visitors perceived level of enjoyment during the AAI (n = 24).

Figure 6: To show patients/visitors perceived benefit as a result of the AAI (n = 24).

Qualitative data

The qualitative data was compiled from the comments made in the free text responses. Comments were independently grouped into emergent themes for the 2 groups of responders. There were 6 emergent themes for each group. The emergent themes and some quotes are detailed in Tables 1 and 2. 5 of the emergent themes were shared by both groups. These were mood, distraction, dog and handler, AAI and recommendations. The themes that were not shared between the 2 groups of responders were infection control (patients/visitor responses) and impact (staff responses).

Table 1: To show the emergent themes and comments from the patient and visitor comments.

*Shared themes for staff and patients/visitors.

Table 2: To show the emergent themes and comments from the staff comments.

*Shared theme identified from patient and visitor comments.

Discussion and conclusion

The primary findings of the evaluation identified that patients, visitors and staff did not have any concerns or issues relating to a visiting dog and handler to provide AAI to patients on a general intensive care unit.

The project also highlighted that greater than 95% of all responders would recommend a similar service to other wards and hospitals. There were 5 common emergent themes identified from the qualitative data. They were AAI, distraction, mood, recommendations and dog and handler. The was an overwhelming number of positive comments made in these emergent themes. All of these findings concur with those of Uglow (2019) and highlights the potential need for developing an AAI service more widely. Provision of an AAI to patients when they move from ICU to a general ward would promote continuity of humanised care. This would potentially facilitate patient engagement and self-interest in their rehabilitation and well-being (Wilson et al. 2019).

There were no objections to the provision of the service from any ICU staff, patients or visitors. Potential risks were identified during a pre-visit telephone call on the day of the visit. One staff member highlighted a mild dog allergy and the handler ensured that there was no contact between the dog and this staff member.

The results of the questionnaire highlighted that visitors and patients reported that 79.1% of interactions with the dog were when the patient was inactive or resting. Staff reported a similar figure of 86.4%. This could be attributed to the fact that most dog visits took place during visiting hours (1–4pm), a time when patients are typically less active. It is also worth recognising that when a patient is sitting out in a chair they are perceived as being inactive, however, this may not be the case if it is part of the patient’s progressive rehabilitation plan. This requires future consideration, by identifying if there are influential factors for maximising the benefit of the AAI, for example, patient position or time of day.

Following the RCN guidelines that the dog’s front paws can be placed on the bed if a single use, disposable sheet is used to protect the patients bed sheets means that patients can still interact with the dog if they are not able to get out of bed. Future recommendations would involve recording the patient and dogs’ position during the AAI which would enable more detailed evaluation of the impact of the interaction.

It is also recommended that responders are separated into 3 distinct subgroups (patients, visitors and staff) for comparison. This would also allow for more detailed evaluation and analysis in future studies.

Only one responder identified interactions as an AAI. It would be interesting to explore the knowledge and understanding of staffs’ perception of an AAI. By exploring this in detail it could highlight training needs to facilitate appropriate use of the AAI and maximise the impact of both informal and targeted AAIs to achieve individual goals for the patients.

The survey responses also highlighted recommendations for service development. It was identified by one patient or visitor response that patients without any visitors may benefit from an interaction with the dog more than someone who had visitors. In contrast to this, however, a member of staff identified that a visit from the dog whilst the visitors were present facilitated interaction between the patient, visitors and staff. Another comment from a staff member suggested that long term ICU patients may benefit more from the dog visiting than short stay patients. The dog handler also has some anecdotal evidence and reflective thoughts on this. It seems that perceptions are personal and individual to each patient or staff member, but it also highlights the need to explore this in more detail in the future. The provision and detail of the information through displaying posters and giving explanations to patients, visitors and staff is another area that could be explored, following the comment from a member of staff who regarded this as important. There are plans to explore this with input from the ward patient advisor with a view to including information on the trusts website.

Although self-reported data or visual analogue scales are not considered as robust levels of measurement, they are frequently used in research to measure and evaluate the impact of interventions. For example, pain rating scales are routinely used in medical assessments to plan and prescribe appropriate analgesia. Another limitation of this evaluation is one of bias. The sampling method was convenience sampling. This may increase the likelihood of selection bias. Whilst responders were informed that the questionnaire was anonymous, the presence of the handler on the ward whilst the questionnaire was completed may cause observation bias, known as the Hawthorne effect. This occurs when responders are aware that they are being observed or involved in scientific study and this has a potential to influence the answers or responses given. Finally, there is a risk of confirmation bias because the researcher was also the dog handler and may therefore be looking for information or patterns in data to confirm pre-conceived ideas. These issues of bias need addressing in any future research plans.

The findings of this evaluation align with the findings of Uglow (2019) and contributes to the currently limited but developing research and evidence base of AAIs in ICU.

This evaluation identifies that a service providing AAIs to adult patients in ICU is safe and feasible. It highlights additional perceived benefits for visitors, staff and patients. In addition, a number of recommendations for service development and future research have been highlighted.

Key points

• AAIs in ICU are safe and feasible.

• Contributes to the developing research base of AAIs in ICU.

• There is an overwhelming perceived benefit and enjoyment of AAI for patients, visitors and staff.

Declarations

The evaluation received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Acknowledgements

Thank you to those patients, visitors and staff who have supported the service and completed questionnaires.

Hosey, M.M., Jaskulski, J., Wegener, S.T., Chlan, L.L., & Needham, D.M. (2018). Animal-assisted intervention in the ICU: A tool for humanisation. Crit Care, 22(1). https://ccforum.biomedcentral.com/articles/10.1186/s13054-018-1946-8.

Intensive Care Society. (2019). Guidance for animal assisted intervention (AAI) in a critical care setting. https://www.ics.ac.uk/Society/Guidance/PDFs/Animal_assisted_intervention_guidance.

Lasa, S. M., Ferriero, G., Brigatti, E., Valero, R., & Franchignoni, F. (2011). Animal-assisted interventions in internal and rehabilitation medicine: A review of the recent literature. Panminerva Medica, 53(2), 129–136.

Lundqvist, M., Carlsson, P., Sjödahl, R., Theodorsson, E., & Levin, L. Å. (2017). Patient benefit of dog-assisted interventions in health care: a systematic review. BMC Complementary and Alternative Medicine, 17(1), 358. https://doi.org/10.1186/s12906-017-1844-7.

Morrison, M. (2007). Health benefits of animal assisted interventions. Journal of Evidence Based Integrative Medicine, 12 (1), 51–62. https://doi.org/10.1177/1533210107302397.

NICE. (2018). Clinical guideline 83: Rehabilitation after critical illness in adults (2018). https://www.nice.org.uk/guidance/cg83.

North, R. (2018). The implementation of pet therapy in critical care: A service improvement project. Journal of the Intensive Care Society, 20(2_suppl), 1–253. https://journals.sagepub.com/doi/full/10.1177/1751143719835452#_i748.

Pets as Therapy. (2020). About us. https://petsastherapy.org/?gclid=EAIaIQobChMIv4Cv6NXW8gIVwrHtCh3b6wftEAAYASAAEgJxX_D_BwE.

Puthucheary, Z., Brown, C., Corner, E., Wallace, S., Highfield, J., Bear, D., Rehill, N., Montgomery, H., Aitken, L., & Turner-Stokes, L. (2021). The post-ICU presentation screen (PICUPS) and rehabilitation prescription (RP) for intensive care survivors part II: Clinical engagement and future directions for the national post-intensive care rehabilitation collaborative. Journal of the Intensive Care Society, 2021(Feb). https://doi.org/10.1177/1751143720988708.

Robson, C. (1993). Real world research: A resource for social-scientists and practitioner researchers. Blackwell Publishers.

Royal College of Nursing. (2019). Working with dogs in healthcare settings. https://www.rcn.org.uk/professional-development/publications/pub-007925.

Uglow, L. (2019). The benefits of an animal assisted intervention service to patients and staff at a children’s hospital. British Journal of Nursing, 28(8), 509–515. https://doi.org/10.12968/bjon.2019.28.8.509.

Wilson, M. E., Beesley, S., Grow, A., Rubin, E., Hopkins, R. O., Hajizadeh, N., & Brown, S. M. (2019). Humanising the intensive care unit. Critical Care, 23(1), 32. https://doi.org/10.1186/s13054-019-2327-7.