Cold is an unpleasant sensation at best, but has been linked to increased anxiety and pain (Aléx et al, 2014). Cold can lead to accidental hypothermia, where, in restricted physiology, core temperature drops below 35°C. Hypothermia leads to worse clinical outcomes including death (Aléx et al, 2013a).
The authors undertook a structured literature search on the PubMed database, and followed up references cited in articles. The results of this search showed there was a paucity of published literature on the topic of ground ambulance patients' thermal comfort and ambulance care preventing hypothermia.
In two papers, Aléx et al (2013b) interviewed patients of a Swedish ambulance service, who reported they felt cold during transport. Haverkamp et al's (2018) literature review and Lapostolle et al's (2012) HypoTraum study found that all prehospital patients were at risk of becoming hypothermic.
To date, no study reporting the patient's thermal comfort/satisfaction in the Irish context has been published and there is anecdotal belief that ambulances are cold (Advanced Life Support Group, 2019). A question was developed to test the authors' hypothesis that patients and their escorts transported to an Irish emergency department (ED) by the statutory provider, the National Ambulance Service (NAS), would feel cold.
Setting
Ireland is situated the north-western seaboard of the European continental shelf. The weather is mainly influenced by the Atlantic Ocean and the Gulf Stream giving it a mild, temperate climate, mainly without the extremes of temperatures experienced by other countries at similar latitude (Met Éireann, 2022).
The NAS is the statutory prehospital emergency and intermediate care provider for the state, and works in cooperation with the Dublin Fire Brigade, the Irish Air Corps, the Irish Coast Guard and private and not-for-profit providers. NAS receives more than 300 000 emergency 999/112 calls each year (NAS, 2022).
Each ambulance is crewed by two registered practitioners licensed by independent statutory agency the Pre-Hospital Emergency Care Council (PHECC) at various clinical levels (emergency medical technician, paramedic and advanced paramedic) (PHECC, 2008).
In this paper, an escort is either a family member who was invited to accompany the patient for moral and/or emotional support, or a non-NAS medical professional with knowledge of and/or responsibility for the patient's clinical needs.
Method
A convenience sample of 96 respondents from 86 ambulance journeys was surveyed for 77 consecutive hours between 09:00 on 2 March and 14:00 on 5 March 2020, in the ambulance triage area of the University Hospital Limerick (UHL) ED. It had been intended to obtain 100 respondents but, because of the outbreak of the COVID-19 in the area, the fieldwork was stopped and a 96% recruitment rate was achieved.
All NAS passengers were informed of the survey and invited to consent to participate. Emotionally and/or clinically distressed patients and prisoners were excluded.
Participants' basic demographics were recorded, and their age distributed to the following cohorts: children aged 5–12, teenagers aged 13–17, young adults aged 18–40, middle-aged people aged 41–64, seniors aged 65–74 and elderly aged ≥75 years. A visual evaluation of the participants' clothing (London-Loughborough EPSRC Centre for Doctoral Research, 2011) and supplementary insulation as well as patients' acute complaint category (medical/trauma/both) was performed by the research team.
Participants' level of thermal comfort/discomfort was expressed according to the American Society of Heating, Refrigeration and Air-Conditioning Engineers Standards (ASHRAE) seven-point model of thermal comfort (International Standards Office, 2005) at the beginning and at the end of the ambulance journey. This Likert scale ranges from +3, representing hot/baking to −3, representing cold/freezing, with 0 representing the most comfortable temperature and a further differentiation of ‘neutral/comfortable/I did not notice’.
Ambulance practitioners were asked for the length of the journey, to identify their vehicle's patient compartment heater control type and state whether the heater had been turned on. During this survey, the fleet of NAS ambulances' patient compartments were fitted with two different types of thermostat control, independent from the front cab. One was analogue (low to high heat) (Webasto, Stockdorf, Germany) (Figure 1a). The other, a digital model (Eberspächer Climate Control Systems, Esslingen, Germany) allows users to choose a targeted setting and it automatically combines the heater with the air conditioning system (Figure 1b). A secondary on/off switch, but not a full control, was located in the front cab.


Overall, this project endeavoured to determine the passengers' levels of thermal comfort and if this aspect of service quality needed to be improved. Studying patients and escorts may bring to light differences in thermal perception or needs between healthy and ill passengers.
During the period, the NAS transported 163 patients to the target ED, giving a response rate of 58.9%. Notably, the number of escorts is not routinely recorded by the NAS, so a response rate for them could not be calculated. The reasons for non-participation were as per study protocol (no consent, not physically or emotionally stable or being a prisoner) but were not specifically recorded.
The data were electronically gathered with Google Forms (see Appendix), and automatically transferred to a Microsoft Excel spreadsheet for analysis. All percentages were rounded to the nearest first decimal.
The air temperature data for the nearest weather station (Shannon Airport) was retrieved from the website of the national weather service, Met Éireann's (www.met.ie).
Approval from both the NAS and the UHL research and ethics committees was received.
Results
The sample was composed of 50 women (52.1%) and 45 (46.9%) men, with one (1%) who declined to disclose their gender. The sample comprised of 22 (23%) escorts and 74 (77%) patients.
Of the patients, 57 (77%) had a medical illness as a reason for the visit to ED. The remaining 16 (21.6%) sustained a traumatic injury and one (1.4%) a combination of both trauma and medical complications. Forty-one patients (55.4%) were aged above 65 years, including 26 (27.1%) aged over 75 years. Escorts were most likely to be female and aged 18–64 years (Table 1).
Respondents | Patients | Escorts n (%) | Total n=96 (100%) | ||
---|---|---|---|---|---|
Medical n (%) | Trauma n (%) | Both n (%) | |||
Age | |||||
5–12 | 0 | ||||
13–17 | 1 (1.0) | 1 (1.0) | 2 (2.1) | ||
18–40 | 10 (10.4) | 5 (5.2) | 10 (10.4) | 25 (26.0) | |
41–64 | 13 (13.5) | 3 (3.1) | 9 (9.3) | 25 (26.0) | |
65–74 | 12 (12.5) | 3 (3.1) | 1 (1.0) | 16 (16.7) | |
≥75 | 21 (21.9) | 4 (4.1) | 1 (1.0) | 2 (2.1) | 28 (29.1) |
Gender | |||||
Female | 27 (28) | 7 (7.3) | 1 (1.0) | 15 (15.6) | 50 (52.1) |
Male | 29 (30.2) | 9 (9.3) | 7 (7.3) | 45 (46.9) | |
Prefer not to say | 1 (1.0) | 1 (1.0) |
Of the respondents, 64 (66.6%) rated their thermal comfort at 0 (neutral/comfortable/didn't notice) at the beginning of their journey and 65 (67.7%) rated 0 at the end of their journey.
Two (2.1%) escorts and 15 (15.7%) patients reported finding the patient compartment cold (−1 (n=13); −2 (n=3); and −3 (n=1)) at the beginning of their journey, and four patients and one escort still felt cold at the end of the journey. One man in the ≥75 years age group felt colder at the end than at the beginning of his transfer (0 at the beginning; −1 at the end) (Table 2).
Feeling cold | +3 n (%) | +2 n (%) | +1 n (%) | 0 n (%) | −1 n (%) | −2 n (%) | −3 n (%) |
---|---|---|---|---|---|---|---|
Beginning | Patient | 12 (12.5) | 3 (3.1) | ||||
Escort | 1 (1.0) | 1 (1.0) | |||||
End | Patient | 3 (3.1) | 3 (3.1) | 5 (5.2) | 4 (4.1) | ||
Escort | 1 (1.0) | 1 (1.0) | 1 (1.0) |
In contrast, three (3.1) escorts and 12 (12.5%) patients (medical n=6; trauma n=6) found the patient compartment slightly warm (+1), warm (+2) or hot/baking (+3) (n=12, n=3 and n=1 respectively) at the beginning of the journey. One reported a drop of 1 point, three an increase of 1 point, one an increase of 2 points, and 11 reported no change at the end of the journey (Table 2 and Figure 2).

Patients were likely to be dressed in light or mid-season clothing, while escorts wore heavier clothing (Table 3).
Clothing | Naked / Underwear n (%) | Light clothing n (%) | Mid-season n (%) | Winter coat n (%) |
---|---|---|---|---|
Patient | 6 (8.1) | 31 (41.9) | 30 (40.5) | 7 (9.5) |
Escort | 0 | 7 (31.8) | 6 (27.3) | 9 (40.9) |
Patients were more likely to be covered with one blanket (n=44; 59.5%) or none (n=21; 28.4%) than with two blankets (n=8; 10.9%), three or more blankets with a foil blanket (n=1; 1.3%), a duvet (n=0) or the vacuum mattress (n=0).
There were 86 individual ambulance journeys. Transportation had a median time of 27.5 minutes (range 3–90 minutes)(Figure 3).

The patient compartment heater was switched on in 69 (80%) of journeys. There was no statistical difference in the number of passengers expressing comfort or discomfort between the analogue and the digital heater thermostat. However, a positive trend for passenger reporting to be warmer by the end of their journey was detected (Table 4; Figure 4).
Heater switched on 77 passengers | +3 n (%) | +2 n (%) | +1 n (%) | 0 n (%) | −1 n (%) | −2 n (%) | −3 n (%) | |
---|---|---|---|---|---|---|---|---|
Beginning | Analogue 26 | 1 (3.8) | 4 (15.4) | 17 (65.4) | 3 (11.5) | 1 (3.8) | ||
Digital 51 | 3 (5.8) | 7 (13.7) | 30 (58.8) | 8 (15.7) | 3 (5.8) | |||
End | Analogue 26 | 1 (3.8) | 3 (11.5) | 4 (15.4) | 17 (65.4) | 1 (3.8) | ||
Digital 51 | 2 (3.9) | 8 (15.7) | 7 (13.7) | 31 (60.8) | 2 (3.9) | 1 (2.0) |

Only eight patient-escort pairs and one double escort sharing the same ambulance journey were available to answer the survey. A smaller number (n=8) of those passengers expressed some level of discomfort (any rating other than 0). The sample was too small to run a t-test.
There were four expressions of extremes of discomfort:
No patient physiological data were recorded, and no escort revealed or appeared to be sick or having a fever.
The air temperature recorded at Shannon Airport weather station was −0.8°C at its lowest and 10.1°C at its highest with a mean of 3.9°C. This was significantly but not unusually lower than the 30-year long-term average mean of 6.9°C.
Discussion
The objective to this study was to explore the thermal comfort of non-distressed NAS passengers. Thermal comfort is defined by ASHRAE as ‘that condition of mind which expresses satisfaction with the thermal environment and is assessed by subjective evaluation’ (International Standards Office, 2005).
The ASHRAE standards are based on a body of evidence measured around the healthy adult working human. ASHRAE declares the limitations of the data regarding our population of ambulance patients (International Standards Office, 2005).
The feeling of comfort is subjective so it is difficult to objectively quantify this on a scale. Schweiker et al (2017) noted that the seven-point scale used in the present study did not take account of the psychological aspect of the subjects, that the relationship between heat and thermal comfort is not linear and that the notion of ‘comfortable’ varies widely from person to person. However, this is the scale predominantly used for the assessment of thermal sensation in the engineering industry.
Being cold has physiological and psychological effects on patients (Aléx et al, 2013a). Thermal comfort has been described as very important part of basic care for the sick and injured; it reduces pain and anxiety and improves overall patient satisfaction (Kober et al, 2001; Aléx et al, 2014).
There are six primary factors which must be addressed when defining conditions for thermal comfort. The first is the metabolic rate of the patient; it can be expected that the metabolism of a critically ill patient, particularly when anaesthesia has been administered, will be affected (Preiser et al, 2014; Cheshire, 2016). The remaining factors are; clothing and insulation (2); air temperature (3); radiant temperature (caused by direct sunlight radiating through the windows) or the stored energy in the ambulance's furniture and walls (4); air speed (5); and humidity (6) (Cheshire, 2016). Ambulance practitioners can influence their patient's thermal comfort with on-board equipment (factors 3, 4 and 5) or judicious blanketing (factor 2).
Two studies by Aléx et al (2013a; 2013b), which involved interviewing patients of a Swedish ambulance service, concluded that the vehicles were cold, the occupants felt uncomfortable and the patient was warming up the stretcher. In their literature review, Haverkamp et al (2018) found that all prehospital patients were at risk of becoming hypothermic with an incidence of between 13.3% and 43%, especially in the severely ill. The French HypoTraum study (Lapostolle et al, 2012) focused on accidental hypothermia in the civilian adult trauma patient using a large multi-centred observation survey. They determined that 14% of patients arrived at the ED with a temperature of <35°C. Watts et al (1999) reported a drop in trauma patients' core temperature even during warm weather.
Beyond comfort, hypothermia (defined as a core temperature below 35°C), particularly in the context of trauma, worsens the clinical picture. Vasoconstriction decreases the delivery of oxygen at distal cellular level, and cells therefore switch to anaerobic metabolism, which generates a higher level of lactate, creating metabolic acidosis and altering the blood clotting cascade. However, this can be reversed by rewarming (Haverkamp et al, 2018).
Neonates born out of hospital are at a high risk of developing hypothermia leading to metabolic deterioration and even death (Kumar et al, 2009; Lunze et al, 2013).
Lapostolle et al (2017) determined that the temperature in ambulances' patient compartments was a factor in the presence or absence of hypothermia in critically ill patients on arrival to the ED.
An important question concerns the desirable temperature required in an ambulance patient compartment. Current research on the thermoneutral zone (TNZ) (the temperature range where body temperature is maintained without a need for changes in normal metabolic rate) and the more elusive correlation with thermal comfort zone (TCZ) of humans suggests that a temperature between 17.5°C and 32.3°C will be comfortable for a healthy person (Kingma et al, 2014; Pallubinsky et al, 2019). However, the lower critical temperature is affected by a number of factors, such as BMI, age and disease and, of course, the amount of metabolic activity, clothing and additional insulation (International Standards Office, 2005; Kingma et al, 2014).

The Wilderness Medical Society's (Zafren et al, 2014) recommendations include that, whenever possible, the temperature inside a transportation vehicle should be raised to at least 24°C to prevent further heat loss.
For neonatal resuscitation (in particular pre-term), the American Heart Association (2015) recommends an environmental temperature of 22°C–25°C, among other techniques, to prevent hypothermia (Perlman et al, 2015).
The District of Columbia in the US published in 2010 a standard for climate control inside its ambulances according to the seasons and weather conditions (Department of Health, District of Colombia, 2010).
A survey in a Belgian hospital highlighted a very high patient satisfaction rate (95%) when the ward's temperature was maintained between 22°C and 24°C, except for in the neurology department (Verheyen, 2011).
The literature search carried out by the authors of the present study yielded no other temperature targets; the advice was generally to simply ‘warm up’ the patient compartment.
Lundgren et al (2011), in their secondary analysis, reported that, despite the ambulance heater's thermostat being set at 25°C, the ambient air temperature was recorded to be an average of 20°C (±3°C).
Short-term physiological adjustment to changed environmental conditions is reported to be achieved in 20–30 minutes in a healthy mobile person (Auliciems and Szokolay, 2007). However, patients with a head injury and/or induced coma and/or immobilisation are not able to rewarm themselves through muscular action and shivering, which can potentially lead to hypothermia (Lapostolle, 2012).
Another factor to consider in the prevention of hypothermia is ‘after-drop’. Described by Danzl and Huecker (2017) and Haverkamp et al (2018), this is a phenomenon where, in a warmer environment, the blood circulating from the core is cooled down at the peripheries and, in turn, reduces the core temperature. Various means of preventing and reversing this have been offered (Goheen et al, 1997; Lundgren et al, 2009; Danzl and Huecker, 2017; Haverkamp et al, 2018).
In vehicles with the digital heater control installed, despite this having an automatic function that regulates the temperature by warming up or cooling down the air, 10 passengers felt too warm at the end of the trip. Anecdotal observations suggest that, once the patient was on board the ambulance, practitioners were turning on the heater to the maximum setting with either control system. A review of practitioner training needs regarding thermal control strategies may be required. This is important for any prehospital care providers, and needs to be addressed in future ambulance designs.
In the design of space for human occupancy, it is recognised that people entering a space may not find the condition comfortable as the effects of previous exposure or activity may affect their perception of comfort for approximately 1 hour (International Standards Office, 2005).
Much is also still unknown about the seemingly complex relationship between the environmental temperature or TNZ and the patient's TCZ (Kingma et al, 2014; Pallubinsky et al, 2019).
The ambulance practitioners' decision whether to switch on the heater or not is subject to an ongoing study.
Limitations
This is a small population study in a single centre during a limited period of time.
Other studies have used objective assessment of thermal comfort by measuring by skin temperature and evaporative heat loss (sweat) sensors (Aléx et al, 2103a; Kingma et al, 2014; Pallubinsky et al, 2019), devices that were not available during this study.
No patients' physiological data, in particular body temperature, were recorded; fever could alter the sensation of warmth or cold.
Different biases are acknowledged. First, there could be a halo effect since the interviews were undertaken in presence of the NAS practitioners, which may have influenced the respondents to give an overall positive view of the service which they had just received. Second, the Hawthorne effect on the staff cannot be excluded as the survey was advertised to them. In addition, some of the ambulance crews conveyed multiple patients to this ED during the study time.
To mitigate for those biases, during the request for consent, research assistants explained and highlighted the quality improvement purpose of this exercise.
Conclusions
Our hypothesis that NAS ambulance passengers are cold during transport seems to be disproved. This study shows that during this relatively cold week in March 2020, when ambulance practitioners had actively switched on the patient compartment heaters, two-thirds of passengers rated their thermal comfort as comfortable, 15% warm beyond comfortable and only 5% were still feeling cold at the end of their transport. To validate those results, larger-scale studies during different climatic conditions are warranted.
With the view to determine the need to improve future patients' experience, other points of interest arose from this experiment, including: practitioners' appreciation of passenger's comfort; attitudes towards thermal comfort; strategies for the prevention of patient's body heat loss; and ambulance patient compartment temperature variations.