32
Int. Journal of Business Science and Applied Management, Volume 11, Issue 1, 2016
Linking Environmental Sustainability and Healthcare: The
Effects of an Energy Saving Intervention in Two Hospitals
Danae Manika
School of Business and Management
Queen Mary University of London
Mile End Campus, Bancroft Building, London, E14NS, UK
Tel: +44 (0)207 882 6541
Email: d.manika@qmul.ac.uk
Diana Gregory-Smith
Birmingham Business School
University of Birmingham
Ash House, Edgbaston, Birmingham, B152TT, UK
Tel: +44 (0)121 414 3344
Email: d.gregory-smith@bham.ac.uk
Victoria K. Wells
Management School
University of Sheffield
Conduit Rd, Sheffield, S101FL, UK
Tel: +44 (0)114 222 3271
Email: victoria.wells@sheffield.ac.uk
Lee Comerford
Global Action Plan
9-13 Kean St, London, WC2B 4AY, UK
Tel: +44 (0)20 7420 4444
Email: lee.comerford@globalactionplan.org.uk
Lucy Aldrich-Smith
Global Action Plan
9-13 Kean St, London, WC2B 4AY, UK
Tel: +44 (0)20 7420 4444
Email: lucy.aldrich-smith@globalactionplan.org.uk
Abstract
Set in a real organisational setting, this study examines the challenges of implementing
environmentally sustainable behaviour in healthcare. It evaluates the success of a real energy saving
behaviour change intervention, based on social marketing principles, which targeted the employees of
two National Health Service (NHS) hospitals. It also explores the intervention benefits for three key
stakeholders: the organisation/hospitals, hospital employees and patients. A rich secondary dataset
containing actual workplace behaviour measures (collected via observations) and self-reported data
from employee interviews and patient questionnaires is used for this purpose. The intervention
encouraged three employee energy saving actions (called TLC actions): (1) Turn off machines, (2)
Lights out when not needed, and (3) Close doors when possible; which led to energy savings and
carbon reduction for the two hospitals. Hospital employees reported a greater level of work efficiency
as a result of engaging in TLC actions, which increased the ‘quiet time’ periods in both hospitals.
Indirectly, employees’ TLC actions also improved patients’ quality of sleep (which in turn is positively
associated with greater patient hospital experience satisfaction). These findings shed light on the
benefits of social marketing interventions targeting energy saving behaviour change for multiple
stakeholders in healthcare organisations. They also illustrate connections between environmental
sustainability and social and political pillars of corporate social responsibility. Additionally,
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
33
organisational culture was highlighted as a key challenge in changing practices. To encourage long-
term sustainable behaviour, this study recommends a pre-intervention assessment of infrastructure and
equipment, the communication of expected benefits to motivate higher involvement of employees, the
need for internal green champions and the dissemination of post-intervention feedback on various
energy saving and patient indicators.
Keywords: Environmental sustainability; Healthcare organisation; Energy saving intervention;
Hospital patient experience; Energy data; Corporate Social Responsibility
Int. Journal of Business Science and Applied Management / Business-and-Management.org
34
1 INTRODUCTION
Corporate Social Responsibility (CSR) has driven a number of organisational practices related to
sustainability and research on the adoption of environmental sustainability for businesses and its effects
has received increasing recent attention in academia (Cramer et al. 2006; Lueg et al. 2015; Walker et al.
2015). However there is considerable scope for further examination (Lo et al. 2012a; Young et al.
2013). Strategically, sustainability within CSR practices and organisations have been motivated by
reducing cost, increasing operational efficiencies, building competitive advantage and increasing
reputation, which can result in favourable consumer responses, attractiveness to investors, employee
engagement and commitment amongst many others (Lindgreen & Swaen 2010; Aguinis & Glavas
2012). While multiple sectors are engaged in sustainability, and motivated by any number or
combination of these strategies (Sharma & Sharma 2011), it is clear that one size does not fit all in
terms of sustainability practices (Manika et al. 2015) and what may work in one industry is not certain
to work in another. Indeed, sustainability practices may be problematic in certain industries due to their
particular features, products/services and nature of the industry.
Healthcare is a “business unlike other businesses” (McCurdy 2002: 532) and where sustainability
choices could be affected by its unique features such as service orientation, its status as a public/social
good and its environment with distinctive features, such as room layouts, sound level, lighting, and
temperature (Leino-Kilpi et al. 2001). Additionally, the strategic focus and main motivator for
sustainability and CSR practices within healthcare, especially in the case of the UK National Health
Service (NHS), is cost saving. In the NHS, this strategic focus has developed due to a ‘plague of
reorganisations focused on attempts to control resource consumption, the lack of financial resources
and increasing complexity and size (Tudor 2013). While the challenge and importance of sustainable
hospitals has been highlighted in the popular press (HFMA 2013; Hamilton 2008), the effects of
adopting eco-efficient initiatives in healthcare, has been researched very little (Siebenaller 2012).
Academic research on environmental sustainability in healthcare has focused on recycling and waste
management (Tudor et al. 2007; 2008; Tudor 2013), while energy saving in the workplace has received
academic attention mainly in other industries (Pérez-Lombard & Pout 2008). Therefore, the primary
objective of this study is to fill this research gap in the academic literature and evaluate the success of
an energy saving intervention in two NHS hospitals in the UK.
Most energy saving schemes within healthcare have been focusing on technical solutions, such as
low watt light bulbs, retrofit insulation, double glazing windows, and improving heating controls,
among others, to reduce energy consumption in buildings and associated costs (Morgenstern et al.
2016). However, such schemes may have potential negative consequences for patient care provision
(Wicks 2002) as they result in delays in daily operations, additional costs, and disruptions associated
with new infrastructure (Grose & Richardson 2013). Additionally, managers are reluctant to implement
them due to a lack of trust in their effectiveness and uncertainty about the impacts on the reputation of
their organisations (Neven et al. 2014). However, changes in user behaviour in non-domestic buildings
have been increasingly recognised in academia and practice as having potential for energy savings
(Banks et al. 2012; Jeffries & Rowloands-Rees 2013). Therefore, this paper examines a behaviour
change social marketing intervention encouraging energy saving actions among employees, which
could potentially help hospitals and the NHS become more environmentally sustainable, while also
reducing operational costs. Little is known to date about the effectiveness of such interventions
(Morgenstern et al. 2016).
This study is set in a real organisational setting and uses a real intervention (called TLC)
encouraging three employee energy saving actions: (1) Turn off machines, (2) Lights out when not
needed, and (3) Close doors when possible. Within hospitals, lighting usage accounts for the largest
percentage of energy consumption (36%), followed by the use of medical equipment (34%) (Saidur et
al. 2010) and in the NHS specifically, 22% of CO2 emissions are a result of energy usage in buildings
(Tudor 2013). Thus, energy saving actions such as, turning off machines, lights out when not in use,
and closing doors to stabilise temperature (i.e., TLC actions), in a healthcare setting can reduce carbon
footprint and associated costs with energy consumption.
The healthcare system however, includes various key stakeholders with diverse needs (Vallance
1996; Pouloudi 1997) and “for health care organisations, a significant ethical challenge is to determine
how to fulfil institutional responsibilities to patients, physicians and other health care
professionals….and the community” (Gallagher & Goodstein 2002: 433), while also reducing
operational costs (Siebenaller 2012). Desjardins (2010) notes that potential and existing connections
between environmental sustainability and social and political pillars of CSR (in this case patient
welfare and wellbeing) are worthy of attention, and provide a different strategic focus for healthcare
organisations than the current focus on cost savings. Therefore, the potential wider environmental
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
35
responsibility effect on activities and the integration between the pillars of CSR must be carefully
considered and understood (Enderle 2010). Aside from the direct benefits of energy saving behaviour
change social marketing interventions for the hospitals/NHS (i.e., energy savings and cost reduction),
such interventions encouraging TLC energy saving actions among hospital employees could also
directly benefit employees who engage in these actions. For example, TLC actions could result in noise
reduction, and increase quieter times within the hospitals, thus allowing employees to work more
efficiently and ultimately increase employee satisfaction with the workplace. TLC energy saving
actions that hospital employees engage in also have the potential to indirectly benefit patients. Aside
from the fact that there is a positive link between hospital employee satisfaction and patient experience
(Peltier et al. 2009), TLC actions themselves carried out by employees could improve patient
experience indicators such as quality of sleep due to a reduction of bright light disturbance (Lei et al.
2009) and as a result increase patient hospital satisfaction (Naidu 2009). Thus, a secondary objective of
this study is to explore the benefits of such energy saving behaviour change social marketing
interventions on three key healthcare stakeholders: the hospitals/NHS, hospital employees and patients.
A rich secondary dataset containing actual workplace behaviour measures (collected via
observations) and self-reported data from employee interviews and patient questionnaires, allow us to
explore these potential benefits of the TLC intervention for hospital employees and patients, going
beyond prior studies that focused on organisational benefits of environmentally-friendly initiatives.
To the authors’ knowledge, this is the first study that uses a social marketing approach to examine
an environmental intervention with healthcare. Additionally, this study goes beyond cost saving as a
strategic focus. Through this approach, the present research links the environmental and social
dimensions of CSR. Several practical recommendations are made regarding the implementation of
energy saving CSR initiatives and measures, reflecting national and global endeavours for reducing
carbon emissions (Gerstlberger et al. 2014), along with the consideration of organisational factors and
non-financial incentives needed to enhance employees’ engagement with energy saving behaviour.
2 LITERATURE REVIEW
2.1 Employee Environmental Behaviour, CSR and Social Marketing Interventions
While the environmental behaviour of households has been studied extensively, the
environmentally sustainable behaviour of employees within organisations, and the use of social
marketing campaigns/interventions delivered during working hours has been studied very little (Lo et
al. 2012a). However, current work in this area suggests that ‘one size does fit all(Manika et al. 2015)
and that each type of industry differs in their motivations for and potential consequences of an
intervention. The literature has also focused on a range of behaviours with waste
management/recycling being the most popular (Ludwig et al. 1998; Marans & Lee 1993; McDonald
2011). Moreover, studies outside the care-related industries have researched climate control (Lo et al.
2012b), computers, lighting and energy usage (Scherbaum et al. 2008; Carrico & Riemer 2011)
amongst others. However, caution should be exercised in assuming that the antecedents and
concomitants of any particular behaviour are the same or even similar (Vinning & Ebreo 2002; Steg &
Vlek 2009). For example, past analyses have highlighted that recycling is not strongly related to energy,
water conservation (Berger 1997) or household purchasing behaviour (Ebreo & Vinning 1994).
Studies on employee environmental behaviour have also focused on a wide range of antecedents
and barriers, both individual and organisational (Hoffman 1993) including: attitudes (Scherbaum et al.
2008; Young et al. 2013), support and incentives (Smith & O’Sullivan 2012; Young et al. 2013),
knowledge and awareness (Rothenberg 2003), norms (Carrico & Riemer 2011), self-efficacy (Smith &
O’Sullivan 2012), organisational commitment (Andersson et al. 2005), organisational focus (Tudor et
al. 2008) and the environmental behaviour of the organisation (Manika et al. 2015), amongst others.
While studies have taken place in a number of industry types such as general office environments
(Grensing-Pophal 1993), industrial and retail firms (Shippee & Gregory 1982), council/government
(Gregory-Smith et al. 2015), academia (Ludwig et al. 1998), tourism (Chou 2014) and even
comparisons across industries (Manika et al. 2015; Walker et al. 2015), there are very few studies on
organisational practices related to sustainability in healthcare.
To date two studies have focused on waste reduction and recycling sustainability practices within
the UK National Health System (NHS). Initially, Tudor et al. (2007) used self-reported (i.e. survey-
based) and actual behaviour measures (i.e. waste bin data) to assess sustainable waste practices in the
NHS. They found that employee environmental behaviour is complex and that waste management
beliefs and perceived benefits of recycling were significant predictors of waste bin practices, unlike
subjective norms, behavioural control and awareness. Aside from the fact that the Theory of Planned
Behaviour was not fully supported, Tudor’s et al. (2007) study only focused on one type of NHS
Int. Journal of Business Science and Applied Management / Business-and-Management.org
36
stakeholders (i.e. employees), which can be seen as a limitation within a healthcare context. Tudor et al.
(2008) further assessed sustainable waste practices in the NHS, using not only questionnaires and waste
bin analysis, but also participant observation and interviews. A major finding was that organisational
factors were found to drive employee behaviour, while they also act as barriers to behaviour change.
Particularly, organisational focus was a key predictor of behaviour as it impacted on attitudes and
beliefs of staff resulting in a high degree of apathy and a belief that sustainability issues were
secondary to the core work priorities. On the other hand, it was found that the strong bureaucratic
organisational structure and the low priority of sustainability played a significant part in this and that
organisational culture in terms of group dynamics, awareness and norms (unlike in their earlier work),
did predict behaviour. Based on the findings of both studies, a key lesson learned is that any policies
regarding sustainable behaviour in healthcare must address issues around the structure and culture of
the organisation as well as individual variables such as beliefs, attitudes and motivations.
Given that environmental behaviour and sustainability policies in healthcare have mainly focused
on waste management and cost saving (Tudor 2013), this paper contributes to limited prior research on
energy saving initiatives and specifically, behavioural social marketing interventions targeting hospital
employees; these have been studied very little (Morgenstern et al. 2016). Social marketing is an
approach to achieve and sustain behavioural goals on a range of social issues and provides a
mechanism for tackling social problems by encouraging people to adopt certain behaviours (Lee &
Kotler 2011). Social marketing interventions and campaigns have been used to encourage
environmental behaviour change (Kennedy 2010; McKenzie-Mohr 1994; McKenzie-Mohr et al. 2011).
Behaviour change social marketing interventions encouraging energy saving actions among hospital
employees is a strategy that does not need a new infrastructure and, without much disruption of daily
operations, could potentially help hospitals and the NHS become more sustainable and
environmentally-friendly, while also reducing operational costs.
2.2 Hospital Employees’ Perceptions of Energy Saving Interventions and Related Research
Questions
Beyond the energy saving benefits for the NHS and the hospitals, the perceived benefits of social
marketing interventions promoting behaviour change among hospital employees, should also be
explored. Healthcare is different to many industries as “healthcare is an extraordinarily people-centric
industry…the patient consumes services to his or her physical body, nearly all treatments and
procedures are administered by people” (Peltier et al. 2009: 2). In this way there are similarities with a
range of other service organisations, such as hotels and hospitality. Here employees are often the main
target for behaviour change interventions and CSR initiatives due to the close relationships between
employees and consumers (Chou 2014; Coles et al. 2011) and individual behaviour is often seen as
being at the centre of change processes (Arena & Chiaroni 2014). Hospital employees are, therefore,
key to the successful provision of healthcare services and, thus, healthcare organisations need to ensure
that they respond to medical staff’s suggestions and perceptions quickly to ensure quality of care
(Mwachofi et al. 2011). Peltier et al. (2009) also note that there is a positive link between hospital
employee satisfaction and patient experience. Therefore, employees’ perceptions of the energy saving
behaviour change social marketing intervention are vital within healthcare, not only for engaging in
energy saving actions and reducing carbon emissions and costs, but also for ensuring that patient
satisfaction with the hospital experience (i.e. quality of care) is not negatively affected as a result of
such initiatives. This has parallels with the suggestion that high quality service standards required in the
services industry are likely to be a key determinant of the uptake of energy saving behaviours (Wells et
al. 2016).
This study also responds to calls for further research on employees (Rupp et al. 2013; Akremi et al.
2015) by exploring hospital staff’s perceptions of such energy saving interventions in terms of their
perceived benefits for employees, patients and the organisation. Hospital employees, like any other
employees, are assumed to take notice of CSR actions (Rupp et al. 2013) but their reactions to CSR
initiatives are considered dependent on whether they perceive the initiative to be important to them or
not (Glavas & Godwin 2013). Promislo et al. (2012) also note that beliefs about ethics and social
responsibility, including CSR initiatives, can affect individual well-being.
Thus, based on the aforementioned literature and the focus on the TLC energy saving intervention
among hospital employees, the following research questions are explored:
RQ1: To what extent were hospital employees aware of the TLC energy saving intervention and
the energy saving actions that were encouraged?
RQ2: To what extent were hospital employees personally involved with the TLC energy saving
intervention?
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
37
RQ3: What were the perceived benefits of the TLC energy saving intervention for employees,
patients, and the organisation itself, from hospital employees’ perspective?
RQ4: What were the perceived challenges of the TLC energy saving intervention, from the
hospital employees’ perspective?
RQ5: To what extent did the hospital employees perceive the TLC intervention to be successful
and what were the perceived intervention outcomes?
These research questions reflect common stages used to assess the development and success of
social marketing interventions (Lee & Kotler 2011): awareness (of the intervention; its importance also
noted in Young’s et al. (2013) employee pro-environmental behaviour framework), interest
(engagement/involvement of the audience), perceptions of benefits and barriers to action (challenges),
and behaviour change.
Lastly, given that perceptions may be inaccurate (Akremi et al. 2015), this study also benefits
from measures of actual behaviour via observations and energy data contained within the secondary
dataset used in this paper. These can be regarded a superior method since past research has noted the
gap between self-reported and actual behaviour (Barker et al. 1994).
2.3 Could Energy Saving Actions Affect Hospital Patients’ Experience Indicators?
As noted in the introduction, hospital employees’ energy saving actions encouraged by
behavioural social marketing interventions like the one examined in this paper (i.e., Turn off machines,
Lights out when not needed, and Close doors when possible) have the potential to indirectly affect
patients’ hospital experience. For example, turning off lights when not needed could save actual energy,
as well as enhance patients’ quality of sleep due to a reduction of bright light disturbance (Lei et al.
2009). This research focuses on four patient experience indicators, potentially affected by energy
saving TLC actions: 1) quality of sleep, 2) level of privacy; 3) thermal comfort; and 4) overall
satisfaction with hospital experience. These indicators are commonly included in hospital patient
experience surveys used worldwide (CMS 2014; Jenkinson et al. 2002). Below relevant prior literature
on patient experience indicators and how these may be affected by TLC energy saving actions carried
out by employees is discussed and associated hypotheses are advanced.
Quality of sleep is important as sleep aids patients’ recovery and may affect patients’ mood,
memory and cognition (Lei et al. 2009). Hospital patients generally require more sleep due to their
health status (Lei et al. 2009). Among the potential factors, which may affect quality of sleep include:
noise from machines, night-time nursing, temperature, bright lights (Lei et al. 2009) and the presence
of other people (Pimentel-Souza et al. 1996). Patients in intensive care units especially are
significantly affected by sleep disturbances caused by both environmental and non-environmental
factors (Bihari et al. 2012) and specifically noise from phones and medical equipment alarms were
found to be key disturbing factors for sleep in this patient cohort. Bihari et al. (2012) note that sleep
disturbance is multifaceted, meaning it can vary from complete awakening to sleep fragmentation and
arousal, all of which can lead to poor sleep quality.
Given the limited control patients have over the hospital environment, they may experience loss of
privacy, which can also disrupt patients’ sleep patterns (Leino-Kilpi et al. 2001) (see Parrott et al. 1989
for a review). Lei et al. (2009) suggested future research should examine interventions that may
enhance quality of sleep, by minimising sleep disturbing factors. Our study fills this gap and focuses on
energy saving actions that hospital staff can take to reduce energy consumption, which may also
enhance quality of sleep, including privacy, through quieter times.
Thermal comfort, which influences the energy consumption of a building (Djongyang et al. 2010),
has received considerable attention in healthcare literature with studies focusing on environmental
parameters (i.e., indoor temperature, humidity in hospitals), and on thermal discomfort and sensation of
patients and staff (Khodakarami & Nasrollahi 2012). Patients with worse health expect a warmer
indoor environment, as this can help with the healing process (Hwang et al. 2007). Therefore, patients’
thermal sensation is affected by their health status (Verheyen et al. 2011). For people affected by
illnesses, the optimal temperature is normally higher than the one for healthy people (Hwang et al.
2007). Moreover, a comfortable thermal environment has been found to contribute to stabilization of
patients’ moods (Hwang et al. 2007).
Another factor that can affect patients’ thermal comfort is represented by the so-called
acclimatisation effects, which relate to the differences between home and hospital thermal levels, as
perceived by patients (Hwang et al. 2007). While the tendency might be for patients to counteract
discomfort from the indoor ward climate by adding clothing insulation (Hwang et al. 2007) this might
not always be possible in hospital environments and some patients might not take these adaptive steps
by themselves due to health issues, disabling conditions or lack of knowledge. The literature also points
Int. Journal of Business Science and Applied Management / Business-and-Management.org
38
out that the type of hospital rooms (single bed vs. multi-bed/bay) rooms and the number of beds in a
ward, which may differ from one hospital to another, may also affect thermal sensations of patients
(Yau et al. 2011). Additionally, the seasons and related temperature variations may affect the thermal
comfort of hospital patients (Hwang et al. 2007).
One of the key challenges to ensuring thermal comfort to hospital patients is the fact that
temperature settings need to take into account requirements for different hospital users (e.g. patients,
medical staff) who may have different needs in terms of what is considered a confortable environment
for them (Verheyen et al. 2011). Moreover, Verheyen et al. (2011) note the need to control for
temperature at room level and even to ensure individual adjusting that would take into consideration
each patient’s health and physical strength, where possible.
Therefore, the above literature highlights the need for improving thermal comfort, with closing
doors being one of the measures that can be taken to stabilise temperature. Closing doors was one of
the actions included within the TLC environmental intervention examined in this paper.
Lastly, overall patient satisfaction with the hospital experience can enhance hospital image and
benefit the healthcare provider’s long-term success (Naidu 2009). Patient satisfaction is an evaluation
of distinct healthcare dimensions (Linder-Pelz 1982), and is affected by many variables (see Naidu
(2009) for a review). Patient satisfaction is considered challenging to measure and explain due to being
a “multi-dimensional healthcare construct affected by many variables” Naidu (2009: 366). Privacy
(Silvestro 2005) and comfort (e.g. thermal comfort, sleeping comfort) (Naidu 2009) have been found to
affect significantly patients’ satisfaction. This is consistent with Butler’s et al. (1996) study that
concluded patients’ service quality perceptions are primarily affected by quality of the facility (e.g. the
hospital room, ward) and the staff performance. Both of these two factors are variables affecting the
quality of sleep, privacy and comfort of patients. Thus, we expect that patient satisfaction could
indirectly be affected by energy saving actions, through improvements in quality of sleep, privacy and
comfort of patients.
Based on the factors included and measured in the secondary dataset associated with the TLC
energy saving intervention examined in this paper, and the aforementioned literature review, we
hypothesise that:
H1: a) Patients’ perceptions of quality of sleep, b) privacy, c) thermal comfort and d) overall
satisfaction with the hospital experience will improve after the energy saving intervention, as a result of
hospital employees engaging in energy saving actions.
The literature review on patients’ hospital experience indicators also supports the following:
H2: Patients’ perceptions of a) quality of sleep, b) privacy, and c) thermal comfort will be
positively and significantly related to overall satisfaction with hospital experience.
H3: Patients’ perceptions of a) privacy and b) thermal comfort will be positively and significantly
related to perceptions of quality of sleep.
These hypotheses (H2-H3) are expected to hold both in pre and post-intervention data, even
though they have not been empirically tested before. Thus, to investigate this further we propose an
alternative hypothesis (i.e., H4) and test it via a multigroup SEM analysis with the intervention as the
grouping variable: Group 1: Pre-intervention & Group 2: Post-intervention.
H4: H2 to H3 will be moderated by the energy saving intervention.
This concludes the summary of prior literature, which has explored the benefits of an energy
saving behaviour change social marketing intervention for three key healthcare stakeholders: the
organisation/hospitals, hospital employees and patients. Next the methodology will be discussed.
3 METHODOLOGY
This paper uses an energy saving social marketing intervention conducted within a real (i.e. non-
laboratory) setting represented by two Barts Health Trust hospitals (part of the NHS). The data used in
this paper is drawn from a rich secondary dataset containing actual workplace behaviour measures
(collected via observations) and self-reported data from employee interviews and patient questionnaires,
which were used to explore the benefits of the intervention for the organisation, hospital employees,
and patients. The intervention was developed and carried out by Global Action Plan (GAP) as a leading
environmental charity, which also collected the secondary data analysed in this paper. Figure 1 gives an
overview of the TLC intervention, which is detailed subsequently, and the secondary data available
from GAP with the related timeline.
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
39
3.1 The TLC Intervention
The TLC intervention designed and delivered by GAP, encouraged three energy saving actions
among hospital employees: Turn off machines, Lights out when not needed, and Close doors when
possible. These actions were selected by GAP and were seen to potentially reduce the hospitals’ energy
consumption. It was also considered that these actions could easily be carried out by employees and
would imply minimal interference with medical treatments and hospital requirements. The intervention
was delivered via multiple communication platforms. Face to face discussions were carried out with
employees using electronic tablets as props to help hospital employees become familiar with energy
saving actions. Posters and stickers were placed on doors throughout the hospital, and pens and t-shirts
were distributed; as reminders of energy saving actions.
Besides being part of the same organisation (Barts Health Trust), which regulates aspects of the
NHS hospitals’ daily operations and infrastructure, both hospitals which received the TLC intervention
were located in London, each hospital had a minimum of six buildings, a capacity of more than 300
beds with both single and multi-bed (bay) rooms and had an Accident and Emergency Unit. Due to
these similarities, the two hospitals are used in this paper as one organisation and one sample for the
analysis. Both hospitals were simultaneously exposed to the same TLC intervention.
Figure 1: Timeline and Secondary Data Used
3.2 Secondary Data Used and Associated Analyses
The secondary data used in this paper were initially collected by GAP employees who used a
concurrent longitudinal mixed methods approach to gather data from different stakeholders. This
methodological approach allowed the triangulation of diverse perspectives on the benefits of the TLC
intervention (Foss & Ellefsen 2002), while being “more flexible, integrative, and holistic” (Powell et al.
2008: 306). As indicated in Figure 1, this secondary dataset included: energy data in aggregate form
and observations (i.e. lights turned on and doors left open unnecessarily) before and after the
intervention, employee self-reported data after the intervention, and patient self-reported data before
and after the intervention, which respectively shed light on the benefits of the TLC intervention for the
organisation, the hospital employees and the patients. Figure 1 also contains the timeline for the
particular data collection carried out by GAP.
This secondary data were not designed nor collected with an academic approach in mind. This has
restricted the analyses and findings reported here. However, in addition to previously highlighted
contributions, this organisation-situated intervention overcomes key weaknesses related to laboratory
academic research (i.e. lack of realism, artificiality, and generalisability; Levitt & List 2007). Moreover,
Int. Journal of Business Science and Applied Management / Business-and-Management.org
40
actual workplace behaviour measures (i.e. observations of energy saving actions and energy data)
contained in this secondary dataset enhance the contribution of this study. Below, we provide
additional information on how the specific data used were collected by GAP and how we analysed the
data in connection to the listed research questions and the proposed hypotheses.
3.2.1 Using energy data and observations to examine the impact of the intervention on
energy consumption
To achieve the primary objective of this paper (i.e., RQ1 to evaluate the success of an energy
saving social marketing intervention in a healthcare setting), energy data in aggregate form and
observational data (i.e., lights turned on and doors left open unnecessarily) before and after the TLC
intervention, collected by GAP, were used. Energy data in aggregate form serves as a measurement of
actual environmental workplace behaviour to examine whether or not the intervention was successful
in reducing energy consumption. Such measurements improve the study’s reliability, given the discord
between self-reported and actual measures of behaviour as noted in past environmental research (Chao
& Lam 2009; Huffman et al. 2014), and help reduce the issue of common method variance in cross-
sectional survey research (Rindfleisch et al. 2008). Given the longitudinal nature of the secondary
dataset, our findings overcome sources of common method biases, such as measurement context effects
(Podsakoff et al. 2003). Thus, the energy data used in this study provides a distinctive contribution to
the paper and allowed us to calculate the energy savings as a result of the intervention and associated
cost savings.
In addition, observational data of employees’ actual environmental behaviour: 1) doors left
unnecessarily open and 2) lights left unnecessarily switched on, were collected by trained GAP staff
pre and post-intervention at several times during the day and night, at approximately the same time,
each day/night, to ensure consistency and comparability. This data was used in this paper to examine
the success of the TLC intervention in changing employees’ energy saving behaviour (in addition to
subsequent analyses dividing wards of hospitals in low and high energy saving adopters).
3.2.2 Using employee data from post-intervention interviews to explore hospital employees’
perceptions of the TLC intervention (RQ1-RQ5)
A total of 14 interviews with employees were collected after the intervention, by GAP, which
contained information regarding the level of awareness of and involvement with the intervention and
the perceived benefits of the intervention, as well as recommendations about future interventions. Thus,
this employee data were appropriate for the investigation of RQ1 to RQ5 regarding employees
perceptions of the benefits of the energy saving intervention for employees, patients and the
organisation.
The employee data included 4 male and 10 female participants (representative of the fact that 10.1
times more women work as nursing and midwifery professionals than men in Europe and the US
(OECD 2006). The interviewees had various roles such as: ward manager, healthcare support officer,
nurse, discharge coordinator, housekeeper, education centre coordinator, and office manager clinical
lead. Their age ranged between 23 and 60 years old and working experience within the hospital varied
from 2 to 23 years. This cohor provided an adequate representation of hospital employees. The
interviews were recorded using the Recordium iPad app by GAP and carried out as a short intercept
interviews due to the busy nature of the wards and employees’ job tasks (a method increasingly used in
health-related research; Tse et al. 2014).
The academic team transcribed and coded the recordings of the interviews using a semi-inductive
approach, a common approach in health-related research (e.g. Wells et al. 2004; Fortin et al. 2010),
with some themes related to the research questions (RQ1 to RQ5) and other new themes also emerging
from the data (Thomas 2006).
3.2.3 Using hospital patient data from pre and post-intervention questionnaires to explore
the indirect benefits of TLC actions on patient experience indicators (H1-H4)
To examine the indirect benefits of TLC actions carried out by hospital employees, as a result of
the energy saving intervention, on patient experience indicators as per hypotheses H1 to H4, the
academic team used the pre and post-intervention patient data collected by GAP via questionnaires.
The questionnaires examined patient experience indicators, which could be affected by employees’
TLC actions and thus are appropriate for examining H1 to H4.
The pre-intervention questionnaire included 70 hospital patients (Hospital 1: n=30; Hospital 2:
n=40) and the post-intervention questionnaire included 88 hospital patients (Hospital 1: n=29; Hospital
2: n=59). All questionnaires were administered in paper and pencil format by GAP staff. Some were
completed by patients and others completed with the help of the charity’s representatives, when
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
41
assistance was needed. Verbal consent was given and questionnaires were filled in anonymously;
ensuring compliance with ethical procedures, increasing individuals’ participation and reduction of
social desirability bias (c.f. Richman et al. 1999). Different patients were used for the pre and post-
intervention data collection (see Table 1).
Table 1: Patient Sample Demographics and Nights in the Hospital
Pre-Intervention
Sample
Post-Intervention
Sample
Frequency
Frequency
Percentage
Gender
(N=66)
(n=85)
Male
23
41
48.2%
Female
43
44
51.8%
(N=67)
(n=87)
Age
<18
3
4
4.6%
18-25
5
6
6.9%
26-35
13
6
6.9%
36-45
7
9
10.3%
46-55
8
10
11.5%
56-65
11
15
17.2%
66-75
11
16
18.4%
76+
9
21
24.2%
Nights In
Hospital
(N=65)
(N=86)
1-5 nights
37
40
46.6%
6-10 nights
8
21
24.4%
More than 10 nights
20
25
29.0%
Even though the patients before and after the intervention were not the same, which poses some
limitations, the use of distinct samples before and after a pro-environmental intervention has been used
before to examine its effects (e.g. Gregory-Smith et al. 2015) and is acceptable under certain conditions.
Given the hospital setting where this secondary data came from, having different hospital patient
participants with the same characteristics before and after an intervention is acceptable, since there is a
quick turnaround time in hospital admissions and discharge after treatment. In addition, the
intervention was aimed at hospital employees not patients, which limits some of the obstacles and
limitations of not having matched samples before and after the intervention. As noted, we expect that
the TLC actions themselves, not the intervention, would indirectly affect patient experience indicators
as per H1.
To ensure that potential differences in patient experience indicators before and after the
intervention were not due to the influence of patients’ individual/demographic variables, it was
important to demonstrate that the two groups were comparable (Rubin & Babbie 2011; Gregory-Smith
et al. 2015) in terms of age, gender and number of nights in the hospital. No significant differences
were found between the patients that completed the pre-intervention questionnaire and the those that
completed the post-intervention questionnaire in terms of gender (χ2(1)=2.73, p>.05), age
(F(1,152)=3.42, p>.05), and nights in the hospital (F(1,149)=.36, p>.05). These results show that the
patients before and after the intervention had similar characteristics, and thus could be used to examine
H1 to H4.
Both pre- and post-intervention questionnaires contained the same questions (continuous variables
measured on a 5-point Likert scale - see Tables 2 and 3). The hospital name, ward and room type
(single or bay/multi-bed room) was also recorded. All multi-item scales included in the questionnaires
had a Cronbach’s Alpha above .78, signifying reliability. Because the questionnaires were designed by
the charity, not all the variables were measured as multi-item scales. This approach is increasingly
accepted in the academic literature and appropriate under certain conditions such as experiments in
organisations (see Manika et al. 2015) and in service intensive industries front line employees will have
little time away from their role and hence shorter questionnaires are often the only option. Composite
mean scores were calculated by the academic team for the multi-item scales. The dataset also contained
information on whether or not patients talked to hospital employees about quality of sleep, thermal
comfort and privacy.
Int. Journal of Business Science and Applied Management / Business-and-Management.org
42
Table 2: Variables, Measures and Cronbach’s Alpha
Pre-Intervention Sample
Post-Intervention Sample
Variables
N
M(SD)
Composite
Descriptives
&Cronbach
Alpha
N
M(SD)
Composite
Descriptives
&Cronbach
Alpha
Quality of Sleep
Composite
On the scale of one (1-Extremely disturbing) to five (5-Not at all disturbing), please select the number that best describes
the level of disturbance you experienced, during the night whilst visiting the hospital from the following:
Noise from machines
65
3.69 (1.22)
a=.79
N=61
M=3.78
SD=.90
81
3.86 (1.11)
a=.79
N=68
M=4.14
SD=.71
Noise from outside your room
63
3.97 (1.10)
73
4.18 (.91)
Noise from fellow patients
69
3.26 (1.44)
77
3.78 (1.26)
Noise from employees at night
67
3.79 (1.27)
75
4.12 (1.01)
Light from the corridor
65
4.17 (1.15)
75
4.31 (.94)
Regarding your level of comfort due to room temperature levels, on a scale of one (1-Strongly disagree) to five (5-
Strongly agree), please indicate your level of agreement with the statements below:
Thermal Comfort
The room temperature made me feel
warm enough
65
3.69 (1.25)
n/a
76
3.78 (1.01)
n/a
Perceived Privacy
Composite
On a scale of one (1-No privacy at all) to five (5-A lot of privacy), please select the number that best describes the level
of privacy you experienced during the following times.
During discussions with doctors
there was...
54
3.89 (1.21)
a=.92
N=49
M=3.93
SD=1.04
71
4.14 (1.03)
a=.89
N=65
M=4.09
SD=.87
During examinations there was...
55
4.38 (.91)
72
4.49 (.75)
During personal time during the day
there was...
53
3.77 (1.32)
74
3.85 (1.06)
During personal time during the
night there was...
51
3.90 (1.27)
71
4.17 (1.12)
During visiting time there was...
51
3.71 (1.35)
71
3.82 (1.10)
Satisfaction with
Hospital
Experience
Composite
On the scale of one (1-Strongly disagree) to five (5-Strongly agree), please select the number that best describes your
level of agreement with the following statements, related to privacy, quality of sleep and room temperatures.
I am satisfied with the service
provided during my stay at the
hospital
55
4.07 (1.03)
a=.92
N=48
M=3.87
SD=1.05
74
3.95 (1.10)
a=.94
N=66
M=3.84
SD=1.08
My expectations have been met
55
3.87 (1.09)
72
3.89 (1.10)
Compared with other hospitals, the
level of satisfaction was high
50
3.66 (1.17)
66
3.71 (1.16)
All KaiserMeyerOlkin values for each multi-item scale were between .5 and 1, indicating the
appropriateness of the Exploratory Factor Analyses. Bartletts tests of sphericity were significant (p
.001) with changes in eigenvalues, indicating a one-factor solution for each scale. Factor loadings
were significant and close to each other, therefore, all multi-item scales were reliable and valid, for
both the pre and post-intervention data. Composite scores of the latent variables quality of sleep,
perceived privacy and satisfaction with the hospital experience were then used for all sub-sequent
analyses to examine H1 to H4.
Table 3: Patients who Talked to Hospital Employees about Quality of Sleep, Room Temperature
and Privacy
Pre-Intervention Sample
Post-Intervention Sample
N
Frequency
Percentage
N
Frequency
Percentage
Did you ask any employee(s) for any changes (e.g.,
medications, extra pillows, changing the bed
position) to help increase the quality of your
sleep?
Yes
65
34
52.3%
85
38
44.7%
No
31
47.7%
47
55.3%
Did you ask any employee(s) for any changes (e.g.,
extra blankets, turn up or down heating) to help
increase the quality of your thermal comfort?
Yes
67
23
34.3%
75
21
28.0%
No
44
65.7%
54
72.0%
Did you talk to any employee(s) about any privacy
concerns that you experienced during your visit?
Yes
54
3
5.5%
71
1
1.4%
No
51
94.5%
70
98.6%
H1 was examined via chi-squares and t-tests computed on SPSS 22. It should be noted that data
for single rooms and bay/multi-bed rooms were explored separately given that closing doors (C) is not
permitted in bay/multi-bed rooms. H2 and H3, were examined using a structural equation modelling
approach (SEM) (Mplus 7 software) with observed variables (i.e., the composite scores of the latent
variables) as per Manika et al. (2015) before and after the intervention separately. This was done to
explore how patient experience indicators relate to overall patient satisfaction, and which indicator is
the most important predictor of satisfaction with hospital experience. H4 was then examined using the
combined pre and post-intervention patient data and a multi-group SEM analysis to test if relationships
Danae Manika, Diana Gregory-Smith, Victoria K. Wells, Lee Comerford and Lucy Aldrich-Smith
43
between patient experience indicators (H2 and H3) vary before and after the intervention (H4). The
overall SEM model examining H2 to H4 is depicted in Figure 2. The aforementioned analyses
controlled for demographics and number of nights in the hospital (given that socio-demographic factors
may affect patient experience indicators Haiyan et al. 2011).
Figure 2: SEM Model Testing H2 To H4
4 RESULTS AND DISCUSSION
As discussed in the literature review and the methodology the benefits of an energy saving
behaviour change social marketing intervention are examined for three healthcare stakeholder groups:
the organisation/hospitals, the employees and the patients. This investigation allows us to examine the
links between the environmental sustainability and social and political pillars of corporate social
responsibility. Each section below reports and discusses the results based on the secondary dataset used
in this paper and the TLC intervention examined. The results are organised in terms of the three
stakeholders: the organisation/hospitals, hospital employees and patients, respectively.
4.1 Benefits of the Energy Saving Intervention for the Hospitals/Organisation based on
Observations and Energy Data
Observation data of employees’ actual environmental behaviour, as provided by the charity, are
summarised in Figure 3. After calculating the total number of doors and lights left open and switched
on unnecessarily in each ward after the intervention, this number was divided by the total number of
doors and lights observed in hospital wards, respectively. This led to the calculation of doors and lights
performance indicators, which were subsequently averaged to create a combined indicator of the
energy saving actions adoption rate for each hospital ward. The combined indicator ranged from .16
(highest performance) to .50 (lowest performer), while the average was .33; illustrating variability in
wards’ energy saving adoption rates. This result infers that the success of the intervention in motivating
employee energy saving actions varied by ward.
The secondary dataset also included a measure of actual environmental workplace behaviour
(energy data), based on calculations by GAP staff pre- and post-intervention. Table 4 shows estimated
energy savings of 764,820 KWh. According to the Energy Saving Trust (2014) this is equivalent to
£103,403.66 and 367.11 tCO2 (based on an average rate of 13.52 pence/kWh of electricity and 0.48
kgCO2/kWh). These savings provide some evidence of the success of the intervention. However,
caution should be shown when interpreting these results because they were based on “on the spot”
observations (rather than using data collected by energy meters) and because of the limited ability of
this type of measurement to control for other factors that influenced employees’ behaviours.
Nevertheless, this proxy measure of actual behaviour, along with the observational measures,
strengthens the contribution of this research and supports the success of the TLC intervention in
lowering energy consumption and associated costs for the two hospitals.