Table 2

Summaries of the studies’ designs and key stress outcome findings

Study	Study design and methods	Key findings
Clow and Fredhoi15	Studied self-reported stress and arousal, and salivary cortisol levels in a group of London city workers during a lunchtime visit to an art gallery. Measurements were taken before and after the 35–40 min gallery visit to explore pre-post intervention changes.	Self-reported stress and salivary cortisol levels both decreased over the intervention. There were no differences in arousal levels.
D'Cunha et al24	Evaluated the psychophysiological effects of attending the National Art Gallery of Australia Art and Dementia programme. People living with dementia attended the group-based, 6-week programme which involved viewing and discussing artworks, led by an art director. Measures of salivary cortisol and interleukin-6 were taken at baseline, at the end of the programme and 12 weeks later.	Waking salivary cortisol levels increased from baseline to postintervention, but decreased at follow-up. No changes in evening cortisol or interleukin-6 were observed. The ratio of waking to evening cortisol increased from baseline to postintervention indicating a more dynamic diurnal cortisol rhythm.
de Jong14	Three groups of participants (advanced art history students, advanced fine art students and laboratory workers as controls) viewed projections of 12 paintings considered to be ‘beautiful’ and 12 paintings considered to be ‘ugly’ in a random order while their heart rate, respiration rate and skin conductance was measured continuously.	The fine arts and art history students showed a greater change in skin conductance than the laboratory workers. Respiration and skin conductance were higher during the ‘beautiful’ paintings than the ‘ugly’ paintings in all groups. The fine arts students had faster heart rate during the ‘beautiful’ paintings compared with the ‘ugly’ paintings, however, for the other two groups, this result was reversed.
Eisen et al10	The third phase this study investigated which type of art was most effective in reducing stress in paediatric patients. On arrival to the hospital, patients were randomly allocated to one of three rooms; a room with a nature artwork, a room with an abstract artwork or a room with no artwork. Self-reported stress, blood pressure and respiratory rate were taken at baseline and after 2 hours of exposure to the artworks.	Overall, there were no significant differences between the groups on stress, blood pressure or respiration. However, subanalyses showed that significantly more males than females in the 8–10 age group were positively affected by the nature artwork, as demonstrated by decreased self-reported stress, blood pressure and respiratory rates.
Karnik et al21	Installed a diverse collection of artworks in the public spaces and clinic rooms of a hospital. Patients were retrospectively contacted with a survey which included evaluating whether the art installations changed their self-reported stress levels.	61% of the patients that reported seeing the artworks stated that the artworks somewhat or significantly reduced their stress levels.
Krauss et al16	Participants viewed six Flemish expressionism artworks in an art museum, while heart rate and skin conductance were continuously measured. Participants were randomly assigned to either receive descriptive information about the artworks (described the artwork in a declarative way) or elaborative information about the artworks (described the context and deeper meaning behind the artworks).	There were no significant differences in heart rate, heart rate variability or skin conductance between the two groups. However, in both groups heart rate was lower, and skin conductance and heart rate variability higher when viewing the artworks, compared with baseline.
Kweon et al22	Conducted an experiment investigating the effects of artwork posters on stress and anger levels in an office setting. Students were asked to complete a series of stress and anger provoking computer tasks in one of four different mock office conditions; an office with nature posters, abstract posters, both nature and abstract posters or no posters. Levels of self-reported stress were measured across the experiment.	Males had the highest stress levels in the office with no posters, and the lowest stress levels in the office with mixed art posters. On the other hand, females had the highest stress in the office with all abstract posters and the lowest levels in the office with all nature posters. However, these results were only significant for males and not females.
Law et al25	Conducted a pilot study to investigate whether nature artworks could improve recovery from a laboratory stressor. Participants were randomised to either view a 30 min digital slideshow of landscape artworks or digitally scrambled versions of these artworks after being exposed to a laboratory stressor. Saliva samples were taken at baseline, after the stressor, during the art viewing and after the art viewing to measure cortisol and alpha-amylase.	Salivary cortisol levels decreased more rapidly while viewing the scrambled images compared with the landscape artworks. There were no changes in alpha-amylase across the experiment or between groups.
Mastandrea et al17	Students visited an art museum and were randomly assigned to visit one of three art exhibitions for 5 min; a figurative art exhibition, a modern art exhibition or a museum office as a control condition. Blood pressure and heart rate were measured before and after the visit.	Systolic blood pressure decreased in all groups; however, this decrease was only significant in the figurative art group. Heart rate also decreased in all three groups, however, there was no significant differences between groups.
McCabe et al9	Evaluated the effects of the Open Window art intervention on stem-cell transplantation patients. The Open Window is a virtual window which is installed in a hospital room, where the patients can switch through nine art channels with different artworks. Patients were randomised to either a room with the Open Window or not. Self-reported distress was measured at admission, the day before transplant, 7 days after transplant, prior to discharge and 60 days, 100 days and 6 months post-transplant.	Results demonstrated no significant differences in levels of distress between the two groups at any of the time points.
Pearson et al23	Examined the impact of nature-themed window murals on physiological measures in paediatric patients. Paediatric patients were assigned to hospital rooms with either a fish-themed window mural, a tree-themed window mural or no window mural. Patients’ blood pressure and heart rate were taken retrospectively from the patients’ medical records.	Those patients with the window murals had significant improvements in heart rate and systolic blood pressure, with the tree-themed mural having the greatest effect.
Siri et al18	Examined the effects of viewing original physical artworks and their digital reproductions within a museum context. Cardiovascular variables were measured via ECG continuously in healthy volunteers while viewing two real abstract paintings and their digital reproductions.	Results showed that there was a significant difference in heart rate between viewing the two real paintings, but no difference was found between the digital reproductions, or between the real and digital reproductions. No differences in heart rate variability were found.
Tschacher et al19	Monitored the physiology of visitors to an art museum using an electronic sensor glove which recorded physiological data and locomotion activity while they viewed the artworks. Afterwards, they were asked to rate the aesthetic qualities of some of the artworks.	Heart rate variability increased while viewing artworks that were deemed beautiful, high quality and surprising/humorous. Skin conductance variability increased, and heart rate decreased while viewing more dominant artworks (artworks experienced as dominant and stimulating by the viewers).
Wikström et al20	Investigated whether visual stimulation could improve the health of elderly women living alone. The women were randomised to either an intervention or control group. The intervention group were shown a selection of pictures, including artworks, and asked to discuss them, whereas the control group discussed current events. Blood pressure was measured at baseline, immediately after the intervention and 4 months later.	The intervention group had significantly lower systolic blood pressure than the control group after the intervention and at follow-up.