Mixed methods evaluation of well-being benefits derived from a heritage-in-health intervention with hospital patients

Background This study sought to determine the effects of a heritage-in-health intervention on well-being. Benefits of arts-in-health interventions are relatively well-documented yet little robust research has been conducted using heritage-in-health interventions, such as those involving museum objects. Methods Hospital patients (n = 57) participated in semi-structured, 30–40 minute facilitated interview sessions, discussing and handling museum objects comprising selections of six artefacts and specimens loaned from archaeology, art, geology and natural history collections. Well-being measures (Positive Affect Negative Affect Scale, Visual Analogue Scales) evaluated the sessions while inductive and deductive thematic analysis investigated psycho-educational features accounting for changes. Results Comparison of pre- and post-session quantitative measures showed significant increases in well-being and happiness. Qualitative investigation revealed thinking and meaning-making opportunities for participants engaged with objects. Conclusions Heritage-in-health sessions enhanced positive mood and social interaction, endorsing the need for provision of well-being-related museum and gallery activities for socially excluded or vulnerable healthcare audiences.


Background
Arts-in-health interventions have received substantial attention in recent years (Cox et al., 2010;Clift et al., 2009;Sonke, Rollins,Brandman &Graham-Pole, 2009;Staricoff, 2004Staricoff, , 2006Wreford, 2010)and encompass awidevariety of cultural activities aiming to enhance individual and community welfare, healthcare deliveryand healthcare environments. Arts CouncilEngland (2007,pp. 5-6),found a"considerableand growingevidence base of the effectiveness of arts interventions in healthcare and in promoting well-being" that included improving the "mental, emotional and spiritual state of Health Service users" and "help medicalstaff, caregivers,patientsand families to communicate more effectively with each other by offering opportunities for social interaction, involvement and empowerment". It is increasingly accepted that health, well-being and quality of life are reliant upon interconnections between physical, psychological and social functioning. Although this view is broadly in keeping with the World Health Organization (1948) definition; "healthis acomplete state of physical, mental and social well-being, not merely an absence of disease q 2013 The Author(s). Published by Taylor &F rancis. This is an Open Access article distributedu ndert he termso ft he Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.T he moral rights of the named author(s) have been asserted. or infirmity", the word" complete" can be debated in the light of an ageing population, better screening techniques and an increase in the diagnosis of chronic disease. The WHO definition underestimates the humanc apacity to adapt to physical, emotional and social change,and to experience well-being in the presence of disease or disability. The definition of well-being is even more ambiguous and controversial (Carlisle &Hanlon, 2007). The New Economics Foundation( aU KG overnment think-tank)d efine well-being as "the dynamic process that givesp eopleas ense of how their lives are going, through the interaction between their circumstances,activities and psychological resources or 'mental capital'" (NEF,2 009,p .3 ). This definition is adopted here sincem useumsa nd galleries appeartoidentify with NEF's view of well-being (Chatterjee &Noble, 2013)byendorsing NEF'sfive "actions" to improve well-being in everyday life: "be active", "connect", "keep learning", "give" and "take notice" (NEF, 2008).

Theoretical Framework
The multi-disciplinary studyr eported here used am ixed-methods approach to assess the well-being benefitso fh andling and discussing museum objects with ar ange of hospital patients. Thestudy sought to quantify the effectsofaheritage-in-health intervention using clinically accepteds cales derived from psychological and medical practice. It used qualitative thematic analysis of audiorecordings to provide in-depthunderstandingofthe processesi nvolved in engaging with heritage objects and to see if and how this engagement could lift mood and enhancewell-being.Theoretical bases drawn from artsin-health,psychology and educational research provided aconceptual framework for the study. Simmons (2006,p p. 2-4) suggested that arts-in-health practices are grounded in two theoretical approaches,"dual coding" (Baddeley &Hitch, 1974;Paivio, 1971Paivio, ,1986 and the "contiguity effect" ( Clark &P aivio, 1991). Both theories were derived from psychological research into memoryand are seen as reliant upon the interaction between sensory modalities. WithinP aivio's( 1971)c oncept of dual coding, verbal and visual Arts &H ealth materialisconnected in ashort-term storeor"working memory" during encoding and are subsequently integrated with materialr etrieved from long-term memory (Paivio,1 986). Within the contiguityeffect,performanceisenhanced when verbaland visual material is coordinated, not presented separately, ap rocess attributed to the formation of better connections in the brain (Clark &P aivio, 1991).
Similarly, Baddeley (1986Baddeley ( , 1992h ypothesized at wo-component workingm emory store comprising auditory memoryor"echoic store" and visual memoryor"iconic store". Each storeh as limited capacity; hence,a ne ncoding strategy that draws simultaneously upon both stores should demonstrate a"modality effect"inreducing the cognitive load of one and exploiting available capacity in the other. As an alternative to the concept of memorys tores, Craik and Lockhart( 1972)a dvocated a" levels of processing" approach where "deeper" encoding of information leads to the formation of moreconnections in the brain than "shallower" encoding. Their two-stagem odel consisted of "maintenance rehearsal", where material is retained only long enough to use it and "elaborative rehearsal", where material is processed more deeply for subsequent retrieval from memory. Given that many arts-in-health interventions combine conversation with visual exploration, they may draw upon the modality effect in tapping into available capacity.
In addition to hearing and vision, heritage-in-health interventions involve sensessuch as touch and smell, theoretically implicating amodel of multiplecoding rather than dual coding. Amultiple coding concept is relevantfor olderhealthcare recipientswith sensory decline (e.g. stroke, macular degeneration, etc.) because if oneo rm ore of the sensesi s compromised, it may be important to maximize communicationand social contact during the session through otheravailable cognitive channels. Spector, Orrell and Woods (2008) showed that twice-weekly cognitive stimulation therapy (CST)w ith older adults diagnosed with early stage dementia living in residential care led to increases in two measures of cognition (Mini Mental State Examination and Alzheimer'sD isease Assessment Scale -C ognition) and in aparticipant-rated qualityoflife measure (Quality of Life Scale in Alzheimer's Disease), when compared with no treatment. CST employed sensory stimulation using avariety of stimuli such as social historyobjects and sound to prompt discussion and reminiscence.I ts authors believe CST impactsu pon cognitive processing and neuronal growth by tapping into multipleb iological, psychological and social factors.
Educational researchinto stimulating and integrating sensory modalities,particularly the deployment of VAK (visual, auralorkinaesthetic/tactile) preferences associated with the Montessori Method, amethod of educating children that stresses the development of initiativea nd natural ability ( Kilpatrick, 1914) demonstratedw ider appeal and assimilation of learning from the multiple presentation of material. Educational theory suggests learning is acognitive process by which skill or knowledge is acquiredassociated with behavioral change and positive effects on mood (Uljens, 1997). Furthermore, Hein (1999)s uggests that learnersa re active rather than passive in their acquisitiono f information. Constructivist theories, originating from the early twentieth century workof Piaget (Piaget, Eames &Brown, 1982)and Vygotsky (Vygotsky &Kozulin, 1986)suggest that children learn by building upon knowledge already acquired through their prior experience of the world (Pass,2004). Although these educational theories are associated with childhooddevelopment and acquisition of knowledge, the more recent emphasis on lifelong learning has necessitated reference to constructivist frameworks and models of adult learning (e.g. Falk &D ierking, 2000;H ein, 1999)b ased on theories originally conceptualised for younger learners. Field (2009) Brown and Wragg (1993) determined that the main reasons for asking questions were cognitive, affective and social, and dealing with knowledge, feelings and relationships. Therole of materialobjects in "meaning-making" is relevant to the study. Rowe (2002) attributes this to their ability to function as a" mental representationo fp ossible relationships among things, events, and relationships. Humans bring their own knowledge, experiences and values to objects and makemeaning" (Baumeister, 1991,p15). Material objects alsoelicitasense of identity and play arole in the development of self-awareness through multisensory interaction (Vygotsky, 1933(Vygotsky, [2002;s ee Camic, 2010;R omano, McCay &Boydell, 2012). The workofCamic, Brooker and Neal (2011)on"foundobjects" (referred to as material objects that are found or discovered, are notusually purchased,hold no intrinsicfi nancial value and have personal significance) showed that the use of such objects in psychotherapy helped to enhance engagement, increasecuriosity, reducedifficult feelings, evoke memories and provide asense of agencythroughincreased physical activity and environmental action. Furthermore, several authors have suggested that museum objects triggerm emories, ideas and emotions in ways that otheri nformation-bearing materials do not (Chatterjee &N oble, 2013;K avanagh, 2000;L anceley et al., 2012;Phillips, 2008). Pearce (1995)arguedthat museum objects function as symbols of identity, relationships, nature, societyand religion, and Dudley (2010)suggested that multisensory museum object encounterselicit ideas and meaning-making opportunities. Froggett et al. (2011)c onducted an evaluation of health and well-being programmes run by several museumsi nt he North West of England. Thes tudy determined that wheni ndividuals interact with museum objects the intrinsic, physical and material properties of the objects triggers ensory, emotional and cognitive associations, memories and projections. This is further exemplified by Newman and McLean(2006), Ander et al. (2012)and Lanceley et al. (2012)i ns tudies focused on assessing the impact of museum object encounters on well-being.
Theabove psycho-educationaltheories acted as aconceptual framework in which the perception of well-being derived from the object handling sessions was assessed.T he study examined quantitative and qualitative changes in psychological well-being resulting from handling and discussing museum objects in on-to-one, facilitated sessions. The researchq uestion asked if standardized psychological measures of well-being and happiness would be improved as aresult of amuseum object handling session and whether qualitative methods could be used to investigatethe aspects of these sessions that led to the predicted benefits. The aim of the researchw as to describe typical features of this intervention, consider the factorsthat influenced the patients' contributions to the sessions and examine the relationship of these factorst oi mmediate, post-session, psychological well-being outcomes, in relation to the psycho-educational theories explored above.

Participants
The study was conducted with volunteer inpatients from al arge, central London NHS Foundation Trust hospitalo ver a6 -monthp eriod. Participants were of mixed gender, ethnicitya nd social background and spoke English sufficiently well to understand the patient information leaflet. All participants gave their informed consent to take part prior to inclusionint he study and for audio recording.

Materials
The choice of quantitative measures was based upon ar eview of scales for evaluating psychological well-being, quality of life and perceived health status in healthcare settings (Thomson, Ander, Menon,L anceley &C hatterjee,2 011). The review indicated that the most suitableself-reportmeasures for assessing well-being at patients' bedsides were the Positive Affect Negative Affect Scale (PANAS) (Watson, Clark&Tellegen, 1988), and the Visual Analogue Scale (VAS) (EuroQol Group,1990). Six boxes were compiled that each contained six museum objects displayed in conservation materials. Objects comprised archaeological and ethnographic artefacts, etchings and printing plates,fossils, mineral samples and zoology specimens that varied in their tactile, visual and kinaesthetic properties (e.g. Egyptian bronze figurine, Neolithic hand axe, 1950s print, fossilized shark's tooth, piece of agate and turtle carapace).

Design
Amixed-methods approach to data collection and analysiswas used. The studyexamined changesinmood pre-and post-session usingthe PANAS to measurepsychological wellbeing (10 positive and 10 negative mood adjectives each rated from 1to5)and two VAS scales to assess subjective well-being and happiness (each estimated outo f1 00). Qualitative methods were used to investigate the processes that may lead to engagement with the objects and to well-being benefit. Quantitative, multivariate analyses of variancew ere carried out on the PANASa nd VAS scores from adult participants ( n ¼ 57) in four inpatient groups: Acute and Elderly Care ( n ¼ 11), GeneralO ncology( n ¼ 16), Gynaecological Oncology ( n ¼ 16) and Neurological Rehabilitation ( n ¼ 14), using the statisticals oftware package SPSS (Statistical Packagefor the Social Sciences) 17.0 (2007).
Content and thematicanalyses carriedout on the recorded discoursefrom 16 sessions with participants, selected to represent the 4inpatient groups and considered typical of the data overall, were entered into the qualitative analysis using the qualitative analysis software NVivo8 (QSR International, 2008). Data were first subjected to content analysis to summarize the use of positive and negativemood adjectives during the object handling session (Krippendorff,2 004). The analysis was performed using the keyword search function in NVivo and involved examining the frequency with which PANASadjectives, alternateforms of thesewords or synonyms occurred during the session.
As econd-stage thematica nalysis was used to bring out individual, personal ways in which patients engaged with the objects and how each session was facilitated. All transcripts were independently coded by one researcher (HP) and concerned particular responses and reactions. Codes were groupedinto more detailed themes to understand the interaction more fully (Braun&Clark, 2006;Patton, 1990). Analysis was both inductive andd eductiveb ecause thes emi-structured format of thes essionse nsured that predetermined areas were covered while allowing emergence of newc oncepts from the participants. Ac oding manual was produced in which the codes,t heir definitions and relationship to themes,with text examples, were documented (Table 1) in accordancewith acceptedanalytic practice methods (Joffe, 2011). Two researchers (AL; HC) who were not involved in the sessions, tested the coding manual for validity and inter-rater reliability using the same transcript(Appendix1)and discussed any differences. Agreement was high, but where minor discrepancies arose, discoursewas reread and discusseduntil agreement was reached. There was agreement after scrutiny of 16 interactions that no new codes were emerging and that data analysis had reached saturation (Holloway &Wheeler, 2010).

28
H.L. Paddon et al. Being correct Patients are keen to be correct about objects -i fnot, maybe this is construed by the patient as unfavourable or that they are stupid, silly, etc. It can make adifference to the amount of participation they have in as ession and whether they enjoy it or not.

Questioning
Facilitators using questions to stimulate conversation based around the objects and also to find out more about the patient; their background and level of knowledge.

Questioning
Patients question the facilitators about the objects, 'how old is it? Where does it come from?' etc. They may also question facts that the facilitator gives them, sometimes rhetorically but in some instances because they doubt the information that has been given to them or want to know more about where the facilitator got the information from. "But how do you know?"

Giving information
Facilitator gives facts about objects and can often, quite literally, hand over object information sheets. Giving information Patient gives facts about his or her life.
Inviting touch In order to get patients interacting and engaging with the objects, facilitators invite the patients to touch the objects. Often they encourage them to feel textures, gauge weight or understand its fragility.
Enjoying session Patient may laugh or joke with facilitator. Indication that they are enjoying the session. They may also overtly acknowledge enjoyment.
Sharing power/ passing over control This is mainly seen when facilitators invite patients to select an object. By doing this at the start of the session in particular, it shows willingness from the facilitator to be led by the patient but also sets them on more equal ground. Patients are more likely to engage with things they select as they often remark "I'm interested in that one".

Agreeing
Agreeing with the facilitator can be as imple acknowledgement of hearing to afirm acknowledgement followed by an observation, fact or question.
( Continued) Arts &H ealth 29 Hearing impaired This is important in elderly and acute illness sessions. Hearing might be impaired because of medical problems, distractions in ward, quiet responses from facilitators.

Seeking validation
Patients who treat the object identification as a guessing game, and even those that aren't, seek validation that they are correct with guesses and facts.
They look to the facilitator to confirm their thoughts. This code links to "confirming patient thinking", "being correct" and "guessing game". Facilitator as expert.
Selecting objects Selected objects because of colour, size, shape (features) or they were curious about what it was. They also selected on basis of knowing (or thinking they knew) what it was. Excitedly jumps from one object to another.
Hearing impaired Some instances occur where facilitators cannot hear. Must always be attentive to patient and ask to repeat if comment not heard (do no just agree).
Correcting statement(s) Must do this if afact is incorrect. While patients like to guess they are not always right. Do it in a" good guess but not quite right" way.
Triggering associations Patients may suddenly remember an event, object, person from their past or something in everyday life.
It may be triggered by sight, touch, hearing or smell. This code has links with remembering/reminiscing and makingo bservations.
Acknowledging patient This refers to the instances where the facilitator does not want to break the flow of info/knowledge coming from patient so simply acknowledges engagement in conversation "yes", "mm", "uh-huh" type comments. Sharing knowledge This is different to giving information because it is not about facts but rather about the patients' personal understanding/interpretation of an object or af act.
They often talk about it from first-hand experience or can give an example; they feel comfortable in their own knowledge of it.

Disclosing feelings
The facilitator may disclose their feelings about objects but also about why they chose what they chose and why they omitted objects from the study. For example, facilitators thought medical objects would be inappropriate given the settings but many commented on wanting to see those collections. Disclosing feelings The patient could feel disgust at looking at an object, at finding out what it is. Equally, they may disclose any number of feelings; fear, happiness, shock, amazement, etc. The object has obviously engaged with their emotions/feelings and they disclose personal connections or inner most interpretations.
( Continued) 30 H.L. Paddon et al. Introducing session/ objects The facilitator takes the opportunity at the start of the session to explain what will happen. They will also tell the patient, when anew object is selected, either by the facilitator or the patient, facts and figures about the object or begin by posing aquestion "what do you think it is?" Guessing game Some patients are keen to guess what the objects are and so it becomes almost competitive (perhaps this is reflexive of their personality?). This code is connected with "being correct", "seeking validation" and "confirming patient thinking".

Making observations
The facilitator may make observations about the object as away of questioning the patient, for example "you can see the wear and tear on it, can't you?" He/she may also make observations about the patient ... as af orm of questioning.
Remembering/ reminiscing This links strongly with "triggering associations" (RETHINK). Remembering or reminiscing about things is often as aresult of an object, or less directly, from aconversation induced by an object. Patients remember facts, events and most often very personal stories.

Referring to aides
The facilitator is not expected to be an expert in all things or in the objects for handling so they may refer to aides like information sheets. This project also used images of things when alive/in use to make observation /engagement/comprehension easier. Making observations By handing objects to patients and giving them time to look at each one, patients begin to make observations. This may be about their features, for example, weight, colour, patterns, dimensions. May also relate to other similar objects.
Demonstrating object use The facilitator may take the object and illustrate how it could be use, which way up it would be, etc. This has links to referring to aides as sometimes objects are compared with photos or similar contemporary objects. Worrying about handling Some of the objects may be perceived as fragile and not fit for handling by the patient; some will not touch because they don't like the look of it. However, all objects have been selected for the purpose and the facilitator works to encourage touch and dispel any fears.
Communicating opinions This is different to sharing knowledge where the patient talks about something he/she has knowledge about. Here they give their opinion for example "you've got to do it! You never know". The sentence will often have a'because' in it as they explain their opinion. They also tend to start with "I think" or "I don't think". 32 H.L. Paddon et al.
It was hypothesized that for the quantitative analysis, participants would show improvements in psychological well-beinga nd happiness betweenp re-and post-session measures. The qualitative analysis investigated the processesbelievedtoaccount for these changes.

Procedure
The researchused astandardized protocol (Appendix2)developed in otherresearch into heritage-in-health interventions (e.g. Chatterjee &N oble, 2009), with as emi-structured interview format to examinet he enrichment potential of museum object engagement. Interview questions were linked to the physical and emotional properties of the objects. Sessions lasting between 30 and 40 min took place during afternoon visiting hours for patients without visitors. Sessions were conducted by female facilitators, one a psychologist, the other amuseum professional, engaged as researchers on the project. Both facilitators obtained UK Criminal RecordsBureau clearance for working with vulnerable adults and were appropriately trained to undertakethe work in ahospitalenvironment, e.g. infection control procedures. The studyw as approved by the hospital Medical Ethics Committee (Ethics Committee approval MREC 06/Q0505/78) and the study was performed in accordance with the ethicalstandards laid down in the 1964Declaration of Helsinki and its later amendments.

Quantitative Analysis
Two sets of multivariate analysis of variance( MANOVA)w ere conducted in SPSS: analysis(i) comparedpre-and post-session measures for pooled patient groups (one-way, repeatedm easures analysis); analysis( ii) comparedp re-and post-session measures for separatep atient groups (two-way, mixed analysis). Dependentv ariables were pre-and post-session PANASp ositive and negative adjective scores and VAS wellness and happiness scores; each of these measures was analysed separately. Meansa nd standard deviations (SDs) ( Table 2) wereused to estimate effect sizes (Table 3) based on dividing the meanpre-and post-session differences by pooledSD. Effect sizes (Cohen, 1988), were medium to large for PANAS positive scores and mainlym edium to small for the other measures.
VASs cores are considered ratio data suitablef or parametric testing because assessment is made from zero to 100. PANAS scores use five-point Likert scales normally regarded as ordinal data so homogeneity of variancew as checked prior to undertaking parametric tests, givent he unequals ample sizes. An F -test showed that 2o ut of 20 differencesw ere non-significant, implying homogeneity of variance, and as parametric tests are considered robustt om inor violations ( Howell, 1987), MANOVAs were performed on the data.
Analysis (i) Pooled patient groups.Highlysignificant improvements in all PANAS and VAS measures for pre-and post-session scores comparisons.
Analysis (ii) Separate patient groups.N os ignificant differences betweenp atient groups or interaction of patient groups with other variables (Table 4). Most patient groups demonstrated similar levels of improvement on all measures despite somes tarting from

Qualitative Analysis
The content analysiss ought to determinew hich words were used with the highest frequencyt oe xpress emotion during the object handling sessions, in particular whether participants used adjectives from the PANASword lists,alternate forms of these words or synonyms (Table 5). Findings showed that participants tended not to mention PANASa djectives directly but insteadu sed related words to account for their feelings. The word "interested" occurred with the highest frequency, but rather than use the word in this form, participants tended to say "I findt his interesting" or "I am fascinatedb yt his". Overall, however, PANASadjectives and related words had arelatively low frequencyofoccurrence within participant discourse( Table 6).
Asearch of additional words and phrases from health and well-being literature (Brazier et al., 1992;K obau, Sniezek, Zack, Lucas &B urns, 2010;N EF, 2009;W atson &C lark, 1999)was carried out in NVivo (Table 7) to further scrutinizethe discourse for adjectives used spontaneously by the participants. Of these, three positive words ("amazed", "happy" and "purposeful") and two negative words ("painful" and "tired") were found to occur with the greatestf requencya mong interactions, with "amazed"b eingu sed by the greatest number of participants, althoughnot to the same extent as "interested".
Twenty-sevenc odes labelled aspects of the object handling sessions emerged from thematica nalysis of the data. Conceptual grouping of thesec odes produced seven overarching features or themes of the interactive process (Figure 1). Four clear features emergeds pecific to participants with af urther three emerging to explain the facilitator  (Cohen, 1988). role. Participant features consisted of the "influence of social/physical/environmental contexts", "thinking and meaning-making", "positive interactions" and "self-esteem", whereasf acilitator features comprised "encouraging engagement", "communicating knowledge and information" and "building trust and developing rapport". While features appeared distinct and specific to participant or facilitator, interactional aspects of the sessions strongly implied that the features were interlinked. For instance, "questioning" occurred as afeature of "thinking and meaning-making" by the participant, but it alsoa pplied to the facilitator role, where questioning techniques were used to encourage engagement and build trust and rapport with participants. Each feature contributed to the process of ah andling session, but the balance of features and their frequencyo fo ccurrence appeared to affect theo utcome.F or example, the "communicating opinions"a specto fp articipant interaction occurred in each of the 16 sessions and 321 times overall. Reporting the qualitative results of the study in their entirety would be lengthy and beyond the remit of as ingle journal article. Instead,t he article focuses on participant themes and highlights important features of the participants' role in the sessions.
Thinking and Meaning-Making Using the above criteria, "thinking and meaning-making" emerged as the mostimportant featureofthe patient roleinthe handling sessions.From the initial codes generated by the qualitative analytic approach, both overt and subtle properties of thinking and meaningmaking began to emerge.Participants were able to reminiscence about events, places and people. In this instance, handling aflint hand axe that the participant (PT)discovered was from Sweden, triggered the recall of av isit there.
PT: My husband is ateacher and he spent some time with agroup of Swedish teachers as sort of an exchange. One of them had invited us over. We became very good friends ... Not for long, it was only four or five days that we went there. We didn't see much but what Idid see, it was all very fresh and clean. Avery clean place. Very expensive too.
Arguably, remembering this visit had little to do with the flint axe, apartfrom the fact that it was from Sweden; however,u singp rior knowledge and acquirede xperience, the  participant developed an interpretation for the object connectedw ith av isit to friends in another country. Remembering eventsofthis sort allowed participants to feel sufficiently comfortable to communicate opinions such as the cleanliness of the country and the cost of living there. Fifteenofthe 16 participants discussed memories or reminisced about life experiences and events while handling the objects.These includedchildhoodmemories of days at the beach,l earning experiences during school years, artistic tendencies of family members, workinga broad, employment skills and hospital and museum visits. While few participants madeexplicit mention of "learning",somemade connections to learning from the objects and/ort he facilitator (FT). In the following extract, an inpatient on a neurological rehabilitation ward is finding outa bout af ossil vertebra of am arine reptile called an "ichthyosaur".
PT: Triassic. What does the "assic" stand for then?
FT: Oh yeah, Id on't know actually. Id on't know what the sic stands for. Ik now that the Jurassic, that's named after the region where lots of these rocks are from, the Jura in France. Id on't know ... Id on't think Tri is ar egion like Jura is. It hink Tri might be something to do with maybe there's three layers of it or three eras of the Triassic. And then Ihave no idea about the "sic" bit.
PT: Jura is aplace in France?
Much of the conversation generated in each handling session took the form of learning, whetheritwas constructing new meanings for objects,sharing facts and ideasoragreeing with the other person's opinions.L earning, meaning-making and thinking tended to be confirmed by the reiteration of points. Participants were seemingly reiterating points to gain assurancefrom the facilitator that their ideas, opinions and knowledge were correct and that they had understood the information imparted to them. Linked with meaningmaking, thought and the reiteration of points is the aspect which describes the process of object identification as a"guessinggame". Theguessing game aspect, apparent in all 16 sessions,w as often provoked by open questioning from the facilitator such as "What do you think it is?" Participants used priork nowledge to guess whatt he object was.O ther skills were used to identify objects and were interrogated through different senses. The aspect "making observations"spoke most obviouslytothe visual sense when participants interacted with objects. In the following extract, taken from asession with apatient in an oncology ward, thinking and meaning-making was apparent.T he participant noto nly imparted facts gained from observing the object at close range,but also gave opinions and constructed meaningfrom acquired knowledge; in this case agreater understanding of the basic anatomyofatooth.
PT: If you break it on there, it's going to come out ayellow colour. Ours isn't. Our tooth, you cut from here, you only going to get yellow in here, but this ...
FT: But this has got more colours.
PT: This has got more colours, yeah. FT: Iwonder if they are little blood vessels?
PT: It could be. And this one moves and this one comes up, so must have some sort of connection.
The aspect" makingo bservations"i mplies visual examination. However, the data illustrated how visual scrutiny of objects was complemented by the sense of touch. In the extract below, the patient keenly observed aN eolithic axe head, noting the human interaction of the maker with the stone. Touching the object allowed the participant to explore the material used to manufacture the object. The participant's enthusiasm for and description of the axe head conveyed amazement and awe. Theparticipant talked about its physical properties,the skill of the person who madeitand the techniques involved.The encounter is typical of ap atient -object interaction where the opportunity for hands-on object engagement heightened the participant experience and encouragedasharing of knowledge, ideas and feelings.
PT: Yeah, when touching Ithink, whoever made the work on it, whoever done it, the way they done it, obviously at first might just be abig stone, and somebody cut it maybe with the sand and water, cut it through, must have an idea in mind he's going to make this shape of axe or whatever. Must have askill to do these things and the way he's created this thing is amazing because, look here, they just look like, from here you cannot tell if it ashell or stone or what it is. But when you hold in your hand, my initial reaction was it could be marble, but even in astone, it's different when you can hold it up. You can actually feel what it is, how created, how it been done.
Theability of participants to closelyobserve and touchmuseum objects led to another major aspectofthinking, that of questioning. Ranked secondint he list of patient codes, questioning gave patients an opportunity to learn morea bout the objects, to query their observations or find out about how objects were made and used. Theposing of questions by patients indicated engagement and as timulation of curiosity. Questions increased in number if participants were intrigued by an object. An intriguing object was more likelyto augment the communication of personal opinions and feelings. This communication tended to occur further into the handling session, as trust and rapport built up betweenparticipant and facilitator. There are many examples that illustrated disclosure as all of the participants revealedtheir feelings and communicated their opinions multiple times over the course of a handling session. Revealing feelings and communicating opinions were two universal aspects that contributedgreatly to the featureofthinking and meaning-making.
PT: Iwas abit tired and when they told me Iwouldn't be going until tomorrow ... because you don't sleep, Ithought "not another night of not sleeping". Iwas abit tearful before you arrived so I'm quite glad you came to distract me. Thank you very much.
PT: Well everything's if, if, if, if, and I've been here since something like the 15th September. Then Ihad to go home and now I'm back. Was so ill with pain and stuff and now I'm still abit ill, they're supposed to be giving me chemotherapy but nothing's happening so who knows, it's like, who knows.
PT: Um ... just like the patterns on them as well actually. It's quite restful looking at it and tactile as well because you have the rough side and the smooth side.
Them ost frequent participant aspect to emerge, however, was agreeing. Agreeing indicated that the participant was focused on the session, listening to the information, concurring with what the facilitator was saying and demonstrating understanding. Agreeing was potentially indicative of patients thinking about an object, although in some 40 H.L. Paddon et al.
cases it was apparent that aparticipant's desire to satisfy the facilitator led to an expression of agreement regardless of whether they actually agreed or understood about the object. This tendency to behave in as ocially desirable manner probably occurred as ar esult of participants' previous experience in social situations and the physical and environmental context.I nfluence of social,p hysical and environmental context was another feature affecting the participant roleinthe handling sessions.Although not always obvious, these contextsp otentially had an effect on levels of participant involvement,s elf-esteem and confidence.
PT: Oh yes it does look lava-ish doesn't it?
Discussion It was hypothesized that participants would show improvements between pre-and postsession measures of well-being and happiness.T he study demonstrated statistically significant, overall enhancement of psychological well-being as determined by the PANASm easures, and subjective well-being and happiness as determined by the VAS measures. Positive PANAS, wellness and happiness VAS scores increased, and negative PANASs cores decreased in line with predictions, although there weren os ignificant differencesb etweent he four patient groups.T he average increase in positive mood was greater than the average decrease in negativemood supporting the view of Watson et al. (1988)t hat the two PANASs cales werei ndependent and orthogonal. Generally, participants reported low levels of negativem ood pre-session leaving little room for improvement post-session. Effect sizes carried out on pre-and post-sessiond ifferences showed arange from small to large although were generally small for the VAS measures, indicating that additional participants and equal sample sizes would be neededtoincrease the statistical powerofthe study.
If touchenhanced well-being, it implies the presence of ashort-termmemory (STM) tactile representationadditional to the verbal and visual representations proposed by dual Arts &H ealth coding (Paivio,1971(Paivio, ,1986)adding weight to triple or multiplecoding models. Craikand Lockhart's (1972) levels of processingm odel is relevanti nt hat tactile qualitiess uch as texture, shape and weight couldh ave enhanced the kinaesthetic experience of sessions leading to deep and elaborate encoding. Spector et al. (2008)s uggestedt hat cognitive stimulation therapy (CST)increased cognitive processing and laid down new connections in the brain as aresult of encounters with novel stimuli and social interaction. It is possible that museum object handling and discussion centred on the objects broughtabout asimilar level of cognitive processing. Although it was beyond the remit of the current researchto ascertain whether neuronal formation occurred; af uture study could employ brain scan techniques (e.g. functional MRI) to examinethispossibility.
Results from the quantitative and qualitative analysis inferred that object handling sessions contributedp ositivelyt oi ndividual participant well-being and that am ixed method approach afforded amore nuanced view of the impact of object handling.While the content analysisfound that the frequencyofoccurrence of PANAS words, alternative word forms and synonymsw as low within the 16 participant sessions, other words related to health and well-being ("amazed", "happy", "tired", "pain" and "purpose") takenf rom relevantliterature occurred frequently within the transcripts. The NEF (2009)identified key indicatorst om easuret he well-being of communities at an ationall evel. The qualitative analysis, conducted as part of this research project, uncovered similar indicators at an individual levelrepresented as features in the context of facilitator -participant interactions. For example, NEF explained that for people to experience personal well-being they need to be engagedinactivities, take the opportunity to learn new things and feel that their life has meaninga nd purpose. The research findings presented here (Figure1 )s howed that participants were engaged, contributed to meaning-making and interacted positively.
In keeping with constructivist models of adultlearning (Falk&Dierking, 2000;Hein, 1999), participants added to their existing knowledge of objects and linked memories, taking the opportunity to questioni nformation in order to constructn ew meanings. In keeping with Browna nd Wragg's (1993) worko nt he acquisition of learning skills, the majority of questions asked by participants were cognitive in that they were keen to gain knowledge. Ther esearch showed that objects taken outside the museum space evoked emotion as well as the recall of events, people and places. Mack (2003,p.8)referred to the "role of archive material in triggering memories" and Phillips (2008,p .2 04) found that when used in reminiscence sessions,m useum objects, specifically coins and medals, promoted "learning, creative thought, skills development and greater confidence". Reminiscence through objects enabled 15 of the 16 participants in the current study to recall and talk about aw ide range of memories. Thee xperience of remembering past events adds supportt ot he notion that museum objects accrue multi-layered identities "ranging from conceptual, through the factual, the functional and the structural, to the actual identity" (Maroević ,1 995,p .2 4). Participants contributed to these identities by finding meanings in the objects often attributed to personal experiences. The processes of remembering and reminiscing demonstrated how meaning-making couldcontribute to the beneficial effect of the session and be used "in positive and constructive wayst hat help build self-esteem and bolster as ense of identity" (Kavanagh, 2000,p .1 18).
These outcomes are in line with other similar studies which show that museum objects function as symbols of identity, relationships and society, and that museum object encounters elicit ideasand meaning-making opportunities (e.g. Ander et al., 2012;Dudley, 2010;F roggett et al., 2011;L anceley et al., 2012;N ewman &M cLean, 2006). Some authors have arguedt hat meaning-making is important for adjusting to stressful events, such as bereavement (e.g. Gillies&Neimeyer, 2006)a nd illness (e.g. Lanceley et al.,42 H.L. Paddon et al. 2012;see references in Park, 2010). The implication of meaning-making in the healthcare setting is explicated by Park (2010,p .2 37) who stated that "meaning-making playsa central role in the coping and adjustmentofmostpeoplefacing major life stressors". Thus, addressing meaningmay be afruitful approach to clinical interventions aimed at helping people recover from thesehighly stressful experiences. Throughout the object handling sessions,t here were many examples of participants engaging with objects in amultisensorymanner. While meanings are normally developed and built upon through the visual sense in am useum environment, object handling in healthcare settings provided participants with the opportunity to experience museum objects through other senses, specifically touch. Arguably, the chance to interact through visual, auditory and tactile senses in an interesting and engaging way with museum objects triggered recall of long-term memories of eventsand associated meanings. The outcomes of object handling sessions with the patients selected for this study, those in chronic and acute care settings, were not intendedaseducational or learning experiences, although the presentation of museum objects and related verbal material was in keeping with learning theoriesc oncerned with thei ntegrationo fi nformation andV AK preferences. Furthermore, some of the elements that emergedf rom the qualitative analysisp ointed towards a"community of learning". This patient -facilitator "community" was transitory, but exhibited some of the features of learning communities, notably the facilitation of "information exchange,k nowledge sharinga nd knowledge constructiont hrough continuous interaction, built on trust and maintained through as hared understanding" (Daniel, McCalla &S chwier, 2007,p .2 96).

Limitations and future work
Ak ey limitation of this study was the fact that it was not longitudinal. In af uture longitudinal studyo f6months or over, individual case histories couldb ee xamined in depth for sustained effects in coping and resistance to negativel ife experience. Sample size was also an issue, so afuture studyshould consider arandomized controlled trial with agreater number of participants where the object handling intervention couldbecompared with care as normal.

Conclusions
The evaluation of aheritage-in-health intervention conducted across four patient groups in the same hospital suggestedthat museum object handling sessions produced beneficial and therapeutic effects on patient well-being and happiness.Similar increases in psychological well-being across the three positive emotion scales (positive PANAS; wellness and happiness VAS measures) implied that findings were not an artefact of the study but represented real improvement over the duration of the object handling session, although it could notb ea scertained whether these effectsw eres hort-term or sustained. Specific features of the current study such as meaning-making and links to previously stored memories could be used as abasis for further analysis of verbal discoursefrom healthcare interventions. Findings added weight to the need for provision of arts-and heritage-inhealth activities for communities of hospitalized adults temporarily or permanently excluded from gallery and museum visits. As an on-pharmacological intervention, the results of theseo bject handling sessions have shownt hat meaning-making and thinking have the potential to help patients cope and take part in apositive experience during their hospitalstay.