Art Therapy for Elderly Women Diagnosed with Alzheimers: A Positive Person-Centred Approach Increases Ease in the Care Process

“Art and art therapy is regarded as a haven, getting a clearer view, clearing the way emotionally and enhancing and enlivening the self.” These narratives resulting from experienced storylines from Collie et al. [1] realistically sum up the clinical benefits experienced by people in pain when they become involved in art. Such storylines also show existence being affirmed, better wellbeing achieved through artistic expression and “meaning making being achieved through physical acts of making” Collie et al. [1]. This is really a therapeutic purpose of art therapy. Art via arttherapeutic processes can give to people an empowering healing process that makes meaning making easier and positive arts interventions promote psychological well-being Darewych et al. [2]. Art and aesthetics, by stimulating creativity and embodiment, also re-educate people and facilitate social connections Byers et al. [3-8]. In France, increasing numbers of suffering persons have used a non-pharmacological supportive aesthetic-based activity in healthcare called modern art therapy. Modern art therapy refers to beauty. It has been defined by its founder Richard Forestier as “the exploitation of the artistic potential of an individual in a humane and therapeutic goal” Forestier [9]. This therapy also defines art as “a humane and voluntary action, aesthetics oriented” and emphasizes the physical engagement in the art making as a key factor through the therapeutic process to recover self-esteem. It is then regarded as a sensory stimulation intervention Lusebrink [10] to help patients participate in the therapeutic process by an increase in feeling pleasurable and aesthetic sensations Forestier et al. [9,11]. This art-based therapy also relies on the preserved abilities of the individuals, rather than attempting to correct the disabled part.


Introduction
Background "Art and art therapy is regarded as a haven, getting a clearer view, clearing the way emotionally and enhancing and enlivening the self." These narratives resulting from experienced storylines from Collie et al. [1] realistically sum up the clinical benefits experienced by people in pain when they become involved in art.
Such storylines also show existence being affirmed, better wellbeing achieved through artistic expression and "meaning making being achieved through physical acts of making" Collie et al. [1]. This is really a therapeutic purpose of art therapy. Art via arttherapeutic processes can give to people an empowering healing process that makes meaning making easier and positive arts interventions promote psychological well-being Darewych et al. [2]. Art and aesthetics, by stimulating creativity and embodiment, also re-educate people and facilitate social connections Byers et al. [3][4][5][6][7][8]. In France, increasing numbers of suffering persons have used a non-pharmacological supportive aesthetic-based activity in healthcare called modern art therapy. Modern art therapy refers to beauty. It has been defined by its founder Richard Forestier as "the exploitation of the artistic potential of an individual in a humane and therapeutic goal" Forestier [9]. This therapy also defines art as "a humane and voluntary action, aesthetics oriented" and emphasizes the physical engagement in the art making as a key factor through the therapeutic process to recover self-esteem. It is then regarded as a sensory stimulation intervention Lusebrink [10] to help patients participate in the therapeutic process by an increase in feeling pleasurable and aesthetic sensations Forestier et al. [9,11]. This art-based therapy also relies on the preserved abilities of the individuals, rather than attempting to correct the disabled part.
People with neurological diseases continue to express themselves through art Miller [7] and to manage aesthetics Byers [3] as the degenerative illness progresses. Aesthetic experience lies in the articulation of unconscious processes Byers [3]. Neurodegenerative diseases provoke stress and distress, and a multisensory stimulation such as art therapy has a significant effect on the resulting anxiety and quality of life Baker et al. [4,[16][17][18][19][20]. In this way, the purpose of aesthetics-based art therapy sessions for such affected people is to steer the person through her/his self-realization, creating pleasurable sensations, feelings of beautifulness and ultimately recovering her/his self-esteem.
As such, people diagnosed with neurodegenerative diseases improve their self-esteem and well-being by using creative activities by means of acts of making, which produce pleasure and reduce anxiety Byers et al. [3,5,[21][22][23][24][25]. This is likely because neurodegenerative diseases have an impact on the human memory and decrease cognitive faculties, but emotions and affect are less impacted by degeneration Baran et al. [26][27][28][29][30][31]. Fundamental properties of artistic technique can then be used to help people participate in the therapeutic process, and people affected by such diseases remain sensitive to subjective art-based experiences Camic et al. [5,[32][33][34][35][36]. Systematic reviews have also suggested that creative art therapy is effective for the treatment of behavioural and emotional challenges resulting from the diseases Cowl et al. [4,37,38], and an emotional enhancement of memory effect has been seen in people with neurodegenerative diseases Baran et al. [26,39,40]. It also appears that positive emotions broaden the scope of action and build physical resources Fredrickson [41]. In addition, non-pharmacological interventions have the potential to improve psychological aspects such as cognitive improvement via neuroplasticity Ehresman et al. [4,42,43].
To date, art-therapy-oriented studies that have focused on the self-esteem of affected persons have shown an increase in global self-worth Chancellor et al. [1,23,25,[44][45][46][47]. Trust and confidence being part of self-esteem, art therapy is helpful in building greater trust and self-disclosure Hartz et al. [47] and creative processes are identified in its acts to inherently give people the authority or power to do something promoting empowerment and selfassertion Franklin [46]. Cognitive behavioural therapies show that the patient is not the only factor involved in the therapeutic process. The therapists, as well as the relationship, are regarded as key determinants (more specifically, the therapist's attitudes and behaviours) Lambert [48]; Lambert, Okiishi et al. [49][50][51]. In a systematic review of the clinical effectiveness of art therapy among people with non-psychotic mental health disorders, the patienttherapist relationship appears to be an important determinant in successful treatment Uttley et al. [52]. Also, some studies precisely demonstrate that art therapy's effectiveness is mainly therapistbased with regard to the used approach Cavazos [53], and the therapist's variability in the alliance is seen to be more important than the patient's variability for improved patient outcomes Del Re et al. [54]. Also, in psychotherapy, the accounted variance of the therapist is seen to be eight times more than the variance of the patient in treatment Lambert [48].
Overall, quantified common factors, which include the patienttherapist relationship, account for 30% of the variance in patient outcomes Lambert et al. [55]. In particular, personal attributes of the therapist, such as confidence and trust within the therapeutic frame, are regarded as essential to the therapeutic success Ackerman et al. [56] and some studies show that a built and reliable physician-patient working alliance requires affective dimensions such as trust Fuertes et al. [57][58][59][60]. In addition, trust in physicians has been shown to be improved by using a patient-centred approach Berrios-Rivera et al. [61], and the alliance is similarly found to strongly correlate with patient centeredness Huff et al. [59]. This is because the therapeutic bond comes first through building mutual trust Berdondini et al. [27,62]. In the same way, an authentic partnership approach including trust is more and more required in dementia care Dupuis et al. [63]. People with neurodegenerative disease maintain an affective memory, which allows them to know what attitude to adopt, depending on the people present Poncet-Jeanne [64]. Consequently, it can be expected that trusting the therapist will allow a decrease in the anxiety of the patient, who is reassured and who thus acquires a certain emotional availability.
Thus, the quality of the relationship that the carer will establish with the patient will be a central element in the care of these people Poncet Jeanne [64]. This is particularly true when the session occurs in a non-judgmental atmosphere Collie [65,58]. Nonjudgmental absorption in what is being observed by the therapist requires empathy that enhances the therapeutic alliance Ackerman et al. [56,66,58]. Regularly, studies have demonstrated that patientperceived empathy has increased patient trust Bachelor et al. [67-69, 58,70,71]and active participation Raglio et al. [72,25]. A study conducted in the UK by Evans-Jones et al. particularly underlined the importance of empathy for cognitive-behavioural therapies Evans-Jones et al. [73]. Empathy has also been noted to lower patients' anxiety and distress Derksen et al. [69]. An empathic behaviour is part of a person-centred approach. Established by Carl Rogers in 1961 [74], this type of approach diverges from the traditional model of the therapist as expert and moves instead toward a non-directive, empathic approach that empowers and motivates the client in the therapeutic process Rogers [74]. A goal of a person-centred approach is to assist patients in becoming more autonomous and confident. In this context, the purpose of the therapist is to create a growth-promoting atmosphere to solve problems by being an empathic and attuned therapist Cavazos et al. [53,75]. Such approach is seen as effective for people with dementia Clisset et al. [76][77][78][79].

Objectives and Hypothesis
This case study aimed to address the influence of the empathic therapist's behaviour in developing interpersonal trust through the therapeutic alliance during modern art therapeutic-based sessions.
Clinical observations regarding the use of a person-centred approach were supposed to demonstrate that trust is essential in order for the patient to develop self-confidence and then commit to the activity. In particular, trust is expected to be the initial lever acting at the beginning of the relationship, and therefore in all the therapeutic process.

Research Model
Self-confidence has already been regarded by some researchers as suitable for expressing self-esteem and personal autonomy. To accurately define self-esteem, Bernhard et al. [80] specifically established a tripartite model based on the sum of three pertinent components, namely, love of self, self-confidence and assertiveness.
As a patient's trust is an additional level that can be used by the therapist's behaviour to emphasis confidence, it was added in Bernhard et al. [80] model to define a new one, expressed as follows: Consequently, a research model was stated as such:

1.
The therapist establishes a confidence-based relationship by developing a patient's trust, which is initiated and reinforced with the help of an empathic person-centred approach.

2.
Trust connects the patient with the therapist and with her/himself. This mirrored confidence reactivates his/her selfconfidence.

3.
Self-confidence makes the patient want to commit to the activity.

Purpose of the Study
Modern art therapy identifies two main behaviours: a contemplative mode when the patient observes the therapist acting and an active mode when the patient actively participates in drawing and/or painting. In both modes, an aesthetic intention is present: both modes initiate a reinforcement loop linked to the contemplation and the desire of mimesis Forestier [9]. Modern art therapy emphasizes the physical engagement in the art making as a key factor through the therapeutic process to recover self-esteem.
Thus, the intention of the art-therapist is to initiate or facilitate the action during the session because the final objective is to decrease anxiety via making and feeling. However, modern art-therapy does not recommend any positive approach based on empathy such as the one developed by C Rogers [81]. It primarily focuses on the care effect of art on self-esteem, aesthetics-wise. The purpose of this study is to show how a positive (Roger's) approach can facilitate this process by first establishing client trust that facilitates engagement in aesthetic activities. The reinforced empathic behaviour, part of a person-centred approach is used in this way. In addition, the arttherapist can ask the patient to express the pleasure that results from the artistic activity and that contributes to the increase of selfesteem Forestier [9].

Ethical Approval
Ethical approval prior to the implementation of the study was obtained from the carers, mainly the families. The signed forms were stored in the files of the patients and were available on request.

Settings, Protocol and Therapeutic Strategy
The duration of the study was set for four months. Weekly sessions were planned. One-to-one contact was preferred. An important aspect of the therapy was to provide a safe environment in order to encourage the patient to commit, especially for people with high levels of anxiety Turnbull et al. [83]. Consequently, individual art therapy sessions were planned for each participant and conducted by the corresponding author of this study graduated in modern art therapy. Note that when a participant refused to enter the room at the last minute or to sit down at the desk, or went back, the session was not recorded. The first session was regarded as an opening session and resulted in a discussion. It was also an attempt to complete the anamnesis often partially described in the medical file. The therapeutic strategy was first mimesis oriented during the starting sessions in order to connect any aesthetic intention. After asking the patient "What would you me to draw for your pleasure, what could I draw that could make you happy?", the art-therapist started to draw or paint and engaged the discussion in the same time to build interpersonal trust. A special focus was put on the emotional enhancement of memory by using souvenirs linked to family members or great personal details of their past life.
The patients were also verbally and progressively led to develop their visual creativity by soliciting their imagination with the help of personal details of their lives and the use of preferred colours as preferences, because colours were observed as stable, despite the disease Miller et al. [7,84] and were preferably seen to contribute to the sensual aspects of art-making amongst older women Reynolds.

Methods of Analysis
An observational grid was adequately established to fit with the modified Bernhard et al. [80] model. Nine items were identified per component (36 items in total). Behavioural items were preferred.
The Appendix 1 describes the resulting grid and the Appendix 2 the glossary that defines each item which, was independently quantified using a 5-point Likert scale ranging from 1 to 5. The scores were quoted by the therapist from his field observations just after the session. Note that when an item was not observed during the session, it was not rated (previous calculations showed that an added default score set at "3" to complete the matrix can significantly reduce the experimental variability). Such raw quoted data were used for the variability study and data were averaged per component and per session to calculate and monitor the self-esteem progression ( Figure 1).

Figure 1:
Overall tendency of the self-esteem progression for the whole population.

Statistical Analysis
All statistics were calculated with a 95% confidence level in probability (Minitab PC 16 version). A p-value above 0.05 was regarded as not significant. No shift over time of the data was observed (time series analysis). The random distribution of the residuals was also checked for each calculation (homoscedasticity).
The impact of some interesting field variables was studied: the within person variability (first name), her age, her MMSE value, the duration of institutionalization, the living unit, the moment of the activity (morning versus afternoon), the number of sessions and the mode (active, i.e., physical versus contemplative mode).
Some variables were considered as having a weak influence over the duration of the study: the daily therapeutic treatment

Results
Except for one woman who remained in the contemplative mode, all patients became active during the course of the sessions.
Between four and eight sessions (five on average) were required for a patient to become active. The bodily commitment was remarkable and observable via a positive gap of the quoted items. The clinical variability of the main field variables on self-esteem was studied via ANOVA (n = 3289). The mode was coded as a dichotomous variable (Y/N). Table 1 summarizes the results. Overall, the expressed variability is slightly above 100% (106%), which means there was a residual collinearity between the chosen variables. The two most influential identified factors were the mode type and the number of sessions. As previously described, action is key (F-value of 996). The living unit factor had a weak but noticeable impact, as it was the third pertinent factor (F-value of 81). This influence was confirmed by the staff in particular for one unit, in which the most dependent patients were preferably placed. Patient-related variables were seen to have a similar influence. The time of the session (morning versus afternoon) had no influence, which could facilitate the organization.    such an approach adversely impacts assertiveness during the course of the sessions. The MMSE impact is similar whatever the approach is (a gap of five sessions). Table 4: Influence of a person-centred approach on the number of sessions for given values of the components of self-esteem.

Empathy
The present results reinforced previous outcomes regarding the impact of trust in the therapeutic relationship, which can be facilitated by a patient-centred approach Rogers [74] [86,[87][88][89][90][91][92][93][94][95][96][97]. This approach has also been successfully used to take care of patients with depressive symptoms Bickmore et al. [98]. The important point is that the feelings of trust, genuineness, and empathy are all lived experiences and that they derive their entire meaning from their individual embodied, spatial and temporal context Morstyn [85]. With an empathic approach, the therapist can tune his/her answer regarding the patient faculties and capacities. For Rogers, the concept of trust focuses more on trusting the potential within the individual even before it manifests in his/her outward behavior Rogers [81].
Rogers defined facilitative attitudes, which are congruence, empathy, level of regard and unconditional regard, and the more the therapist will develop these attitudes, the more the patient will trust the therapist (Rogers, 1961). For this study, the empathic approach is expressed by frequent eye gaze, listening and attentiveness to the talk of the patient, and sometimes humour.
In order to foster trust, non-verbal behaviour, active listening and paraphrasing is also used Fuertes et al. [74]. However, therapist  (Table 4).

Five Sessions to Commit
At the beginning of the relationship, the bond can be regarded as fragile and it develops through a series of passages -first through building trust, which occurs only when the patient realizes that the therapist is on her/his side (no need for supervision) and she/he can accept what interesting experience can come out of it Quattrini [90]. The patient at this time can usually accept what the therapist suggests, even if the suggestion is not necessarily appealing at first. There were opportunities during this survey for the therapist to signal his desire to take care of the patient during the contemplative stage. A significantly higher level of trust can be observed in the relationship to respond when the other person takes the initiative to signal his/her desire to be trusted Swinth [91]. Thus, with the therapeutic bond elaboration coming first in trust/safety conditions, the patient is able to take her/his eyes off the therapist and think about her/his well-being Berdondini et al. [27]. A productive collaboration is not necessarily a natural or spontaneous process. It requires a series of efforts for both patient and therapist, meaning several sessions Berdondini et al. [27]. In average, five sessions are required here to build a sufficient and identifiable trust-based relationship. Then, the contemplative period of time can be regarded as a "builder of trust".
During a previous follow-up involving mild cognitive impairment people receiving in-home care, between four and five sessions were required for these patients to operate Deygout [92]. This number of five sessions follows certain reproducibility.
Some authors have described the first critical alliance phase as being the initial development, which usually takes place within the first five sessions. During this phase, trust must be established Horvath et al. [93]. Also, in a non-directive teaching experience with students, Rogers set unstructured courses without informing the class about it, just answering questions and waiting for student initiatives Rogers [74], pp 299-304. By the fifth session, something definite had happened. Students spoke to one another and the selfconscious group became an interacting group, a brand new cohesive unit, carrying on in a unique way. It took the class four sessions to realize that they were wrong and that they were very likely to develop mutual trust and self-confidence. Interestingly, Rogers also reported the case study of Mrs. Oak in a storyline regarding the experiencing of the potential self. From an early part of the fifth interview, verbal material emerged that described her awareness of experience, which was previously discussed Rogers [74], p 77.

Absence of Judgment and Congruence
Rogers recommended an "unconditional positive regard" her trust in the process develop. Immigrant students performing art therapy-based activities had a similar feeling. They remarked that when they earned respect from their peers, they were more likely to obtain self-esteem; they suggested that self-esteem firmly depends on the way peers see you Chen [45]. Thus, another way to see the present research model would be to consider the empathic behaviour of the therapist as a mirror that reflects the patient's trust and transforms it into self-confidence: "The therapist directs the looking; without the therapist, you might stick with what you were already thinking and a lot more can came out with the therapist's direction" Collie [65]; Patricia's narratives p 136.

Aesthetics and Affect
Common visual aesthetic experience within the art therapy relationship deeply involves both patient with severe dementia and therapist due to an aesthetic countertransference to the work; therefore, such patients become able to express and communicate aesthetic pleasure resulting from the activity Byers [3]. By making artistic choices that fitted with the environment of the patients, the therapist also expected to reactivate their imaginary world.
The present observed behaviour showed that it was sometimes not possible to dissociate the artistic aspects from the relational ones confirming the previous Byers's hypothesis that aesthetics can bring a point of resonance into the therapeutic relationship Byers [3]. Probably motivated by the established interpersonal trust, most of the patients verbally expressed their will to share their feelings of sympathy with the therapist. Interestingly, when in 1873 the German philosopher Robert Vischer introduced the term of empathy ("Einfühlung"), it was to express the "aesthetic sympathy" he felt when he was presented with art. It is very likely that the neurodegenerative disease reinforced the patients' dependency, but it was obvious that the patients liked to share what they aesthetically felt. Aesthetics is not only technical knowledge.
Such intimacy is seen in the helping relationship, when the art therapist and the patient are engaged in a creative activity such as art therapy. Intimacy in a therapeutic context depends primarily on trust and may involve empathy and sympathy Kossak [94].
This intimacy is seen to make the link with the increase of positive feelings when a pleasant situation is shared, because the sharing amplifies the intensity of the experience Boothby et al. [95]. Modern art therapy encourages this sharing for educational purposes (assertiveness) and results have suggested that more and less trained therapists can be equally liked and trusted Horvath et al. [93]. In a study assessing strategies to raise self-esteem in female juvenile offenders, an art psychotherapy approach was seen to show a significant increase in personal connections (characterized, among other things, by trust) compared to a plastic art-based strategy that only addressed technical knowledge Hartz et al. [47].
The present study also focused the patient and therapist experiential collaboration on emotion-focused therapy, which offered a mutual involvement of equals and required the patient's agreement and commitment, and her coaching and even teaching Berdondini et al. [27]. Felt and expressed positive emotions engage in active imagination and healing Darewych et al. [2,96]. Similarly, the skill to build up a trusting and long-standing patient-physician relationship encouraged physicians to resonate with the patient emotionally Derksen et al. [69], even if more and more new forms of trust relations may be emerging in the context of health care delivery, reflecting a change in motivations for trust from affect-based to cognition-based trust Rowe & Calnan [97]. Anyway, for patients affected by neurodegenerative disease, affect-based trust was preponderant. The determined impact of the centred-person approach was found to be significantly higher than the MMSE impact (Table 4).

Risk Versus Trust
Mutual trust implies a mutually perceived promotive interdependence, i.e., a cooperative expectation from the one who trusts and a cooperative intention from the trusted one; perceived trust can also increase with the communication level in the relationship Loomis [98]. Such, an open and non-judgmental collaboration, which enhances mutual trust, can involve the patient and the therapist encountering one another through taking risks Berdondini et al. [27]. Berdondini referred to the notion of bond established by Bordin as part of the therapeutic alliance Bordin [62]. The very ideology of patient-centred therapy tends to mute the responsibilities of the therapist and highlight those of the patient Bordin [62]. In any case, the patient takes some risks of sharing her/his experience and in developing a collaborative experience. Risk is crucial to the operation of trust, which results from cooperative behaviour. To trust is to let another think about and take action to protect and advance something that the "truster" cares about Baier [99]. Deutsch also defined two processes through the development of trust for the trusting person Deutsch [100]: the formation of expectations and the translation of those expectations into behaviour under risk.
Similarly, Luhmann defined confidence as "an anticipation of the future based on prior experience and which aims to reduce the complexity of the future world" Luhmann [101]. Complexity therefore characterizes a potential for undetermined events and confidence aims to reduce uncertainty and consequently the feeling of risk. Luhmann precisely defined in German the word "vertrauen" to mirror English, which dissociates the notions of confidence and trust. He translated self-confidence as "insured confidence" and trust as "decided confidence" Luhmann [102]. Decided confidence presupposes a risky situation that one thinks to avoid by deciding

From Trust to Self-Confidence
Carl Rogers's philosophy is based on a trust in an inherent impulse toward growth in every individual (Rogers, 2016).
Proposed theories of trust concern the question of whether trust is a belief, an expectation, an attitude or an emotion; theoreticians have proposed cognitive, will-based or affective accounts of trust Baier et al. [99,25,[103][104][105].Baier proposed a will-based account of trust involving the competences and the goodwill of the trustee Baier [99]. However, like Jones, the present results reinforce affective accounts of trust, depicting it as a behavior and emotion. It is likely that people diagnosed with severe-stage neurodegenerative diseases preferably rely on trust and self-confidence via their affect and later via their connotation in order to switch to the active mode. Self-confidence involves an affective attitude in addition to one's competence Jones [25]. For Jones there is a parallel between trust and self-confidence. She distinguishes "be self-confident" and "trust yourself" because of an important difference between the two: willingness. With self-confidence, people are only worried about their capacities rather than their will. On the other hand, will is part of trust and one needs to trust oneself -for instance, when one is worried about the possibility of self-sabotage Jones [25]. During the opening session, some patients already expressed their willingness to actively collaborate, meaning they were in the "trust yourself" concept but not yet in the "be self-confident" one.
This way to encompass trust can be seen as an optimistic attitude to oneself. In the standard case, the confident expectation that the one trusted will respond directly and favourably to the thought that the "truster" is counting on them is itself grounded in the attitude of optimism. Trust is optimism about the goodwill and competence of another, the way optimism leads people to anticipate a favourable outcome. This attitude of optimism makes the difference between trust and reliance Jones [25]. Overall, women of the University Hospital of Dijon diagnosed with neurodegenerative diseases made an optimistic choice in according trust.

Attachment
Neurodegenerative disorders do not prevent the fundamental aspects of a personality to manifest. One case expressed this between-people variability by showing a patient regularly refusing to attend the sessions (e.g., entering the room) or remaining in a contemplative mode. There was an opportunity to meet the patient's daughter's and discuss her mother's relationships.
She declared that she had never trusted people throughout her life, even her close family. She had always been a lonely woman, without any attachment. Since John Bowlby, the literature has described attachment theory and close relationships following two dimensions: attachment anxiety and attachment avoidance Bowlby [106]. As such, patient attachment is a mix between on one hand the fears of rejection and abandonment and on the other hand the fear of intimacy, of getting close to others Benett et al. [106][107][108].
The role of patient attachment shows that more securely attached persons are more likely to experience positive adjustment, and therefore their rehabilitation and control of their health-related quality of life is better Benett et al. [107]. It is likely that during this survey, the therapist faced the four attachment categorisations defined by Bartholomew

Strengths & Limitations of the Study
Regarding the methodology, the observed normal distribution of the data confirms that there was a sole field factor influencing the observations. With the additional and exponential influence of two or three field determinants, data sets would have led to a skewed distribution. It can be supposed that this unidimensional model is due to the way the therapist quoted the observations. This is likely why there is some consistency with previous observations collected by the same therapist but under different conditions (different geographical locations and time of the study, mild cognitive impairment with home-based people; Deygout [92]. This is also consistent with other studies in which the therapist's trust was regarded as unidimensional from the patients' perspectives and a unidimensional conceptual model was found more representative and confirmed by the fact that global items were preferred and account for the largest category Hall et al. [113][114][115][116][117][118]. Anyway, a further generalisation of the 36-item grid could be done via the establishment of its inter-rater reliability (Kappa coefficient of concordance), obtained, for instance, from two trained observers rating the same individuals Kinney et al. [22]. This would quantify the bias due to the subjective part of the therapist's quotation. week, which limited this bias.

Implications for Art Therapy Practice
This descriptive study aimed to increase knowledge of the influence of trust that Alzheimer's patients place in the practitioner in art therapy through the care process. As the result of the observed behavioural outcomes, trust can significantly contribute to facilitate the commitment in the activity and therefore the physical engagement in the art making. This is particularly noticeable at the beginning of the process at least during the first sessions when the built interpersonal trust strengthens the therapeutic relationship.
For the art therapists working with Alzheimer's disease people, this study recommends first increasing the empathic approach of practitioners to help the patient to develop trust. It also suggests better efficiency in the healthcare process when using a person-centred approach such as that established by Carl Rogers. Empathy as well as absence of judgment facilitate aesthetic expression and creativity.

Conclusion
Severe-stage Alzheimer's patients require specific interpersonal styles when they face a practitioner in art therapy. A trust-based relationship established by the therapist can be a gateway to reactivate their self-confidence and provoke a physical engagement essential to the art-therapeutic activity, which ultimately will lead to a decrease in anxiety. Illustrated by the model proposed in the present study, Carl Rogers' person-centred approach appeared to be suitable for this purpose. In particular, empathy was regarded as essential to reinforce the patient's trust establishment in the therapeutic relationship. The Rogers approach, as well as his dynamics of change theory, can apply to severe-stage neurodegenerative disease people and the reactivation of patients' self-confidence. Such an approach is recommended at the beginning of the care process, and then must be released when the patients express assertiveness.