For the follow-up of chronic diseases, it is necessary to develop indicators that can easily be assessed, such as the measure of the health-related quality of life (HRQOL), which is governed by specific guidelines for implementation [1, 2]. HRQOL as an indicator provides essential information to the clinicians to estimate the efficiency of their therapeutic and preventive actions [3, 4]. Patients affected by chronic disease have a particular profile, due to their recourse to regular care and the necessity to adapt to their disease, and this can have consequences on HRQOL assessment. Thus, chronic diseases can generate psychological distress  and can be associated with a lower HRQOL . The conceptual framework for our study is a variation on Broffenbrenner’s ecological model , proposed by McLeroy , and explains the multiple levels of influence on health outcomes at both individual and environmental characteristics in HRQoL. The McLeroy model indicates five levels of influence: (a) intrapersonal factors (characteristics of individual such as personality traits, knowledge, attitudes, behavior, self-concept, skills, etc.), (b) interpersonal factors (formal and informal social support systems, including the family, work group, and friendship networks), (c) institutional factors (social institutions, organizations such as schools and healthcare facilities), (d) community factors (relationships among institutions and informal social networks in a defined area), and (e) public policy (local, state, and national laws and policies). For our proposed model, we considered only the influence at the individual level.
Some determinants of HRQOL such as gender, type of disease, age or socio-demographic characteristics (e.g. level of education, professional activity…) have been clearly identified in the literature. According to the original conceptual model of Wilson  reviewed by Ferrans , characteristics pertaining to both the individual and the environment can have an impact on the five major domains of HRQOL, namely biological and physiological factors, symptoms status, functional status, general health perceptions, and overall HRQOL. The effect of psychological characteristics has also been often evoked, but still warrants further exploration . In this study we focus on the personal characteristics, particularly anxiety and optimism. Optimism and trait anxiety are characteristics inherent to every individual, and do not change over time or according to events with which the individual is confronted. Scheier and Carver theorized that the “disposition” towards optimism could be called “dispositional optimism” and proposed the notion of a measure for optimism [12, 13]. They defined it as a relatively stable feature of the personality, which has important consequences on the way a person regulates their actions in the face of difficulties or stressful situations. For anxiety, Spielberger distinguished the notions of « state » and « trait » anxiety. He characterized trait anxiety as relatively stable individual differences in the tendency towards anxiety . This tendency would be consistent according to different types of stressful situations and would be stable over time [14, 15].
Psychological factors could be considered as items determining the quality of life [6, 16]. Furthermore, HRQOL and anxiety or optimism can influence how patients accept a diagnosis, and can be used as an outcome to evaluate the efficiency of a particular therapeutic approach [17, 18]. However, the relation between these factors has never been specifically addressed, and available data in the literature do not yield a consensus regarding the role of these psychological factors in HRQOL.
Many studies have underlined the importance of optimism in the evaluation of HRQOL [4, 19–23] but its impact remains controversial. Optimistic patients may have coping strategies characterized by better acceptance of the disease, and this can contribute to a lower risk of certain chronic diseases and as a result, better HRQOL [24–27]. Using negative coping has been reported to be associated with low levels of optimism and a high level of anxiety . Anxiety is thus associated with a lower HRQOL [29–34]. Accordingly, pessimistic patients could have a lower HRQOL, exacerbated by anxiety or depression [21, 35]. On the other hand, an association between optimism and anxiety or depression and HRQOL may no longer be significant after adjusting for anxiety, depression and socio-demographics variables . Thus, the role of each of these factors has often been taken into account separately in the evaluation of HRQOL, and few studies have evaluated both simultaneously . Furthermore, the results of studies published to date are divergent regarding the role of each factor. If these two factors are related to HRQOL, they are thus potential confounding factors, and it is therefore necessary to know the effect of each trait and to take it into account in the evaluation of HRQOL.
On the basis of Wilson’s  and Ferrans models , the objective of this study is to clarify the relationships between trait anxiety, optimism and HRQOL. We hypothesized that the model has three causal pathways that contribute to the outcome variable, HRQOL: the anxiety-trait as a predictor (α), the impact of optimism as a moderator (β), and the interaction of these two (α and β) . The moderator hypothesis is supported if the interaction is significant. There may also be significant main effects for the predictor and the moderator, but these are not directly relevant conceptually to testing the moderator hypothesis. We aimed to evaluate the relation between HRQOL and the traits optimism and anxiety among patients after hospitalization in relation to their chronic disease.