Continuity of care has been garnering more attention in recent years due to the increase in health care complexity, high specialisation and the involvement of a number of services, as well as an increase in patients with chronic diseases and multiple conditions [1,2]. According to the Reid et al. conceptual framework, continuity of care is defined here as ‘the degree to which patients experience care over time as coherent and linked’ [3,4] and it is the result, from the patients’ perspective, of a combination of adequate access to care, good interpersonal skills, good information flow and uptake between professionals and organisations, and good care coordination between professionals to maintain care consistency [3]. Three types of continuity are identified [3,4]: (1) relational: patients’ perceptions of an ongoing, therapeutic relationship with one or more providers, (2) informational: patients’ perceptions of the availability, use and interpretation of information on past events in order to provide care which is appropriate to their current circumstances, and (3) managerial: patients’ perceptions of receiving different services in a coordinated, complementary and unduplicated way. While relational continuity is related to the continuous caring relationship with professionals, both informational and managerial continuity are related to the perception of interaction among providers [3].

In order to monitor and improve continuity of care, it is important to measure it [5], and its assessment should involve the analysis of relational, informational and managerial continuity of care from the patients’ perspective [3,6,7]. Most available instruments are addressed to specific populations [8], such as patients with diabetes [9,10], cancer [11,12], mental illness [13,14] and patients with an unspecified chronic disease [15] or aimed at users attended to in primary care settings or as inpatients [16,17]. The first generic tool developed to address continuity of care across care levels as perceived by health care users, regardless of morbidity, was the CCAENA© questionnaire (Cuestionario Continuidad Asistencial Entre Niveles de AtenciÓn)[18]. This tool is divided into two sections: the first reconstructs the care trajectory for a specific episode, and the second, which is the object of this paper, consists of Likert scales that measure the patients’ perceptions of the three types of continuity. Two subsequent generic tools were designed to explore patients’ perceptions of continuity of care: the Nijmegen Continuity Questionnaire [19] and the questionnaire elaborated by Haggerty et al. [20]. While the CCAENA questionnaire is focused on the perception of the interaction between providers from different care levels, the other tools also include the perception of interaction of professionals from the same care level. The CCAENA questionnaire allows us to explore patients’ trajectories through the health care services in order to identify aspects of health care supply related to continuity of care where improvements could be made. Furthermore, it is the only available questionnaire of this kind that has been designed in a Spanish-speaking country.

The initial validation of the CCAENA questionnaire indicated that this is a useful instrument to assess continuity of care from the patients’ point of view [18]: face and content validity were high, comprehensibility was considered adequate and the interviewer burden was acceptable [18]. Moreover, items showed an adequate internal structure (construct validity), and scales reached acceptable levels of internal consistency. As a result of this testing some changes were made, such as a modification of the items’ scoring system, the positive formulation of all items and the elimination of redundant items [18]. Recently, the questionnaire was applied in a large survey of 1500 users with the aim of analysing patients’ experiences and perceptions of the three types of continuity of care [21,22]. In this article we present a secondary aim, which is to provide additional evidence on the psychometric properties of the scales of the CCAENA questionnaire.

A cross-sectional study was carried out by means of a survey among patients of the Catalonian health care system. The three selected areas were Baix Empordà (rural and semi-urban), the city of Girona (urban) and the Ciutat Vella district of Barcelona (urban). A single provider supplies both primary and secondary care services in Baix Empordà (Serveis de Salut Integrats del Baix Empordà - SSIBE; a public entity under private law) and in Girona (Institut Català de la Salut - ICS; a public entity under public law). In Ciutat Vella, two entities supply primary care (ICS and Institut de Prestacions d'Assistència Mèdica al Personal Municipal - PAMEM) and a different entity provides secondary care (Parc de Salut Mar). The population of 18 years or over served by these organisations in the study areas is 74,144 in Baix Empordà, 83,312 in Girona and 99,093 in Ciutat Vella [23].

Of the patients contacted, 77.5% refused to take part in the study. However, there were no statistically significant differences between the final sample and the population of study in terms of sex and age. Information on the sociodemographic or health characteristics of the population of study is not available for comparison.

Over half of the respondents were female (55%), had an education on a primary or secondary level (65.4%) and were born in Spain (78%). About half of the patients (57%) perceived that their health status was either good or very good, and 24% reported to be suffering from just one medical condition (Table 1).

The CCAENA© questionnaire is a generic tool aimed firstly at exploring patients’ trajectories in health care services to identify the elements of (dis)continuity experienced in the transition across care levels [21] and secondly at measuring the degree of continuity of care perceived by users using Likert scales [22]. The questionnaire was previously validated using an expert group, two pretests and a pilot test with a sample of 200 health care users [18]. The results presented are an extended validation of the scales of the questionnaire. Two hypotheses related to the psychometric characteristics of the CCAENA questionnaire and their ability to discriminate between groups were tested in a large sample of patients from different age groups, suffering from diverse medical conditions and attended to in different care settings of the Catalan health care system.

The first hypothesis, i.e. that the structure of the scales would reproduce the classification of the types of continuity of care defined by Reid et al. [3], is partially supported by the factorial analysis, which identified two factors that represented relational continuity (patient–primary and secondary care provider relationship) and one factor representing the perception of continuity across care levels (transfer of medical information across care levels and care coherence across care levels). These scales also demonstrated an adequate level of internal consistency and the multidimensionality of the scales was supported by the inter-factor correlation. The fact that items composing the ‘continuity across care levels’ scale, which are related to informational and managerial continuity, load on the same factor has a theoretical foundation in recent studies which argue that the most relevant distinction of patients is between relational continuity and ‘seamless care’, which includes aspects of both informational and managerial continuity of care [5,38]. In the validation process of the Nijmegen Continuity Questionnaire, a similar structural distribution of items was observed [19]. We also explored the psychometric properties of the two subscales that at a conceptual level compose the ‘continuity across care levels’ scale: the ‘transfer of medical information across care levels’ and ‘care coherence across care levels’. This subdivision is supported by the conceptual framework that guided the study [3] and by the reliability and multidimensionality analyses, which demonstrate that the internal consistency of each scale is adequate and the scales of informational and managerial continuity are sufficiently uncorrelated to consider them separately.

The second hypothesis, i.e. that the scores would be able to differentiate perceptions between patient groups, is partially supported by the discriminant analyses: scales discriminate patients according to the health care area where they were attended to and according to some of their individual characteristics - i.e. sex and age. Only the ‘continuity across care levels’ scale and its ‘transfer of medical information across care levels’ subscale were able to discriminate patients according to their health status. However, evidence linking continuity of care and morbidity is inconclusive [15,30,34]. Therefore there are two possible explanations: on the one hand, that there are differences in relational and care coherence perceptions according to the health status of patients but scales lack sufficient discriminant ability to detect those differences; on the other hand, it is also possible that perceptions of these types of continuity do not vary according to health status, as suggested by some research [15,30]. This issue should be further explored in future research.

The test suggested that it was advisable to remove four items related to care coherence and four items related to accessibility across care. However, the information provided by these items is highly relevant to our understanding of patients’ continuity of care perceptions and they have been maintained in the questionnaire. The structural behaviour of items related to care coherence could be explained by the fact that its theoretical construct is composed of a number of attributes that are not necessarily correlated among themselves as previously observed [18], i.e. perception of care coordination, communication between services, no duplication of tests or exams, adequate sequence of health care and appropriate follow-up of patients. The lack of internal consistency of items related to accessibility across care levels could be explained by the fact that accessibility to primary care and accessibility to secondary care do not necessarily need to be correlated. The possibility of splitting the scale into two different scales (items 26 and 29; items 27–28) was rejected because the literature recommends that scales have at least three items [39].

In conclusion, this research is in line with those of the previous validation process [18], and suggests that the CCAENA questionnaire is an adequate tool for measuring patients’ perceptions of continuity of care. Providers and researchers interested in improving continuity of care across care levels could apply the CCAENA questionnaire to identify areas for improvement.