A cross-sectional study was carried out by means of a survey of users of the Catalan public healthcare system in three selected areas: Baix Empordà, the city of Girona and the Ciutat Vella district of Barcelona. Healthcare areas were selected to explore the potential influence of healthcare factors on CC experiences, specifically the different management models for PC and SC levels: in Baix Empordà and Girona both PC and SC are managed by a single entity, under private law in the former case and under public law in the latter. In Ciutat Vella, PC and SC are managed by independent entities under public and private law. The population served by these organizations in the study areas is 74,144 in Baix Empordà, 83,312 in Girona and 99,093 in Ciutat Vella. All three areas have introduced some coordination mechanisms between care levels, such as shared clinical guidelines and protocols. In terms of information systems, in Baix Empordà, patients have a single electronic medical record for both care levels, whilst in the other two areas electronic records are shared but different.

The study population consisted of patients of 18 years of age or over who had received PC and SC for the same condition in the three months prior to the survey and were assigned to the selected healthcare areas.

Sample size was calculated to analyse the association model between variables at 95% confidence level, to fulfil the de Moivre theorem of expected frequency greater than five as well as to express the fit and likelihood statistics as a chi-square distribution. The sample size required was approximately 400 patients per healthcare area. The final sample size was 1500 and the sample was distributed across the three areas according to the size of served population.

A simple random sample without replacement was selected from the records provided by the study area PC centres and hospitals which permitted the identification of the patients that had used both PC and SC in the three months prior to the interview for the same condition. A list of substitutes which included individuals of the same sex and age group was used to replace any refusals. Patients who had not been attended to by medical professionals or who could not understand or communicate effectively in Spanish or Catalan were excluded.

The CCAENA questionnaire was applied, which is designed to comprehensively evaluate patients’ experiences of CC between care levels. This tool, previously validated,25 is divided into two sections. The first reconstructs the care trajectory for a specific condition in the previous year for relational continuity and the last three months for informational and managerial continuity; and identifies the elements of (dis)continuity experienced in the transition between PC and outpatient, hospital and emergency SC. This paper focuses on the transitions between PC and outpatient SC (see Table I in the Appendix on the online version in this article). The second section, is for patients to evaluate CC via Likert scales.

Data were collected by face-to-face interviews conducted by trained interviewers, from January to May 2010. Interviews were carried out mainly in the patients’ PC centre (93.7%), and some in their homes (6.1%) or other locations (0.2%). Patients were contacted first by mail and then invited to participate by telephone.

Of the patients contacted, 77.5% refused to take part in the study. There were no statistically significant differences between the final sample and the population of study in terms of sex and age. The highest rate of refusals was observed in Barcelona (80.7%), followed by Baix Empordà (77.6%) and Girona (70.3%).

• •

  Dependent variables: elements of the three types of CC in the transition between PC and outpatient SC, as reported by patients. Regarding relational continuity, the stability of the medical team was defined as the reported number of general practitioners (GP) and SC physicians who had treated the patient for the same condition (one or more than one physicians during the last year). The transfer of medical information (informational continuity) was evaluated through the patient's reports of the health professionals’ awareness of: a) the reason for their visit (yes/no), and b) the instructions received at the other care level (yes/no). Lastly, consistency of care (managerial continuity) was evaluated by means of patients’ reports of the way they accessed SC and PC (referral from the other care level or other means, e.g. referral from the same care level or a follow-up visit).

• •

  Individual and contextual independent variables: the individual variables were those that the literature indicates as potential factors influencing patients’ experiences of care5,6,9–17: socio-demographic characteristics (sex, age, education level and immigration status) and declared morbidity (perceived state of health and duration of the condition which prompted them to visit the health services). The contextual variable refers to the healthcare area in which the patient receives medical care.

A descriptive analysis was conducted of patients’ experiences of continuity in the transition from PC to outpatient SC and vice versa, calculating frequencies and 95% confidence intervals. Following this, a series of logistic regression models were generated in order to evaluate the relationship between variables. Robust covariance adjustments (employing the healthcare area variable) were used to account for correlated observations due to clustering.26 Adjusted ORs and their corresponding 95% confidence intervals were obtained. Statistical analyses were carried out using Data Analysis and Statistical Software (STATA), version 11.

The study was conducted in accordance with the current European and Spanish legislation on ethical research. Informed consent was obtained from every interviewee participating in the survey and confidentiality of data was assured through anonymous analysis. The protocol was approved by the Ethical Committee for Clinical Research Parc Salut Mar (2009/3414/I).

In the last year and due to the same condition, 79.7% of interviewees reported to have been seen by the same GP and 84.9% by the same SC physician (taking into consideration only one medical specialization). When asked about the main professional responsible for managing their condition, 40.7% of interviewees identified the SC physician and 45.0% identified both professionals (see Table II in the Appendix on the online version in this article).

Most interviewees (82.0%) reported that the SC physician was aware of the reason for their visit and 81.5% that he/she was aware of the tests carried out at PC level. Likewise, 81.3% of patients reported that the GP was aware of the instructions received at SC level and 84.7% of the treatment prescribed (see Table II in the Appendix on the online version in this article).

With regard to consistency of care, 71.6% of patients reported that outpatient SC consultations that followed a PC visit were due to a referral from that care level, whilst only 36.0% of patients reported that PC consultations that followed an outpatient SC visit were due to a referral from the SC level. Most patients (87.6%) reported that the GP resolved doubts regarding their visit to the SC professional, and 72.0% and 68.6% of interviewees considered that professionals were coordinated regarding the transitions between PC and SC. With regard to accessibility between care levels, 41.6% and 19.0% of interviewees respectively held the opinion that waiting time to be seen at the referral level, SC or PC, was long or excessive (see Table II in the Appendix on the online version in this article).

• •

  Healthcare area: the healthcare area was associated with all types of continuity. Patients from Girona were more likely to report to have been seen by the same GP in the last year, to experience informational transfer across care levels and referral to the other care level (SC and PC) than patients from the other healthcare areas; likewise patients from Baix Empordà, as compared with those from Ciutat Vella. Additionally, patients from Girona and Ciutat Vella were more likely to report to have been treated by the same SC physician than those of Baix Empordà (Table 2).

  Table 2.

  Continuity of care perceived by patients, results of the logistic regression analysis.

  	Relational continuity				Informational continuity				Managerial continuity
  	Same GP consulted in last year (n=1495)		Same SC physician consulted in last year (n=1435)		SC physician was aware of reason for visit (n=743)		GP was aware of instructions from SC (n=645)		Visit to SC as consequence of a referral from PC (n=998)		Visit to GP as consequence of a referral from SC (n=693)
  	%	aOR	%	aOR	%	aOR	%	aOR	%	aOR	%	aOR
  Study area
  Baix Empordà	84.2	Ref.	80.7	Ref.	81.5	Ref.	85.8	Ref.	73.4	Ref.	37.1	Ref.
  Girona	83.5	1.00	85.8	1.46a	90.7	2.47a	89.0	1.57a	80.7	1.57a	45.5	1.47a
  Ciutat Vella	73.3	0.53a	87.3	1.65a	73.1	0.73	72.1	0.43a	62.4	0.52a	28.8	0.67a
  Age group
  18- 35 years	71.7	Ref.	82.9	Ref.	76.5	Ref.	71.1	Ref.	70.4	Ref.	33.3	Ref.
  35- 50 years	77.6	1.48	83.5	1.09	73.7	0.81	81.0	1.83a	72.5	1.50	34.4	1.23a
  51- 65 years	79.1	1.50a	84.0	1.22	84.2	1.47	80.3	1.60	70.4	1.42a	35.4	1.28
  > 65 years	85.0	2.01a	86.9	1.63	86.7	1.47	85.0	1.71	70.8	1.27	37.1	1.25a
  Sex
  Female	78.8	Ref.	83.7	Ref.	81.2	Ref.	80.4	Ref.	68.8	Ref.	35.9	Ref.
  Male	80.7	1.12	14.0	1.15	81.5	1.00	81.7	1.06	74.0	1.30a	35.1	0.98
  Country of birth
  Spain	80.8	Ref.	84.5	Ref.	82.1	Ref.	81.9	Ref.	70.1	Ref.	34.5	Ref.
  Other	75.5	1.05	85.5	1.24	78.9	1.23	76.9	1.06	74.3	1.55a	40.2	1.67
  Level of education
  Incomplete primary education	84.5	Ref.	84.6	Ref.	90.0	Ref.	87.6	Ref.	72.5	Ref.	43.1	Ref.
  Primary education	83.4	0.97	85.3	1.16	84.2	0.62	84.4	0.73	73.6	0.92	37.5	0.80a
  Secondary education	76.6	0.75a	84.9	1.09	78.2	0.49	78.7	0.54a	69.6	0.67a	32.6	0.60a
  University education	76.2	0.80	83.7	0.96	76.3	0.43	72.6	0.43a	69.1	0.62	31.7	0.57a
  Self-perceived state of health
  Very good or good	77.9	Ref.	85.3	Ref.	78.8	Ref.	80.2	Ref.	74.7	Ref.	31.8	Ref.
  Fair, poor or very poor	81.1	1.06	84.2	0.90	83.9	1.04	81.5	0.89	67.7	0.64a	38.3	1.25
  Duration of medical condition
  <1 year	79.4	Ref.	86.7	Ref.	77.4	Ref.	75.5	Ref.	75.8	Ref.	36.0	Ref.
  ≥1 year	79.8	0.87	83.4	0.76	84.5	1.33	84.0	1.53a	67.5	0.61a	35.5	0.87

  GP: general practitioner; PC: primary care; SC: secondary care; aOR: odds ratio adjusted for the other variables in the table; Ref., reference category.

  a

  Statistically significant OR at 95% significance.
• •

  Individual factors: age was associated with all types of continuity of care, but not all continuity elements. Age was particularly associated with the probability of having been seen by the same GP in the last year, the consulted GP being aware of SC instructions, and reporting a visit to PC or SC as a consequence of a referral from the other care level; older patients were more likely to report these elements of continuity than younger ones. Sex and immigration status were associated with managerial continuity; male patients and immigrants were more likely to report a visit to SC via a referral from PC than women and the native population. Patients with lower education levels were more likely to report to have been treated by the same GP, that their GP was aware of the instructions from SC, and to have visited one care level due to referral from the other than more educated patients (Table 2).

• •

  Health-related factors: patients who perceived their state of health as good or very good were more likely to report a visit to SC due to a referral from PC than patients with a fair, poor or very poor self-perceived health status. Patients with a medical condition that had lasted more than one year were more likely to report that the GP was aware of the instructions given in SC than those whose medical condition had lasted less than one year. The inverse relationship was observed regarding the probability of consulting SC as a consequence of a referral from PC (Table 2).

Transitions between PC and outpatient SC were mostly reported as connected and consistent; however, some noteworthy elements of discontinuity were identified. In terms of informational continuity, the results highlighted gaps in the transfer of information. Even if the figures indicate better informational continuity than in other contexts,27,28 they should receive attention, due to the fact that lack of communication across levels is the most frequent cause of patients experiencing poor quality healthcare.29,30 In terms of managerial continuity, elements related to inaccessibility across levels and inconsistency of care arose: a significant percentage of patients considered waiting times inadequate, as described by other studies24,31; over a quarter of patients perceived that care received was not coordinated; and, they reported a low percentage of visits to PC and SC due to referrals from the other care level. Low referral rates may contribute to the perception of lack of coordination,27 and future research should examine whether this low level of referral is consequence of the patients’ lack of information or retention of patients at the SC level.

In terms of relational continuity, it is remarkable the high proportion of patients that identified the SC physician as the main professional responsible for managing their condition. This could be related to low back-referral figures from SC24 or to the patient attributing higher technical competence to SC professionals. Nevertheless, and according to other studies, this does not seem to be jeopardising the patient-GP relationship.24

Patients attended to in organizations in which PC and SC were managed by the same entity experienced more healthcare elements related to continuity, and the most favourable of these reports was given by patients who were attended to at the services managed by a single entity under public law. The implementation of specific coordination strategies might serve to explain the differences observed.

In terms of the individual factors related to experienced CC, an inverse relationship was identified between informational and managerial CC reports and level of education. Available studies suggest that more educated patients have higher expectations,12 judge quality more critically32 and elicit more information.33 Foreign-born patients were more likely to report being referred from PC to SC, reasons for that could be: native population accesses care more often in non-conventional ways; professionals perceive a greater need for care coordination when treating foreign-born patients to reinforce their decisions; or finally, the work overload attributed to foreign-born patients34 encourages professionals to refer them more frequently. However, aspects mediating these results need to be explored.

Patients with a longer duration of medical condition experienced higher levels of informational continuity, in contrast to other studies.16,18 This different result could be due to the patients’ seeing more frequently their physicians, favouring the perception that they have the information they need about their case, or to the existence of specific programs for patients with chronic conditions

In conclusion, patients experienced in the transition between PC and outpatient SC high levels of healthcare elements related to CC, particularly relational continuity. However, identified elements of discontinuity indicate a margin for improvement in healthcare provision. This study is the first to show an association between experiences of all types of CC and healthcare setting. However, this relation needs to be further explored in order to determine which organizational elements might be contributing to enhance CC. Further research is also needed to understand reasons underlying the differences observed in CC experiences between patients. To make progress on those areas, services and professionals perspectives should also be explored. Finally, this research was conducted before two important events of the Catalan healthcare system took place: the reduction in the healthcare budget and the split of the major public provider into different public companies to manage separately PC and SC in the area of study. As such, it will provide a base-line to analyse the potential impact of these measures on healthcare provision.