Design of the study
This is a cross-sectional study performed within the framework of the international Quality and Costs of Primary Care in Europe (QUALICOPC) project coordinated by the Netherlands Institute for Health Services Research (NIVEL), aiming to evaluate the quality of PHC in 31 European countries as well as Australia, Canada, and New Zealand [18]. The project took place between March 2010 and February 2014. A set of four questionnaires was developed by the QUALICOPC study team: (1) one for general practitioners (GPs) (“PC Physician” questionnaire), (2) one for patients about their experiences during one specific GP consultation (“Patient Experience” questionnaire), (3) another for patients about the values of PHC they consider important (“Patient Values” questionnaire), and (4) a practice questionnaire about the structure of the PHC setting (“Fieldworker” questionnaire) [19]. The original questionnaires were translated from English to Polish with a formal forward-back translation process. The detailed development of Polish questionnaires, including validation, is available elsewhere [20, 21]. In our study, we used data from two QUALICOPC questionnaires: “Patient Experience” and “Patient Values.”
The study was approved by the Jagiellonian University Bioethics Committee (approval number KBET/104/B/2011).
Sample
The details of the recruitment of Polish physicians and patients has been described previously [21]. In brief, we approached participating patients through their GPs. A nationally representative sample of 220 primary care physicians was selected from the database of the Polish National Health Fund (the exclusive health insurance company of Poland) by a stratified random sampling procedure. The fieldworker visited the participating PC physician and handed her or him the “GP” questionnaire. Then, in the waiting-room, the fieldworker distributed questionnaires to ten consecutive consenting adult patients visiting the selected physician. The first nine patients completed the “Patient Experience” questionnaire and the tenth one filled out the “Patient Values” questionnaire. The questionnaires were filled out immediately after the appointment with the GP and returned in a sealed envelope to the fieldworker. To realize an aimed response of 2200 patient participants, the fieldworkers had to invite a total of 4663 patients (a response rate of 47%). We excluded two incomplete surveys from data analysis. The final sample included 1979 patients completing the “Patient Experience” questionnaire and 219 patients filling out “Patient Values” questionnaire. To check the representativeness of the participating patients, we compared their age, gender, education level, and employment status to national statistics. Our study population included more women and more patients with secondary and higher-level education than the general population.
Measures
Items from both questionnaires were assigned to areas of care, according to the key provided by the international study coordinator of QUALICOPC [18]. In this paper, we focus on patients’ experiences and expectations with four primary care areas: “Accessibility” (ACCS), “Continuity” (CONT), “Quality of care” (QUAL), and “Equity” (EQ).
The “Patient Experience” questionnaire asked whether certain performances occurred during the just-completed appointment with the GP. The answering formats of the performance items were “yes” or “no.” In the “Patient Values” questionnaire, patients rated the importance of each of the statements contained in the patient experience survey on a four-point Likert scale from “not important” to “very important.” To compare the questions about experiences with questions about values and expectations, we developed for each studied PC area: a satisfaction index (SI), a value index (VI), and a met expectation index (MEI).
We rescaled all questionnaire items to a uniform scale ranging from − 1 (extremely negative) to + 1 (extremely positive). In the case of dichotomous experience questions (Yes/No), the values (+ 1) or (− 1) were assigned, respectively. For importance questions measured on the Likert scale, values from (− 1) (for “not important”) to (+ 1) (for “very important”) were given; the others were given intermediate values (− 0.5) or (+ 0.5). The satisfaction and value indexes were calculated as arithmetic means (μ) of experience and importance questionnaire items representing particular PC areas. The met expectation index (MEI) for each PC area (A) was computed as a difference between respective satisfaction index (SI) and value index (VI):
$$ {\mathrm{MEI}}_{\mathrm{A}}={\mathrm{SI}}_{\mathrm{A}}-{\mathrm{VI}}_{\mathrm{A}} $$
Additionally, we computed MEI for each experience questionnaire item and a corresponding value questionnaire item. The MEI could range from (− 1) to (+ 1). The values equal to or near 0 indicated complete fulfillment of patient expectations. Negative values pointed to the areas that are very important to patients, but with which they had relatively unsatisfactory experiences.
Socio-demographic data were collected from all patient participants, including gender, age, education level, country of birth, mother’s country of birth, employment status, health status, and presence of chronic disease(s).
Statistical analysis
Statistical analysis was performed with Statistica 13.1 software (Dell Inc.). To present the results, we used descriptive statistics. To investigate the associations between the values indexes and patients’ characteristics, the Mann-Whitney U test was used for comparing two groups, the Kruskal–Wallis test for more groups, and Spearman’s rank correlation coefficient for two qualitative variables. An α level of p = 0.05 was considered for tests of statistical significance.