Methods
Study population
This was a cross-sectional study conducted in Turkey from October 2011 to January 2012. The sampling method was a self-administered survey of heads of households by using a multistage sampling technique: first, we divided Turkey into seven geographical regions (Aegean, Black Sea, Central Anatolia, Eastern Anatolia, Marmara, Mediterranean, and South-eastern Anatolia); then, we randomly selected one province from each geographical region; two districts from each selected province; two municipalities from each selected district; two quarters from each selected municipality; six blocks from each selected quarter; and two heads of houses from each selected block. Trained interviewers were recruited to explain the objectives and conditions of the study to respondents. Each eligible respondent received one version of the questionnaire during the weekend days and collected back a week later by interviewers. Respondent or a member of the respondent household to be included in the study, must have been under one of the health insurance schemes; used at least two or more types of health care services during the last ten years; was at least 18 years old or older when health reform process began since 2003 and willing to participate. All healthcare providers, health management personnel, politicians, media workers and mentally unstable were excluded. Supervision during data collection phase was ensured in all stages. Out of 672 distributed questionnaires, 482 completed questionnaires were used for analysis, making a response rate of 71.7%.
Questionnaire
A self-administered modified questionnaire was employed to collect the public opinion. The questionnaire had two parts: sociodemographic items on age, gender, marital status, education, area of residency, happiness, health status, and occupation. For the purpose of statistical analysis, we categorized some of the independents variables into two categories. The second part contained 17 items designed to assess people's opinions about the healthcare reforms. Five aspects were measured: accessibility (five questions), availability of resources (three questions), quality of care (four questions), and opinion regarding the public attention paid to the healthcare reforms by politicians and mass media (three questions). Two questions asked for people’s preferences about the old and the new healthcare system and whether they prefer health insurance coverage now or that available a decade ago.
A five-point Likert-type scale was used to score the closed comparative statements. Each statement had response categories ranging from (1) “strongly agree” to (5) “strongly disagree.” Negatively worded questions were reverse scored (so that 1 = 5, 2 = 4, etc.). For the purposes of cross-tabulation and logistic regression analysis, and to assess the people’s opinion toward health reform process, we needed to effectively dichotomize the number of respondents into two contextual groups: high and low (positive and negative opinion) on each dimension and on overall scale. Therefore, dummy variables for (0) negative and (1) positive opinion were constructed and summed from the seventeen items as originally scored (1–5) (range 17–85). Decision was made to dichotomize the summary score based on a median split (cut-off point) into (0) for low or negative opinion toward health reform process and (1) for high or positive opinion toward health reform process as two dependent variables.
Ethics
This study was approved by ethics committee of National University of Malaysia- Medical Center (UKMMC), code number (FF-175- 2011). All respondents gave their written informed consent.
Data analysis
Normality tests were done and all the quantitative data were found to be normally distributed. Data collected were analyzed using Statistical Package for Social Science (SPSS) program version 16.0. Cross-tabulation (Chi-square test) was used for dichotomized characteristics of respondents and people’s opinion. Multiple logistic regressions were performed to identify significant contributing factors for people’s opinions in this study.
Discussion
Our study brought out opinions of the Turkish people on HTP process employing opinion polls survey. We employed a survey, which questions the subjective satisfaction of people with various aspects of care. In the survey, five aspects of care were measured: accessibility, availability of resources, quality of care and public opinion regarding the attention paid to the healthcare reforms by politicians and mass media, and measuring the preference.
The important specialty of the survey is that questions in the survey do not question people’s opinion about only the current healthcare system but their opinion comparing the current system to the one a decade ago (2003). Therefore, the results of the survey provide comparison of the opinion of the general public about the system before and after the HTP reforms, and measure the success (or failure) of the reforms from people’s perspective.
The overall respondents’ opinion was positive (69.3%) when the current situation was compared with that a decade ago. At the same time, 77.6% of them preferred current situation than that in the past. The results were not surprising, because Turkey has been engaged in health sector reforms since 2003. One of the important goals of the reforms is to establish a healthcare system that is responsive to patients. Before the reforms due to the lack of enough health personnel, there was overcrowding in public hospitals, long waiting times, poor quality, poor responsiveness, and low patient satisfaction with the health system [
18]. Performance-based payment system (P4P) and the family medicine (FM) system were among the key interventions to address these problems [
8]. The P4P system links the individual bonus payments of the health personnel to their performance and encourages them to provide productive and qualified services [
18]. An aggregate amount of bonus payments is adjusted by the institutional performance multiplier, which is given to the MoH hospitals by the MoH according to institutional performance audit results considering equally weighted five topics: a) access to examination rooms, b) hospital infrastructure and service processes, c) patient and caregiver’s satisfaction, d) institutional productivity, and e) institutional targets [
19]. Currently, a human-oriented service principle adopted FM system covers the whole country. The main aims of FM system are to provide primary healthcare services to people in need with an easy access to health service utilization and to implement a reasonable referral system that is expected to avoid excessive workload and help to allot adequate time for patients in secondary healthcare level [
8,
20,
21].
In our study when we asked people’s opinion on the availability of health personnel and facilities we observed that 73.7% of them though that were enough doctors and hospitals currently as compared to a decade ago. Actually, the number of hospitals and primary healthcare institutions clearly increased when compared with that before the HTP [
9]. The HTP reforms such as the rights to choose physician and type of hospital (including the private one), which has been implemented since 2004, as well as the implementation of P4P that provided good incentives for many specialist doctors who left their private clinic and joined hospitals. However, the number of physicians per hundred thousand had not changed significantly. Turkey ranks at the bottom of the WHO European Region. The fact that in our study most of the participants were from the urban regions (63.7%), where there was the highest proportion of doctors and specialists, when compared with the rural regions may partly explain our findings [
9].
Healthcare system reform is rarely evaluated from people’s perspectives in most of the developed and developing countries. This reflects lack of interest of political parties in managing public expectations and preferences. In contrast, the expanded oversight role of modern mass media for criticizing and clarifying health system reform procedures in addition to portraying trends in mass opinion in the last two decades has impact on public preferences, especially among the low and medium politically aware ones [
22,
23]. Bostan et al. [
16] showed that “the level of the expectations of the patient was high on the factor of receiving information” [
16]. In fact, health system reform in Turkey has gained special attention from the highest governmental authorities and various media coverage [
9,
24]. This care has touched simple citizens who expressed positively by three quarters in our study.
Furthermore, we have analyzed people’s view about the new healthcare system when compared with pre-reform system by socio-demographic factors. In multivariate analysis, by employing chi-square tests, we observed a statistically significant relationship between characteristics of respondents and their opinions. We have found that the elderly, married females, those who believe that people are happier now than 10 years ago, and those who live in urban area have more positive opinion on the changes. As commonly found in the literature, older people have a critical opinion. They are the major recipients of healthcare services and their judgment is mostly linked to their experience [
7,
25,
26]. As our analysis found out, the elderly has more positive opinions, it is an important indicator of the success of the HTP process from the people’ perspectives.
Turkish family consists of four persons on the average. In the new legislation, healthcare and medication is free of charge up to eighteen years old. Pregnant women are encouraged to contact maternal healthcare centres with monthly incentives [
9], in addition to many special programs for people with special needs, elderly, and chronically ill patients. These services significantly reduced the economic burden on the family [
9]. Similarly, in Croatia, older people, women, and those with lower education or lower income have a negative opinion toward the patients’ copayments for various health services [
26].
In our study we observed that the unemployed, low educated those who perceived themselves as unhealthy and those who live in rural area showed less positive opinions on the HTP changes. In Turkey, the need for healthcare across regions and social classes is not equally distributed. People who live in rural regions have disadvantaged socioeconomic conditions; usually, they have lower education and income level. The mortality and morbidity rates tend to be significantly higher among lower income. In fact, the inverse care rule (access to care inversely related to need for care) still exists in Turkey even after the HTP reforms since significant differences in the number of health staff remain between the least developed regions and other regions [
8,
18,
27,
28].
Although HTP reforms united different public social security schemes under one umbrella, it lags behind in integrating the unemployed or informally employed into system [
7,
29]. These reasons may explain why unemployed, low educated respondents and those who live in rural area have less positive opinions about the HTP reforms and reported more difficulties in access. Also in Turkey, people with high education level have lower probability of having out of pocket expenditure as they have better health status [
7,
29]. Thus, people with low education level have higher probability of having out of pocket expenditure as they may have worse health status. Imply that the results support the original hypothesis that low educated people would be less satisfied with the healthcare system.
We also compare our findings with the results of the life satisfaction surveys published periodically by TURKSTAT. TURKSTAT claims that overall satisfaction with health services among Turkish citizens increased from 39.5% in 2003, just before the beginning of the HTP reforms, to 66.5% in 2007 and 75.85% in 2011 [
14]. Study done by Ali Jadoo et al. [
7] indicated a high level of satisfaction toward national health insurance in Istanbul city [
7]. These results are parallel to our findings as we have found that a vast majority (70%) of respondents have positive opinion on current health system when compared with the system a decade ago. Similarly, Romanian healthcare system has undergone a reform process and have been evaluated a decade later by Bara et al. [
25], which found that more than 74% of Romanian people preferred the current healthcare system than that a decade before [
25]. Balabanova and McKee [
30] evaluated the public perspective toward reforming healthcare financing in Bulgaria. They concluded that "people prefer a universal health insurance system that is equitable, transparent and accountable to most stakeholders, as an antidote to the former tax-based model" [
30]. Koch [
23] noted that people’s opinion on the government-provided health insurance may be changed greatly over a relatively short period of time [
23]. In the case of Turkey, although universal health insurance system is a recent one, the results appeared through reduction of the gap between public satisfaction in Turkey and other European countries [
9] and through preference of the current health insurance system on the past by 76.6% of our study participants. Although the prominence of arguments advanced by political elites may affect the structure of preferences of the community, it seems that the presence of comprehensive health insurance has a positive impact on people's opinion. In contrast, for example, one-third of the uninsured in United States were dissatisfied with the quality of care they received [
31]. The absence of universal healthcare coverage in United States has created serious problems for many Americans who do not have health insurance [
32], such as delayed treatment for a serious illness [
31‐
33].
Finally, in 2011, Ankara Doctors Chamber (Ankara Tabip Odası, ATO) conducted a health services satisfaction survey. This survey has been employed on 290 patients who have received outpatient services from four hospitals. The results show that 65% of the patients are not satisfied with the current healthcare system [
17]. First of all, ATO’s survey results state the view of the patients to the current system, but it does not elicit views of the Turkish people on health system reform process going on since 2003. Second, this survey was restricted to outpatients only, and third, as in our analysis, unhealthy people have less positive views on healthcare system. Thus, ATO’s result does not represent the general idea of public, but it probability has a sample selection bias.
Limitation of study
First, despite use of multistage sampling method to collect the data nationwide and the benefit of household survey as a good source of information, we think that to elicit public opinion; the sample has to be broader including all the provinces. Second, the questionnaire was modified with close comparative statements with five-point Likert-type scale; this may be considered a bias as the public have limited place for their expression. Third, we tried to add more healthcare aspects (accessibility, quality of care, availability of resources, attitude of politicians, and media); however, other aspects such as continuity of care have to be included. Fourth, the idea of comparing the health system situation over a long period of time (i.e. 10 years), depending on the patient's memory, may be punctuated by the loss of important events that most probably will affect the assessment of health reform process from patient’s point of view and considered a recall bias.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
SAAJ: conceived study, collected, coded and analyzed the data, and wrote the first and final draft of the article. SA: advised and contributed to the study design and data analysis. SNS: advised and contributed to the data analysis and writing of article. AMN: advised in the study design. All authors have read and approved the final manuscript.