Predictors of Patient Satisfaction; Quality Dimensions and Demographic Characteristics

Background: The healthcare industry is increasingly growing in a competitive atmosphere. One of the essential key issues for the survival of healthcare organizations is patient satisfaction. This study aimed to investigate the impact of health service quality and demographic characteristics on patient satisfaction with outpatient departments at teaching hospitals aliated with Tehran University of Medical Sciences in Iran. Method: This cross-sectional study was conducted in 2018. A sample of 400 patientsreferring to outpatient departments at teaching hospitalswasrandomly selected. A valid and reliable questionnaire was used to collect data which were then analyzed by using SPSS 23. Results: The mean scores of service quality and patient satisfaction were 3.73 ± 0.51 and 3.61 ± 0.97 out of 5, respectively. Moreover, patients’ demographic characteristics; age, marital status, residence area, and service quality dimensions; admission process, physician consultation, service costs, accessibility, and appointment were identied as the most effectivefactors on patient satisfaction. Conclusion: In order to increase patient satisfaction, physicians and reception staff are advised to provide patients with useful information and cost-effective services. Nonetheless,it seems necessary for teaching hospitalsto establish plans which facilitate payment, appointment, and examination process.


Background
The healthcare industry is increasingly growing in a competitive atmosphere. Patient satisfaction is one of the key issues for the survival of healthcare organizations (1) since satis ed patients are willing to continue using healthcare services from quality institutions and recommend them to others (1,2). Patient satisfaction refers to an individual's assessment of his or her healthcare experiences, expectations, and quality of care (3); it also re ects patients' judgment of their interactions with service providers (4). Patient satisfaction is becoming one of the essential constructs of healthcare services and is associated with completing treatment and an increased likelihood of getting better (5), In addition, it has gained importance to managers during planning, solving organizational problems, and recognizing the overall level of a health unit performance (6). Customer satisfaction is not an adequate requirement for re-visit intention; however, it is considered to be a valuable prerequisite for patients' loyalty intention (7).
Service quality, in the health sector, consists of technical (outcome) quality and functional (process) quality. The former refers to the skills and the accuracy of medical diagnosis and procedures, whereas the latter focuses on the provision of healthcare services to patients, such as admission processes, physical environment, and waiting time, inter alia (8). Traditional evaluations of patient satisfaction mainly focused on technical and physical features of healthcare delivery and patients' feedbacks were not taken into consideration (9), but nowadays technical quality of care has been found to be less important than other factors (10). Assessing medical services in terms of clinical effectiveness can present a number of constraints; therefore, the necessity for e ciency, cost reduction, and high-quality services requires healthcare organizations to improve their skills and gain patients' trust in health service providers (9).
Patient satisfaction is a multidimensional concept which contains in uential factors, such as demographic characteristics, waiting time, information provision, technical competence, interpersonal factors and physical environment (11). Over time, the concept of patient satisfaction has been correlated with health quality and considered as a part of quality consequences (12).
Several systematic review studies indicated a signi cant association between patient satisfaction and elements of health service quality which has been recognized as one of the key predictive factors of patient satisfaction. It was also found that waiting time and doctor-patient relationship, as health service factors, had the greatest impact on patient satisfaction (11,13,14). In addition, a signi cant association was found between patients' satisfaction and their demographic characteristics, such as age and health status. For instance, older patients and healthier patients were usually more satis ed (13,15).
The measurement of the relative weight of quality dimensions can also lead to more effective administrative activities, resource allocation, and decision making and guarantee patient satisfaction (16). However, healthcare systems in most developing countries are not e cient and face serious nancial problems and, therefore, have di culties to resolve issues related to patient satisfaction.
Satis ed patients may demonstrate favorable behaviors which are important for the success of healthcare providers in the long run (17).
Most studies in Iran have mainly focused on the assessment of the relationship between inpatient service quality and patient satisfaction (18) without evaluating hospital or outpatient services. While outpatient departments are regarded as one of the most important parts in health systems (19) because they refer most patients to inpatient departments and patients judge the overall hospital services on the basis of the services they receive in clinics (20). Therefore, the current study aimed to investigate the impact of health service quality and demographics characteristics on patient satisfaction with outpatient departments at teaching hospitals a liated with Tehran University of Medical Sciences (TUMS) in Tehran.

Method
This cross-sectional study was conducted over the rst half of 2018. There are 16 teaching hospitals a liated with TUMS in Tehran (seven general and nine specialized hospitals); however, only four hospitals (two general and two specialized) were randomly chosen. The study population consisted of 400 patients referring to outpatient departments. They were recruited through a multistage systematic random sampling. The budget share of each hospital and the number of questionnaire distributer were determined in proportion to its size (number of outpatients). Outpatient departments work from Saturday to Thursday; consequently, a systematic sampling technique was employed to provide the patients with an opportunity to participate in this study. The participants were required to have a sound perception of quality, so the patients of 18 years and above were included in the study. A questionnaire was completed by the patients just after visiting their doctors and prior to leaving the hospital; however, patients (n = 14) refusing to take part in the study were substituted by other patients.

Data collection tool
Data were collected by using a validated and reliable questionnaire (21) with seven themes on sociodemographic and eight main themes on hospital's outpatient service quality including 37 sub-themes, i.e., accessibility (3 items), appointment (2 items), waiting time (2 items), admission process (3 items), physical environment (6 items), physician services (11 items), disclosure of information to patient (7 items) and cost of services (3 items). It also included seven items on patient satisfaction. The patients were required to indicate the degree to which they agreed with the items by using a ve-point Likert rating scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Data analysis
Data were analyzed by SPSS 23. After the descriptive tests, linear regression analysis was applied to measure the effect of service quality variables and demographic characteristics on patient satisfaction (22).

Results
According to Table 1, 52% (n = 206) of the respondents were from specialized hospitals. About 55% (n = 221) of the patients were male and 73% (n = 290) were married. According to the ndings, 86% (n = 344) of the participants lived in urban areas, and 66% (n = 265) were primary and secondary school graduates.
Finally, 46% (182) of the patients reported their health status to be fair. As highlighted in Table 2, the lowest mean score (2.64) was pertinent to patient waiting time, while the highest mean scores were related to physician consultation (4.17) and service costs (4.15). Regarding the sub-themes, the lowest and highest mean scores were related to Q 7 (Delay and waiting in the clinic to see the doctor) (2.20) and Q 27 (Observing the patient's privacy) (4.45), respectively. Moreover, the mean score and the standard deviation for the overall service quality were 3.73 and 0.51, respectively. Overall Service quality 3.73 0.51 Table 3 presents the mean and standard deviation scores of the overall satisfaction with service quality.
The highest and lowest mean scores were related to Q 1 (Totally, I am satis ed with the clinic and its services) and Q 4 (This clinic and its services are very close to the ideal clinic in my mind), respectively.
The mean score and standard deviation for the overall satisfaction were 3.61 and 0.97, in that order. According to Table 4, the linear regression analysis demonstrated a positive correlation between patients' satisfaction and their age, marital status (married), and residence area. The highest unstandardized coe cient was related to patients with low economic status (b = 1.34). The most signi cant unstandardized coe cient was observed between service quality dimensions, admission process, physician consultation, service costs, accessibility, and appointment and patient satisfaction. The R square value for all variables was 0.662.

Discussion
The current study aimed to investigate the relationship between service quality dimensions and demographic characteristics and patient satisfaction. The total mean score of service quality (3.73) was less than that of the study in Nigeria where the overall mean score of service quality across all dimensions was 4.20 (23). This result was almost in accord with another study in Iran in which the overall service quality mean score was 3.89 (5).
As stated by Cohen, R 2 values greater than 0.25 represent a signi cant variance in the model (24). The R 2 value in this study was 0.662 meaning that approximately 66.2% of the variance of satisfaction were de ned through the service quality dimensions and demographic characteristics as the independent variables. Therefore, the regression model has relatively good predictive power. The results are in line with those of the studies previously conducted in Iran and South Korea (5,25).
The overall patient satisfaction mean score was 3.61 out of 5 that is relatively high; however, other studies in the Central and Eastern European countries and Uganda reported low consumer satisfaction with the healthcare system (10). Satisfaction levels with healthcare services can be associated with patients' expectations (2,10,14), i.e., patients with lower expectations who are uncritical or lack knowledge may be more satis ed with services than patients with higher expectation (2).
All questions obtained good mean scores; however, the highest mean score (3.8 out of 5) was related to Q1, i.e., "Totally, I am satis ed with the clinic and its services". Whereas in other similar studies in Iran and Ethiopia the highest mean score was related to Q7, "I will recommend this clinic to my friends and relatives" (5,26).
Based on the regression results, the admission process was the most signi cant predictor of patient satisfaction. It means that the behavior of reception staff (courtesy, friendliness, and respect) can increase the overall patient satisfaction. The nding is in accord with those of previous studies where patient satisfaction was positively associated with quick and easy admission (11) and scheduled admission (27). Likewise, the results of a study in Turkey referred to admission process as one of the most critical healthcare shortcomings causing patient dissatisfaction (28).
Physician consultation was the second determinant of patient satisfaction. The result can be attributed to patients who were not aware of medicine and medical procedures; consequently, gave higher scores to this item. Similarly, other studies approved the signi cant relationship between physician consultation and patient satisfaction meaning that doctor-patient relationships, effective communication, and empathy during the consultations play an important role in patient satisfaction (5,14,16).
A signi cant relationship was observed between service cost and patient satisfaction (coe cient = 0.26), that is, patients are satis ed if they perceive that out-of-pocket payments are reasonable in terms of value and quality. The result matches the ndings of previous studies where service costs were found to be one of the important determinants of patient satisfaction (5,25,29) Accessibility and appointment were recognized as two other factors affecting patient satisfaction. These ndings are reinforced by those of previous studies in which accessibility was an important and determining factor in patient satisfaction (13,30). The importance of getting an appointment (faster and easier) has been demonstrated in another study (11).
The regression analysis of patients' demographic characteristics and their satisfaction indicated that only age, marital status, and residence area had signi cantly affected patient satisfaction. Although previous studies described the relationships between patients' demographic characteristics and their satisfaction with services, there are not su cient sources in literature to narrate the main reasons of the relationships which can be either due to differences in patient values and expectations or the differences in the treatment they receive (31).
Regarding the patients' age, it was found that the younger patients were more satis ed with services than the older ones. This can be attributed to old patients' frequent healthcare visits, their focus on details of provided services or facilities which meet their special needs (e.g. wheelchair). This is consistent with the ndings of other studies in which the older patients reported lower satisfaction with their hospital services (31,32) Moreover, a statistically signi cant association was found between patients' residence area and the overall satisfaction score. The respondents in urban areas were more satis ed with the services. Perhaps, patients from rural area, after travelling long distances, had higher expectations of services in the urban clinics. These results are in contrast with those of a study in Nigeria where the patients living outside the site of the clinic were more satis ed than those living within the location (33). However, a study by Atkinson and Haran in Brazil indicated that urban populations were more likely to be satis ed with services (34).
As for marital status, single patients were found to be more dissatis ed with provided services, and a negative coe cient (b= -1.31) was observed between the single and married patients. This is in contrast with the ndings of the study by Quintana et al. who conducted a cross-sectional study to assess predictors of patient satisfaction with regard to their socio-demographic variables and found that single or divorced respondents were more satis ed with health services (35).
Finally, a limitation of this study is that the data were collected through a self-report questionnaire in hospitals which may lead to a bias which could have been avoided if the patients were required to answer the questions the day after refereeing to hospitals.

Conclusion
The ndings indicated that admission process, physician consultation, service costs, appointment and accessibility (service quality dimensions), age, marital status and residential area (demographic characteristics) were the most important determinants of patient satisfaction.
In order to increase the patient satisfaction, it is recommended that physicians and reception staff provide better and more useful information and cost-effective services to patients. Moreover, patients were mostly dissatis ed with waiting time; therefore, the teaching hospitals are advised to establish easy payment schemes, use ticket machines, and schedule the timely presence of doctors to examine patients.
The ndings can be valuable for hospital managers to have a better understanding of their patients' special needs and improve their service quality.
Abbreviations TUMS Tehran University of Medical Science Declarations Ethics approval and Consent to participate The ethics approval for the research was granted by the Deputy of Research Affairs, The school of Allied Medical Sciences, Tehran University of Medical Sciences. Prior to the investigation, the patients who participatedvoluntarily were all informed of the aims of the study. Verbal informed consents were takenfrom the patients because according to ethical principles of Iran verbal consents written consent are not required for studies including non-invasive clinical techniques and verbal consents will su ce. To ensure the con dentiality of the information, the required data were collected anonymously. The participants did not have to write their names, phone numbers, and their address in the questionnaire. The participants had the right to withdraw from research at any time.

Consent for Publication
Not applicable.

Availability of data and materials
The data that support the ndings of this study are available from the corresponding author.