Patient satisfaction with the quality of nursing care

Abstract Aim To evaluate patients’ satisfaction with the quality of nursing care and examine associated factors. Design A cross‐sectional, descriptive survey study. Methods The sample was composed of 635 patients discharged from a private hospital. Data were collected using “Patient Satisfaction with Nursing Care Quality Questionnaire” with a total of 19 items, and a questionnaire designed to record socio‐demographic characteristics and medical histories between January 1–May 31, 2015. Results Patients were more satisfied with the “Concern and Caring by Nurses” and less satisfied with the “Information You Were Given.” Patients (63.9%) described nursing care offered during hospitalization as excellent. Patients who were 18–35 years old, married, college or university graduates, treated at the surgery and obstetrics–gynaecology units, and patients who stated their health as excellent and hospitalized once or at least five times were more satisfied with the nursing care. According to this study, the nurses needed to show greater amount of interest to the information‐giving process.

consumers will change the healthcare facility they applied for treatment and care (Ksykiewicz-Dorota, Sierpińska, Gorczyca, & Rogala-Pawelczyk, 2011;Shinde & Kapurkar, 2014). Patients who are more satisfied with their care are more likely to follow medically prescribed regimens and thus contributing to the positive influence on health (Buchanan, Dawkins, & Lindo, 2015;Dzomeku, Atinga, Tulukuu, & Mantey, 2013;Fröjd, Swenne, Rubertsson, Gunningberg, & Wadensten, 2011). More satisfied patients are more liable to recommend the hospital to family and friends (Buchanan et al., 2015;Mohanan et al., 2010). Patients' opinions are the best source that can tell the providers of what is important, that is why this information can be used in healthcare planning and evaluation (Abdel Maqsood et al., 2012;Alsaqri, 2016;Merkouris et al., 2013;Villarruz-Sulit, Dans, & Javelosa, 2009). All these changes and developments in the healthcare field require restructuring of all healthcare services, including nursing, through questioning the quality of treatment services (Şişe, 2013).

| Background
Patient satisfaction is a concrete criterion for evaluation of health care and therefore quality of nursing care (Alhusban & Abualrub, 2009;Shinde & Kapurkar, 2014). It provides crucial information for healthcare managers by providing important resources for processes such as those involved in measuring patients' expectations and satisfaction with nursing care quality, improving nursing service quality through identification of areas of failure and planning and implementing necessary training (Abdel Maqsood et al., 2012;Gadalean & Cheptea, 2011;Geçkil, Dündar, & Şahin, 2008). Evaluation of health care involves defining the objectives of care, monitoring healthcare inputs, measuring the extent to which the expected outcomes have been achieved and assessing the extent of any unintended or harmful consequences of the intervention (Alsaqri, 2016;Sitzia & Wood, 1997;Tang, Soong, & Lim, 2013).
Nursing care is one of the major components of healthcare services (Buchanan et al., 2015;Merkouris et al, 2013;Mohanan et al., 2010;Sitzia & Wood, 1997). Patients' satisfaction with nursing care has become an established as the most important predictor of the overall satisfaction with hospital care and an important goal of any healthcare organization (Goh et al., 2016;Laschinger et al., 2005;Mohanan et al., 2010;Reck, 2013). Measuring patients' satisfaction with nursing care could be effective in improving nursing service quality by facilitating the creation of standards for care while monitoring both results and patients' perceptions of quality (Akın & Erdoğan, 2007;Senarath & Gunawardena, 2011;Tang et al., 2013). The nurses have a central role in offering emotional and psychological support to patients and their families in all settings, such as supporting the patient through diagnosis and ensuring optimum care given to them. Besides the provision of technical care, nurses must have the qualified professional knowledge, attitudes and skills, providing the informational, emotional and practical supports (Akhtari-Zavare et al., 2010;Buchanan et al., 2015;Goh et al., 2016).
If healthcare organization managers are able to identify patient expectations, they could accordingly adjust the performance of services that they offer to meet these expectations (Freitas et al, 2014;Fröjd et al., 2011;Milutinovic, Simin, Brkic, & Brkic, 2012).
The surveys in health services concerning health satisfaction are carried out to evaluate the patient satisfaction, to learn patient's expectations, their suggestions and feedbacks, make the quality improvement constantly in all service periods, to search the effects of socio-demographic and treatment periods on patient satisfaction (Buchanan et al., 2015;Özer & Çakıl, 2007;Sitzia & Wood, 1997). That is why patient satisfaction should be measured constantly using valid, reliable assessment instruments to assess care quality, identify variables that affect care and determine which items should be prioritized and which require alteration in the service based on patients' responses (Buchanan et al., 2015;Merkouris et al., 2013). A good assessment instrument measuring the factors that determine patient satisfaction should be developed to improve nursing service quality (Freitas et al., 2014;Laschinger et al., 2005). Therefore, the findings of nursing management research should be used as an indicator of the contribution made by nursing to the patient care process and this could aid the advancement of the profession in terms of scientification (Alsaqri, 2016;Freitas et al., 2014;Goh et al., 2016).

| Research questions
• What is the satisfaction level of patients about the quality of nursing care?
• Is there any relationship between patients' satisfaction with the quality of nursing care according to their socio-demographic characteristics and medical history?

| Design
The study used a descriptive, cross-sectional research design.

| Setting and samples
Participants included 635 hospitalized patients receiving internal medicine, surgery and obstetrics and gynaecology services at a private hospital between January 1-May 1, 2015. The sampling criteria were as follows: patients aged 18 years or older, patients who were discharged, hospitalized for at least 2 nights at the time of data collection, able to speak and understand Turkish, not too confused or ill to complete the questionnaires and agreeing to participate in the study.
The response rate of this study is 92.8%. The survey was not administered to all patients who had not planned their discharge (those were decided or wished to be discharged suddenly) or were transferred to another hospital. Incompletely filled out surveys were not included in the study.

| Ethical considerations
Prior to data collection, the research protocol was reviewed and approved by the relevant scientific ethics committee (IBU Clinical Research Ethical Committee, Ethical Approval Number: 01.11.2014/25-168). Permission to conduct the research was also obtained from hospital administrators. Written approval to use the Patient Satisfaction with Nursing Care Quality Questionnaire (PSNCQQ) and translate it into Turkish was obtained from Laschinger, who developed the scale. All patients provided written informed consent.

| Patient Satisfaction with Nursing Care Quality Questionnaire
The PSNCQQ was designed to measure the extent of anticipated need, assess patient satisfaction following short-stay hospitalization and determine the influence of socio-demographic, personal and other factors at a minimum level. The scale was developed using the Patient Judgements of Hospital Quality Questionnaire, which was developed by a multidisciplinary research team at the Hospital Corporation of America (Laschinger et al., 2005;Reck, 2013).
The scale consists of 19 items pertaining to features of a wide range of nursing activities including nurses' attention, kindness, respect, courtesy, skills, competence and fulfilment of patient needs.
As it is short and it can be completed easily, it has very good psychometric properties that can be used by managers in quality improvement activities (Fröjd et al., 2011;Laschinger et al., 2005). Each item consists of a "signpost," which is a phrase designating its content and a "descriptor," which is a detailed question. The scale also includes a general perceptions section consisting of four additional questions designed to measure satisfaction with the overall quality of care and treatment received during hospitalization, the overall quality of nursing care, thoughts on overall health and the likelihood that the patient would recommend the hospital to relatives and friends (Laschinger et al., 2005;Milutinovic et al., 2012).
The scale was designed for application by administrators in areas requiring improvement, to provide patient-oriented outcomes and for the identification of strong and weak aspects of the nursing care process. Items were based on factors identified as important elements of patient satisfaction with nursing care. The PSNCQQ can be incorporated into existing hospital quality monitoring systems to monitor patient satisfaction. In addition, the PSNCQQ can be used as an evidence-based indicator given its contribution to the patient care process as a result variable, to evaluate changes in departmental and institutional processes. This feedback provides useful information to nurse administrators (Abdel Maqsood et al., 2012;Laschinger et al., 2005).
Participants' responses are provided using a 5-point Likert-type scale. Total possible scores range from 19-95. Lower total scores indicate greater satisfaction with nursing care. The scoring of the scale was: 1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor.

| Data collection
Data were collected using the PSNCQQ, which measures healthrelated properties considered to affect patient satisfaction and a questionnaire, designed in the light of related literature, to record socio-demographic characteristics and medical history. A questionnaire consisting of 16 items pertaining to variables affecting patient satisfaction was developed according to these characteristics. Income levels were measured by the patients' self-perception of their economic status and lifestyle. It was presented in four options: low, moderate, high and very high. Perceived health was measured by a self-reported question was graded by six variables prior to their admission as excellent, good, fair, poor, very poor and unsure.
The data were collected by the researcher. The patients completed the questionnaires prior to their discharge from the hospital.
The patients who agreed to participate in the study were provided with an explanation about the purpose of the study and they signed informed consent forms. Those who refused to participate reported that they did not have the time or were just not interested in participating. Data were collected by face to face interviews from illiterate patients. About the parametric tests, t tests were performed to analyse independent variables with two categories, one-way ANOVAs were performed to analyse independent variables with more than two categories and Pearson's correlation coefficients were used to analyse relationships.

| Validity and reliability analysis
The PSNCQQ was translated into Turkish and the linguistic, and conceptual equivalence of the items was established. Back translation was performed to ensure language equivalence between the English and Turkish versions of the scale. The original scale was translated into Turkish linguists who were highly competent in both languages. Five bilingual experts consisted of a doctor, two nursing faculty members, a nurse manager and a linguist.
Expressions used in the scale were analysed individually and in combination and optimal expressions were selected by forming a pool of 19 items. Back translation from Turkish to English was performed by two trained linguists (English teachers) with knowledge and experience in both languages. The back-translated and original versions of the PSNCQQ were compared and found to be highly similar in meaning and reorganized based on the characteristics of the country. After then, the expert met and reviewed to determine the scope of the validity of the scale. Eight experts (nursing academicians specialized in medical nursing, surgical nursing and nursing administration) provided opinions about meaning and content sufficiency. A pilot study was then conducted from 1-31 December 2014 to determine whether there were any unclear questions in the scale. The data from the pilot study were then excluded from the final data analysis. According to the results of the pilot study, small changes were then made to the expressions in some scale items to increase their understandability.
In this study, the coefficients for correlations between average PSNCQQ item scores ranged from 0.80-0.89, which demonstrated an appropriate level of reliability. Cronbach's α for the PSNCQQ, calculated to determine internal consistency and uniformity, was 0.98, which was very high. In Laschinger et al.'s (2005) study, the coefficients for the correlations between PSNCQQ items ranged from 0.61-0.89 and were described as high and Cronbach α was 0.97, which was described as excellent. Therefore, the results obtained in the current study were similar to those reported by Laschinger et al. (2005). In view of this, the Turkish version of the PSNCQQ could be considered to possess excellent psychometric properties, which were similar to those reported for the original scale.

| Socio-demographic characteristics and medical history
The mean age of the sample age was 47.94 (SD 19.66) years and 37.6% were aged between 18-35 years. The most of participants were women (77.3%), married (74.5%) and college or university graduates (33.2%), at moderate-income level (52.1%) and housewives (31.3%). Of the patients, 2.2% were illiterate. More than half of the sample (61.6%) were admitted to the service directly from the patient admission department and had been hospitalized once in the preceding 2 years (66.6%). The average duration of the current hospitalization was (4.38 SD 5.75) days (Table 1).

| PSNCQQ scores
Analysis of PSNCQQ scores revealed that the item for which satisfaction levels were highest (1.38 SD 0.66) was the "Concern and Caring by Nurses: Courtesy and respect you were given; friendliness and kindness" item. The item for which satisfaction levels were lowest (1.74 SD 0.86) was the "Information You Were Given: How clear and complete the nurses' explanations were about tests, treatments and what to expect" item. Overall, patients' PSNCQQ scores ranged between 1-4.05, with an average score of 1.61 (SD 0.65). This indicated that the level of satisfaction with nursing care was high ( Table 2).
Analysis of PSNCQQ scores for perception-related items showed that 61.4% and 63.9% of participants' responses for the "Quality of the care and service provided during your stay at the hospital" and "Quality of the nursing care provided during your stay at the hospital" items, respectively, were "excellent." In addition, 87.9% of patients stated that they would recommend the hospital to their family and friends.

| Comparison of PSNCQQ scores according to patients' socio-demographic characteristics
The mean PSNCQQ score of patients at the age of 56 years or older was significantly higher (1.75 SD 0.68) in relation to those observed for patients aged 18-35 years (1.50 SD 0.61) and aged

| Comparison of PSNCQQ scores according to patients' medical histories
The mean PSNCQQ score of the patients hospitalized in the inter-

| D ISCUSS I ON
The results of this study revealed similarities and differences with the existing national and international literatures. This issue has been discussed as follow.

| Discussion of finding about the PSNCQQ scores
Measures of patient satisfaction can assess communication in the consultation such as information transfer, patient involvement in decisions and reassurance (Goh et al., 2016;Shinde & Kapurkar, 2014). Effective and continuous interaction and communication are critical determinants in patients' satisfaction, hospital stay and recovery (Koç, Sağlam, & Şenol, 2011;Mohanan et al., 2010;Negarandeh, Bahabadi, & Mamaghani, 2014;Villarruz-Sulit et al., 2009). Health professionals' communication skills play a pivotal role in ensuring that patients feel valued and cared for. The allocation of sufficient time for talking and listening to patients and providing information is a prerequisite for patient satisfaction, as it ensures that patients are less stressed and more engaged and well adjusted (Koç et al., 2011). There is evidence that the health professionals are perceived as communicating well when the patient feels he/she shows individualized interest, understanding and reassurance (Sitzia & Wood, 1997 Information provision and education are important factors affecting patient satisfaction (Abdel Maqsood et al., 2012;Koç et al., 2011;Villarruz-Sulit et al., 2009). Nurses and other healthcare professionals play a key role in providing support and information. Nurses care for the patients on a 24-hr basis and should be empowered to provide requisite information and instructions to the patients (Alhusban & Abualrub, 2009;Shinde & Kapurkar, 2014). Patient education has been linked with positive clinical outcomes such as improved adherence to a therapeutic regime, reduced anxiety and enhanced ability to cope with symptoms (Sitzia & Wood, 1997). It is known that re- Sitzia and Wood (1997) stated in their review study that older people tend to be more satisfied with health care than younger people are.
Similarly, according to Shinde and Kapurkar (2014) older respondents were more satisfied, probably because they were more social and accepting than younger or they had more respect and care for providers.
On the contrary, we found that patients aged 56 years or older were less satisfied than other age groups. This can be related to the fact that the nurses did not pay more attention to elderly patients. Another possible reasons can be that levels of satisfaction could differ according to cultural values or the patients did not held positive attitudes towards events, based on age-related increases in tolerance and maturity levels.
Similarly, in our study, Sitzia and Wood (1997) found that patient gender did not affect satisfaction values and a conclusion reached also in the reports that significantly more men than women. In other studies (Alsaqri, 2016;Arslan & Kelleci, 2011), no relationships were found between gender and patient satisfaction levels. However, while some of these studies (Akın & Erdoğan, 2007;Alhusban & Abualrub, 2009) reported that women's levels of satisfaction with care were higher relative to those observed in men, others (Koç et al., 2011;Milutinovic et al., 2012;Shinde & Kapurkar, 2014) showed higher satisfaction levels in men relative to those observed in women. In addition, in a study conducted by Dzomeku et al. (2013), 38% and 30% of hospitalized men and women, respectively, were completely satisfied with their nursing care. While the reason for these differences can involve cultural characteristics, they can also occur because, relative to men, women pay more attention to hygiene and care and are more anxious.
In this study, college or university graduates were more satisfied relative to those who were literate patients. However, in some other studies (Dzomeku et al., 2013;Geçkil et al., 2008;Milutinovic et al., 2012;Özsoy et al., 2007), literate individuals and primary school graduates reported greater satisfaction with nursing services relative to that reported by college or university graduates. In addition, Sitzia and Wood (1997) indicated that greater satisfaction was associated with lower levels of education. Patients with lower levels of education being most satisfied, similarly, showed that higher educational attainment was strongly associated with dissatisfaction. Some studies (Akın & Erdoğan, 2007;Shinde & Kapurkar, 2014) revealed that the level of education was not associated with patient satisfaction. These study findings indicated that patients expect more from nursing and care as their education levels increase. This can occur because patients with high educational levels possess more information about treatment alternatives and expect higher care standards and therefore are more critical in this regard.
Patients with high incomes tend to anticipate an improvement in their symptoms and expect to receive care from highly qualified staff and they become dissatisfied if they receive care that does not meet their expectations. Patients with low incomes had low health, get lower health care, had less continuous relation with doctors and have difficulties in getting appointments (Shinde & Kapurkar, 2014).
Some studies (Akhtari-Zavare et al., 2010;Arslan & Kelleci, 2011;Özsoy et al., 2007) reported that satisfaction with nursing care did not differ significantly according to income. In our study, patients with high incomes were more satisfied relative to those with moderate incomes. We can say that these patients received care in the direction of their expectations.

| Discussion of finding about to the PSNCQQ scores according to patients' medical histories
Patients who were hospitalized in surgery and obstetrics and gynaecology units were more satisfied relative to those hospitalized in the internal medicine unit. Shinde and Kapurkar (2014) found that the gynaecological ward had a significantly higher percentage of patients' satisfaction with nursing care than the surgical wards. In a study conducted by Alhusban and Abualrub (2009) The results of the present study revealed that the patients who hospitalized once or at least five times in the preceding 2 years were more satisfied relative to those hospitalized twice in the preceding 2 years. Alsaqri (2016) showed that there was a statistically significant difference between previous admissions and patient satisfaction levels. The same study demonstrated patients with a history of admission to hospital during the last 2 years found nurses more caring. It seems that more lengths of stay in hospital increase patients' opportunities for receiving more nurses' care and observing their caring behaviours. Similarly, in these studies (Koç et al., 2011;Milutinovic et al., 2012) satisfaction levels reported by patients who had been hospitalized previously were higher relative to those who had not. In contrast, in a study conducted by Arslan and Kelleci (2011) satisfaction levels reported by patients with previous hospital experience were lower relative to those without previous hospital experience. The result of another study (Akın & Erdoğan, 2007) found no statistical relationship between satisfaction with nursing care and the numbers of hospitalization. According to these results, we can say that patients' expectations can vary according to previous experience in similar situations and as the number of admissions increase, they can compare their care with that received previously. Also, the positivity or negativity of patients' previous experience can be reflected in their approach to current care.
A study (Alsaqri, 2016) indicated that people who perceived themselves as being healthy were more likely to be satisfied with access to care. According to Alsaqri (2016), patients who perceived themselves to be in excellent or good health are more likely to be satisfied with their health care. Also, it is indicated in the same study that, a person's health prior to arrival at hospital, whether through accident, a chronic condition or a voluntary procedure may affect the patients' expectations about the care. In addition, Laschinger et al. (2005) reported that patients with good health status postdischarge report greater satisfaction than those with poor health status. Similarly, in our study, patients with very poor, poor, fair or good health were less satisfied relative to those of patients with excellent health. This may be due to the fact that healthier people do not need as much medical care and they interact with healthcare providers less frequently. They have less opportunity to experience problems with access to health care and therefore may express more satisfaction with access.

| Study limitations
The sample was restricted to patients from the general surgery, obstetrics and internal medical units. In addition, the study was conducted in a single private hospital in Turkey. Therefore, the results cannot be generalized to all hospitals. Future studies should include more than one hospital in both the private and public sectors and the nursing care provided in private and public hospitals should be compared.
Test-retest reliability analysis should have been performed to strengthen the results of the study. Therefore, patients should be surveyed for a second time in 2 weeks of discharge and the results should be tracked and addressed in future studies. Although methodological problems, such as poor return rates and an inability to collect tracking data for all participants occurred in the study, the results could be considered useful because of the stability criterion for patient satisfaction surveys.

| CON CLUS ION
The results revealed that nurses should inform patients about each application and procedure and provide necessary explanations about illness, diagnosis and treatment to ensure patient satisfaction and the provision of high-quality nursing care. The results also showed that nurses should provide care in a framework of respect, favour and courtesy towards patients by emphasizing the importance of communication. Besides these, the patients were highly satisfied with overall quality of hospital care, nursing care and reported that they would recommend this hospital to their families and friends.
Nurse managers could contribute to the quality service provision by evaluating the patient satisfaction with nursing care for the development and improvement of nursing care based on patients' expectations. Data obtained from this evaluation should be considered in determining training requirements for nurses and in-service training programs should be organized to develop nurses' knowledge and skills in care planning. The PSNCQQ is considered useful for nurse administrators in improving nursing care. The scale could allow managers to determine the attitudes of individuals with whom they work and those whom they manage and exert some degree of control over employees' behaviour.

ACK N OWLED G EM ENTS
The authors thank the patients who participated into the study.

CO N FLI C T O F I NTE R E S T
The authors declare that there was no conflict of interests.

AUTH O R CO NTR I B UTI O N S
AK, ZD: Study design. AK, ZD: Data collection and analysis. AK, ZD: Manuscript preparation.