metricas
covid
Buscar en
Journal of Healthcare Quality Research
Toda la web
Inicio Journal of Healthcare Quality Research Results of mapping patients expectation using SERVQUAL
Información de la revista
Vol. 35. Núm. 6.
Páginas 381-390 (noviembre - diciembre 2020)
Compartir
Compartir
Descargar PDF
Más opciones de artículo
Visitas
1263
Vol. 35. Núm. 6.
Páginas 381-390 (noviembre - diciembre 2020)
Original Article
Acceso a texto completo
Results of mapping patients expectation using SERVQUAL
Resultados del mapeo de las expectativas de los pacientes realizado con SERVQUAL
Visitas
1263
D. Mehrotraa, S. Bhartiyab,
Autor para correspondencia
shalinibhartiya69@gmail.com

Corresponding author.
a ASET & IT, Amity University, Uttar Pradesh, Noida, U.P., India
b VSIT, VIPS, New Delhi, India
Este artículo ha recibido
Información del artículo
Resumen
Texto completo
Bibliografía
Descargar PDF
Estadísticas
Figuras (5)
Mostrar másMostrar menos
Tablas (5)
Table 1. Features and facilities of healthcare services.
Table 2. Demographic details of the respondents (number of respondents=349).
Table 3. Rotation sum of squared loadings (rotation method: Varimax).
Table 4. Rotation component matrix categorizing variables into five factors.
Table 5. Criteria and sub-criteria finalized for the ranking of hospitals.
Mostrar másMostrar menos
Abstract
Background

Understanding expectations of the patients towards the quality of services that hospitals providing is very important in quality in healthcare. With every health care organization claiming its services to be the best, benchmarking and comparing them to ascertain the real best is very difficult but much required in this competitive world. Patients have a varied need depending upon the resources and criticality of disease. The aim of the study was to identify the factors linked with patient's expectations contributing to the quality of the service delivered by the hospital.

Materials and methods

Multi-speciality hospitals were visited to survey indoor and outdoor patients of different age groups. Omega coefficient and Cronbach α test were used to test the questionnaire's validity and reliability. The factor analysis technique was used to identify factors determining patient expectations. The SERVQUAL framework was used to categorize evaluation criteria.

Results

The resultant factors show variations in prioritizing service demands by the patients. Such a system would enable the patients to select a hospital capable of delivering best quality services.

Conclusion

This study identifies the factors that contribute in delivering high quality service by the hospital. It gives a futuristic view to design a framework for evaluating the quality of the service delivered by a hospital.

Keywords:
Expectation
Factor analysis
Healthcare
Quality of services
Satisfaction
SERVQUAL
Resumen
Antecedentes

Comprender las expectativas de los pacientes sobre la calidad de los servicios prestados por los hospitales es muy importante para la calidad de la atención sanitaria. Como cada organización de atención sanitaria defiende que sus servicios son los mejores, la evaluación comparativa para comprobar que esta afirmación es cierta es muy difícil, pero es a la vez muy necesaria en este mundo competitivo. Los pacientes tienen diferentes necesidades, dependiendo de los recursos y de la gravedad de la enfermedad. El objetivo de este estudio fue identificar los factores vinculados a las expectativas de los pacientes, que contribuyen a la calidad del servicio prestado por el hospital.

Materiales y métodos

Se realizaron visitas a hospitales de múltiples especialidades, para entrevistar a los pacientes internos y externos de diferentes grupos de edad. Se utilizaron el coeficiente omega y la prueba α de Cronbach para probar la validez y fiabilidad del cuestionario. Se utilizó la técnica de análisis factorial para identificar los factores determinantes de las expectativas de los pacientes, y el marco SERVQUAL para categorizar los criterios de evaluación.

Resultados

Los factores resultantes muestran variaciones a la hora de priorizar las demandas de servicio por parte de los pacientes. Dicho sistema permitiría que los pacientes seleccionaran un hospital capaz de prestar servicios de la mejor calidad.

Conclusión

El presente estudio identifica los factores que contribuyen a la prestación de servicios de alta calidad por parte del hospital, y aporta una visión futurista para diseñar un marco de evaluación de la calidad del servicio prestado por el hospital.

Palabras clave:
Expectativa
Análisis factorial
Atención sanitaria
Calidad de los servicios
Satisfacción
SERVQUAL
Texto completo
Introduction

The biggest challenge for any service industry is to deliver the best service so as to maximize customer's satisfaction. “Service quality” is an indicator of how much an organization successfully satisfies the needs of its customers. The service-quality of an organization is best judged by how often a customer acquires services of that organization. In concern with the healthcare industry, quality of service is a critical factor for benchmarking it amidst its competitors. Though hospitals provide nearly similar type of services but the quality of service delivered, varies significantly. The two major causes of such variation are, ‘poor quality services delivered by a hospital’ or ‘uncertainty in patient's satisfaction for similar services delivered by the hospital’. Casual or improper care of patient may lead to permanent physical damage or even death.

‘Expectation’ and ‘Satisfaction’ are highly variable, and vary from person to person. Today, people are quite aware about health plans, services and legal rights. With, increased literacy rate and readily accessible information on the internet, patient demand transparency in the treatment provided by the respective health care unit. It is a big challenge for hospital administrators to satisfy the patient's queries. Thus, it becomes important to identify the factors influencing service quality in hospitals and investigate intra-relationships between these factors. The objective of the study is to understand the key factors1 that patient find critical in providing quality health services. The two types of quality possessed by a hospital are: technical and functional quality. Technical quality is the extent to which diagnoses and procedures are accurate whereas functional quality means the form in which the services are provided.

Various generic tools and frameworks are available to measure organization's service quality. One of the most famous framework2 and standard and reliable tool is Service Quality Measurement Scale (SERVQUAL). SERVQUAL model presented by3 determines gap between perception and expectation of the consumers. A ten-dimensional concept of the service quality was initially proposed. Later, a five-dimensional tool-RATER (Reliability, Assurance, Tangibility, Empathy and Responsiveness), was proposed using exploratory factor analysis from a questionnaire of 97 items. It has been identified as a valid indicator that determines patient satisfaction4 with the quality of healthcare services provided. Meeting patient's expectation requires identification of most crucial service quality dimensions. Hence, a method is sought for detecting the most critical service quality factors5 that would enable the hospitals to dynamically assess and modify the service parameters as per the patient's need.

The study objective was to identify the most important factors for patients to evaluate the quality of services provided by hospital. Some common features that patients look for, were collected through literature review and pilot survey. Based on the expert opinion and result of pilot survey a questionnaire was designed to find the stakeholder opinions. The opinions were mapped to SERVQUAL framework using the factor analysis. As there can be large number of facility and features that patients look for in health services, the factor analysis proved helpful for dimension reduction.

SERVQUAL has been exhaustively utilized to measure quality of hospital services in the past researches. This work extends the SERVQUAL analysis to identify the dependency and variations of demographic data such as ‘Age’ and ‘Frequency of visit’. A micro-level study done for each factor and variation in their likings is evaluated. The study shows the variations obtained on analysing the resultant under various sub categories of these parameters. The analysis reveals that within each factor the priority of the people changes.

Literature review

The literature study revealed various approaches to assess the quality of service provided by the hospitals. It focused on two important aspects. Firstly, we tried to understand the features and facilities of the healthcare services. Specialization-based assessment6 of service quality indicates ‘very low’ satisfaction of the patients and illustrates the need for a robust quality assessment tool. Another research7 stresses on the need to devise a solution for outpatients as lengthy waiting time reduces their satisfaction level in availing the services of the hospital. A hypothetical study8 assessed the quality of health services using partial-least square structural equation method. But, the study was limited to only one hospital. Patient's behavioural intentions and satisfaction with the service quality were tested9 on three dimensions – process, interaction and environment. Exploratory factor analysis10 was applied on a cross-sectional data collected from patients using surgery, obstetrics and medical services. Service quality and determinants of customer satisfaction in Turkey's hospitals were investigated through logistic regression analysis (Table 1).11

Table 1.

Features and facilities of healthcare services.

Research paper  Method used  Criteria 
Sanigest International White Paper(Hospital Compare) (Cercone and Brien, 2010)  Statistical Analysis  • Process of Care• Outcome of Care• Patient's Hospital Experience• Medicare Payment & Volume 
Health Insight (Health insight policy research)  Statistical Analysis  • Examining process of care measures 
Leapfrog Group(Binder, 2008)  Literature Review  • Computerized Physician order Entry System• ICU Staffing• High risk treatments• Leapfrog safe practice scores 
WHO – Performance Assessment Tool for Quality Improvement in Hospitals (PATH)(Groene et al., 2008)  Literature Review  • Clinical Effectiveness & Safety• Patient Centric• Production Efficiency• Staff Orientation• Responsive Governance 
CHKS Top Hospitals Program(Top Hospitals, 2017)  Literature Review  • Clinical Effectiveness• Health Outcomes• Efficiency• Patient Experience• Quality of Care 
National Health Services Choices Hospital Score Card(Papanicolas and Smith (Eds.), 2013)  Literature Review  • Waiting time• Length of stay in hospital• Risk of Readmission• Experience of Surgical Department with specific procedure• Patient rating of care received• Patient experience during treatment• Survival rate for elective procedure• Survival rate for emergency procedure• Cleanliness of hospital 
Evaluation of service quality by using fuzzy MCDM: A case study in Iranian health care centers(Afkhama, Abdi and Komijan 2012)  SERVQUAL  • Empathy• Reliability• Responsiveness• Assurance• Tangible 

Secondly, the SERVQUAL framework applied in healthcare environment. Evaluation of Hospital Service Quality based on Fuzzy Multiple Index approach12 presents 22 performance criteria through a 5-factor system. It involves hospital's whole situation, the security and condition of therapy service, hospital's outfit service measure and affiliated service establishment. SERVQUAL13–15 has been extensively utilized to evaluate and analyze patient's perceptions with the actual delivery of treatment services. These studies were used to design the questionnaire for gathering patient's opinions and analysing overall quality services of hospitals. Service quality gaps in hospitals were analyzed and validated17 by a total unweighted SERVQUAL score. Research18 found that hospital's physical environment, service cost and information shared with the patients play significant role in determining patient satisfaction.

A strong relationship between patient's expectations and service quality was identified through rigorous literature study. Hospitals’ maintaining the quality of services as per the patient's expectations has been a topic of research and debate. Making the patient aware about the technical and medical complexities is sometimes difficult for the hospital administrators. This leads to a communication gap and affects the expectation and satisfaction level of the patient. Moreover, the patient, categorized into various dimensions based on age, frequency of visit, societal status, location, etc., may have varied expectations. Hence, it becomes important to identify the factors meeting patient's expectations and contributing to the high quality service delivery by the hospital.

Methods

It was important to understand the environment and the expectations of the people living in that environment. Every country has different environment, hence the people have different need and expectations from the services, the study considered the Indian perspective and economy.

Questionnaire

The questionnaire included 50 questions initially and was reduced to 23 viable questions. Expert opinions were considered to refine the questions in terms of words, vocabulary and meaning. The questionnaire was divided into two parts. The first part collected demographic details of the patient, whereas the second part included questions related to different quality service dimensions in a hospital. Pilot study through 50 respondents selected randomly. It enabled to identify the gaps in conveying the meaning that the researchers wanted their respondents to interpret. The language was edited accordingly before distributing the questionnaire to the patients. The feedback of 349 patients was collected on a rating of 5 point Likert scale. The scale ranges from 1 to 5, indicating, “Strongly disagree”, “Disagree”, “Neutral”, “Agree” and “Strongly agree”.

Each question needed to be interpreted and assigned to the SERVQUAL factor. Patients, while filling the questionnaire were not aware of the categorization of factors. There had been no verticals shown to the patients and no grouping of questions was done at the initial level. It is only after the factor analysis that patient's opinions could be categorized based on highest loadings in each factor. Using these questionnaires, the criteria and sub-criteria (Table 5) that can significantly contribute in ranking of hospitals based on their services, were identified.

Surveyed hospitals and study variables

The hospitals undertaken for this study are the multi-speciality hospitals located in National Capital Region of India. Medium sized hospitals having the Admission, OTs, Laboratory, OPD and Emergency services were considered to collect the opinion of patients.

Patients at various locations were interviewed. Patients were categorized according to indoor and outdoor patient; gender; age-range; and frequency of visit to the hospital. The questionnaire was dispensed and collected from various patients in OPDs, IPDs and Laboratory waiting queues.

Statistical analysis

Factor reduction technique using a statistical tool, SPSS19 maps 23 factors into the 5 standard broad factors of SERVQUAL model i.e. Tangibility, Reliability, Responsiveness, Assurance and Empathy. The factor analysis identifies inter-dependence of various factors that determine the quality of healthcare services. This study combines multiple factors to produce a reduced component matrix arranging each criterion in one of the five factors identified in SERVQUAL framework.

Table 2 lists the distribution of data obtained from the population considered in this study. The demographic data is divided into 4 major categories, ‘Gender’, ‘Age’, ‘Frequency of visit (number of times respondents visited the hospital in an year)’ and ‘Patient type (i.e. indoor or outdoor)’. The factors considered hugely affect the opinions of the respondents while selecting the hospital of their choice. Many other implicit factors are associated in their decisions, i.e., speciality of the hospital, social status of self, to name a few.

Table 2.

Demographic details of the respondents (number of respondents=349).

Demographic factor  Categories  Frequency 
Gender  MaleFemale  209140 
Age  18–25 years26–35 years36–50 years51–65 years65 years and above  5357639086 
Frequency of visit (number of times respondents visited the hospital in an year)  1–10 Times10–20 Times20–30 Times30–40 TimesMore than 40 times  27701287846 

As this questionnaire is based on multiple Likert questions that form a scale, the reliability of scale must definitely be ascertained. Cronbach alpha (α).20 Alpha limits the reliability analysis to identify the internal consistencies at the factor level. Hence, a better tool, Omega (ω)16 resolves the limitations of alpha by calculating a more precise measure of internal consistencies for the scales. Reliability test, in this study, was conducted using both the methods and a comparative result is shown in Table 2. Alpha (α) value of 0.816 obtained on the input data set signifies it to be reliable for further analysis. Limitations of alpha were overcome through Omega analysis that came out to be 0.827 at 95% confidence level. It proves more realistic estimate of true reliability of scale.

Before performing factor analysis, it is important to find the suitability of data used for the process of factor extraction. Kaiser-Meyer-Olkin (KMO) measure21,22 and Bartlett's Test of Sphericity23 are applied for sampling adequacy. The KMO measure is based on the idea of partial correlation. The KMO index can range from 0 to 1, but a value >0.50 is considered suitable for factor analysis.24,25 In Bartlett's Test of Sphericity the value of ‘p’ (Pearson's coefficient) less than 0.05 is considered significant for factor analysis. A significant value of Bartlett test indicates that the correlation matrix is not an identity matrix. There exist some relationships between the variables included in the analysis. The approximate of Chi-square is 3164.048 with 253 degrees of freedom, which is significant <0.001 level of significance. The KMO statistic of 0.726 is also large (greater than 0.50). KMO index and Bartlett's Test of Sphericity obtained thus signifies the suitability of data for factor analysis.

The aim of factor reduction is to reduce the number of variables into limited number of factors. The purpose is to handle the challenges of computation time and managing the huge amount of data collected for the analysis. This will help in simplified study by removing redundant variables and give a hierarchical model of the variables by grouping them under certain factors. There are a number of methods to obtain the factors like Principal Axis Factoring (PAF), Principal Component Analysis (PCA), and maximum likelihood, canonical, image factoring and alpha factoring.25,26

Results

PCA is applied for factor reduction due to its simplicity and intuitiveness. PCA reduces the dimensionality of the problem to provide easy interpretation. It is applied to summarize each related factor with unique characteristics. However, by SERVQUAL framework, it is required to extract n factors that would summarize the given factors based on relativity. PCA enables to visualize variables especially when in higher dimensions by reducing the plotting space and increasing clarity of observation in the obtained factors.

Factor analysis needs selection of rotation method. As the purpose is to minimize the number of variables to produce independent factors that best fit the data, Varimax method is selected. The Varimax rotation tries to maximize the variance of each factor. In concern with the missing values, data set for this study has no missing value. The Eigen values are the variance of the factors. With these five factors, cumulative variance is explained. The cumulative variance represents the cumulative percentage between present and preceding factors. Initial number of factors were equal to the total number of factors (i.e. 23) considered for this study. Only those factors, having Eigen value >1(Kaiser's criteria i.e. eigenvalue >1 rule)20,27 were considered, therefore, reducing them to 5. The result obtained was divided into 3 columns. The first column (Total) explains the Eigen value such that the first factor always account for the highest Eigen value reducing with each successive factor. Column 2 (% of variance) contains the percent of total variance accounted for by each factor. Third column (Cumulative %) explains the cumulative percentage of variance accounted for by all preceding and current factors, that in this case is 57.353% (Table 3).

Table 3.

Rotation sum of squared loadings (rotation method: Varimax).

Total  % of variance  Cumulative % 
3.426  14.896  14.896 
2.877  12.508  27.404 
2.767  12.029  39.433 
2.493  10.839  50.272 
1.629  7.081  57.353 

Scree plot represents the Eigen value against the factor number. It helps in determining the optimum number of principal components. Eigen value of each principal component is divided by all the Eigen values to obtain the proportion of variance. As visible in Fig. 1, there is significant difference in the amount of total variance in first four columns. Thereafter, from column 5, the graph is almost flat which specifies that each successive factor accounts for much smaller amount of total variance. It concludes that the last huge fall occurs at fifth component, hence using the first five components to be taken as ultimate solution.

Figure 1.

Cronbach alpha for reliability test of the sample.

(0.23MB).

Rotation component matrix represents rotated factor loadings that are the correlation between variables and the factors. The matrix categorizes the variables in one or the other rotated factors that have been extracted through factor analysis. Table 4 showcase the core factors that highly correlated with utmost one rotated factor and less correlated on other rotated factors. The factor loadings <0.3 have not been displayed for the sake of better readability. The results in the table are post-rotation. The pre-rotation failed to give concrete results as the variables were loaded highly onto first factor leaving all other factors weakly visible.

Table 4.

Rotation component matrix categorizing variables into five factors.

Number  Factors  Rotated component matrix
   
Latest_Equipment        .408   
InfraStructure        .522   
Location  .648         
Fee  .720         
Cafetaria          .732 
Personnel_Appearance      .554     
Staff_count          .704 
Behaviour_staff      .447     
Knowledge_staff    .471       
10  Less_Waiting  .487         
11  Ambulance_Response    .666       
12  Ease_admission    .747       
13  Register_onphone      .753     
14  Conveynance_workinghrs    .846       
15  Feedback        −.711   
16  Staff_interest  .739         
17  staff_Response      .565     
18  Claiming_insurance        .590   
19  Error_free record      .686     
20  Cleanliness    .459       
21  Acccuracy_time        .576   
22  Sharing    .339       
23  Securing  .709         

After rotation, the variance of each factor was identified. Factor 1 accounts for 14.896% of the variance; Factor 2 accounts for 12.508% of the variance; Factor 3 accounts for 12.029% of the variance; Factor 4 accounts for 10.839% of the variance and Factor 5 accounts for 7.081% of the variance. In total, all the 5 factors account for 57.353% of the variance in measuring the service quality of hospitals.

Discussion

The factors obtained after dimension reduction were mapped to five broad categories of the SERVQUAL model. Table 5 lists the categorization obtained from factor analysis. The variables that load high on factor 1 relate to ‘EMPATHY’ which means the “Ability of Healthcare Provider to understand the feelings and need of the patient”. ‘EMPATHY’ holds highest share of total variance and plays a major role in achieving patient's satisfaction. Patient assesses a hospital on the availability of various facilities, equipment and easiness of their accessibility. Factor 2 has been labelled ‘RELIABILITY’. The variables falling under factor 2 are concerned with ‘Reliability of Healthcare Services’ extended by the hospitals. The variables loading high on factor 3 are related to ‘RESPONSIVENESS’ that indicates ‘behaviour skills of the hospital staff’. Factor 4, labelled as ‘ASSURANCE’ loads the variables that can ‘enhance trust and confidence’ of patient in obtaining services from the hospital. Finally, the variables that load high on factor 5 labelled as ‘TANGIBILITY’ relate to ‘Physical and Visible Aspects’ affecting the decisions made by patients while selecting the hospitals. ‘TANGIBILITY’ gets a low weightage in total share of the variance because if all the above expectations are met, its weightage in measuring the quality of services reduces significantly.

Table 5.

Criteria and sub-criteria finalized for the ranking of hospitals.

Division of questions   
  EMPATHY – Empathy is the service provider's ability to understand and share the feelings of the customers. 
T1  Location of the hospital 
T2  Fee of the hospital 
T3  Less waiting time 
T4  Interest of the hospital personnel in solving your problem 
T5  Secrecy of patient's data 
  RELIABILITY – ability to perform service with accuracy and dependably. 
RL1  Knowledge of staff to answer the questions 
RL2  Ambulance Service Response time 
RL3  Ease of admission and meeting the doctor 
RL4  Convenient operating hours and availability of doctors 
RL5  Cleanliness and hygiene 
RL6  Information shared by the hospital on its website 
  RESPONSIVENESS – This is the service provider's willingness to help the customers and respond to their needs. 
RP1  Personnel Appearance 
RP2  Behaviour of staff and doctors 
RP3  Convenience of reserving registration service on phone 
RP4  Response to patient's personal requests 
RP5  Error free record 
  ASSURANCE – Assurance is the service provider's ability to encourage trust and confidence 
A1  Latest Medical Equipment 
A2  Hospital Infrastructure 
A3  Taking and giving value to feedback 
A4  Ease of claiming insurance 
A5  Giving accurate dates and times 
  TANGIBILITY – something that is real and definite. In healthcare terms, it may be defined as the physical qualities, equipment and staff, etc. 
E1  Quality of hospital's eatery 
E2  Number of Staff 

The crux of this study reveals that correlations among the variables fit into five common factors whose reliability has already been proven in SERVQUAL framework. The coefficients between the variables and principal components are simply the ‘weights’28 assigned to variables to indicate their importance. The findings can be utilized to rank the hospitals by assigning weights to each factor as per the user's perception and satisfaction. Ranking would ensure the quality of services provided by the hospitals. It would be beneficial not only to patients but also the health professionals. The trust and confidence of the stakeholders would further help in brand building.

Factor analysis resulted in distribution of 23 variables into five major factors with each variable contributing highly to utmost one factor. The study is further extended to observe a change in decision based on the categorization of patients, namely, gender, in-door/out-door, age, and frequency of visit. The analysis found a significant variation in patient's opinion depending on their age and the number of times they require visiting any hospital. The other two factors, gender and in-door/out-door patient, did not show significant variation in the opinion. Mean of variables in each factor calculated for ‘Age’ and ‘Frequency of visit’ to obtain the cumulative average scores. Data set was classified on the basis of ‘Age’ and ‘Frequency of visit’ into different groups.

Age-wise observation classifies respondent's age into five categories (Table 2). The data set divided into five sub-sets where each sub-set corresponds to one age-category. Further, variables were arranged as per the loadings obtained through factor analysis. Means of each variable was then calculated and added to the overall average of each age range. Graph in Fig. 2 clearly states that the patients in the category – ‘middle-aged’ and ‘senior citizens’, ranked ‘RELIABILITY’ higher than any other factor. They have ranked ‘EMPATHY’ and ‘ASSURANCE’ at second and third place respectively. Conversely, young patients gave higher weightage to ‘RESPONSIVENESS’ and ‘TANGIBILITY’. This indicates difference in needs and satisfaction of patients corresponding to a specific age group. At micro level, coefficient of variation identifies significant variation of the factors influenced by the factor ‘Age’, i.e., as age grows, their expectations also change.

Figure 2.

KMO and Bartlett's test.

(0.06MB).

Variation of opinion within each age group of respondents was observed using trend analysis. The purpose was to identify any change in patient's opinion with the advancement of age. Trend chart (Fig. 3) represents the opinions of the patients on five questions that are highly influenced with age. ‘Location of hospital’, ‘Ambulatory Services’, ‘Fee’, ‘Claiming Insurance Easily’ and ‘Accuracy of Time and Date of Next Visit’, were those five questions. The result shows that senior patients gave priority to ‘Location of hospital’, ‘Ambulatory Services’, ‘Accuracy of Time and Date of Next Visit’ over ‘Claiming Insurance Easily’. The younger patients on the other hand find ‘Claiming Insurance Easily’ as an important service over others.

Figure 3.

Scree plot reflecting the relevant factors obtained on rotation.

(0.07MB).

Visit-wise observation classifies the data set into five categories based on ‘Frequency of Visit’ during a year with each sub-set corresponding to one category. Further, variables were arranged as per the loadings obtained through factor analysis. Mean of each variable was calculated and added to the overall average for each subset of ‘Frequency of Visit’. Fig. 4 observes that the patients with larger number of visits to the hospital rank ‘RELIABILITY’ higher than any other factor. ‘EMPATHY’ and ‘ASSURANCE’ were ranked at second and third place respectively. Conversely, patients with lesser need of healthcare services consider ‘RESPONSIVENESS’ and ‘TANGIBILITY’ much important over other factors.

Figure 4.

Age-wise significance of each factor.

(0.12MB).

The trend chart (Fig. 5) shows the variation in opinions of patients that are frequent visitors to the hospitals over the ones who visit less often. The important questions considered for this analysis include, ‘Location of hospital’, ‘Ambulatory Services’, ‘Fee’, ‘Hospital Infrastructure’, ‘Accuracy of Time and Date of Next Visit’ and ‘Quality of Hospital Cafeteria’. The chart shows that apart from ‘Ambulatory Services’ all other questions were differently prioritized by the patients. The ones who visit the hospital less often are more considerate towards ‘Hospital Infrastructure’ whereas the frequent visitors gave higher priority to ‘Location of hospital’, ‘Fee’ and ‘Quality of Hospital Cafeteria’.

Figure 5.

Visit-wise significance of each factor.

(0.42MB).

Skewness in Figs. 3 and 5 is a clear indication of dynamic changes in patient's expectations in correspondence to increasing/decreasing needs of hospital services. Patients, for instance, with frequent needs to visit hospitals expect higher ‘EMPATHY’ (i.e., Location, Fee, Less_waiting, Staff_interest and Secrecy). The facilities vary in hospitals of different locations. In fact, multiple branches of same hospital vary in the quality of services delivered. Patient is more empathetic on the availability and reputation of services delivered by a hospital, wherever located it may be. ‘TANGIBILITY’ (i.e., staff_count and Cafeteria) are highly important for patients who have lesser needs of visiting the hospitals.

The facts clarify that patients view and assess hospital services differently even when the quality of services delivered remain same over time. A comparison with the PCA result is its best indicator. The PCA emphasizes ‘EMPATHY’ as the highest loading factor having highest weight over other loadings. Whereas, the analysis obtained on ‘Frequency of Visits’ and ‘Age-wise distribution of patients’ emphasize ‘RELIABILITY’ having highest correlation over other loadings. Therefore, a framework or a model that could consider the variation of weights obtained in a given variables and provide a dynamic measurement to assess the service quality can be developed.

The factor analysis widely used in multivariable analysis helps to reduce variable into a smaller set by finding latent relations between them. It helps to makes the interpretation easy and computation simpler for complex systems. The analysis is based on patient expectations from the hospital administration and health providers while undergoing treatments.

Funding

This work is not funded by any organization.

Conflicts of interest

The authors confirm that this paper has no conflicts of interest.

References
[1]
P. Manolitzas, E. Grigoroudis, N. Matsatsinis.
Using multicriteria decision analysis to evaluate patient satisfaction in a hospital emergency department.
J Health Manage, (2014), pp. 16
[2]
A. Parasuraman, V. Zeithaml, L.B. Leonard.
A conceptual model of service quality and its implications for future research.
J Market, 49 (1985), pp. 41-50
[3]
A. Parasuraman, V. Zeithaml, L. Berry.
SERVQUAL: a multiple-item scale for measuring consumer perception of service quality.
J Retail, 64 (1988), pp. 12-40
[4]
E. Babakus, W.G. Mangold.
Adapting the SERVQUAL scale to hospital services: an empirical investigation.
Health Serv Res, 26 (1992),
[5]
A. Ameryoun, S. Najafi, B. Nejati-Zarnaqi, S.O. Khalilifar, M. Ajam, A. Ansarimoghadam.
Factor selection for service quality evaluation: a hospital case study.
Int J Health Care Qual Assurance, 30 (2017),
[6]
A. Ziari, M. Ansari, A. Valinejadi.
The gap between the service quality and patients’ expectations in amir-al-momenin hospital of Semnan university of medical sciences in 2016, Semnan, Iran.
Koomesh, (2018), pp. 221-227
[7]
A. Alkuwaiti, A. Maruthamuthu, S. Akgun.
Factors associated with the quality of outpatient service: the application of factor analysis – a case study.
Int J Healthcare Manage, (2018),
[8]
B. Mohamed, N.A. Azizan.
Perceived service quality's effect on patient satisfaction and behavioural compliance.
Int J Health Care Qual Assurance, 28 (2015), pp. 300-314
[9]
E. Zarei, M. Arab, S.M.G. Tabatabaei, A. Rashidian, A.R. Forushani, R. Khabiri.
Understanding patients’ behavioral intentions: evidence from Iran's private hospitals industry.
J Health Org Manage, 28 (2014), pp. 795-810
[10]
A.J. O’Malley, A.M. Zaslavsky, R.D. Hays, K.A. Hepner, S. Keller, P.D. Cleary.
Exploratory factor analyses of the CAHPS® Hospital Pilot Survey responses across and within medical surgical, and obstetric services.
Health Serv Res, 40 (2005), pp. 2078-2095
[11]
H. Zaim, N. Bayyurt, S. Zaim.
Service quality and determinants of customer satisfaction in hospitals: Turkish experience.
Int Bus Econ Res J, 9 (2013), pp. 51-58
[12]
H. Akdag, T. Kalayci, S. Karagoz, H. Zulfikar, D. Giz.
The evaluation of hospital service quality by Fuzzy MCDM.
J Appl Soft Comput, 23 (2014), pp. 239-248
[13]
A.A. Aaron, R.A. Atinga.
Service quality in healthcare institutions: establishing the gaps for policy action.
Int J Health Care Qual Assurance, 26 (2013), pp. 481-492
[14]
Y.P. Pai, S.T. Chary.
Measuring patient-perceived hospital service quality: a conceptual framework.
Int J Health Care Qual Assurance, 29 (2016), pp. 300-323
[15]
P.A. Madhura, L.S. Iyer.
Assessment of SERVQUAL model in hospitals located in Tier II Cities of India.
J Health Manage, 20 (2018), pp. 28-37
[16]
R.P. McDonald.
Test theory: a unified treatment.
Lawrence Erlbaum, (1999),
[17]
A. Chakravarty.
Evaluation of service quality of hospital outpatient department services.
Med J Armed Forces India, 67 (2011), pp. 221-224
[18]
B. Pouragha, E. Zarei.
The effect of outpatient service quality on patient satisfaction in teaching hospitals in Iran.
Mater Sociomed, 28 (2016), pp. 21-25
[19]
A. Field.
Discovering statistics using SPSS.
4th ed., Sage Publications Ltd., (2009),
[20]
H.F. Kaiser.
The application of electronic computers to factor analysis.
Educ Psychol Measure, 20 (1960), pp. 141-151
[21]
H.F. Kaiser.
A second-generation little jiffy.
Psychometrika, 35 (1970), pp. 401-415
[22]
H.F. Kaiser, R. John, Little Jiffy, I.V. Mark.
Educ Psychol Measure, 34 (1974), pp. 111-117
[23]
M.S. Bartlett.
Tests of significance in factor analysis.
Br J Psychol, 3 (1950), pp. 77-85
[24]
B. Williams, T. Brown, A. Onsman.
Exploratory factor analysis: a five-step guide for novices.
JEPHC, 8 (2010), pp. 1-13
[25]
J. Hair, R.E. Anderson, R.L. Tatham, W.C. Black.
Multivariate data analysis.
4th ed., Prentice-Hall Incorporation, (1995),
[26]
B.G. Tabachnick, L.S. Fidell.
Using multivariate statistics.
5th ed., Pearson Education Incorporation, (2007),
[27]
B. Thompson.
Exploratory and confirmatory factor analysis: understanding concepts and applications.
Appl Psychol Measure, 31 (2007), pp. 245-248
[28]
B. Thompson, L.G. Daniel.
Factor analytic evidence for the construct validity of scores: a historical overview and some guidelines.
Educ Psychol Measure, 56 (1996), pp. 197-208
Copyright © 2020. FECA
Descargar PDF
Opciones de artículo
es en pt

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?

Você é um profissional de saúde habilitado a prescrever ou dispensar medicamentos

Quizás le interese:
10.1016/j.jhqr.2023.07.001
No mostrar más