Dissatisfactions that Leads to Least Online Rating of Hospitals by its Customers
Received Date: Jan 25, 2018 / Accepted Date: Feb 02, 2018 / Published Date: Feb 05, 2018
Word of mouth publicity has been casting substantial impact on hospital’s business. This has become even more impactful with increasing use of online rating and reviews. A lower average rating can potentially affect the hospital’s business negatively. Average rating gets considerably lowered with customers giving least rating to a hospital. This study attempts at identifying components that leads a customer to give least rating to a hospital. The study analyses 669 descriptive reviews accompanying a rating by qualitatively analysing and grouping them in component of dissatisfactions (CoD). Each CoD was then tested for their association with least online rating to identify significant ones. Out of 5 CoD, 3 were found significant (Medical Care, Conduct and Money making attitude) while remaining 2 were not (System and facilities and Expensiveness). Amongst CoD that were found significant, no significant difference was found in between them in their strength of association with least online rating.
Keywords: Dissatisfaction; Healthcare; Hospital; Online rating; Online reviews; Word of mouth
Word of mouth (WOM) publicity casts substantial impact on hospital’s image and business. This has become even more impactful, in recent years, with increasing use of online rating and reviewing mediums. As pointed out by Negroponte and Maes [1,2], the advent of the Internet and the growth of the World Wide Web, have given consumers an entirely new realm in which they can communicate and thus influence each other. Increasing use of electronic WOM has also been witnessed for hospitals in India. Indian customers usually post their experiences with hospitals on platforms such as mouthshut.com, consumercomplaints.com and google reviews. There is no platform specifically for hospitals, but google reviews seems to be the most used for posting rating and reviews on hospitals. Data from google reviews shows that number of online ratings for hospitals in India has increased by 162% in 2016 as compared to 2015 and by 473% as compared to 2011 (i.e. within 5 years). In last 2 years few corporate hospitals like, Columbia Asia, Narayan Hrudalaya and Apollo Hospitals have started to responds to online customer voices. This indicates that hospitals are now considering online ratings and reviews as an important part for their publicity and image. Growing popularity of online rating is making it necessary for hospitals to understand what can lead to a lower rating by customers.
As of date, there is no empirical research to understand what causes customers to give a specific rating to hospital. This research study attempts at identifying those components, with which customer’s dissatisfaction may increase the chances of hospital being rated low. The study is done on customer ratings of Indian hospitals, using secondary data available online on Google reviews.
Numerous researches have established importance of WOM on hospital’s publicity and image. Ferguson, Paulin et al.  found WOM recommendation as a powerful marketing tool for hospitals. In a research study done in Malaysia, Yeoh et al.  found that most medical tourists in Malaysia were influenced by friends, family, relatives and doctor’s referral. Lauer [5,6] also stated that WOM and physician referral drive healthcare providers’ choice. Traditionally WOM reach of a customer was limited to his/her family, friends, relatives and neighbours. But the advent and increasing popularity of internet had overcome this limitation. It has given rise to a new format of word mouth i.e. Online Customer Reviews (OCR). OCR enables prospective customers to know what other customers have to say about a particular product/service. Effect of OCR on business has been studied by many researchers and in most studies it was found effective. Kostyra et al. , found that valence, moderated by volume and variance, affects customers’ choice for a product. Cheung et al.  reported that comprehensive and relevant OCR influence information adoption by prospective customers. Chevalier and Mayzlin  stated that higher volume of OCR and better valence improves relative sales for Amazon. Ho-Dac, Carson and Moore  reported that Positive (negative) OCRs increase (decrease) the sales of models of weak brands.
No literature could be found that specifically studies effect of OCR on hospital business. Few studies have been conducted on online physician reviews. Gao et al.  stated that online physician rating is rapidly growing and becoming commonplace. Reimann and Strech [12-14] identified 13 dimensions of patient’s experiences and satisfaction that influence rating on physician rating sites.
Most studies support that valence has a significant influence on prospective customers’ choice. 14 studies identified by Kostyra et al.  in his literature review supports this against 2 that do not support it. Also, there are few studies that report that significance of negative reviews. Chen et al. , reported that negative reviews are more influential than positive. Anderson  found that dissatisfied customers do engage in WOM more than satisfied customers.
Literature studied indicates that low valence on OCR can have a negative effect on hospital business. Least online rating can significantly affect the valence. This calls for a need to know what causes customers to give low rating. There is already a dearth of study on relationship between OCR and hospital business and there is no study that identifies the link between customers’ experience of hospital and his/her rating decision. Thus there is a gap in understanding this phenomenon, which this study tries to address. If hospitals can understand what their customers are intolerant towards, when it comes to OCR, appropriate preventive measures can be taken.
User experience and rating
As identified through literature, valence has a significant influence on business. A dissatisfied customer is more likely to give low rating which can reduce valence and ultimately affect the business. Applying these evidences in a hospital setting, it can be stated that a dissatisfied customer can give low rating to hospital which can affect its image and business. However, a question arises as to what type of experiences in a hospital setting is dissatisfactory enough for customers of to give least online rating. To know this, descriptive reviews can be decomposed and those components of hospital can be identified with which customers’ dissatisfaction can result in least online rating.
OCR of hospitals on Google review
On Google, OCR for hospitals has two parts: Quantitative and descriptive. In descriptive OCR, the customer in a free form provides a written description of his experience with the hospital. There is no limit on the length of the description. On an average descriptive reviews were found to be of 93 words in length with standard deviation of 57 words (n=150). It is not mandatory for customers to provide descriptive review. Quantitative review on other hand requires customers to reflect their impression of the hospital on a 1 to 5-point star rating scale with 1 star reflecting worst impression and 5 stars reflecting best impression. These ratings are given for each reviews and an average rating (valence) is also given for every hospital reviewed on Google. Unlike descriptive review, providing quantitative review is mandatory for customers. This gives rise to two types of online reviews: reviews having both parts and reviews having only quantitative components.
For this study, we have considered only those reviews which have both quantitative and descriptive parts in it.
The study is conducted using exploratory sequential design mixed method . Initially the qualitative reviews of sampled hospital were qualitatively analysed to code them in different components of hospital to which customer mentioned their dissatisfaction with. Each component was then statistically tested to measure its association with least online rating given by customers.
Sampling and data
Data for analysis was sampled from customer posted reviews of different hospitals on Google reviews. Initially, a list of hospitals matching the criteria (Table 1) was prepared for four regions of the country (North, South, West and Centre and East). From this list, 39 hospitals were selected through stratified random sampling. From the sampled hospitals, OCR on Google reviews of the sampled hospitals was collected. Sampling plan was designed to reflect the online rating across India. All reviews matching the criteria (Table 1) of the sampled hospitals were collected. The sample size, regional distribution and the selection criteria for hospitals and reviews are given in Table 1.
|Region||Hospitals||Online written reviews (OWR)|
|West & centre||11||117|
Table 1: Sample size, regional distribution and selection criteria.
*Reviews with 5 star rating were excluded on the basis of assumption that these customers are completely satisfied with the hospital. This puts them in an altogether different group making them non-suitable for comparison with those who gave a lower rating. It is also assumed that customers who gave 2, 3 and 4 star rating did have some concerns for not giving highest rating, thus rendering them suitable for comparison with 1 star rating customer.
Descriptive reviews were qualitatively analysed to decompose them in different codes, termed as Component of Dissatisfaction (CoD). These were those components of the hospital with which descriptive reviews reflect its dissatisfaction. At the beginning, a list of 17 anticipated CoD was developed. The list was then continuously modified as the descriptive reviews were assessed for coding into different CoD. Many CoD that were identified as overlapping, interrelated or closely connected were merged under a broader CoC. Finally, 5 CoC that were distinct, were kept in the list. These are described in Table 2.
|Medical Care||Qualification; experience and capabilities of doctors; skills of nurses in patient care; diagnosis; treatment outcome; patients’ death; harm to patient||“Doctor messed it up and finally my father died”
“I don’t believe doctor said that patients in hospitals are prone to get infection. This is just to cover up their mistake”
“Nurse didn’t knew how to apply IV and pricked my baby 4 to 5 times”
“Even after 2 days of admission, they couldn’t tell what was the problem with my mother”
|Conduct towards patient/caretaker||Behaviour of staff; attention to patient; care and support provided; sensitivities to needs; helping attitude||“They don’t even listen and return patient saying that doctor do not meet without appointment”
“Inconsiderate doctors and staff”
“They don’t even tell us what is happening with our patient”
|System and facilities||Location; infrastructure and maintenance of facility; Administrative processes; staff’s efficiency; waiting time; hospital’s policies; accessibility||“Service is very weak and slow” “The front staffs are inefficient. They should at-least hire someone who is good in speaking English language” “No body picks the phone. I had to do multiple rounds to get my investigation reports”|
|Attitude towards money||Rationality of prescriptions; investigations and treatment; referrals; billing policies; Focus on money||“They do cheating if patient is insured” “If you had a simple disease the will loot by doing no of test and they do not do x ray rather prefer CT scan” “They just see patient as money making opportunity”|
|Expensiveness||Cost of services; Total bill||“Expensive hospital”
“The final bill amount was very high. Difficult to afford” “Looks good externally, but will cut your purse down bit by bit”
Table 2: Components of dissatisfaction (CoD) and their description with examples.
Each descriptive review was read and statements that described any form of dissatisfaction were highlighted. Based on the description (Table 2), each CoD that reflected from highlighted statements were noted in front of each descriptive review. Each review was re-examined by a second person to verify if the CoD noting was appropriately done.
Based on identified CoD (Table 2), hypotheses were developed for each one of them, as describe below,
H1: Customers with medical care as CoD are more likely to give least rating to the hospital compared to others.
H2: Customers with conduct towards patient/caretaker as CoD are more likely to give least rating to the hospital compared to others.
H3: Customers with systems and facilities as CoD are more likely to give least rating to the hospital compared to others.
H4: Customers with money making attitude as CoD are more likely to give least rating to the hospital compared to others.
H5: Customers with expensiveness as CoD are more likely to give least rating to the hospital compared to others.
The components that are found significant after testing for above hypotheses will be compared with each other to test whether they vary in their association with 1 star rating. The hypothesis for this is:
H6: The significant CoDs differ with each other in their association with least rating tendency.
Qualitative analysis gave the data on distribution of descriptive reviews amongst 5 components of hospitals. Based on this data, quantitative analysis was directed at testing the hypotheses stated above. As 1 star is the least possible rating that can be given, the analysis was focussed on the frequency of 1 star under different components. Frequencies of other star ratings, even if found statistically significant were not considered as those were not the objectives of this study.
For hypothesis 1, 2, 3, 4 and 5, Chi square test of independence was used to test if 1 star rating is independent of presence/absence of the component under testing. For each component that was under testing, the frequency of each rating was totalled, after controlling other components to prevent their effects on rating. Then the frequency of each rating was summed from those descriptive reviews that did not have the CoD under testing present. These frequencies were plotted on a 4 × 2 matrix table (4 columns for rating and 2 rows for presence or absence of a component). Using Post Hoc Chi square test of association, residual value and its p value of the cell describing presence of component under test and 1 star rating was calculated.
For hypothesis 6, Chi square test of association was used to test if the 5 component are similar in their association with rating given to hospital. In this frequency of each rating for each component (after controlling other components) were plotted on a 4 × 5 matrix table (4 columns for rating and 5 rows for presence of components). Using Chi square test, residual value and p-value of each cell that depict 1 describe 1 star rating for a component was calculated.
Results and Discussion
From qualitative analysis frequency and distribution of components of hospitals were found. All 670 descriptive reviews were analysed for coding in component out of which 77 reviews were found nonconclusive. These were excluded from the study and from remaining 593 reviews, in 114 no CoD was present, i.e. they did not mention any dissatisfaction/concern with the hospital. 479 descriptive reviews had mentioned 1 or more CoD present with 163 reviews mentioning more than 1 component and 316 reviews mentioning any one CoD. The distribution of quantitative rating of the hospital amongst descriptive reviews having presence of a CoD is given in Table 3.
|Component||1||2||3||4||Total for component|
|Medical care||93 (40)||8 (4)||4 (3)||0 (0)||105 (47)
|Conduct toward patient/caretaker||127 (40)||17 (2)||5 (3)||8 (4)||157 (49)
|Inconvenience with facilities and systems||113 (58)||33 (24)||21 (15)||25 (21)||192 (118)
|Money making attitude||106 (43)||11 (3)||5 (2)||0 (0)||122 (48)
|Expensive||30 (9)||14 (5)||19 (12)||30 (28)||93 (54)
|Total for rating||469 (190)||83 (38)||54 (35)||63 (53)||669 (316)|
Table 3: Distribution of quantitative rating to hospital by presence of CoD.
Numbers in parenthesis represents the frequency after controlling for other components.
The percentage in last column is the percentage of QR in which concern for the component was mentioned.
To test whether or not presence of a CoD in descriptive review is positively associated with 1 star rating, post hoc Chi square test of association was used. Since the objective was to see association in only one cell (i.e. the cell that describe presence of a component and 1 star rating), adjusted residual value and its p value of only that cell was considered. Also, as we were specifically looking for an association only in positive direction any negative association reflected through residual value have been considered as no positive association. The result from statistical analysis is given in Table 4. The findings from Table 4 supports hypothesis is H1, H2 and H4 and do not support hypothesis H3 and H5.
|Component||Statistical values||Hypothesis tested|
|Medical care||Residual value: 5.9
|Conduct towards patient care taker||Residual value: 5.5
|Inconvenience with facility and system||Residual value: 1.1
|Money making attitude||Residual value: 6.6
|Expensive||Residual value: -4.5
P not applicable
Table 4: Residual value and its P value of 1st cell in each CoD from post hoc chi square test of association.
To test hypothesis H6 the hypothesis H1, H2 and H4 were statistically tested using Chi square test of association. No significant difference was found between medical care, conduct and money making attitude components and 1 star rating. (Chi square 8.941, p=0.177).
Least online rating of hospital varied with the type component reflected in customer’s descriptive review. Out of 5 components tested 3 were found to be significant for 1 star rating. The conclusions from the study is presented below
Dissatisfaction with medical care, conduct of staff/doctor towards patient/caretaker and attitude towards making money were significant in resulting in least online rating to the hospital. These 3 components did not vary and were equally likely to result in least online rating.
Dissatisfaction with system/facilities was not significant in resulting in 1 star rating.
Being perceived as expensive, when other components are absent, was not significant in resulting in 1 star or even 2 star rating.
The research findings have an implication for healthcare providers and managers of hospitals. By taking care of the significant CoDs hospitals can prevent themselves from getting least rating online, which in turn will prevent low average rating. This in turn will prevent their image and business from the negative impacts of online rating.
There are some important limitations with the study that must be kept in mind while taking its conclusions.
1. The data coming out of qualitative analysis is dependent on what was written by customers in a free form. While this gives an advantage of being free from lead bias, it could be possible that there are certain experiences that were not written, but had played a role in customer’s rating decision. Experiences/feedback not written in QR was not reflected in the data from qualitative analysis.
2. The study also did not capture the customer’s level of dissatisfaction, which can have an effect on their rating decision. Assessing level of dissatisfaction from unstructured QR was not possible and hence the study only classified them into two levels, i.e. whether a component was present or absent.
The study also did not factor the effect of positive experience that may have moderated customers rating decision.
- Amblee N, Bui T (2011) Harnessing the influence of social proof in online shopping: The effect of electronic word of mouth on sales of digital microproducts. Int J Electron Com 16: 91-114.
- Negroponte N, Maes P (1996) Electronic word of mouth. Wired Magazine 4: 1-2.
- Ferguson RJ, Paulin M, Leiriao E (2007) Loyalty and positive word-of-mouth: Patients and hospital personnel as advocates of a customer-centric health care organization. Health Mark Q 23: 59-77.
- Yeoh E, Othman K, Ahmad H (2013) Understanding medical tourists: Word-of-mouth and viral marketing as potent marketing tools. Tourism Management 34: 196-201.
- Tu TH, Lauer JR (2008) Word of mouth and physician referrals still drive health care provider choice. Res Brief 9: 1-8.
- Chintagunta PK, Gopinath S, Venkataraman S (2010) The effects of online user reviews on movie box office performance: Accounting for sequential rollout and aggregation across local markets. Mark Sci 29: 944-957.
- Kostyra DS, Reiner J, Natter M, Klapper D (2016) Decomposing the effects of online customer reviews on brand, price and product attributes. Int J Res Mark 33: 11-26.
- Cheung CM, Lee MK, Rabjohn N (2008) The impact of electronic word-of-mouth: The adoption of online opinions in online customer communities. Internet Res 18: 229-247.
- Chevalier JA, Mayzlin D (2006) The effect of word of mouth on sales: Online book reviews. J marketing res 43: 345-354.
- Ho-Dac NN, Carson SJ, Moore WL (2013) The effects of positive and negative online customer reviews: do brand strength and category maturity matter? J Mark 77: 37-53.
- Gao GG, McCullough JS, Agarwal R, Jha AK (2012) A changing landscape of physician quality reporting: Analysis of patients’ online ratings of their physicians over a 5-year period. J Med Internet Res 14: 38.
- Liu Y (2006) Word of mouth for movies: Its dynamics and impact on box office revenue. J mark 70: 74-89.
- Mudambi SM, Schuff D (2010) What makes a helpful review? A study of customer reviews on Amazon.com. MIS Quarterly 34: 185-200.
- Reimann S, Strech D (2010) The representation of patient experience and satisfaction in physician rating sites. BMC 10: 1.
- Chen Y, Wang Q, Xie J (2011) Online social interactions: A natural experiment on word of mouth versus observational learning. J market res 48: 238-254.
- Anderson EW (1998) Customer satisfaction and word of mouth. J Servic Res 1: 5-17.
- Creswell JW, Clark VLP (2007) Designing and conducting mixed methods research.
Citation: Raza A (2018) Dissatisfactions that Leads to Least Online Rating of Hospitals by its Customers. J Health Med Informat 9: 301. DOI: 10.4172/2157-7420.1000301
Copyright: ©2018 Raza A. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Select your language of interest to view the total content in your interested language
Share This Article
- Total views: 791
- [From(publication date): 0-2018 - Oct 24, 2018]
- Breakdown by view type
- HTML page views: 741
- PDF downloads: 50