Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Journal of the Academy of Hospital Administration

A Patient Satisfaction Survey in Public Hospitals

Author(s): Prasanta Mahapatra*, Srilatha. S**, Sridhar.P***

Vol. 13, No. 2 (2001-07 - 2001-12)

Key Messages:

  • Corruption by all categories of staff was the greatest cause for dissatisfaction, followed by general cleanliness, poor utilities etc.
  • Significantly high level of dissatisfaction was also noted regarding patients' assessment of technical quality of doctor's work and time spent by doctor.
  • Such patient satisfaction surveys form an important tool for any managerial interventus.


A patient satisfaction survey was conducted in 25 District or Area Hospitals managed by the Andhra Pradesh Vaidya Vidhana Parishad (APVVP). The study obtained feedback from patients and, in case the patient could not be interviewed, the attendant, using a modified version of the Patient Satisfaction Questionnaire-III originally developed by Ware and others (Hays, Davies and Ware; 1987). The study refers to the period from May to July, 1999. Altogether 1179 persons were interviewed, including 237 attendants, at the rate of about 40-50 patients per hospital. In each hospital, patients were identified by stratified random sampling. Stratification was on the basis of sex and wards.Most patients had already stayed for more than three days in the hospital and were drawn from all areas of hospital service including surgical, medical and maternity wards. Female and male patients of different ages are equitably represented in the sample. Majority of patients were poor and illiterate.

Overall, the level of patient satisfaction in APVVP was about 65% of what could be achieved. Corruption appears to be very highly prevalent and was the top cause of dissatisfaction among patients. Other important areas of hospital services contributing to patient dissatisfaction were poor utilities like water supply, fans, lights, etc., poor maintenance of toilets and lack of cleanliness, and poor interpersonal or communication skills.

Keywords : Patient satisfaction, Questionnaire, Public hosptial


Measurement of patients' satisfaction with services provided by the concerned hospital is important from two angles. Firstly, patients constitute the hospital's direct clientele. Thus overall satisfaction of the patient is an important aspect of the service itself, apart from other dimensions like technical quality of medical care, effectiveness of treatment etc. Secondly, patient satisfaction provides an indirect measure of the other dimensions as well. There is increasing evidence to suggest that patient's satisfaction is usually correlated with effectiveness of treatment1. Recognising the need for operations research to continuously improve quality of service, hospitals and health care organisations are increasingly turning to measurement of patients' satisfaction 2.

A Patient Satisfaction survey (IHS, 1999) was done in District and Area Hospitals of the Andhra Pradesh Vaidya Vidhana Parishad (APVVP)3. APVVP is the largest provider of first referral hospital services in Andhra Pradesh. This study reports findings of this survey, about the Patient Satisfaction and concerns with respect to first referral public hospitals in AP.


A. Choice of study set and sampling of patients:

All District (19) and six Area hospitals managed by APVVP were taken up for the study. To build up the sampling frame of patients, the number of inpatients whose length of stay in the hospital is 4 days or longer was counted. If this yielded more than 50 patients, then this was used as sampling frame. If the yield was less than 50, patients with stay of 3 days were included recounted and checked if the resultant number reached 50. If it was not yet 50, then patients with stay of two days or longer were included, even if the total list did not exceed 50. In the present study, most of the sampled patients had stayed in the hospital between 3 to 30 days. If the total list was less than 50, no further sampling was needed, and every one in the list was taken up for the survey. If the total number of persons in the resultant sampling frame (persons with stays of more than 4,3 or 2 days as the case may be) exceeded 50, then only 50 persons were chosen randomly. The list was stratified by major wards and random samples were drawn from each ward, proportionate to its bed strength. The distribution was almost balanced among different services and wards. The survey was conducted in May to July 1999.

B. Patient satisfaction questionnaire:

The Patient Satisfaction Questionnaire - III (PSQ-III) developed by Ware and others, and reproduced in Wilkin and others (1992) formed the core part of the questionnaire used in this study4. PSQ-III serial numbers mentioned below refer to the questionnaire from this publication. About ten questions from the PSQ-III were dropped since pilot testing for this study showed that people had difficulty in either grasping the meaning of the question or did not feel comfortable to give an answer. We added eight questions to factor in the quality of nursing services and general cleanliness dimensions. Seven of these are structured questions similar to the PSQ-III items. The eighth item is an open ended question to enable the respondent give a more spontaneous and qualitative assessment about hospital services. The resultant questionnaire is shown in table-4. Note the question identification scheme, since it would help in interpretation of the results. Each question has been given an identity. Questions borrowed from the PSQ-III start with the alphabet M and those added by Institute of Health Systems (IHS) start with I. Thus MPSQxx means the question is originally from the PSQ-III and IPSQxx means it was added by us for this study. xx represents the serial number of the questions. For questions added by IHS serial number starts from 52.

Table-I: Examples of negative and positively framed questions and corresponding system of scoring of items in the patient satisfaction questionnaire.

Negatively framed question:

You are usually kept waiting for a long time when you need doctor's attention/consultation (MPSQ7).

Choice-> Strongly
Agree Uncertain Disagree Strongly
Score-> 1 2 3 4 5

Positvely framed question:

You have easy access to the medical specialists in the hospital (MPSQ25)

Choice-> Strongly
Agree Uncertain Disagree Strongly
Score-> 5 4 3 2 1

To minimise framing effect, PSQ uses both negative and positively framed questions for the same issue. The same approach was retained for this study. The order of presentation of questions was randomised to avoid, any bias due to sequencing of items. table-4 shows the serial numbers actually assigned to each item, in the survey instrument. Table-I shows an example of a pair of negative and positively framed questions, and the system of scoring for data analysis. Note that the scoring system is reversed for the positively framed question, so that a higher score represents more satisfaction with the hospital's services. Thus a score of 1 or 2 would mean poor performance. Score of 3 means the respondent was uncertain. Scores 4 and 5 means good performance, by the hospital.

Response to each item in the PSQ is obtained with the help of a rating scale, consisting of five categories ranging from strong disagreement to strong agreement. On theoretical grounds it is difficult to predict the properties of a statistic, derived by adding scores across, more than one item. The strategy of balancing negatively framed questions with positively framed questions on the same issue, would, to some extent work to cancel out framing biases, if any, when scores for a pair of items on the same issue but with different frames of presentation are added. Even then, we can not be sure of the validity, reliability of a composite score obtained by adding up multiple items. Ware and others tested the validity and reliability of the composite scores for each sub scale and the single composite of scale of overall satisfaction, with the help of other intuitively appealing and commonsense measures of consistency with level of satisfaction. The PSQ sub scale and composite scores were found to have high validity and reliability. The individual items, however, did not show the same degree of validity and reliability. The key point, being made here is that, summation of scores to arrive at single and sub scale composite scores of patient satisfaction, has empirical support about its validity and reliability.

The items in case of PSQ-III. Table-4 gives sub scale assignment of each item. Access - Availability - Convenience (AAC), Communication (COM), Financial Aspects (FIN), General Satisfaction (GS), Interpersonal Aspects (IPA), Technical Quality (TQ), Time spent With Doctor (TWD). Translation to Telugu was done following translation retranslation procedure by leplege and Verdier, 19955.

C. Data collection:

Interviewers consisted of IHS faculty and research assistants. The surveyor would meet the nurse on duty, explain and reassure that purpose of the study is to constructively estimate the quality of services and to understand how they can be improved. If the patient was unable to participate, the surveyor would look for the attendant who has been with the patient regularly. In case of minors like children, the mother or such other attendant who has been with the patient most of the time were sought for. Each respondent was reassured that the study is designed to improved the quality of services in hospitals, privacy and confidentiality with be maintained and that only statistical information with be used. The last question in the questionnaire has both closed and an open ended component to record special remarks made by the respondent. Data entry was done twice using EpiInfo, and both sets were checked for consistency, using check facility. Discrepancies were rectified by referring to the original interview forms.

Description and profile of the sample:

Altogether 1179 patients were surveyed. Of these 942 patients were able to take the interview. For the rest 237 patients a relative was interviewed. Patients in the sample are from all age groups. The sample is fairly balanced between females and males. There were more females in the reproductive age group of 15-44. Attendants had to be interviewed mostly for children and the elderly. Most of the patients are from poor socioeconomic background (Table-2) with 77% having white ration card. These cards are issued to household whose annual income is less than a cut off level of Rs. 11000 per anum. It was observed that 64% were illiterate, and 28% were daily wage earning labourers.

Table-2: Distribution of respondents by socioeconomic status of the household (n = 1179)

Ration card Occupation Educational status (1179)
Card type % Occupation % Literacy/Education %
White *(Poor) 77.27 Labourer 27.91 Illiterate 64.21
Pink 9.92 House wife 18.15 Primary 23.41
Not available 12.81 Agriculture 10.43 Secondary 9.67
*House holds income less Others 35.96 College 2.71
than Rs. 11000/annum Service 7.55

Results and analysis for all 25 APVVP hospitals:

Overall level of satisfaction with services stood at 65%. The average composite satisfaction score was 153 against a maximum possible score of 235. Sub scale wise scores (table-3) shows lowest level of satisfaction in the General satisfaction sub scale. This sub scale includes items about general cleanliness, toilets facilities, and overall satisfaction with medical care. Next to the lowest level of satisfaction scores is obtained by the time spent with doctor and technical quality sub scales. Both sub scales have items designed to measure patients assessment of doctor's work. Level of satisfaction with communications, access, availability and convenience, and interpersonal aspects was comparatively better. The highest level of satisfaction obtained by any scale was 76%, and considering that this kind of surveys usually tend to generate positive assessment from patients, the level of satisfaction scores for APVVP hospitals appears to be low.

Table-3: Patient satisfaction scores for all 25 APVVP hospitals (1179 patient responses).

Service component Number of items Maximum possible score Actual score Level of Satisfaction
Composite score 47 235 153 65%
Accress availability and convenience 9 45 33 73%
Communications 5 25 19 76%
Financial aspects 6 30 20 67%
General satisfaction 8 40 18 45%
Interpersonal aspects 10 50 36 72%
Technical quality 7 35 22 63%
Time spent with doctor 2 10 6 60%

Responses by Patient's Assessment of Hospital Service:

General satisfaction questions: The item MPSQ33 (Table-5) stated that "There are some things about medical care you receive that could be better." Patients' response to this item yielded a higher frequency of poor scores. But its positively framed complement item (MPSQ11) received positive scores much more frequently. Taken together, some of the poor assessment through MPSQ33 and better assessment through MPSQ11 can be attributed to framing effect. IPSQ57 was about patients bed and surroundings. There are quite many poor assessment scores for this question. MPSQ42 proposed that the medical care received by the patient was excellent. This was a positively framed item, without a balancing negatively framed item. Since, this item was positively worded, one would expect higher satisfaction scores for this item. But only 78% of respondents gave a good feedback.

Technical quality: Note the very high level of dissatisfaction expressed through items MPSQ12, MPSQ41 and MPSQ2 (Table-4). All three questions are about patients perceptions regarding the technical quality of doctor's work.

Time spent with doctor: Here again patients assessment of case by doctor's was poor.

Financial aspects: These questions ask patients about the financial burden of hospital care. So effect of things like money spent on buying medicines, paying fees and brides to various hospital staff would show up through these items. Apart from the generally high level of dissatisfaction expressed through almost all items in this sub scale, the response to item MPSQ19 is of interest. This item asked if the patient, some times, goes without medical care, because it is too expensive. The response to this item shows a very high frequency of poor assessment (Table-4).

Table-4: Percentage distribution of responses by patient's assessment of hospital services.

Sub Scale, Question Id, and item followed by pair identification in parenthesis % Good


MPSQ40 If you have a medical question, you can reach a doctor for help without any problem. 78.4
MPSQ1 You could get admitted to this hospital without any trouble. 94.2
MPSQ37 This hospital is conveniently located 69.3
MPSQ16 It's hard for you to get medical care on short notice.11 69.0
MPSQ51 You are able to get medical aid whenever you need it.11 87.1
MPSQ25 You have easy access to the medical specialists in the hospital.12 42.3
MPSQ7 You are usually kept waiting for a long time when you need doctor's attention/consultation12 79.2
MPSQ28 In this hospital people have to wait too long for emergency treatment.13 78.0
MPSQ5 It is easy for you to get medical aid in an emergency.13 82.0


MPSQ13 During your medical visits you are always allowed to say everything you think is important. 88.2
MPSQ6 Doctors are good about explaining the reason for medical test.14. 77.1
MPSQ18 Some times doctors use medical terms without explaining what they mean.14 61.2
MPSQ38 Doctors sometimes ignore what you tell tem.15. 79.0
MPSQ43 Doctors listen carefully to what you to say.15. 87.6

Financial aspects

MPSQ24 You had to pay more than you could afford for medical investigations and other expenses. 71.8
MPSQ4 You are worried sometimes about having to pay large amounts of money for medicines from outside. 70.9
MPSQ32 The amount you have to spend for medical needs is reasonable16. 77.7
MPSQ10 Some times it is a problem to cover your share of the cost for a medical visit16. 45.3
MPSQ14 You feel confident that you get the medical care you need without being set back financially17. 70.1
MPSQ19 Sometimes you go without the medical care you need because it is too expensive17. 14.8

General satisfaction

MPSQ42 All things considered the medical care you received is excellent. 78.1
IPSQ57 You feel comfortable with your bed and surroundings. 79.4
IPSQ54 General cleanliness in the hospital is adequate1. 81.1
IPSQ55 Toilets are not maintained well in the hospital1. 61.1
MPSQ33 There are some things about medical care you receive that could be better2 48.7
MPSQ11 The medical care you have been receiving is just about perfect2 87.4
MPSQ49 You are dissatisfied with something's about the medical care you receive3 71.1
MPSQ3 You are very satisfied with the medical care you are receiving.3 81.8

Interpersonal aspects

MPSQ9 The doctors who treat you have a genuine interest in you as person. 90.7
MPSQ39 When you are receiving medical care they should pay more attention to your privacy. 16.5
MPSQ34 Your doctors treat you in a very friendly and courteous manner4 87.4
MPSQ17 The doctors who treat you should give you more respect.4 49.1
IPSQ56 Nursing care you are receiving in the hospital is adequate.5 81.3
IPSQ53 Nurses act too businesslike and impersonal5 77.2
IPSQ52 Nursing care you are receiving is excellent.6 82.2
IPSQ58 Nurses are aloof and discourteous.6 82.2
MPSQ47 Doctors always do their best to keep you from worrying.7 88.3
MPSQ29 Doctors act too businesslike and impersonal towards you 7 92.0

Technical quality

MPSQ8 You think the hospital has everthing needed to provide complete medical care. 68.5
MPSQ41 Doctors rarely give you advice about ways to avoid illness and stay healthy. 19.0
MPSQ12 Sometimes doctors make you wonder if their diagnosis is correct. 0.0
MPSQ45 You have some doubts about the ability of the doctors who treat you.8 86.3
MPSQ31 Doctors never expose you to unnecessary risk.8 80.5
MPSQ15 They are careful to check everthing when treating and examining you,9 66.5
MPSQ2 Doctors need to be more thorough in treating and examining you.9 31.7

Time spent with doctor

MPSQ46 Doctors usually spend plenty of time with you.10 74.6
MPSQ35 Those who provide you medical care sometimes hurry to much when they reat you.10 11.9

The questionnaire included an open ended question (IPSQ59) linked to MPSQ33. This later item stated that "There are some things about medial care, that could be better." As a corollary to this item, respondents were requested to give their suggestions through the open ended item (IPSQ59). In practice, it turns out, many patients used this question to vent their feelings and experiences with the hospital. Except for one or two apprecitive remarkes, almost all remarks were deprecatory. Surveyors got an impression, that patients and respondents, who felt strongly about the unsatisfactory experience with the hospial, made a special remark through the open ended item IPSQ59. Altogether 432 (37%) of the 1179 respondents made a special remark. These remarks included about 683 pieces of clearly identifiable statement or comments. Almost all (92%) of these comments were adverse. This gives a clear indication of the very high proportion of dissatisfied and frustrated clientele. These remarks meant that the respondent had to spend more time on the interview, since the surveyors would take this up only after completing all other items. Choosing to make adverse remarks, despite the added transaction cost of a prolonged interview, would suggest, that the respondent felt very strongly about it. We made a content analysis of these remarks and classified the comments into a categories representing various concerns expressed by the patients. some of the top concerns of most patients ordered according to the frequency of remarks on the respective theme, are given below (table-5).

Table-5: Top concerns and adverse remarks expressed by patients in 25 APVVP hospitals.

Area of concerns Frequency Percentage
Corruption by hosptial staff (Corrptn) 126 20%
Utilities like water supply, fan, light, etc, (utlts) 109 17%
Toilet and cleanliness (Toilets) 108 17%
Communication and interpersonal skills (ComIP) 73 12%
Supply of drugs (Drugs) 55 9%
Food supply (Food) 51 8%
Linen availability and cleanliness (Linen) 34 5%
Staff shortage (Staff) 11 2%
Diagnostic and General facilities (D&GF) 9 2%
Miscellaneous (Misc) 53 8%
All aderse remarks (All) 629 100%

Corruption by all levels of hospital staff including doctors, nurses, and other supporting staff, lack of utilities like water supply, fans, lights etc., poor maintenance of toilets and very poor general cleanliness, very poor communication, interpersonal skills and lack of respect for feelings of the patient were evidently top concerns in the minds of patients.


This patient satisfaction survey is the first of its kind for public hospitals in India. While the survey revealed depressing feedback, the motivation of APVVP top management to identify areas of concern and measure patient satisfaction is a step in the right direction. There would not be any scope to improve the services, unless such bold steps at measuring client satisfaction is pursued. We feel, repeating such studies at regular interval of say six months will be useful guide for mangerial intervention.


  1. Wilkin David; Hallam Lesley, and Doggett Marie-Anne.Measures of need and outcomes for primary health care. New York: Oxford Medical Publications; 1992.
  2. Nelson Eugene C. and Batalden Paul B. Patient-based quality measurement systems. Quality Management in Health Care. 1993; 2(1): 18-30.
  3. IHS: Institute of health Systems; APVVP Patient Satisfaction Survey, 1999. Report Series # RP02/1999; Institute of Health Systems, HACA Bhavan, Hyderabad AP 500004 India.
  4. Hays RD, Davies AR, Ware JE; Scoring the Medical Outcomes Study Patient Satisfaction Questionnaire: PSQIII.MOSmemorandum, Rand Corporation, Santa Monica, 1987, Unpublished; reproduced in Wilkin and others, 1992 (p233-234) cited here.
  5. Leplege A. and Verdier A. The adaptation of health status measures: methodological aspects of the translation procedure. in: Shumaker Sally A. and Berzon Richard A., Editors. The international assessment of health related quality of life. theory, translation, measurement and analysis.Oxford, New York: Rapid Communications; 1995.

* Director, Institute of Health Systems, HACA Bhavan, Opp. Public Gardens,Hyderabad, 500004, A.P., India.
** Faculty, Quality Assurance, Institute of Health Systems, HACA Bhavan, Opp. Public Gardens, Hyderabad 500004, A.P., India.
*** Faculty, Health Information Systems, Institute of Health Systems, HACA Bhavan, Opp. Public Gardens, Hyderabad, 500004, A.P., India

Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica