Archives of Medical Research 35 (2004) 82–86

ORIGINAL ARTICLE

Patient Ratings of Care at a Rheumatology Out-Patient Unit

Patricia Clark,a,b Pilar Laviellea,c and Carolina Duartec

aUnidad en Epidemiologıa Clınica, Hospital de Especialidades Bernardo Sepulveda, Centro Medico Nacional Siglo XXI (CMN-SXXI),Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico

bFacultad de Medicina, Universidad Nacional Autonoma de Mexico (UNAM), Mexico City, MexicocDepartamento de Reumatologıa, Hospital General de Mexico (HGM), Mexico City, Mexico

Received for publication January 6, 2003; accepted July 18, 2003 (03/002).

Background. Our objective was to determine patient satisfaction with services renderedby health care personnel at the rheumatology out-patient consultation.

Methods. We carried out a cross-sectional survey at a tertiary-level care hospital in MexicoCity. A questionnaire on satisfaction with quality of medical care was applied to 347adult patients who attended the out-patient rheumatology clinic at least twice and who agreedto participate. Questions in the instrument covered three aspects of quality of care, includingstructure, process, and results.

Results. In overall assessment, 83.5% of patients reported satisfaction with care. Structurereceived worst score and process, the best. Outcomes (e.g., disease improvement) receivedintermediate scores. Correlations within these areas were strong but correlation withoverall satisfaction was not high. The majority of patients stated that physician competencewas the most valuable aspect of the visit; the least valuable was waiting time.

Conclusions. Specific strategies need to be implemented to improve certain areas. Overallsatisfaction as a measurement renders no specific and useful information due to complexitywith regard to different dimensions of satisfaction. � 2004 IMSS. Published byElsevier Inc.

Key Words: Satisfaction, Patient satisfaction, Evolution of satisfaction.

Introduction

In the last decade, the concept of patient satisfaction hasbeen recognized as an important variable in health care(1,2). Patient satisfaction—defined as the patient’s inter-pretation of the quality of their interaction with a healthcare provider or health care delivery system (3)—reflectsperception of the manner in which a particular health serviceis functioning.

Positive associations have been found between patientsatisfaction and physician interaction skills (e.g., communi-cation, courtesy, commitment to patients) and with variablesrelated to the structure and process of health services (e.g.,accessibility, waiting times, etc.) (3–7). It is important tonote that satisfaction is related to relevant and practical

Address reprint requests to: Dr. Patricia Clark, Blvd. Virreyes 1010,Lomas de Chapultepec, 11000 Mexico, D.F., Mexico. Phone/FAX: (�52)(55) 1998-1094; E-mail: clarkp@prodigy.net.mx

0188-4409/04 $–see front matter. Copyright � 2004 IMSS. Published by Elsedoi : 10.1016/ j .arcmed.2003.07.006

issues that pertain to an out-patient clinic, such as betterlevel of compliance with appointments, treatments, and med-ical orders (8) and the possibility of predicting actual utiliza-tion of health services (9–11).

There have been some studies concerning patient satisfac-tion and health care in rheumatology in Europe and the U.S.(12). However, none has been carried out in Latin Americaor specifically in Mexico, where health care delivery systemsdiffer from those previously described due to socioeconomicand cultural factors. Previous studies revealed that type ofhealth organization and type of payment arrangement influ-ence quality of service, patient perception of this quality, andpredict what proportion of patients will leave their physicianswithin the next several months. This becomes particularlyrelevant for patients with chronic diseases such as diabetes orrheumatoid arthritis, in which disability and complicationsnecessitate medical supervision for long periods of time(13,14). Therefore, the objective of the present study was

vier Inc.

Patient Ratings at a Rheumatology Out-Patient Unit 83

to explore the level of patient satisfaction with regard toservices received at the Hospital General de Mexico (HGM)Out-Patient Rheumatology Clinic.

Methods

Survey design. We conducted a cross-sectional study fromOctober 1998 to February 1999. We used a purpositive con-venience sample of adult patients with different rheumato-logic diagnoses who were cared for at the out-patientrheumatology clinic at the HGM.

Site. The HGM is the largest tertiary-care institution affili-ated with the Mexican Ministry of Health and serves mainly alow-income population without access to other types ofmedical care. Within the same institution through the generalout-patient consultation service, primary-care health servicesare offered to these patients as well to the population re-questing this service.

Patient selection. A non-probabilistic, purpositive consecu-tive sample of patients attending the HGM RheumatologyService Out-Patient Clinic were included in the study. Pa-tients were selected if they had at least two visits to theclinic for us to be assured that the patient had a conditionthat needed to be treated by rheumatologists. We includedpatients willing to participate independently of their rheuma-tologic condition. To include a wider spectrum of patients,we selected two periods of inclusion of 4 weeks initiated at1 month from each other, taking into account that timebetween regular appointments is from 3 to 6 months in mostcases; in addition, we were able to interview a wider repre-sentation of patients.

The questionnaire. The satisfaction questionnaire used inthis study was derived from a similar questionnaire usedin a previous study (12). Face validity of this modified instru-ment was tested by a group of rheumatologists, psycholo-gists, and medical students. There were 57 questions; themajority were presented on a 5-point Likert scale (rangingfrom 5 � excellent to 1 � poor). Some questions were di-chotomous (yes or no), and two were open-ended. Questionscovered three aspects of quality of care: structure, pro-cess, and results.

Questions with regard to structure concerned three differ-ent areas: physical environment, interaction with health careproviders, and waiting time. Questions on process pertainedto physician professional competence or skills, informationprovided on the disease, being seen by a physician-in-training, and choice of physician (i.e., whether or not theappointment was always conducted by the same physician).Questions concerning results requested information on levelsof disability and pain and coping with disease in relation tothe patient’s previous appointment. Additionally, we had oneoverall Likert-type satisfaction question and two open-ended

questions that sought to determine the most and least valu-able aspects of the appointment on the day the questionnairewas applied.

Procedure. A pilot study was performed with 80 rheumaticpatients to evaluate whether the instrument was understand-able and to test questionnaire reliability. Good reliabilitywas found in both areas evaluated, i.e., in structure (physicalenvironment, waiting time, etc.) and in process (profes-sional competence or physician skills, information providedon the disease, etc.); 0.52 and 0.56 were found, respectively(Spearman rho test-retest). Test-retest was performed with adifference of 3 weeks. The instrument was self-administered,although medical students were trained to help patients eitherby explaining questions when they were not understoodor (in the case of illiterate or disabled patients) by helpingfill out the questionnaire.

Data analysis. Univariate techniques were used to describegeneral group characteristics and the dimensions of struc-ture, process, and results. Spearman correlations were usedbetween overall satisfaction and each dimension.

Results

General characteristics. We interviewed 347 patients ofwhom 84% were women. Mean patient age was 41.6 years(standard deviation [SD] 15.0 years). Mean educational levelwas 7.14 years (SD 4.32 years), and the majority of patientswere married; only 16.6% of patients were employed. Themost frequent diagnosis was rheumatoid arthritis and the leastfrequent, fibromyalgia. Patient demographics, classificationof diseases, functional status (HAQ), and mean duration ofdisease in years are shown Table 1.

Mean time spent filling out the questionnaire was 29.9min (SD 11.46 min), and 27.8% of patients received helpin completing the questionnaire. Median time spent in transitfrom the patient’s domicile to the hospital was 2 h, whilemean time from arrival at the hospital to consultation was2.9 h (SD 1.33 h). A total of 67% of sample patients thoughtthat time spent prior to consultation was excessive and that areasonable waiting time prior to seeing a physician andreceiving care should be a mean of 1.18 h (SD 0.83 h)at most. Fifty two percent of referred patients reside incommunities outside of Mexico City. Eighty nine percentof patients have no affiliation with any health insurancesystem and have no health insurance; the Hospital Generalis the only health service available to them.

Patient levels of satisfaction with various aspects of thearthritis clinic are shown in Table 2. The most relevantfeatures are the high rating for the process, including compe-tence skills and humanistic approach of rheumatologistsin contrast to low scoring for structure, which reflectsorganization, administrative personnel, and physical facili-ties. With regard to results, patients reported fair improve-ment of disease (pain and disability) with high scoring

Clark et al. / Archives of Medical Research 35 (2004) 82–8684

Table 1. Demographics

Variable

Marital status (%)Single 27.9Married 56.6Divorced 8.3Widowed 7.2

Working status (%)Employed 16.6Homemaker 59.3Unemployed 11.6Student 5.5Retired 0.6Other 6.4

Annual family income (U.S. dollars)�1,500 76.11,600–6,400 16.86,500–10,200 5.5�10,300 1.6

Disease (%)Rheumatoid arthritis 45.3Systemic lupus erythematosus 15.7Gout 5.6Osteoarthritis 3.8Fibromyalgia 1.8Other 9.5Do not know 18.3

HAQ (0–3)(Mean, SD) 0.35 (0.45)

Disease duration in years 6.4(Mean, SD) (6.3)

for coping (in comparison with the patient’s previousappointment).

Variables included within the three major dimensions(structure, process, and results) were correlated with overallsatisfaction. Results are shown in Table 3. In general, correla-tions with overall satisfaction were not high, but correlationswithin the different dimensions had higher associations. Allitems related to medical skills had high correlation amongthem (history taken, physical examination, knowledge, infor-mation provided, and medical personnel interest perceivedby patients); these correlations ranged from 0.56 to 0.77.Items related to results also showed high relation amongthemselves (pain, disability, and coping, range 0.60 to 0.79).There was no correlation for items related to health providersand physical environment.

Seventy three percent of patients identified their physicianas the primary source of information concerning their dis-ease. When patients were asked whether they required moreinformation or knowledge on their condition, 98% respondedpositively. Areas in which they wished additional informa-tion or knowledge were prognosis, treatment (medications),and diet. Fifty eight patients were referred to other ancillaryservices within the hospital: the most frequently used ser-vices were ophthalmology (11.2%), orthopedics (6.6%), andcardiology (4.3%). Seventy seven percent of patients did not

Table 2. Satisfaction with various aspects of the Arthritis Clinica

Median (25th–75th

percentile)

Structureb

Waiting room 2.0 (2–3)Other facilities (bathroom, shops, etc.) 2.0 (2–3)Cleanliness 3.0 (3–4)

Interaction with health providersSecretary/receptionist 3.0 (2–3)Nurse 3.0 (3–4)Waiting time 2.0 (3–5)

Processb

Medical skillsHistory taken 3.5 (3–4)Physical examination 3.0 (3–5)Knowledge of disease 4.0 (3–4)

Humane approachCourtesy 4.0 (3–5)Commitment 4.0 (3–4)Information given on disease/education 3.0 (3–4)Being seen by a physician in traininga 4.0 (3–4)

Resultsc

Level of disabilityd 3.0 (2–4)Paind 3.0 (2–4)Coping with diseased 4.0 (2–4)Overall assessmenta 4.0 (4–4)

a5 � very satisfied, 4 � satisfied, 3 � indifferent, 2 � dissatisfied, 1 �

very dissatisfied; b5 � excellent, 4 � very good, 3 � average, 2 � fair,1 � poor; cComparison with last consultation; d1 � worse, 2 � no change,3 � slight improvement, 4 � moderate improvement, 5 � greatimprovement.

seek additional care (i.e., a private physician or alternativemedicine) during their treatment at the HGM. In overallassessment with regard to satisfaction, 83.5% reported beingsatisfied and 95.5% of patients would recommend the Rheu-matology Service at the HGM to family and friends. The twoqualitative questions regarding the most and least valuableaspects of patient appointments were answered as follows:nearly one half of patients (48.6%) stated that the physicianand the physician’s competence were the most valuable as-pects of the visit, while the least valuable aspect was thewaiting time they had to spend prior to their appointment.

Discussion

The goal of the present study was to determine patient satis-faction with the Out-Patient Rheumatology Clinic at theHospital General de Mexico using the model developed byDonabedian in which structure, process, and outcome arethe three basic foci. Structure refers to the institution’s orga-nization for delivering care and the conditions under whichcare is provided. Process refers to professional activitiesassociated with providing care, while outcome refers to thechange in a patient’s current and future health status thatcan be attributed to preceding care (15).

Patient Ratings at a Rheumatology Out-Patient Unit 85

Tabl

e3.

Cor

rela

tion

coef

ficie

ntw

ithdi

ffer

ent

aspe

cts

ofsa

tisfa

ctio

n

Ove

rall

Phys

icia

nH

isto

ryPh

ysic

alIn

form

atio

nPh

ysic

ian

inIn

stitu

tiona

lPh

ysic

alsa

tisfa

ctio

nco

urte

syC

omm

itmen

tta

ken

exam

inat

ion

Kno

wle

dge

give

ntr

aini

ngPa

inD

isab

ility

Cop

ing

heal

thpr

ovid

ers

envi

ronm

ent

Ove

rall

satis

fact

ion

1.00

0-

--

--

--

--

--

-Ph

ysic

ian

cour

tesy

0.28

6a1.

000

--

--

--

--

--

-C

omm

itmen

t0.

385a

0.77

1a1.

000

--

--

--

--

-H

isto

ryta

ken

0.26

2a0.

694a

0.70

2a1.

000

--

--

--

--

-Ph

ysic

alex

amin

atio

n0.

324a

0.56

1a0.

602a

0.61

6a1.

000

--

--

--

--

Kno

wle

dge

0.37

3a0.

618a

0.63

6a0.

659a

0.56

8a1.

000

--

--

--

-In

form

atio

ngi

ven

0.38

1a0.

560a

0.62

8a0.

584a

0.65

4a0.

608a

1.00

0-

--

--

-Ph

ysic

ian

intr

aini

ng0.

318a

0.21

4a0.

322a

0.25

9a0.

299a

0.23

4a0.

267a

1.00

0-

--

--

Pain

0.08

90.

089

0.13

2b0.

111b

0.06

70.

233a

0.06

00.

082

1.00

0-

--

-D

isab

ility

0.15

7a0.

096

0.18

9a0.

146a

0.12

6b0.

246a

0.08

30.

164a

0.79

5a1.

000

--

-C

opin

g0.

173a

0.09

60.

168a

0.18

8a0.

153a

0.25

6a0.

120b

0.07

10.

607a

0.66

7a1.

000

--

Inst

itutio

nal

heal

thpr

ovid

ers

0.15

6a0.

331a

0.35

4a0.

373a

0.26

6a0.

249a

0.29

6a0.

113b

�0.

034

�0.

018

0.03

51.

000

-Ph

ysic

alen

viro

nmen

t0.

327a

0.25

7a0.

313a

0.22

3a0.

225a

0.18

3a0.

300a

0.22

1a0.

014

0.00

60.

009

0.25

7a1.

000

a Cor

rela

tion

issi

gnifi

cant

atth

e0.

01le

vel

(tw

o-ta

iled)

;b co

rrel

atio

nis

sign

ifica

ntat

the

0.05

leve

l(t

wo-

taile

d).

Our data showed different ratings of satisfaction in evalu-ation of these three areas. Structure was scored as the worstcomponent, with waiting time and other facilities (bathroom,shops, etc.) receiving the lowest scoring. Process receivedthe best scoring; outcomes scored in the middle. Patientsrated from good to very good all questions related to physi-cian professional performance (e.g., their medical skills,humanistic approach, and information provided on thedisease), and from average to good for outcomes (e.g.,decreased pain and disability and increased ability to copewith the disease).

These results were consistent with the two qualitativequestions regarding the most and least valuable aspects ofpatient visits. The majority of patients rated the physicianand physician competence as the most valuable aspects oftheir visit and rated waiting time spent prior to consultationas the worst aspect of their visit.

Our results were similar to those reported by Sole in hisstudy of satisfaction in patients with arthritis. In Sole’s study,patients reported being least satisfied with waiting time,environment, and accessibility to the clinic and most satisfiedwith physician empathy and technical skills. In addition,there were two other similarities between these two studies(3). The institutions at which these studies were carried outare part of a university and/or governmental institution andare not private systems. Part of the dissatisfaction with thephysical environment might be due to our institution’s lim-ited resources (1). Populations served by these institutionsbelong to a low socioeconomic class and have no access toother health care systems.

While 89% of patients reported being satisfied when theyrated the overall satisfaction question, correlations of thismeasurement with all dimensions of the questionnaire werepoor. We find this interesting and worthy of some comment.

It is noteworthy that high ratings of overall satisfactionwere not exclusive to our study. In the UK, overall satisfac-tion rates for general practice of �90% have been notedconsistently over the past 30 years (15).

These high satisfaction ratings may be due to socialdesirability response bias (in which patients report greatersatisfaction than they actually feel because they believepositive comments are more acceptable to survey administra-tors); ingratiating response bias (when patients use satis-faction surveys to ingratiate themselves with research ormedical staffs), and cognitive dissonance reduction (in whichpatients report that they are satisfied as a way of justifyingthe time and effort they have invested in their treatment atthe unit) (15).

Other relevant issues are the manner in which questionsare asked and the conceptual meanings of the questions. Itwas demonstrated that patients could distinguish among dif-ferent dimensions when they judged quality. It was alsoshown that detailed questions generated greater levels ofdissatisfaction than more general questions, as is the caseof overall satisfaction. This might explain in part why,

Clark et al. / Archives of Medical Research 35 (2004) 82–8686

when overall satisfaction was used as a dependent variablein correlation analysis, results showed only low associations(16–18).

Ware (15) made the distinction between objective satis-faction reports, which referred to providers and care (waitingtimes, physical environment) and satisfaction ratings, whichattempted to capture a personal evaluation of care that cannotbe ascertained by observing care directly and that reflectedthe realities of the care received, the personal preference ofthe patients, and patient expectations (empathy, courtesy).These two clear divisions concerning satisfaction supportedour results. We found high scores for satisfaction ratings(physicians and their competence) for our group. This proba-bly reflects that patient expectations are met with regardto the type of care they will receive when they go to theRheumatology Unit for consultation.

We are aware that our study has population selection bias,as will be true of any study performed in only one hospital.Our patients had low income and the majority had no accessto another health institution; this might influence their re-sponses, biasing results toward the positive. Another consid-eration is long disease duration (average 6.4 years), whichprobably reflects a subsample of rheumatologic patients satis-fied with their attention. Therefore, we cannot be sure ofthe proportion of patients who were not satisfied with thetreatment they received at the HGM who thus did not return.

To summarize, patients are satisfied with their physiciansand dissatisfied with environmental and administrativeaspects of their care. These findings should alert health ad-ministrators to improve the structure (health facilities andadministrative features) of these kinds of hospitals.

Another important part of our survey is related to theinformation and education patients received on different as-pects of their disease, because 99.1% of our patient samplewould like more information concerning their disease. Physi-cians and health staff in direct contact with patients musttake this expectation into account and develop specific strate-gies to provide whatever information is possible.

References1. Reemtsma K, Morgan M. Outcomes assessment: a primer. Bull Am

Coll Surg 1997;82(3):34–39.2. Donabedian A. The quality of care. How can it be assessed? JAMA

1988;260:1743–1748.3. Kirsner R, Federman D. Patient satisfaction. Quality of care from the

patient’s perspective. Arch Dermatol 1997;133:1427–1431.4. Palmer R, Connorton M. Individual and institutional variables which

may serve as indicators of quality of medical care. Med Care 1979;XVII(7):693–715.

5. Larsen D, Rootman I. Physician role performance and patient satisfac-tion. Soc Sci Med 1976;10:29–32.

6. Stiles W, Putnam S, Wolf M, James S. Interaction exchange structure andpatient satisfaction with medical interviews. Med Care 1979;XVII(6):667–679.

7. Korsch B, Gozzi E, Francis V. Gaps in doctor-patient communication.I. Doctor-patient interaction and patient satisfaction. Pediatrics 1968;42(5):855–869.

8. Linder-Pelz S. Social psychological determinants of patient satisfaction:a test of five hypotheses. Soc Sci Med 1982;16:583–589.

9. Strasser S, Aharony L, Greenberger D. The patient satisfaction process:moving toward a comprehensive model. Med Care Rev 1993;50(2):219–248.

10. Haya R, Gandek B, Rogers W, Kosinki M, Mchorney C, Ware J. Patient’sratings of outpatients visits in different practice settings. JAMA 1993;270(7):835–840.

11. Zastowny T, Roghmann K, Cafferata G. Patient satisfaction and the useof health services explorations in causality. Med Care 1989;27:705–723.

12. Sole T, Lipsky P. Satisfaction of patients attending an arthritis clinicin a county teaching hospital. Arthritis Care Res 1997;10(3):169–176.

13. Hill J, Bird H, Hopkins R, Lawton C, Wright V. Survey of satisfactionwith care in a rheumatology outpatient clinic. Ann Rheum Dis 1992;51:195–197.

14. Bartlett E, Grayson M, Barker R, Levine D, Golden A, Libber S. Theeffects of physician communications skills on patient satisfaction;recall, and adherence. J Chron Dis 1984;37(9/10):755–764.

15. Sitzia J, Wood N. Patient satisfaction: a review of issues and con-cepts. Soc Sci Med 1997;45(12):1829–1843.

16. Ross C, Steward C, Sinacore J. A comparative study of seven measuresof patient satisfaction. Med Care 1995;33:392–406.

17. Roghmann K, Hengst A, Zastowny T. Satisfaction with medical care.Its measurement and relation to utilization. Med Care 1979;XVII(3):461–477.

18. Concato J, Feinstein A. Asking patients what they like: over attributesof patient satisfaction with primary care. Am J Med 1997;102:399–406.