O

Pp

Ia

b

c

ARA

KSGTS

PCPPE

p

2

Rev Esp Med Nucl Imagen Mol. 2014;33(2):65–71

riginal Article

ersonalizing the reference level: Gold standard to evaluate the quality of serviceerceived�

. Rodrigo-Rincóna,∗, M. Reyes-Pérezb, M.E. Martínez-Lozanoc

Investigación y Gestión del Conocimiento, Departamento de Salud del Gobierno de Navarra, Red de Investigación en Servicios de Salud en Enfermedades Crónicas (REDISSEC), SpainServicio de Medicina Preventiva y Gestión de la Calidad, Complejo Hospitalario de Navarra, Pamplona, SpainServicio de Medicina Nuclear, Complejo Hospitalario de Navarra, Pamplona, Spain

a r t i c l e i n f o

rticle history:eceived 14 January 2013ccepted 3 March 2013

eywords:ervice qualityold standardhresholdurvey

a b s t r a c t

Objective: To know the cutoff point at which in-house Nuclear Medicine Department (MND) customersconsider that the quality of service is good (personalized cutoff).Material and method: We conducted a survey of the professionals who had requested at least 5 tests tothe Nuclear Medicine Department. A total of 71 doctors responded (response rate: 30%). A question wasadded to the questionnaire for the user to establish a cutoff point for which they would consider thequality of service as good. The quality non-conformities, areas of improvement and strong points of thesix questions measuring the quality of service (Likert scale 0 to 10) were compared with two differentthresholds: personalized cutoff and one proposed by the service itself a priori. Test statistics: binomialand Student’s t test for paired data.Results: A cutoff value of 7 was proposed by the service as a reference while 68.1% of respondents sug-gested a cutoff above 7 points (mean 7.9 points). The 6 elements of perceived quality were consideredstrong points with the cutoff proposed by the MND, while there were 3 detected with the personal-ized threshold. Thirteen percent of the answers were nonconformities with the service cutoff versus19.2% with the personalized one, the differences being statistically significant (difference 95% CI 6.44%:0.83–12.06).Conclusions: The final image of the perceived quality of an in-house customer is different when using thecutoff established by the Department versus the personalized cutoff given by the respondent.

© 2013 Elsevier España, S.L. and SEMNIM. All rights reserved.

Personalización del nivel de referencia: patrón oro para evaluar la calidad deservicio percibida

alabras clave:alidad percibidaatrón orounto de cortencuestas

r e s u m e n

Objetivo: Conocer el punto de corte a partir del cual los clientes internos del servicio de medicina nuclear(MN) consideran que la calidad de servicio es buena (punto de corte personalizado).Material y método: Se realizó una encuesta a los profesionales que hubieran solicitado al menos 5 pruebasal servicio de medicina nuclear. Contestaron 71 médicos (tasa de respuesta del 30%). Se anadió al cues-tionario una pregunta para que el usuario estableciera el punto de corte a partir del cual el encuestadoconsidera que la calidad de servicio es buena. Se compararon las no conformidades, las áreas de mejoray los puntos fuertes de las 6 preguntas que medían la calidad de servicio (escala Likert de 0 al 10) con 2dinteles de referencia: el punto de corte personalizado y el que propuso a priori el propio servicio. Testestadísticos: binomial y t de Student para datos pareados.Resultados: El servicio propuso el valor de 7 como punto de corte, mientras que el 68,1% de los encuestadospropuso un valor superior a 7 puntos (media 7,9 puntos). Los 6 elementos de calidad percibida fueronconsiderados puntos fuertes con el punto de corte propuesto por el servicio de MN, mientras que fueron

3 los detectados con el punto de corte personalizado. El 13% de las valoraciones fueron no conformes conel punto de corte del servicio frente al 19,2% con el punto de corte personalizado, siendo las diferenciasestadísticamente significativas (diferencia 6,44%; IC 95%: 0,83-12,06).Conclusiones: La imagen final de la calidad percibida por los clientes internos de un servicio es diferente sise utiliza el punto de corte que establece el servicio frente al que indica el propio individuo que respondeal cuestionario.

� Please cite this article as: Rodrigo-Rincón I, Reyes-Pérez M, Martínez-Lozano ME. Perercibida. Rev Esp Med Nucl Imagen Mol. 2014;33:65–71.∗ Corresponding author.

E-mail address: mi.rodrigo.rincon@cfnavarra.es (I. Rodrigo-Rincón).

253-8089/$ – see front matter © 2013 Elsevier España, S.L. and SEMNIM. All rights reser

© 2013 Elsevier España, S.L. y SEMNIM. Todos los derechos reservados.

sonalización del nivel de referencia: patrón oro para evaluar la calidad de servicio

ved.

6 ed Nucl Imagen Mol. 2014;33(2):65–71

I

it

amfms

anw“ianiswsv

sata

sooow

mo

o(rw

dp

c

M

topl

nDpffsap

6.1

25.8

43.9

19.7

4.5

0

5

10

15

20

25

30

35

40

45

50

1 2 3 4 5 6 7 8 9 10Scores

% o

f ans

wer

s

Fig. 1. Distribution of the frequencies of the score given to the question “Above what

6 I. Rodrigo-Rincón et al. / Rev Esp M

ntroduction

One of the most relevant elements for improvement in the qual-ty of organizations is knowing the satisfaction and the quality ofhe services perceived by the consumers.1–3

Although the concepts of satisfaction and service quality servicere apparently simple, there is no consensus with regard to theireaning or how to conceptualize the relationship between satis-

action and the quality of the service provided or the most correctethod for their measurement.3 Nonetheless, most institutions use

ome type of tool for their measurement.4

The method most frequently used to measure both satisfactions the service quality is with questionnaires.5,6 Most question-aires use scales following a structure of Likert-type responseith a series of categories of response along the continuum

favorable/unfavorable”. On numerous occasions, the question onlyndicates the meaning of the initial and final points with intermedi-te values remaining unspecified. One example of this is questionumber 3 of the healthcare barometer which asks: “Are you sat-

sfied or dissatisfied with the way in which the public healthcareystem works in Spain?” To answer, the individual is shown a cardith numbers from 1 to 10, with 1 corresponding to very dis-

atisfied and 10 to satisfied,7 without specifying the intermediatealues.

Analysis of the results of questions with this type of scale is notimple. How can the cutoff or reference value to be considered asgood result be determined? Above what score should the institu-

ion consider an aspect as a strong point or at what value is theren area of improvement?

To answer this question different approaches have been useduch as the determination of an objective value from a benchmark8

r a desired value. That is, users are asked about their perceptionf an aspect with the aim of involving the users in the evaluationf a department, but the interpretation of the results is performedith a subjective aim established by the service provider.

To measure the service quality other authors9,10 have used theodel of discrepancies or “gaps” model comparing the perceptions

f the user with respect to their expectations.In the present study we considered an alternative to the setting

f a subjective cutoff point by the Department of Nuclear MedicineDNM). The proposal consisted in having the internal customersequiring tests from the DNM themselves establish the cutoff athich the quality perceived is deemed good.

We compared the strong points, the areas of improvementetected and those discrepant with 2 reference levels, that pro-osed by the DNM and the internal consumers.

The objective was to determine the cutoff at which the internalonsumers of the DNM consider the service quality as good.

aterial and methods

The framework of the sample was made up of professionals fromhe clinical departments of a tertiary level hospital requesting testsr consultations from the DNM. The subjects constituting the sam-le were physicians from other departments who had requested at

east 5 tests from the DNM in 2010.On identifying these professionals they were sent a question-

aire designed to evaluate the quality of service provided by theNM (Annex 1). Two modalities of questionnaire completion wererovided. The questionnaire in paper form was sent to each pro-essional by internal mail of the hospital together with an envelope

or returning the questionnaire. In addition, the professionals wereent an email with a link in order to answer the questionnairenonymously. They were told that the two modalities were incom-atible. Two reminders were sent. The collection period of the

score do you consider that the service quality is good?”.

questionnaires was from June to September, 2011. Of a total of 237professionals, 71 answered (30% response rate).

The questionnaire consisted of 14 items, 6 of which involveditems related to the quality of the services. The scale used for thequestions ranged from 0 (worst possible score) to 10 (best possiblescore).

The reliability of the questionnaire measured with the Cronbachalpha coefficient was of 0.643, with the general alpha value withtypified items being 0.790.

At the end of the questionnaire there was an item asking theprofessional to state at what numerical score they would considerthe service quality as good, considering this score as a personalizedcutoff. Prior to the incorporation of the item to the questionnaire, 5interviews of professionals were undertaken to perform cognitivevalidation of the question and thereby confirm that the statementwas correct and comprehensible.

Prior to the analysis of the results the DNM was requested to seta cutoff at which they considered that the service quality providedwas good. By consensus the department determined the cutoff of 7and this value was denominated the “department cutoff”.

An element evaluated was considered as a strong point of thedepartment if its lowest value of the confidence interval of 95% wasgreater than the reference level, and an area of improvement wasconsidered if the highest value of the confidence interval of 95%was lower than the value of this level.

Using the personalized cutoff the number of discrepancies wascalculated by the difference between the score given to each ques-tion and the value at which the subject considered that the servicequality was good. For example, if an individual gave an item refer-ring to the service quality the reports 8 points and considered that9 was the score that should be obtained to provide good qualityservice, we have a value of −1 point (8 minus 9). All the nega-tive values such as the example indicated were considered to bediscrepant. Likewise, the number of discrepancies was calculatedapplying the value of 7 as the threshold of reference. This value waswhat had been established by the DNM.

The statistical tests used included the binomial method fordependent samples and Student’s t test for paired data.

Results

Table 1 shows the results of the analysis of the items measuringthe service quality.

With regard to the question “Above what score do you considerthat the service quality is good?” 68.1% of the subjects gave a value

greater than 7. That is, the level of reference established a prioriby the service was below the reference level given by many of theprofessionals (Fig. 1).

I. Rodrigo-Rincón et al. / Rev Esp Med Nucl Imagen Mol. 2014;33(2):65–71 67

Table 1Results of the items referring to the quality of service together with the question threshold.

Item Mean (CI 95%) SD Minimum Maximum

Attitude to collaborate in organizational problems 8.34 (7.77–8.91) 1.81 1 10Speed in the performance of tests/consultation 7.96 (7.557–8.36) 1.64 4 10Speed in emitting reports 8.33 (7.97–8.7) 1.48 4 10Reports quality 8.79 (8.46–9.13) 1.27 5 10Information on criteria for not performing tests 7.93 (7.06–8.80) 2.24 1 10Capacity of resolution 8.51 (8.22–8.80) 1.15 6 10

.60–9

.31–9

.68–8

A

stmct

me

A

b

(di

b

piF

Satisfaction with DNM 8.89 (8Recommendation of service to other professionals 8.67 (8Score at which the service quality may be considered good 7.91 (7

nalysis of the results from the mean thresholds of reference

One of the fundamental objectives of the study was to know thetrong points and areas of improvement in the DNM. On analyzinghe results of each of the questions we found that of the 6 questions

easuring the quality of the service all were strong points with theutoff set by the department while 3 were not so on consideringhe mean personalized cutoff (Fig. 2).

No areas of improvement were detected with either of the 2ethods used since the confidence interval was not below the lev-

ls established for any variable.

nalysis of the disagreement with each item evaluated

The number and percentage of discrepancies per question withoth cutoffs are shown in Table 2.

Using the personalized cutoff a total of 62 discrepancy values19.2%) were detected while the mean cutoff of the departmentetected 41 (13%), with these differences being statistically signif-

cant (difference: 6.44%; CI 95%: 0.83–12.06).No statistically significant differences were observed in the item

y item analysis.

The mean values for each item of the variables “difference in

erception with regard to the personalized cutoff” and “differencen perception with respect to the department cutoff” are shown inig. 3.

8.3

8.0

8.3

1

2

3

4

5

6

7

8

9

10

Attitude tocollaborate inorganizational

problems

Speed inthe performance

of tests/consultation

Speed inemitting reports

Evaluated

Sco

res

Fig. 2. Mean values and confidence intervals of 95%

.19) 1.12 6 10

.02) 1.14 6 10

.14) 0.75 6 10

The axis of ordinates was from −10 to +10, being the range ofpossible scores.

No element was given a mean negative value indicating thatthe scores of quality perceived by the professionals were higherthan the cutoff set by themselves or that established by the DNM(value 7). Nonetheless, on comparing the mean values of all the ele-ments evaluated, statistically significant differences were observedon comparing the 2 cutoffs, with the mean differences for threshold7 being greater than for the personalized cutoff (p < 0.05, Student’st test for paired data).

Discussion

The main objectives on undertaking a questionnaire of the qual-ity of service perceived are to determine the strong points and theareas of improvement from the point of view of those surveyed.However, the methodology used for the analysis of the results,and thus, the interpretation of these results is conditioned by thetype of scale of the variables and the cutoff established for evalua-tion.

The analysis of the results indicates that the interpretation ofthe strong points differs based on the method used. Of the 6 items

measuring the quality of service all were considered strong pointsfrom the cutoff set by the DNM while only 3 were considered strongpoints with the personalized cutoff. There were no discrepancieswith regard to the areas of improvement, with none being detected

8.8

7.9

8.5

Report quality Information oncriteria for not

performing tests

Capacity ofresolution

items

Threshold purveyor

Threshold surveyed

of the items referring to the service quality.

68 I. Rodrigo-Rincón et al. / Rev Esp Med Nucl Imagen Mol. 2014;33(2):65–71

Table 2Analysis of discrepancies: individual threshold and threshold established by the department of nuclear medicine.

Item n Individual threshold DNM threshold

No. of discrepancies % % over total ofdiscrepancies

No. of discrepancies % % over total ofdiscrepancies

Attitude to collaborate in organizational problems 41 8 19.5 12.9 5 12.2 12.2Speed in the performance of tests/consultations 67 20 29.9 32.3 15 22.4. 36.6Speed in emitting reports 66 15 22.7 24.2 11 16.7 26.8Reports quality 58 5 8.6 8.1 4 6.9 9.8Information on criteria for not performing tests 28 6 21.4 9.7 3 10.7 7.3Capacity of resolution 63 8 12.7 12.9 3 4.8 7.3Total 323 62 19.2 100 41 13.0 100

n indicates the number of persons answering each item of the questionnaire; % indicates the percentage of individuals considering discrepancy with this item; % over thetotal of discrepancies indicates what percentage of the total of discrepancies corresponds to each item.

0.580.05

0.411.00

0.080.63

1.340.96

1.33 1.790.93

1.51

–10.00

–8.00

–6.00

–4.00

–2.00

0.00

2.00

4.00

6.00

8.00

10.00

Service attitude Test speed Report speed Report quality Inf. not perform test Capacity ofresolution

hold

nces b

ietodcopsc

ptTtgnasq

c1qe

Individual thres

Fig. 3. Quality of service: mean values of the differe

ndependently of the method used. That is, when the cutoff wasstablished by the evaluator, fewer strong points were detectedhan those that would have been detected by the DNM using itswn threshold. Similarly, a greater number of discrepancies wereetected on using the personalized cutoff versus the departmentutoff, with the differences being statistically significant. Likewise,n calculating the difference between the mean values of qualityerceived given and the reference levels, the values obtained wereignificantly higher for the department than for the personalizedutoff in all the items.

The discrepancy as to the areas of improvement and strongoints and the number of discrepancies varied based on how farhe personalized cutoff was from the other threshold established.he problem is that since the user is not consulted with regardo the value at which the quality of service may be considered asood, the error committed in the interpretation of the results isot known. Nonetheless, regardless of the results obtained, fromconceptual point of view the reference cutoff and thus, the gold

tandard, should be that indicated by the subject responding to theuestionnaire.

On the other hand, analysis of the differences or discrepan-

ies is not a new method since this has been used since the980s. The discrepancy method considers that the evaluation ofuality is the result of the divergence between perceptions andxpectations.7,10–12

Threshold 7

etween perceptions and the threshold of reference.

The method proposed in this study differs from the discrep-ancy models such as that by Parasuraman et al.10 in 2 ways. Thefirst is that with the methodology which we used expectationswere not considered. The debate regarding the measurement ofexpectations is explained in other studies.13–15 Nevertheless, thedetractors of the discrepancy model indicate that the inclusion ofexpectations may be inefficient and unnecessary because individ-uals tend to indicate high levels of expectation and thus, the valuesof perception are rarely surpassed.

Secondly, with our method it is not necessary for the sub-ject to provide a level of reference for each item. It is thereforenot necessary to duplicate the number of items made butrather to add one more question to the questionnaire. To avoidduplicity of items, other authors16 have used a questionnaire inwhich the scale of response combines expectations and percep-tions.

With respect to the type of scale, polytomous variables allowrelatively simple classification of the categories referring to the dis-crepancies. Nonetheless, many organizations use an ordinal scale inwhich only the final cutoffs of the scale are set. This option presentssome inconveniences. First, the use of digits – numbers – does not

guarantee adequate psychometric properties for using the usualstatistical tests. Second, there is the problem of defining a referencethreshold of a cutoff at which a strong point or area of improvementmay be considered.

ed Nu

sslTto

pbotp

aTp

ca

I. Rodrigo-Rincón et al. / Rev Esp M

The method described takes into account the opinion of theubjects surveyed when setting the level of reference versus otherystems in which it is the service purveyor who subjectively estab-ishes this value, often a posteriori and after knowing the results.hat is, grant the value of judgment to the individual, which is, onhe other hand, implicit when wishing to obtain the user’s opinionf the quality of service.

In addition, most studies published focus on the evaluation byatients, with studies centered on assessment by professionalseing less frequent. This study did not ask patients about the qualityf service but rather the professionals requesting tests or consulta-ions to the DNM and thus, these results cannot be extrapolated toatients, although the conceptual basis is the same.

The DNM considered the cutoff to be 7 because the profession-ls are, in general, less generous in giving scores than patients.17

he cutoff set by the DNM would have been higher if the qualityerceived was to have been evaluated by the patients.

Comparative data of cutoffs given by users (internal or externalonsumers) in other studies are not provided since we did not findny article applying this type of focus.

□

□

0 1 2 3

0 1 2 3

This questionnaire has the objeDepartment of Nuclear Medicithe specialty of Nuclear Medic

P1- The degree of relationNuclear Medicine (evaluatundergoing tests, number orelation to correct diagnos

P3- Evaluate the attitude resolution of organizationadata of clinical relevance,

P4-The speed of the perfis (from the time of the req

P2- Approximately how m

Please remember that all the qMedicine. To respond, please uThank you for your collaborati

Please put a cross on the optiocross out the incorrect option a

Please circle the option mostThe score ranges from ZERO

LowNo relationship

Less than From

None

Very slow

cl Imagen Mol. 2014;33(2):65–71 69

In summary, the novelty of this study lays in that it proposesthat the users who respond to the questionnaire should establishthe cutoff at which the quality is perceived to be good since the finalimage of quality perceived by the internal consumers of a depart-ment is different if the cutoff set by the department is used versusthat indicated by the individuals responding to the questionnaires.

Author contribution

M. Isabel Rodrigo Rincón participated in all the phases of articlepreparation including the design, data analysis and redaction.

María Reyes Pérez participated in the field work and data anal-ysis.

M. Eugenia Martínez contributed to the conception and design ofthe study as well as the approval of the final version for publication.

Conflict of interests

The authors declare no conflict of interests.

Annex 1. Questionnaire to internal consumers

□ □ □

□ □ □

4 5 6 7 8 9 10

ctive to know your opinion on the global quality of all the actions of the ne. The aim of this questionnaire is to know the “image of competence” of ine to detect aspects which will allow us to continue improving.

ship between the department to which you belong and the Department o ed by the need for this department in your usual practice: number of patients f tests requested, importance the tests requested have for your department in

is or treatment of the patients...):

of the Department of Nuclear Medicine in collaborating with you in the l problems (patient management, channels or circuits of communication for

etc.).

ormance of the tests/consultations of the Department of Nuclear Medicine

any tests do you request from the Department of Nuclear Medicine per year?

uestions are aspects which you would expect from the specialty of Nuclear se the questionnaire below.on.

n which best reflects your opinion as a professional. If you make a mistake, nd place another cross on the option you consider the most adequate.

closely approaching your perception of each of the following statements. (minimum possible score) to TEN (maximum possible score).

Very close relationshipModerate relationship relationship

5 to 9 From 10 to 15 More than 15

ExcellentNA

4 5 6 7 8 9 10

uest to the performance of the test).

Very fastNA

7 ed Nucl Imagen Mol. 2014;33(2):65–71

9 10

9 10

10

9 10

the Department of Nuclear

he test requested for a patient in relation to the criteria for

icine for the patients referred

the results)

Very fast

Very good

Excellent

Very high

NA

NA

NA

NA

9 10

9 10

9 10

9 10

9 10

9 10

_____________________ __

ces available in the Depart-

ear Medicine.

taff of the Department of

er professionals if they were

you consider that the

e Department of Nuclear which

ailable in the Department of

Very abundant

Excellent

Excellent

Very high

Always

NA

NA

NA

NA

NA

NA

0 I. Rodrigo-Rincón et al. / Rev Esp M

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8 9

0 1 2 3 4 5 6 7 8

P5 -The reports, registries or results of the tests/(consultations emitted byMedicine with regard to:

P6- When the Department of Nuclear Medicine decides not to carry out tfrom your department please score from 0 to 10 the information providednot performing the test.

P7- Assess the capacity of resolution of the Department of Nuclear Medby yourself.

P5-1 The speed (from the time of the test/consultation to the emission of

P5-2- The quality you perceive of the reports or results:

Very slow

Very bad

No criteria

Very low

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8

___________ ______ ______ ______ ______ ______ ______ _______________

P8- In your opinion, evaluate from 0 to 10 the quantity of human resourment of Nuclear Medicine to perform their work.

P11- As a whole, evaluate your satisfaction with the Department of Nucl

P10- ¿What IMAGE do you have of the professional competence of the sNuclear Medicine?

P12 Would you recommend the Department of Nuclear Medicine to othin the same situation and could choose a department?

P13- In each of the questions of the questionnaire, Above what score do service quality is good?

P14- Would you like to add any comment concerning any aspect related to thwas not included in the questions above?

P9- In your opinion evaluate from 0 to 10 the technological resources avNuclear Medicine to perform their work.

Very scarce

Very precarious

Very bad

Very low

Never

___________ ______ ______ ______ ______ ______ ______ _______________________________ ______ ______ ______ _________ ______ ______ ______ ______ _____ ___________ ______ ______ ______ ______ ______ ______ ____________________

________________ ____________________________________ __

ed Nu

R

1

1

1

1

1

1

I. Rodrigo-Rincón et al. / Rev Esp M

eferences

1. Garcia Vicente A, Soriano Castrejón A, Martínez Delgado C, Poblete García VM,Ruiz Solís S, Cortés Romera M, et al. La satisfacción del usuario como indicadorde calidad en un servicio de Medicina Nuclear. Rev Esp Med Nucl. 2007;26:146–52.

2. Poblete García VM, Talavera Rubio MP, Palomar Munoz A, Pilkington Woll JP,Cordero García JM, García Vicente AM, et al. Implantación de un sistema deGestión de Calidad según norma UNE-UN-ISO 9001:2008 en un Servicio deMedicina Nuclear. Rev Esp Med Nucl Imagen Mol. 2013;32:1–7.

3. Gill L, White L. A critical review of patient satisfaction. Leadersh Health Serv(Bradf Engl). 2009;22:8–19.

4. Crowe R, Gage H, Hampson S, Hart J, Kimber A, Storey L, et al. The measurement ofsatisfaction with healthcare: implications for practice from a systematic reviewof the literature. Health Technol Assess. 2002;6:1–244.

5. Mira JJ, Aranaz J. La satisfacción del paciente como una medida del resultado dela atención sanitaria. Med Clin (Barc). 2000;114 Suppl. 3:26–33.

6. Reyes-Pérez M, Rodrigo-Rincón MI, Martínez-Lozano ME, Goni-Gironés E,Camarero-Salazar A, Serra-Arbeloa P, et al. Evaluación del grado de satisfacción

de los pacientes atendidos en un servicio de Medicina Nuclear. Rev Esp MedNucl Imagen Mol. 2012;31:192–201.

7. Cuestionario del Barómetro Sanitario utilizado por el Ministerio de Sanidad,Servicios Sociales e Igualdad. 2011 [accessed 30.11.12]. Available in: http://www.msps.es/estadEstudios/estadisticas/docs/BS 2011 Cuestionario.pdf

1

1

cl Imagen Mol. 2014;33(2):65–71 71

8. Rodrigo-Rincón I, Vines JJ, Guillén-Grima F. Análisis de la calidad de la informa-ción proporcionada a los pacientes por parte de unidades clínicas especializadasambulatorias mediante análisis por modelos multinivel. An Sist Sanit Navar.2009;32:183–97.

9. Parasuraman A, Zeithaml V, Berry L. Reassessment of expectations as a compar-ison standard in measuring service quality: implications for future research. JMark. 1994;58:111–24.

0. Parasuraman A, Zeithaml V, Berry L. SERVQUAL. A multiple-item scale for mea-suring consumer perceptions for service quality. J Retail. 1988;64:12–40.

1. Oliver RL. A cognitive model of the antecedents and consequences of satisfactiondecisions. J Mark Res. 1980;17:460–9.

2. Grönroos C. A service quality model and its marketing implications. Eur J Mark.1984;18:6–44.

3. Babakus E, Boller GW. An empirical assessment of the SERVQUAL scale. J BusRes. 1992;24:253–68.

4. Cronin JJ, Taylor S. Measuring service quality: a re-examination and extension.J Mark. 1992;56:55–68.

5. Thompson AG, Sunol R. Expectations as determinants of patient satisfaction:concepts, theory and evidence. Int J Qual Health Care. 1995;7:127–41.

6. Mira JJ, Aranaz J, Rodríguez-Marín J, Buil JA, Castell M, Vitaller J. SERVQHOS, uncuestionario para evaluar la calidad percibida. Med Prev. 1998;4:12–8.

7. de Man S, Gemmel P, Vlerick P, Van Rijk P, Dierckx R. Patients’ and personnel’sperceptions of service quality and patient satisfaction in nuclear medicine. EurJ Nucl Med Mol Imaging. 2002;29:1109–17.