Patient satisfaction: A valid concept?

Pergamon

Sot. SC;. Med. Vol. 38, No. 4, 509-516, 1994 pp. Copyright 0 1994 Elsevier Science Ltd

Printed in Great Britain. All rights reserved 0277-9536194 $6.00 + 0.00

PATIENT SATISFACTION: A VALID CONCEPT!

BRIAN WILLIAMS

Academic Sub-Department of Psychological Medicine, North Wales Hospital, Denbigh, Clwyd LL16 5SS, Wales

Abstract--Over the past 10 yr consumer satisfaction has gained widespread recognition as a measure of quality in many public sector services. This has become manifest in the NHS in the call by the 1983 NHS Management inquiry to ascertain how well the service is being delivered at local level by obtaining the experience and perceptions of patients and the community. Patient satisfaction is now deemed an important outcome measure for health services; however, this professed utility rests on a number of implicit assumptions about the nature and meaning of expressions of ‘satisfaction’. Through a review of past research findings this paper suggests that patients may have a complex set of important and relevant beliefs which cannot be embodied in terms of expressions of satisfaction. Consequently, many satisfaction surveys provide only an illusion of consumerism producing results which tend only to endorse the status quo. For service providers to meaningfully ascertain the experience and perceptions of patients and the community then research must first be conducted to identify the ways and terms in which those patients perceive and evaluate that service.

Key words-patient satisfaction, quality of care, evaluation, consumerism

INTRODUCTION

Over the past 10 yr consumer satisfaction has gained

widespread recognition as a measure of quality in many public sector services. This has been particu- larly true in the National Health Service since the publication of the 1983 NHS Management inquiry recommended that:

The Management Board and Chairmen should ensure that it is central to the approach of management, in planning and delivering services for the population as a whole, to: ascertain how well the service is being delivered at local level by obtaining the experience and perceptions of patients and the community.. [I, p. 91.

The call for an acquisition of ‘experience and perceptions’ has, in practice, been largely translated by Health Authorities and the 1989 White Paper [2] into a call for the measurement of satisfaction. The rationale behind this translation is likely to be 3-fold: firstly, surveys in questionnaire form are considered relatively cheap and easy to conduct; secondly, a distrust of qualitative research and ‘soft’ data; and thirdly, a desire for information in quantitative form in order to facilitate the further Griffiths’ recommen- dation that the service should ‘monitor performance’ against such opinion.

The consequence of this new found enthusiasm has been a proliferation of surveys. These range in scale and sophistication from the ad hoc study which, in practical terms, is little more than a token gesture to consumerism, through to the highly financed CASPE system (Clinical Accountability, Service Planning and Evaluation) which includes service reorientation as an integral element. The validity of the CASPE system has been questioned by Carr-Hill and col-

leagues who asked whether “12 simple questions in a computer-read survey really measure patient satisfaction” [3]. Such questions are seldom asked of the majority of methodologies utilized for measuring patient satisfaction.

The Carr-Hill question reveals an underlying belief which currently predominates the literature, namely that ‘patient satisfaction’ exists in the population, simply awaiting measurement. To be of any practical use we must know what people mean when they say they are ‘satisfied’ with a particular aspect of a service; furthermore to make any relevant changes in service provision we must know why they believe what they do and how they arrived at that view. Survey results must be interpreted, but at this level a number of assumptions are often made concerning what patients actually mean when they say they are ‘satisfied’. Such assumptions in interpretation are themselves the result of assumptions made about the way patients evaluate. Consequently, inferences made from survey results may not reflect the true beliefs of service users.

At this stage a distinction should be made between those who stress the importance of patient’s evaluations but also recognize the methodological and conceptual problems with its collation, and those who advocate patient’s evaluations but assume that such evaluations are accessible through expressions of ‘satisfaction’ in ‘satisfaction surveys’ and assume the meaning of such expressions to be self evident.

In this paper I argue that while patients’ evaluations are important we must jirst gain a clearer understanding of how they evaluate before patient opinion can be accurately collated and interpreted. Consequently, this paper aims to demonstrate the

509

510 BRIAN WILLIAMS

need for further research by showing that patient’s evaluations and beliefs may not be embodied in expressions of ‘satisfaction’ and that the interpretation of such expressions is governed by questionable assumptions about the way in which patients evaluate. This paper does not, therefore, aim to demonstrate beyond doubt that patients evaluate in any particular way; it aims to show that we do not currently know how patients evaluate and because of this inferences made from many satisfaction surveys may not accurately embody the true beliefs of service users.

SATISFACTION: ITS PURPOSE

Patient satisfaction questionnaires have their origin in two separate developments; consequently, many advocates of patient satisfaction have mixed motives.

Improving compliance

The rise of doctor-patient interaction research in the 1950s stemming from work by sociologists such as Parsons [4] and Szasz and Hollender [5], helped place the patients’ perspective firmly on the research agenda. While, it was primarily sociologists who took the initiative in this new area others were quick to see the potential benefits if patient behaviour could be better understood. Consequently, there followed a number of studies investigating the relationship between compliance and the patient perspective.

The utilization of questionnaires for the quantifi- cation of patients’ evaluations facilitated the discovery of a link between three types of compliance and what had come to be seen as ‘patient satisfaction’: appointment keeping [68]; behavioural intentions to comply with recommended treatment 191; and medication use [lo].

The relationship with compliance has been used to argue for the importance of satisfaction as a necess- ary health service goal. Since high quality clinical outcome is dependent on compliance which, in turn, is dependent on patient satisfaction the latter has come to be seen as a prerequisite of quality care. Consequently, this has helped legitimize the importance of the patient’s perspective among health care professionals who are primarily concerned with clinical outcome.

Introducing consumerism

The late 1960s and early 1970s saw the increasing involvement of the consumer in all walks of life. Although it is unlikely that the consumer movement provided the initial impetus for the interaction research mentioned above some interplay between the two developments is undeniable.

It is likely that the discovery of the satisfaction-compliance link bolstered the consumer movement in its demands for the consumer to be seen as the central figure of accountability in all public services. Simultaneously, the increasing acceptance of

consumerist values provided support for further research into patient behaviour and satisfaction.

In the satisfaction-compliance link was evidence that despite the medical profession’s prior reluctance to consign more power to the consumer the ultimate success of its technical ability already lay in the hands of its patients. A dissatisfied patient would not take medication and may well refuse to attend subsequent appointments.

While the consumer movement has pressed for organizational and structural changes the guiding principle has been a belief in the value of the consumer’s opinion. The result of this pressure on the health service is manifest in a shift in the definition of quality utilized. If the patient is to be served then he or she must have a voice in the process of medical care. Satisfaction has therefore come to be seen as a legitimate and desired outcome in itself, not solely as a means of improving compliance. It has become an attribute of quality, a legitimate health care goal [I I]. As Vuori has concluded:

patient satisfaction could be included in Quality Assurance assessments as an attribute of quality care; as a Iegiti- mate and desired outcome. Put simply, care cannot be 01 high quality unless the patient is satisfied (12, p. 1071.

This goes beyond treating patient evaluations as an optional perspective on quality--it is quality. Satis- faction is one of a number of desired end products of health care. Donabedian goes further when he states that “patient satisfaction may be considered to be an

element in health status itself” [13].

Current impetus

The rise in the popularity of satisfaction surveys in the 1980s can be traced to an increase in the evaluation of public sector services and a revival of the consumerist ethos occurring over the same time period. In turn, this rise in evaluation can be traced to two main sources: firstly, a desire for the greater accountability of health professionals which, in part, was a product of the rise in consumerism; and secondly, a desire to gauge efficiency accurately in a service sector industry demanding ever increasing resources.

The evaluative procedures adopted have necessitated the clarification of criteria for measurement and the definition of minimum standards. Determining relevant criteria for the evaluation of health care is of great political importance. Current potential indi- cators range from the relatively esoteric ‘QALYs’ to the more publicly comprehendible lengths of waiting lists. A dearth in the number of relevant criteria has provided a sparsely populated agenda upon which patient satisfaction has been placed.

While the satisfaction-compliance link was an important factor in the early legitimization of the patient perspective it has played little part motivating the more recent proliferation of surveys. For this reason the assumed meaning of satisfaction is to be

Patient satisfaction: a valid concept? 511

found primarily in the writings of those who advocate the concept from more consumerist motives and values. By looking in detail at the reasons why such advocates believe the pursuit of patient satisfaction is in line with their goals it is possible to clarify its assumed meaning.

SATISFACTION: ITS ASSUMED MEANING

Client satisfaction is of fundamental importance as a measure of the quality of care because it gives information on the provider’s success at meeting those client values and expectations which are matters on which the client is the ultimate authority. The measurement of satisfaction is, therefore, an important tool for research, administration, and planning [14, p. 251.

This statement by Donabedian, a renowned ‘guru’ of quality assurance and advocate of patient’s evaluations of care, contains a number of assumptions about the way in which patients evaluate. While Donabedian has not advocated the uncritical use of questionnaires and has called for further research into the concept of satisfaction, this quote illustrates a number of commonly held assumptions which tend to be used by others in the interpretation of satisfaction survey results.

First assumption

Expressions of satisfaction are of conditional utility: they are important because they are assumed to be a function of some prior process (viz. the meeting of ‘client values and expectations’). The importance of satisfaction as a measure of quality is therefore dependent upon the degree to which it actually is the product of the latter. The majority of commentators make the commonsense assumption that a substantial link between satisfaction and the fulfilment of expectations exists.

Critique. While Locker and Dunt [15] have noted the preoccupation of most researchers with identify- ing socio-demographic correlates of satisfaction rather than developing a solid socio-psychological theory of satisfaction a number of studies have been conducted to investigate precisely how patients evaluate. The majority of the models developed have been reviewed extensively by Pascoe [16] and include expectations and values as of assumed central importance.

VaIue-expectancy models

The most well known value-expectancy model has been proposed by Linder-Pelz [17] and was based on the job satisfaction research by Fishbein and Azjen [18]. Linder-Pelz characterizes patient satisfaction as a positive attitude which is related to both his/her beliefs that the care possesses certain attributes and his/her evaluations of those attributes.

Attributes can be seen as distinct dimensions of health care (access, efficacy, cost, convenience, etc). Consequently, satisfaction is based on two pieces of

information: belief strength and evaluations of dimensions of care. Specifically, measures of belief strength (B) about attributes (i) and measures of evaluation (E) of those attributes are multiplied and the products summed. Thus, an equation is derived:

Attitude = 1 B,E,. ,=I

Alternative models

Three alternatives to the value-expectancy models of satisfaction have been identified by Lawler [19]: discrepancy theory, fuhilment theory and equity theory. Discrepancy theory posits that satisfaction is the result of the perceived discrepancy between that which an individual desires and that experienced as a proportion of those desires. Authors vary in their meaning of ‘desires’; some treat the latter as ‘expectations’, others as what is ‘important’ and some as what ‘should be’. Most satisfaction studies have implicitly used a discrepancy approach [2&23].

Fulfilment theory is to be found implicitly in a number of studies [24,25] and defines satisfaction as the difference between rewards desired and those received. Unlik

e advocates of discrepancy theory,

however, fulfilment theorists consider the simple difference between what occurs and what should be/expected/important.

The final group of theories identified by Lawler are equity theories; these propose that satisfaction is perceived equity, or perceived balance of inputs and outputs. Equity theory stresses the importance of evaluating one’s own balance with other’s balances and therefore introduces the role social comparison processes might have in health care evaluations.

The failure of modeliing

Research testing a variety of the theories outlined previously has been conducted by Linder-Pelz [17,26]. A number of hypotheses derived from the original Fishbein and Ajzen theory and the discrepancy and fulfilment models were tested. These hypotheses stated how particular perception-value interactions may determine satisfaction. Data was gathered from 125 first-time patients at a primary care clinic. This included patient’s health care values, expectations, and sense of entitlement to care, which were collected immediately prior to seeing a physician; the individual’s post-visit satisfaction with different aspects of care was also collated.

The Fishbein and Ajzen theory was tested but no support found. In further analyses satisfaction was found to be unrelated to fulfilment but inversely correlated with discrepancy: the better the perceived occurrence in relation to prior expectation, the more satisfaction.

While the effect of expectations was shown to be significant, independent of other variables (see later), it is worth noting that they still only explained 8% of the variance in satisfaction. Furthermore, although

512 BRIAN WILLIAMS

values and perceived occurrences were also found to have independent effects on satisfaction together with expectations these social psychological variables explained 10% of the variation in satisfaction.

The overall conclusion from Linder-Pelz’s work is that very little satisfaction has been explained in terms of expectations and values despite there being some correlation.

An inability to explain the variation of satisfaction levels was also found in an impressive large scale study by Thompson [27] despite considering a whole range of possible variables outside of health service control. The model utilized by Thompson is shown below and necessitated a lengthy questionnaire of 300 variables which was subsequently given to 1357 patients.

Perceptions of the quality of hospital care = fn. [demographic characteristics, personal (non- demographic) characteristics, previous experience, attitudes, expectations, outcome, physical resources of hospital, organizational management, quality of recall, duration of stay, insti- tutional characteristics, disease characteristics]

From factor analysis six factors of patient perceptions of the quality of care were found and indices developed. From the independent variables, a similar process led to the production of seven indices of hospital and personal characteristics. Six multiple regression equations were calculated in the form:

SFY, = Q + i (/I,. SFX,) + c ,=I

where:

SFY, = the ith criterion variable SFX, = the jth predictor variable

a = constant 8, = standardized partial regression

coefficient for the jth predictor variable

t = error term.

Despite the thoroughness of the investigation the percentage of variance explained by the regression equations remained surprisingly low, about 20% for satisfaction with nursing care. Thompson concludes that “other important considerations from the set of unmeasured attributes must provide the bulk of the explanation”.

While there is evidence to suggest that patient’s expectations and values are involved in evaluations they do not appear to be related in any simplistic fashion. One possible explanation may be that expectations are modified throughout the process of care thus influencing subsequent evaluations. To conclude, therefore, we do not currently know the nature of the relationship between expectations and expressions of satisfaction; however, there is little evidence to suggest that satisfaction is largely the result of fulfilled expectations and values. Given the desire

to improve levels of satisfaction one would naturally look to see how service provision could be altered in order to fulfil more patient’s expectations; however, this would be a futile effort is expressions of satisfaction with care have little to do with the fulfilment of expectations.

Second assumption

A second implicit assumption is that an expression of satisfaction with a particular aspect of care necessarily implies that certain attributes of that aspect have, in some way, been approved of or affirmed by the user.

Critique. It would appear legitimate to assume that if, on a questionnaire, a patient indicates that they are ‘satisfied’ with a doctor’s explanation of their health problem that this means the latter has fallen within reasonable proximity of the patients’ expectations regarding medical explanations. However, Linder- Pelz’s study outlined above has thrown an interesting light on this issue which raises the possibility that evaluations may be made independent of the attributes of the aspect of care involved.

In Linder-Pelz’s study the most important an- tecedent social-psychological variable was found to be ‘expectations’. It is important to note that patient expectations were found to have an independent effect on satisfaction (i.e. irrespective of their fulfilment). The practical implication of this finding is consider- able. It

suggests that beliefs about doctor conduct prior to an encounter play a significant role in determining subsequent evaluations of the doctor conduct, irrespective of what (s)he actually did or was perceived to have done. It suggests that patients are likely to express satisfaction no matter what care the doctor gives, at least in the setting of the present study (26, p. 588].

To suggest that evaluations of cart may have littlc, if anything, to do with the care itself undermines the assumed meaning and hence the utility of expressions of satisfaction. Such a conclusion indicates that patient satisfaction may originate in factors outside of the health care system. This is a possibility that has already been recognized within satisfaction studies of the social services. In a review of the subject Shaw makes the following point.

Client evaluations are relative to context. to knowledge of services, to expectations, to help received in past encoun- ters, to help received from other services, to perceptions of the ‘pleasantness’ of the social worker Unless such factors are taken into account, we can never be sure whether the high rate of client satisfaction is related more to factors like knowledge or limited expectations, than the actual helpfulness of the social service contact [28, p. 2801.

A study by Vuori [29] in 1972 suggested a similar occurrence with hospital and general practice patients. He found that hospital patients took technical competence for granted and detected defects in the behavioural aspects of care and communication. The patients in general practice, however, took empa-


thy for granted but were slightly suspicious about the technical quality.

It is this ‘taking for granted’ that is relevant; the expression of satisfaction might simply reflect the latter rather than any active evaluation. What needs to be known is when and under what conditions patients take the quality of particular aspects of services for granted. Without a knowledge of this it is impossible to accurately interpret satisfaction survey results.

This does not indicate that expressions of satisfaction have nothing to do with the object but rather that the assumption that satisfaction is entirely, or even largely, the product of an evaluation per se may, in at least some instances, be untrue.

Third assumption

The third implicit assumption concerns the existence and nature of patient’s values and expectations. If satisfaction is deemed to be related in some way to values and expectations then a quite clear assumptio’n is being made: values and expectations exist. Further- more where expectations do exist and are utilized by patients in evaluations it is assumed that the nature of these does not undermine the meaning and utility of expressions of satisfaction.

Critique. While there are aspects of service provision for which patients clearly have expectations on which they can base their evaluations (e.g. hotel functions or the ‘amenities’ of care) a number of situations are apparent where expectations might not exist. To some degree this overlaps with the second assumption discussed above where ‘taking for granted’ might be seen as either the non-existence of expectations and/or as a reflection of a passive role adopted in relation to health professionals.

Firstly, West [30] has shown that if a service user is coming into contact with the system for the first time then expectations which, for many have been formed through past experience, might be waiting formation.

Secondly, the greater the perceived esoteric or technical nature of treatment the more likely it is that many service users will not believe in the legitimacy of holding their own expectations or of their evaluations. Such a phenomenon would explain why some studies [31] have found that patient’s evaluations of technical matters explain little of the variance in global satisfaction despite the technical aspects of care being the main determinant of clinical outcome. Reports of satisfaction with technical aspects of care may be more accurately interpreted as an expression of confidence in the ability of the medical staff involved rather than the product of an active and critical evaluation. In such a situation it could be argued that expectations do exist but that their nature undermines the assumed meaning of satisfaction and consequently its utility.

Thirdly, if the Parsonian sick role and the passivity it necessitates holds true then the user might deem the

very idea and legitimacy of an evaluation unfounded. In such a situation a patient’s expectations and wishes might be that he or she remains passive and health professionals paternalistic. The fulfilment of such expectations might produce an expression of satisfaction but would be of little consumerist value. In such a scenario patient satisfaction could be said to be primarily a reflection of the role patients adopt in relation to health professionals irrespective of the quality of the care itself.

The arrival of satisfaction surveys has heralded an important change in the perceived importance and legitimacy of public opinion within the NHS. How- ever, there is little evidence to suggest that this ‘enlightened’ idea, though officially part of the new look NHS, has been fully accepted by members of the public. The health service now has to listen to the patient, to heed public opinion. But this same public has traditionally been regarded as ill-equipped to assess health care; passive patients have been regarded as ‘good’ patients and their opinion of the quality of health care deemed illegitimate. Con- sumerism, whether in the form of public partici- pation, formal accountability, democratization or freedom of choice in the marketplace, relies on three fundamental and basic factors:

(1) The existence of consumer opinion (2) A belief (by the consumer) in the legitimacy

of that opinion (3) Willingness to engage in an expression of

that opinion.

In other words consumerism is dependent on a refusal to accept paternalism; it relies on the existence of consumers and not passive patients. Consequently. satisfaction data can only be useful if patients leave passivity and Parsonian roles behind and actively evaluate and criticize. As previously mentioned the expression of satisfaction may not necessarily mean that a critical evaluation has taken place; it might just as well be an expression of the non-existence of opinion and/or an acceptance of medical paternalism. Most survey forms do not allow for the possibility of non-evaluation since respondents are usually offered fixed replies all of which assume an evaluation has taken place. The full acceptance of paternalism is the acceptance of a role which abdicates the right to evaluate and a rejection of consumerism and partici- pation within health care.

While some patients might critically evaluate their care the majority of studies suggest that most service users are very uncritical of it, allowing care to be of extremely poor quality before expressing dissatisfaction. Consequently, reports of overall satisfaction tend to be high, often over 90%. A review by Lebow [32] listed the results of over 50 surveys; the average percentage of satisfied patients was 77.5%. Further- more, satisfaction has been shown to be positively related to age [16] a finding which might be expected if the traditionally passive role predominates among

514 BRIAN WILLIAMS

the elderly as opposed to the more consumerist- oriented role among younger generations.

From the above it would seem reasonable to expect that a passive 80 yr old coming into contact with neurosurgery for the first time is likely to have few expectations. In reality the total non-existence of expectations is unlikely. A more sensible explanation is that the expression of dis/satisfaction is governed primarily by (negative) expectations as to what will/should not happen. The latter will take the form of typical social values and norms which the user might only become conscious of when transgressed. Taking the latter example it would be quite reasonable to assume that the elderly person would expect not to be physically assaulted by the said neurosur- geon. It is likely that expressions of satisfaction will occur assuming no negative expectations have been transgressed. As a result we would expect that a wide range of behaviour to be permissible with only so- cially extreme behaviour causing dissatisfaction.

Nelson and Larson [33] have demonstrated the latter through an investigation of the effect of ‘good’ and ‘bad surprises’ on satisfaction levels. A ‘good surprise’ could be defined as care going well beyond what was expected while a ‘bad surprise’ is equivalent to the transgression of typical social values and norms. The results indicated that the majority of those with no ‘surprise’ expressed satisfaction while those who had experienced a ‘bad surprise’ were more likely to have expressed dissatisfaction.

SATISFACTION: AN UNGROUNDED CONCEPT

It is clear that the process of evaluation and thus the meaning of patient satisfaction data is highly dependent upon the role in which patients perceive themselves in relation to the health care system. Advocates of satisfaction surveys not only implicitly assume that all users do not behave in the traditional non-critical, passive manner but also fail to outline what role they think patients should evaluate accord- ing to. It is easy to say that patients should now act as consumers but what does that mean in practice in a public sector service such as health care? There are, after all, a number of models of consumerism [34] each entailing a different role for the service user. For each role adopted a different evaluation will result irrespective of the care provided.

If we wish to fully understand the views of service users we must first discover what rights and obli- gations they sense they have, in other words what they perceive their role to be. Current satisfaction surveys contain the in built assumption that all service users are acting out the same role and interpret the data accordingly; as we have seen the possibility of the passive acceptance of health care is, in practice, denied. While the existence of patient satisfaction surveys proclaims a concern for the empower- ment and rights of service users the nature of current

interpretation undermines it by denying service users

the right and power not to behave as consumers. The problems of interpretation outlined above

arise not only from the assumption that service users behave in a particular consumer-like manner but also

from the assumption that the concept of patient satisfaction exists in a meaningful way for service users. Since satisfaction surveys are concerned with accessing user opinion the concept of satisfaction can

only be meaningful and useful to researchers if service users think, act and evaluate in terms of it. The above discussion raised the possibility that some patients simply might remain passive and not evaluate and

that others do evaluate but perhaps not in terms of ‘being satisfied’. In sociological terms whatever concept we use it must be grounded in people’s everyday lives [3.5]. This is essential since the utility of any concept used is dependent upon the degree to which it can accurately embody users’ views of service provision.

While the studies mentioned above challenge a number of assumptions undergirding contemporary interpretations of expressions of satisfaction they only suggest that the concept may not be grounded. Other studies, however, have made more explicit their intention of testing the ‘groundedness’ of the concept. In discussing the results of a study in which 95

patients were interviewed Fitzpatrick and Hopkins [36] commented that they “were struck by a lack of fit, in many respects, between patient’s own accounts of their experiences and the assumptions about patients contained in satisfaction research”. They go on to conclude that the conceptual framework derived from patient satisfaction research provides only partial and sometimes misleading insights into the perspectives of the patients studied. The implication is that patients often do not evaluate in terms of being satisfied.

Repeatedly when qualitative methodology is utilized, as in the Fitzpatrick and Hopkins study, little if any support is found for believing that patients think and evaluate in terms of a continuum of satisfaction. Likewise Calnan [37] and Locker and Dunt [ 151 have noted that patients often display a critical nature when given the opportunity, through more open ended questions, to express themselves in their own terms. Consequently, quantitatively measured expressions of satisfaction tends to be high [32] while qualitative reports reveal greater levels of disquiet 137. 3X] T!lc possibility exists thercforc that the reductionism necessitated by quantitative methodology has rendered satisfaction results devoid of much of the meaning they were intended to embody.

In effect the said reductionism may have resulted in diverse opinions ranging from ‘I’ve evaluated the service and I’m happy with it’ through ‘I don’t really think I have the ability to evaluate, but I do have confidence in the staff’ to ‘the service was appalling but I don’t like to criticize, after all they’re doing their


best’ being collapsed into a single category of users all of whom expressed ‘satisfaction’.

The use of quantitative methodology is not inher- ently inappropriate to collating patient’s evaluations. However, questionnaire design must be informed by a knowledge of how and whether patients evaluate if they are to embody user opinion. It has been shown, however, that current design is informed by a set of commonsense but unproven assumptions. This does not mean that current surveys are useless but that they are unnecessarily blunt; if dissatisfaction is expressed then there is likely to be something wrong with the service provision. But what of the 80% of people who express satisfaction? The call in the 1983 NHS Management inquiry [l] was to obtain the experience and perceptions of patients, not to engage solely in ‘troubleshooting’. If future questionnaires are to contribute to this then more qualitative research is required to inform their design and interpretation. As Whitfield and Baker have recently commented:

4.

5.

6.

7.

8.

9.

10.

Poor questionnaires act as a form of censorship imposed on patients. They give misleading results, limit the opportunity of patients to express their concerns about different aspects of care, and can encourage professionals to believe that patients are satisfied when they are in reality highly discon- tented [39, p. 1521.

11.

12.

13.

14.

15.

CONCLUSION 16.

Many of the assumptions on which the utility of satisfaction surveys is based are currently unsubstan- tiated. Patients may have a complex set of important and relevant beliefs which cannot be embodied in simple expressions of satisfaction. Patient satisfaction questionnaires do not access an independent phenomenon but, in a sense, actively construct it by forcing service users to express themselves in alien terms; consequently, inferences made from their results may misrepresent the true beliefs of service users. The original motivation behind satisfaction surveys was to introduce some element of consumerism and accountability to health care; however, through high levels of relatively meaningless expressions of satisfaction an illusion of consumerism is created which seldom does anything but endorse the status quo. If, as suggested in the Griffiths report, service providers should obtain the experience and perceptions of patients and the community then research must first be conducted to identify the ways and terms in which those patients perceive the evaluate that service, and whether some evaluate at all.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

I.

2. 3.

REFERENCES

DHSS NHS Management Inquiry. HMSO, London, 1984. Working for Patients HMSO, London, 1989. Carr-Hill R., Dixon P. and Thompson A. Too simple for words. Hlth serv. J. 15 June, 1989.

27.

28.

29.

30.

Parsons T. The Social System. Free Press, Glencoe, 1951. Szasz T. and Hollender M. A contribution to the philosophy of medicine: the basic model of doctor-patient relationship. Archs int. Med. 97, 585, 1956. Alpert J. J. Broken appointments. Pediatrics 34, 127, 1964. Francis V., Korsch B. M. and Morris M. J. Gaps in doctor-patient communication: patients’ response to medical advice. New Engl. J. Med. 280, 535, 1969. Becker M. H., Drachman R. H. and Kirscht J. P. Predicting mothers’ compliance with pediatric medical regimens. J. Pediat. 81, 843, 1972. Wilson P. and McNamara J. R. How perceptions of a simulated physician-patient interaction influence intended satisfaction and compliance. Sot. Sci. Med. 16, 1699, 1982. Ludy J., Gagnon J. and Caiola S. The patient-pharmacist interaction in two ambulatory set- tings-its relationship to patient satisfaction and drug misuse. Drug inreli. clin. Pharm. 11, 81, 1977. Shaw C. D. Introducing quality assurance. Kings fund project paper No.64, London, 1986. Vuori H. Patient satisfaction-an attribute or indicator of the quality of care? Qua/. rev. BUN. 13, 106, 1987. Donabedian A. The quality of care: how can it be assessed? J. Am. medic. Assoc. 260, 12, 1988. Donabedian A. Explorations in Quality Assessment and Monitoring: Vol I. The Definition of Quality and Ap- proaches to its Assessment. Health Administration Press, Ann Arbor, MI, 1980. Locker D. and Dunt D. Theoretical and methodological issues in sociological studies of consumer satisfaction with medical care. Sot. Sci. Med. 12, 283, 1978. Pascoe G. C. Patient satisfaction in primary health care: a literature review and analysis. Eval. Program Plan. 6, 185, 1983. Linder-Pelz S. Toward a theory of patient satisfaction. Sot. Sci. Med. 16, 577, 1982a. Fishbein M. and Ajzen 1. BelieA Attitude, Intention, and Behavior: An Introduction to Theory and Research. Ad- dison-Wesley, Reading, MA, 1975. Lawler E. E. Motivation in Work Organisations. Wadsworth, Belmont, CA, 1973. Greene J. Y., Weinberger M. and Mamlin J. J. Patient attitudes towards health care: expectations of primary care in a clinic setting. Sot. Sci. Med. 14A, 133, 1980. Fox J. G. and Storms D. M. A different approach to sociodemographic predictors of satisfaction with health care. Sot. Sci. Med. lSA, 557, 1981. Zastowney T. R., Roghmann K. L. and Hengst A. Satisfaction with medical care: replications and theor- etic reevaluation. Medic. Care 21, 294, 1983. Pope C. Consumer satisfaction in a health maintenance organisation. J. Hlth Sot. Behav. 3, 406, 1978. Noyes R. W., Levy M. I., Chase C. L. and Udry J. R.

Expectation fulfilment as a measure of patient satisfaction. J. Obstet. Gynecol. 118, 809, 1974. Larson D. E. and Rootman I. Physician role performance and patient satisfaction. Sot. Sci. Med. 10, 29, 1976. Linder-Pelz S. Social psychological determinants of patient satisfaction: a test of five hypotheses. Sot. Sci. Med. 16, 583, 1982. Thompson A. G. H. The soft approach to quality of hospital care. Int. J. qual. and reliabil. Mgmr 3, 3, 1986. Shaw I. F. Literature review. Consumer evaluations of the personal social services. Br. J. Sot. Work 14, 277, 1984. Vuori H. Patient satisfaction does it matter? Qual. Assur. Hlrh Care 3, 183, 1991. West P. The physician and management of childhood

516 BRIAN WILLIAMS

31.

32.

33.

34.

epilepsy. In Sludies in Everyday Medical Life 35. Glaser 9. G. and Strauss A. L. The Discovery of (Edited by Wadsworth M. and Robinson D.). London, Grounded Theory. Aldine, Chicago, 1968. 1976. 36. Fitzpatrick R. and Hopkins A. Problems in the concep- Cleary P. D. and McNeil 9. J. Patient satisfaction as an tual framework of patient satisfaction research: an indicator of quality care. Inquiry 25, 25, 1988. empirical exploration. Social. Hlth Illness 5, 297, 1983. Lebow J. L. Research assessing consumer satisfaction 37. Calnan M. Lay evaluation of medicine and medical with mental health treatment: a review of findings, Eual. practice: report of a pilot study. Inr. J. Hlth Serv. 18, program Plan. 6, 211, 1983. 311, 1988. Nelson E. C. and Larson C. Patients’ good and bad 38. Gabe J. and Lipshitz-Phillips S. Tranquilisers as social surprises: how do they relate to overall patient satisfac- control. Social. Rev. 32, 524, 1984. tion. Qual. rev. Bull. 3, 89, 1993. 39. Whitfield M. and Baker R. Measuring patient satisfac- Winkler F. Consumerism in health care: beyond the tion for audit in general practice. Qual. Hfth Care 3, 15 1. supermarket model. Policy PO/it. 15, 1, 1987. 1992.

Date post:	25-Aug-2016
Category:	Documents
Upload:	brian-williams
View:	212 times
Download:	0 times

Patient satisfaction: A valid concept?

Documents