Patients’ experiences of their healthcare in relation to their expectations and satisfaction: a population survey

Abstract

Objectives

To investigate patients’ experiences of health services, and how these related to what they had expected to receive, and satisfaction with their care.

Design

Surveys of patients before and after their consultations in general practice and hospital outpatients departments.

Setting

Greater London and Essex

Participants

In total, 833 patients attending 21 hospitals (434 patients; 52%) and 22 general practices (399 patients; 49%) across Greater London and Essex sampled in clinics and a population survey.

Main outcome measures

Patient expectations of care, patient satisfaction.

Results

Compared with younger people, and those in black and ethnic minority groups, older people (aged 65+) and White British people had significantly higher overall realistic expectations of their care (pre-visit realistic expectations score: age 60+: mean 53.26 [standard deviation 13.73]; age <60: 56.20 [15.17]; White British: 54.41 [13.50]; Black and other ethnic groups: 56.90 [16.15]) and greater satisfaction post-consultation (satisfaction score age 60+: 1.71 [0.80]; age <60: 1.97 [0.97]; White British: 1.79 [0.89]; Black and other ethnic groups: 2.01 [0.95]). Pre-visit ideal and realistic expectations of care was not significantly associated with patient satisfaction, although met expectations (post-visit experiences) were. Elements of these which was predictive of satisfaction were communication with the doctor, information conveyed and clinical outcomes. Factors associated with satisfaction included having a sense of control over one's life, being older, female, White British and attending general practice, compared with hospital outpatient clinics.

Conclusions

It is the ability of the system to meet patients’ expectations in respect of the emotional and human features of the consultation, and the clinical outcomes, that matter most to people. This research also questions prevailing stereotypes of older age: it is not the case that older patients are more satisfied with their care because their expectations are lower. In fact, they are higher, but they believe that they are being met.

Introduction

In this paper we describe research on patients’ experiences of health services, and how these related to what they had expected to receive, and their satisfaction with their care. English health ministers have repeatedly expressed their commitment to a ‘patient-centred NHS’,¹ with Andrew Lansley speaking of his determination to put patients ‘first’,² ‘at the heart of care/everything the NHS does’^3,4 and ‘in the driving seat’.⁵ The 2012 Health and Social Care bill has been presented as a means for patients, advised by their GPs, to decide who treats them and how. They will be helped by greatly enhanced information on the performance of providers, including the experiences of those patients whom they have treated previously.

Patient satisfaction is defined here in terms of patients’ evaluations of what happened during their healthcare (i.e. evaluations of their ‘experiences’ of healthcare). Patients’ experiences are their direct, personal observations of their healthcare. Patients’ expectations have been defined as the anticipation that given events are likely to occur during, or as an outcome of, healthcare. Thus, what people expect to receive from their healthcare, compared with their observations of what they received in practice (‘experiences’), are potentially important in influencing patients’ evaluations of their care (‘satisfaction’).^6,7 It is often argued that an excess of perceived delivery (e.g. of healthcare) over what is hoped for, anticipated or expected leads to increased satisfaction, and conversely, that unmet expectations lead to increased dissatisfaction.^6,7

However, expectations have at least two elements.⁶ The first relates to what they would expect in ideal circumstances, in other words what they feel should be capable of being provided in a rich industrialized country. The second reflects what we might term realistic expectations, or what they actually expected to happen, which will be shaped by, among other factors, their understanding of the financial pressures on the health service and media coverage of its performance. But how do patients’ characteristics influence their expectations, and do their expectations affect their satisfaction with the services they receive?

There are many widely held beliefs, supported to greater or lesser degrees by empirical evidence. One is that older people, many of whom will have spent part of their lives in more austere conditions, may have lower expectations and simply be grateful for what they receive, while younger people, whose life experiences have been shaped by the consumer revolution and instantaneous gratification, may be much more demanding. The expectations of individuals from ethnic minority populations who were born abroad may be shaped by having grown up in very poor countries. But are these any more than stereotypical assumptions? Next we describe what we found when we asked patients not only about their experiences of receiving healthcare but also how this related to what they had expected of the service they received, both in terms of what it should be like in ideal circumstances and what they realistically expected it to be like.

Objectives

The objectives of the analyses were to investigate patients’ experiences of health services, and how these related to what they had expected to receive, and satisfaction with their care.

Methods

The study design was based on surveys of patients before and after their consultations in general practice and hospital outpatients clinics.

Sample size and selection

Sample sizes aimed for were a minimum of 100 interviews and 500 self-completed questionnaires (sufficient for assessments of interview bias and multivariable analyses). The large samples needed, within the study timetable, required two methods of sampling: a convenience sample of clinic patients waiting to see their doctors in outpatients or primary care, and a population survey screen to identify and include patients with a pending outpatient or GP appointment. Patients were not randomly sampled, requiring caution when interpreting sample estimates.

The clinic patient surveys

Two UK hospital outpatient clinics (general medicine and cardiology) and six primary care centres participated in the study. Waiting clinic patients were approached by a member of the research team and invited to participate. Patients were asked to sign a consent form if willing to participate, to complete the pre-visit questionnaire while waiting to consult the doctor, and the postvisit questionnaire afterwards. They were asked to return them to the researcher in the clinic waiting area. They were given freepost envelopes to return them in if they failed to complete them at the clinic, and asked to return them within a week. The sample is regarded as a convenience sample because we had no access to a sampling frame as full clinic lists were not accessible to us (due to patient confidentiality), it was easy to miss attending patients (especially as some patients go to different waiting areas or corridors, and some go off with nurses to be weighed, or for urine tests on arrival), and we did not have their personal details if they took the consent form away with the questionnaires. Thus clinic site response rates could not be calculated.

The population patient surveys

This was conducted in Greater London by Ethnicfocus, based on systematic random sampling of postal sectors, by concentration of ethnic group; then a focused enumeration procedure ensuring the representation of people in ethnic minority groups as well as White British. It involves interviewers asking at main sample address about eligibility of those living at the address, to the left and at the addresses to the right. The advantage of this method is to ensure representation of patients in diverse ethnic groups. The questionnaires were given by interviewers to eligible respondents to self-complete (those identified as having a GP or hospital outpatient appointment within four weeks). Interviewers asked them to complete the pre-visit self-administration questionnaire immediately before their clinic visit, and the post-visit questionnaire immediately afterwards. The interviewers revisited them within a week of the clinic visit and collected the questionnaires. Response could be calculated for the Ethnibus survey: 1413 London (inner and outer) households were contacted, of which 318 were eligible (had a hospital/GP appointment within 4 weeks), 255 agreed to participate and completed both questionnaires (80% response rate) and 63 refused. The Ethnibus responders represented a further 19 hospitals and 16 primary care centres.

The questionnaires

We developed two questionnaires, the first was administered to patients in advance of a medical consultation, to explore patients’ (a) ideal and (b) realistic expectations in relation to the same sets of expectation items, the second was administered after they had completed the consultation (within a week of the clinic visit), and asked patients to rate their experiences of these same expectation items, in order to measure whether their expectations for these items, had been met. Thus against each expectation item, patients were asked to rate their:

Pre-visit:

1. Ideal hopes about what would happen during the consultation

2. Realistic expectations of what would happen (‘in reality’)

Post-visit:

1. Actual experiences (to measure expectations met).

This post-consultation questionnaire also included a question asking respondents to rate their overall satisfaction with the consultation, on a 5-point Likert rating scale from ‘Very satisfied’ to ‘Not very satisfied’.⁷ A longer, item-specific satisfaction scale was not used given the length of the expectations scales, and the additional patient burden this would have imposed. These questionnaires, which are publically available, were developed on the basis of a structured literature review and qualitative research to understand what patients considered important, and were piloted.⁷ The 27 item-statements (Box 1 which is available online at available online at Appendix 1: http://jrsm.rsmjournals.com/lookup/suppl/doi:10.1258/jrsm.2012.130147/-/DC1)) were grouped into six subscales addressing the physical environment, finding their way around, communication with the doctor, the content of the consultation, the information given and the outcome of the consultation. In the pre-consultation questionnaire, items were scored on a 5-point scale (strongly agree to strongly disagree); in the post-consultation version, five of these items were reduced to ‘yes/no’ to capture whether certain things had happened at the time of the clinic visit. One represented the highest expectation and five the lowest, so lower scores reflect greater expectations. The questionnaires also asked about health status, quality of life, psychological status and a range of sociodemographic characteristics. The measures of expectations met standard criteria for reliability and validity, with a Cronbach's alpha score (internal consistency) greater than 0.70.

Results

One hundred and twenty-eight patients were interviewed, and the number who self-completed the questionnaires was 705. These 833 patients were attending 21 hospitals (434 patients; 52%) and 22 general practices (399 patients; 49%) across Greater London and Essex. Of the 833 respondents, 59% were women (63% in the GP sample, and 53% in the hospital sample); patients had a mean age of 51.8 (range 18.62–95.15, standard deviation 18.02). Fourteen percent were aged 18–29, 17% 30–39, 16% 40–49, 17% 50–59, 18% 60–69 and 18% were aged 70+; they were comparable by sample (e.g. 33% of the GP sample were aged 60+, as were 30% of the hospital sample). Sixty-two percent were White British, 14% South Asian and the remainder were in other black and minority ethnic groups: (65% of the GP sample and 59% of the hospital sample were White British); 56% were home owners (55% in GP sample and 58% in hospital sample). Patients recruited from clinics and from the Ethnibus survey were similar in their sociodemographic circumstances and their questionnaire responses.

Results are summarized in Supplementary Tables S1–S3 (available online at Appendix 2: http://jrsm.rsmjournals.com/lookup/suppl/doi:10.1258/jrsm.2012.120147/-/DC2). They are disaggregated by age (under 60 versus 60 and above) and ethnicity (white British versus black and minority ethnic groups). Results are also presented by site (general practice compared with hospital patients). There were no differences in expectations between men and women. Some significant differences did, however, emerge when the sample was stratified by age. Although there was no overall difference in what was expected in ideal circumstances (though older people had higher expectations about the clinical outcome), older people had higher overall expectations in realistic circumstances, and specifically so in terms of finding their way around, communication with the doctor, consultation content and the information they would receive. There was no significant difference in their expectations for the physical environment or treatment outcome. Although their pre-visit ‘realistic’ expectations were higher, older people also expressed greater satisfaction with the consultation once it was over. This was true for all of the elements of care except for the clinical outcome, which was not significantly different from that experienced by younger people. In this comparison, the consultation content was scored in terms of what was done; older people reported a fuller consultation.

Overall, white British patients had similar ‘ideal’ expectations to those from black and minority ethnic groups but significantly higher expectations in respect of most of the individual elements (with the exception of the information conveyed and the clinical outcome). They did have significantly higher expectations in what they considered realistic circumstances, although this was not significant for ease of finding one's way, consultation content or clinical outcome. As with older people, despite having higher expectations, white British patients had higher levels of satisfaction, although this time there was no significant difference in the content of the consultation, nor did satisfaction with the information conveyed or the outcome achieved differ.

Patients attending general practice had, overall, higher expectations than those attending hospital, especially in regard to the physical environment and communication with the doctor, both in terms of the ideal and what could realistically be expected. The GP patients had higher ideal and realistic expectations than hospital patients about it being easier to get around inside the building, and that there would be enough space in the waiting room. GP patients had higher realistic expectations about the site of the consultation being easy to find, clean inside, the doctor treating them with respect and dignity; and higher ideal expectations about having a choice of doctors to consult where more than one was on site (not included in scaling due to not-applicable responses). These differences would be expected as GP patients would be more familiar with their consultation site (local GP surgery). GP patients were also more likely, than hospital patients, to have their expectations met about the site being easy to find, finding the doctor: being helpful, treating them with respect and dignity, being knowledgeable/understanding about their condition, clear and easy to understand, involving them in decisions about their treatment, and being given full, clear information about how to manage their condition.

Of course, the key question is whether the level of one's prior expectations (and whether they were met) influence one's subsequent satisfaction with the consultation. The initial results were somewhat counter-intuitive, with those whose expectations were higher expressing higher satisfaction. To try to understand these results, we first explored the predictors of having expectations, whether ideal, realistic or met (postconsultation ratings of experiences for each expectation item). The level of initial expectations was not a significant factor; the only factors that were significant were ethnicity, with those who were white British being more likely to have their expectations met (odds ratio [OR] 1.64; 95% confidence intervals [CI] 1.16–2.42) and housing tenure, where home owners were similarly more likely to have them met than those in rented or other accommodation (OR 1.58; 95% CI 1.07–2.32). A second multivariate analysis looked at the predictors of overall satisfaction (whereby the dependent variable took the value of 1 where the patient was satisfied or very satisfied and 0 otherwise) (Supplementary Table S4 available online at Appendix 3: http://jrsm.rsmjournals.com/lookup/suppl/doi:10.1258/jrsm.2012.130147/-/DC3)). As expected, those whose initial expectations had been met (‘experiences’) were much more likely (five-fold) to be satisfied. However, the levels of the initial expectations, both ideal and realistic, were not significantly associated with satisfaction. Other factors associated with satisfaction included having control over one's life, being older, female and White British, and attending a GP. The latter analysis was repeated for the individual elements to identify where meeting expectations were most important. Those emerging as individually predictive of satisfaction were communication with the doctor, where those whose expectations were met were over six times as likely to be satisfied with the consultation (OR 6.77; 95% CI 3.45–13.29), information conveyed (OR 1.99; 95% CI 1.19–3.33) and clinical outcomes (OR 2.11, 95% CI 1.31–3.42). The other elements did not reach statistical significance at the 0.05 level.

Discussion

Principal findings

This research reported that whether patients’ expectations for healthcare were met determines how satisfied they will be with their consultations. The emotional and human features of the consultation, and the clinical outcomes, mattered most to people. Older patients’ expectations were higher than those of younger patients, and they were also more likely to believe that they were being met. GPs’ patients were also more likely than hospital patients overall to have higher expectations, and met expectations, for their healthcare.

Strengths and limitations of the study

The strengths of the study are that it was underpinned by a systematically conducted narrative review of the literature, which was conducted prior to the empirical research presented here, as well as qualitative research to ensure that, as far as possible, the expectation measures we developed, tested and used were thoroughly grounded in the literature and patients’ views.⁷ The limitations of the study also require consideration. As the clinic samples of patients were based on convenience samples, and it was not possible to calculate response rates, and as the Ethnibus survey patients were not randomly sampled, sample estimates need to be interpreted with caution. The sample of study sites may thus be atypical in various ways, and the patient respondents may not be representative of all ambulatory patient populations. Moreover, due to the length of the expectations scales, patient satisfaction was measured using a single global rating question only, rather than a measurement scale of satisfaction with specific service items which might have been more sensitive.⁸

Comparison with other studies and the meaning of the study

A systematically conducted narrative review of the literature was conducted prior to the empirical research presented here. This confirmed the conceptually weak and fragmented nature of existing research on expectations.⁷ This study confirms classic observational research showing how it is the ability of the system to meet patients’ expectations in respect of the emotional and human features of the consultation, and the clinical outcomes, that matter most to people.⁹ Other elements related to the physical environment and the content of the consultation are obviously important for other reasons, but play less of a role in the degree of satisfaction expressed by patients. What is new is the information that the research gives us on how satisfaction is shaped, or not, by prior expectations. First, it is not what patients expect, but rather whether those expectations are met, that determines how satisfied patients will be with their consultations. Second, contrary to what is often assumed, it is not the case that older patients are more satisfied with their care because their expectations are lower. In fact, they are higher, but they believe that they are being met. This stereotype may owe much to the perception of older people growing up during the Great Depression and the Second World War when they faced deep austerity and multiple deprivations. However, older people today are dominated by the baby boomers, whose expectations were shaped by the seemingly limitless possibilities of the 1960s and the massive growth in consumerism that accompanied it.¹⁰ Interestingly, they are no different from younger people in what they expect in ideal circumstances; rather they are less willing to accept lower standards in the reality that confronts the NHS. This is consistent with a wealth of other evidence on how they differ from earlier generations, viewing themselves as engaged in a process of successful and healthy ageing, rather than one of relentless decline.^11,12

Monitoring patients’ satisfaction with the care they receive, along with details of their experiences of care, is now an accepted component of quality assurance.^8,13 The public, the politicians who represent them, and the health professionals and managers responsible for delivering care will all benefit from information on how satisfied they are with the care they receive, and on patients’ experiences. This information is essential to identify those areas where care is suboptimal and to learn from where satisfaction has increased so that whatever is responsible can be adopted more widely. However, for this information to be acted upon, it is necessary to understand how it is shaped by patients’ expectations of the care they will receive, not least because these have almost certainly changed over time (as they have with regard to many other aspects of life) and are likely to continue to do so, and because they may vary within the population.

Unanswered questions for future research

Qualitative research is needed in order to add insight to the findings that older patients’ expectations for healthcare were higher than younger patients, and they were more likely to believe that they were being met.

Conclusions

These findings have implications for health professionals, managers and politicians. Overall, older people are satisfied with the healthcare they are receiving and as we have shown, this cannot be dismissed as a consequence of their lower expectations. Indeed, it chimes with the finding that satisfaction with the NHS among the general public is now at an all-time high.¹⁴ But there is no room for complacency, given that the delivery of healthcare in England is undergoing profound and unprecedented change, with many services facing severe budget cuts.¹⁵ It will be essential for those who are delivering care in the midst of organizational and, frequently, personal turbulence, to remain focused on what matters most for patients, which means most of all effective communication, adequate information and good outcomes.

DECLARATIONS

Competing interests

None declared

Funding

This project was funded by the NHS R&D Programme Health Technology Assessment (HTA) Programme; project number: 07/58/01.The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Department of Health. The funding body has consented to this publication and mandates open access publication. The sponsors played no role in the design, execution, analysis, interpretation and writing of the study

Ethical approval

This study was approved by Norfolk and Norwich LREC and London MREC

Guarantor

AB

Contributorship

AB, a social scientist, and GR, psychologist, obtained the funding for, and co-managed, the research presented here, and drafted early versions of this paper. Both had access to the raw data. MM, a specialist in European public health wrote the final draft of this paper, with AB, and MM provided the policy context. The source used was the Health Technology Assessment (HTA) report on this topic,⁷ plus additional analyses (odds ratios) undertaken by AB for this paper

Acknowledgements

We thank the patients who participated in the study and their doctors and clinic staff, especially Dr Richard Pearson and Dr Ameet Bakhi, for facilitating outpatient recruitment. We also thank Ethnifocus and their Ethnibus interviewers, Katie O'Donnell, Laura White and Lesley Williamson for the fieldwork, Corinne Ward for study administration, Heather Leishman of Norfolk and Waveney PCT Research Network for her help liaising with local practices and Mable Saili of North Central London Research Network, for valuable help with facilitating honorary contracts. We also thank Sally Brearley, who represented patients’ groups, for her valuable feedback on the questionnaires and design

Reviewer

Richard Baker