The measurement of patients' expectations for health care: A review and psychometric testing of a measure of patients' expectations
Bowling A., Rowe G., Lambert N., Waddington M., Mahtani KR., Kenten C., Howe A., Francis SA.
Background: There is recognition of the importance of measuring patients' experiences, expectations and satisfaction. Objectives: To assess the literature on the concept and measurement of patients' expectations for health care, and to develop and test a measure of patients' expectations, using adult patients in community, general practice and hospital outpatient departments in Greater London, Norwich and Essex, UK. Data sources: Major electronic databases including the British Nursing Index, EMBASE, MEDLINE, PsycINFO and the Applied Social Sciences Index and Abstracts were searched between 2000 and 2009. Review methods: Narrative review, semi-structured exploratory study and surveys of GP patients and hospital outpatients immediately before and after their surgery/clinic visit to measure their pre-visit expectations for their health care and their post-visit experiences (expectations met and satisfaction with visit) (site specific). Results: A total of 20,439 titles and 266 abstracts were identified, of which 211 were included in the review. Most research designs were weak, with small or selected samples, and a theoretical frame of reference was rarely stated. The origin of questions about expectations was often absent, questions were frequently untested and those with reported reliability or validity data had generally mixed results. In the survey data the expectations measures met acceptability criteria for reliability; all exceeded the threshold of α = 0.70, in each mode of administration and sample type. Items and subscales also correlated at least moderately with those variables that they were expected to be associated with, supporting their validity. The item means within subscales were generally similar between samples and all-item-total correlations exceeded the acceptability threshold. Descriptive findings revealed that most patients ideally expected cleanliness, information about where to go, convenient and punctual appointments and helpful reception staff, the doctor to be knowledgeable, clear and easy to understand, to be involved in treatment decisions and to experience a reduction in symptoms/problems. Expectations least likely to be met included being seen on time and choice of hospital/doctor (items requested by the ethics committee). Other items that had low met expectations included helpfulness of reception staff, doctor being respectful and treating with dignity (hospital sample), doctor knowledgeable (hospital), being given reassurance, receiving advice about health/ condition, information about cause and management of condition and information about benefits/side effects of treatment, being given an opportunity to discuss problems, and the three items on outcome expectancies. Previous consultations/experiences of health services and health-care staff/professionals most commonly influenced expectations. Overall, pre-visit realistic expectations were lower than patients' ideals or hopes. Most post-visit experiences indicated some unmet expectations (e.g. cause and management of health/condition, benefits/side effects of treatments) and some expectations that were exceeded. Generally, GP patients reported higher pre-visit expectations and post-visit met expectations. Correlations between subscale domains were strongest between the structure and process of health care, doctor-patient communication style and doctor's approach to giving information, all common indicators of the quality of health care, supporting the validity of the measures. The post-visit experiences subscale significantly predicted single-item summary ratings of overall met expectations and satisfaction. GP rather than hospital patients were also independently predictive of expectations met. Other predictors were having no/little anxiety/depression, older age (satisfaction) and fewer effects of health on quality of life (met expectations). Limitations: The surveys in clinics were based on convenience, not random sampling methods. Conclusions: These findings have implications for establishing the quality of health services and informing their improvement. Awareness of the patient's met and unmet expectations should enable staff to understand the patient's perspective and improve communication. This study examined the perspective of the patient only; it is not possible to examine the extent to which any expectations might have been unrealistically too high or too low. This is a challenge for future research. Funding: The National Institute for Health Research Health Technology Assessment programme and the National Co-ordinating Centre for Research Methodology (NCCRM). © Queen's Printer and Controller of HMSO 2012.