Surveys of satisfaction with the NHS tend to prompt discussion about reform. Research by Irene Papanicolas, Jonathan Cylus and Peter Smith investigates what determines people’s satisfaction with their health system and why it is oftentimes erratic. International comparisons show that overall satisfaction appears to represent something different in each health system, indicating that there is no panacea for improving satisfaction ratings in all countries. In the UK, they find that GP performance is closely linked to overall opinions of the NHS, and so policies that addresses GP responsibilities might have an important influence on satisfaction levels.
The National Health Service (NHS) is a source of pride among the British, even featuring prominently in the Opening Ceremony of the London 2012 Olympics. Yet recent surveys of satisfaction with the NHS reveal a more erratic sentiment among the British concerning their health system’s performance. For example, for the past three years the BBC has reported results of patient satisfaction from the British Social Attitudes Survey. Last year they found a “Record fall in ‘NHS Satisfaction’” despite “NHS Satisfaction [being] ‘at a record high’” the year before. On April 3 of this year, a BBC article reported that now “NHS satisfaction [is] ‘stabilising after [its] record fall’”, (see Dan Wellings here and here for more on the British trends in NHS satisfaction). Each year’s publication of these ratings generates discussion about the degree to which NHS reform is needed. Yet despite large swings in people’s satisfaction with the health system, little is known about what determines peoples’ opinions of the health system, or the extent to which these satisfaction metrics can be used to identify policy priorities.
Previous research suggests that by themselves, individual experiences related to receiving health care are poor predictors of overall satisfaction; institutional arrangements and the attitudes they create may be more suitable. In our recent Health Affairs article “An analysis of survey data from eleven countries finds that ‘satisfaction’ with health system performance means many things”, we too seek to better understand what drives health system satisfaction in a selection of countries. Using survey data from 2010 collected by The Commonwealth Fund, we find that opinions of health system performance vary considerably across countries (Figure 1). Of those countries studied, the United Kingdom had the lowest share of respondents reporting that the entire health system needs to be completely rebuilt—ironic perhaps, since the system has in fact been entirely revamped only three years later with the new Health and Social Care Act coming into effect in April.
Figure 1: Opinions of health systems, selected countries, 2010
To better understand what drives peoples’ opinions and identify whether policy makers can learn from experiences of other countries to make potential improvements at home we use a variety of statistical techniques (see our paper for more details). Based on these, we find that along with socioeconomic indicators—including individual health care coverage, health status, and income— experiences with health service delivery cannot explain more than 13 percent of the variation in health system satisfaction in any country. This seems to indicate that actual occurrences during health care delivery have little to do with individuals’ health system ratings.
In general, we observed strong country-specific patterns that could not be accounted for by any factors measured in the survey. Indeed, respondents in some countries (including the United Kingdom) were more likely to rate their health system highly, even when controlling for many different possible cofounders, compared to respondents in other countries (such as the United States). Whatever these unobserved factors are, they seem to play a large role in determining population satisfaction with the health system.
Our analysis does indicate that there are a few types of experiences that seem to be of universal importance to patients, such as long waiting times for appointments and diagnosis, which were widely associated with discontent across countries. In the United Kingdom, short waiting times, not receiving duplicate tests, and not experiencing a medical error increased the likelihood that someone would rate the health system highly. Other health care experiences, including receipt of conflicting information from providers or a test result not being available appeared to have little to no effect among British respondents.
We also investigated whether other perceptions of the health system, including positive opinions about affordability, effectiveness, and GP quality, were common amongst individuals who rate the health system highly. In all countries aside from the United Kingdom we found that respondents who felt care was affordable were more likely to rate their system favourably. In the United Kingdom however, overall satisfaction was found to be largely independent of perceptions of affordability, perhaps not surprising since all UK residents benefit from free access to NHS health care at the point of service.
British respondents’ overall rating of their health system was much more sensitive to their perception of their GP services (Figure 2). Brits who were very content with their GP had an over 80% probability of rating the entire health system highly. In other countries, GP rating was also associated with overall health system satisfaction, but to a much lesser degree. GPs likely have such a strong influence on opinions of overall health system performance because in many instances, due to the gatekeeping design of the system, they are an individual’s main, if not only point of contact with the NHS.
Figure 2: Probability of respondents’ high overall opinion of health system conditional on ratings or regulated doctors, selected countries, 2010
What can policymakers learn from our study? Interestingly, we found very few commonalities across countries in the factors that seem to drive people’s perceptions of their health system. This would indicate that there is no panacea for improving satisfaction ratings in all countries. In effect, overall satisfaction appears to represent something different in each health system, and thus the relevant policy actions that can improve health system ratings are likely to differ across health systems as well. Although it may be interesting to compare a single indicator of satisfaction across countries, such comparisons are of limited value for drawing policy inferences. Policymakers can nevertheless use this type of analysis to determine priorities for improvement in their own country, and to identify areas where any deterioration in the performance of a country’s system might adversely affect satisfaction. Given our finding that ratings of GP performance are closely linked to overall opinions of the NHS, it will be interesting to see if and how the change in responsibilities of GPs in England will influence health system satisfaction.
Note: This article gives the views of the author, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting.
Irene Papanicolas is a lecturer in health economics in the Department of Social Policy of the London School of Economics. Her research interests include measuring health system performance, international comparisons of health systems and performance based payment systems.
Jonathan Cylus is a research fellow at the European Observatory on Health Systems and Policies and LSE Health, based at the London School of Economics. His research focuses primarily on measuring health system performance and assessing the effects of the financial crisis on health care systems in Europe. Previously, he worked as an Economist in the Office of the Actuary at the Centers for Medicare and Medicaid Services.
Peter C Smith is a professor of health policy and co-director of the Centre for Health Policy in the Institute of Global Health Innovation at Imperial College London. His main work has been in the economics of health and the broader public services, previously as the director of the Centre for Health Economics at the University of York.
Explaining changes in satisfaction with health services is difficult, as the authors of this research note. However, readers may also be interested in analysis by me of the results of the satisfaction (and other health) questions asked by the British Social Attitudes survey (see http://www.kingsfund.org.uk/projects/bsa-survey-results-2011 and http://www.kingsfund.org.uk/projects/bsa-survey-2012). The King’s Fund has sponsored the BSA’s health questions since 2011.
Since the first BSA survey in 1983 there have been near-annual analysis and write ups of the satisfaction results published by the survey organisers NatCen (see http://www.natcen.ac.uk/series/british-social-attitudes ). There is no simple explanation of changes in the public’s satisfaction with the NHS (and its services) over time. It is possible to find associations – such as the colour of the political party in power, life cycle and generational effects, views about problems with the NHS, satisfaction with parts of the NHS (such as General Practice or outpatients), whether recent use of hospital services etc – with satisfaction but these often seem relatively weak.
The most compelling explanation I would suggest for the almost year on year rise in satisfaction with the British NHS from around 2002 to 2010 is the significant reduction in waiting times over that period – a top concern for the public and a change they saw in practice. In addition, the drying up of front page horror stories about the NHS (patients waiting hours on trolleys etc) would also have played a part.
The fall in 2011 from the peak satisfaction in 2010 of 70% very or quite satisfied with the NHS, to 58% is most easily explained by a combination of public worries about the Coalition’s reform plans for the NHS in England (again, lots of media coverage and vocal opposition), Ministerial rhetoric about the need for reform based on poor performance of the NHS and stories about the financial squeeze of the NHS and consequent drive for increased productivity – a word which I suspect the public simply interpret as ‘cuts’. So, an increasing proportion of the public are expressing dissatisfaction with those they perceive to be doing harm to the service rather than with the NHS itself – a surrogate vote if you like. Such explanations do not easily lend themselves to quantification and statistical analysis however.