Friday 7 October 2011

Which is the best health system in the world?

  1. John Appleby, chief economist
  1. 1 King’s Fund, London, UK
  1. j.appleby@kingsfund.org.uk
It may be nice to find your country at the top of healthcare rankings, but the relevance to policymakers is strictly limited, explains John Appleby
According to the World Health Organization, the country with the best health system overall in the world in 2000 was France, with the UK ranked 18th and Burma (Myanmar) coming last at 190th.1 In 2009, according to the EuroHealth Consumer Index, France was ranked seventh out of 33 (mainly European countries) and the Netherlands first (UK trailed in at 14th).2 Meanwhile, last year’s regular 11 country survey of health system performance from the Commonwealth Fund in New York suggested that the UK ranked first in terms of the smallest proportion of members of the public polled thinking the system needed fundamental changes or complete rebuilding (fig 1).3 A parallel Commonwealth Fund survey of seven countries in 2010 ranked the Netherlands top (and the UK second) on a basket of performance dimensions (fig 2).4
Fig 1 Overall public views of the healthcare system, 20103
Fig 2 Performance ranking of seven countries, 20104
Although the authors of these surveys are of course aware of the tremendous difficulty in comparing the performance of different countries’ health systems, the temptation to reduce the comparison to the equivalent of “marks out of 10” is often too great. The Commonwealth Fund resists in its 11 country comparison1 but succumbs in the seven country assessment.4 Different analyses also use different numbers of comparator countries, which makes comparing ranks confusing; coming second out of six is similar to coming 20th out of 60 (that is, both in the top 30%), but the second ranking sounds much worse. Equally, coming last in a ranking of, say, a small number of elite, Western industrialised countries may not be considered as bad as last in the whole world.
Nevertheless, the question, “Which is the best system?” remains a compelling one to try to answer.
Although it may not feel like it at the moment, reforming healthcare systems is not just an English obsession. While 34% of a sample of the UK public think the health system needs fundamental change according to one Commonwealth Fund survey, people in the 10 other countries surveyed reported higher levels of dissatisfaction (fig 1).3 Politicians and policymakers in all countries grapple with changes to their systems to tackle public worries. Scouting around for new policy ideas starts with questions about where to look and then naturally to questions about other countries’ systems and their performance. Has someone else solved the difficult problems we are facing—spending too much, poor patient care, lack of health impact, poor cost effectiveness?
But here comes a central set of difficulties in answering the comparative question: the performance of healthcare systems is multidimensional; it may, in the end, be about health, but it is also about efficiency and effectiveness and affordability and acceptability . . . .5 WHO and the EuroHealth Consumer Index recognise this (as do many such comparative surveys) and construct performance “dimensions” populated with varying numbers of statistics (six dimensions and 38 statistics in the case of the EuroHealth survey2).
Inevitably, such an approach will mean that countries will do better on some dimensions than on others; the UK ranked second best on the distribution of health across its population in WHO’s 2000 health system evaluation but 26th on patient responsiveness. Pulling these measures into one overall number or rank requires some weighting for each individual performance measure—they are unlikely to be of equal importance. But whose values to use? The public’s? Policymakers’?6 And how should we take account of different weightings by different countries’ populations?
Even if this and other problems are answered, the next question for policymakers is why France is first (or is it 7th?) and the UK 18th (or 14th or 2nd). While I make no claims for such a simplistic model, it is interesting to note just how strong the relationship is between where a country is ranked (by either WHO or EuroHealth Consumer) and how much it spends per capita on healthcare: no causation is claimed, but it seems that as spending increases, ranking improves (fig 3). Of course, if all countries increased their spending the rankings may be left unaltered; even if performance is increased in all countries, ranking position may remain the same.
Fig 3 Relation between per capita health spend8 and health system perfomance ranking (data for 24 European countries common to both WHO and EuroHealth Consumer Index analyses1 2). Logarithmic regression produced the highest correlation (R2 varying between 0 (no correlation) and 1 (perfect correlation)) between rankings (response variable) and (log of) per capita spending on healthcare (explanatory variable)
Even less claim can be made for the apparent positive correlation between the WHO health system ranking of countries and the FIFA ranking of international football teams; in 2000 both placed France in first position.7
Frustratingly, given the importance of the policy questions they raise, perhaps the best that can be said for many comparative ranking exercises is that they provoke a ready headline and can generate debate but fail to provide definitive answers.

Notes

Cite this as: BMJ 2011;343:d6267

References

  1.  
  2.  
  3.  
  4.  
  5.  
  6.  
  7.   
  8.