Measuring disparities in health careBMJ 2006; 333 doi: https://doi.org/10.1136/bmj.333.7562.274 (Published 03 August 2006) Cite this as: BMJ 2006;333:274
All rapid responses
Editor - We read with interest the recent article entitled ‘Measuring
disparities in health care’, which proposes that it is more meaningful to
determine an individual or group’s shortfall in the quality of healthcare
from the highest attainable standard as compared to determining
differences between socioeconomic groups. (1) The editorial describes the
use of proportional reductions in shortfall pertaining to differences in
international life expectancies as example of an indicator of improvement,
providing a measure of progress towards boosting health to certain
minimally adequate levels. This methodology which simply describes the
calculation of a relative gap in a measurable health outcome of a studied
group against a defined control group is known as health gap measurement.
Within the UK, the measurement of ‘health gaps’ both within and between
populations has been the focus of the Public Health Observatories,
established in 2000 as a result of the government’s commitment to tackling
health inequality with the publication of the White Paper “Saving Lives:
Our Healthier Nation”. (2) Inequality measures have been used by
economists for decades to assess the equality of income distribution
across a population and later extended for the assessment of health
inequality. Simple inequality measures, (range, rate ratio, etc.) assess
the relative or absolute difference in health status between different
groups in the population. More complicated measures (Lorenz curve and Gini
coefficient) assess the whole distribution of a resource or health outcome
across a population. (3) Both these methods however fail to take into
account a socioeconomic dimension and therefore do not meet the minimal
requirements of a good measure of inequality as defined by Wagstaff et al.
(4) The most critical of these requirements is the sensitivity of the
measure to changes in the distribution of the population across
socioeconomic groups and in this regard the assessment of perceivable
health gaps using validated tools such as slope index of inequality has
been derived. (5)
The traditional measurement of surrogate endpoints of healthcare such
as life expectancy alongside morbidity and disease specific mortality has
arisen partly due to the ease with which they can be measured and
comparisons made. We now understand however, that in order to define
quality of care, inequality in healthcare or indeed inequities in
healthcare, we must look beyond measuring these endpoints alone, as there
exists an intricate interaction between these health-related outcomes,
processes of care and productivity.
The complexities in measuring processes of care will necessitate that
the analysed groups are well defined and their differences measurable.
Distinguishing sociodemographic groups by socioeconomic factors such as
income, deprivation score or employment status, is often well suited to
this task. (6) It is fair to say that shortfall methodology, such as that
exampled by Professor Ruger, could be applicable when comparison is made
of incompletely defined groups. However, when we measure inequalities in
defined groups, such as a specific disease entity, stratification of the
studied population by sociodemography including socioecomical factors may
be more ideally suited.
Researchers, clinicians and policy makers alike are far from defining
‘quality of care’ and therefore being able to abolish inequalities in
healthcare by system-wide reforms. Furthermore decisions such as NICE’s
stance not to recommend Avastin and Erbitux for advanced colorectal cancer
(7) continue to create apparent tiers within healthcare systems.
In order to determine the optimal quality of care, a good
understanding of where the inequalities lie must be appreciated first.
This may often be achieved through comparison of subgroups, however there
are many other facets and contributing factors, including costs,
healthcare policy and socioeconomic considerations that also come into
1. Ruger JP. Measuring disparities in health care. Bmj. 2006 Aug
2. Department of Health. Our health, our care, our say: a new
direction for community services. 2006.
3. Flowers J. Measuring Health Inequalities II. 2003.
4. Wagstaff A, Paci P, van Doorslaer E. On the measurement of
inequalities in health. Soc Sci Med. 1991;33(5):545-57.
5. Low A, Low A. Measuring the gap: quantifying and comparing local
health inequalities. J Public Health (Oxf). 2004 Dec;26(4):388-95.
6. Malin JL, Schneider EC, Epstein AM, Adams J, Emanuel EJ, Kahn KL.
Results of the National Initiative for Cancer Care Quality: how can we
improve the quality of cancer care in the United States? J Clin Oncol.
2006 Feb 1;24(4):626-34.
7. BBC News. Anger at bowel cancer drug ruling. 2006.
http://news.bbc.co.uk/1/hi/health/5262760.stm. Last accessed 21st August
Competing interests: No competing interests
EDITOR—Ruger’s editorial  makes some interesting observations as
to some rather surprising findings by Asch et al  that black and
Hispanic people in the United States scored higher than white people for
quality of care based on a large set of indicators.
It is important to note, however, that the Asch study only included
participants who had visited a health care provider in the previous two
years. Accordingly, disparities in the quality of health care
attributable to access were not measured accurately due to the sample
being biased towards individuals who were fortunate enough to be able to
access health care providers.
Measuring quality by reference to the extent to which virtually every
conceivable investigation, intervention and preventative action is carried
out will mean little to a person from a disadvantaged sociodemographic
subgroup who is unable to access service in the first instance.
1. Ruger JP. Measuring disparities in health care. BMJ 2006;333: 274
2. Asch SM, Kerr EA, Keesey J, Adams JL, Setodji CM, Malik S, et al.
Who is at greatest risk for receiving poor-quality health care? N Engl J
Med 2006;354: 1147-56
Competing interests: No competing interests