Monday February 22 2010
There has been a “dramatic rise” in obesity-related deaths according to the BBC. The news is based on a study looking at 27 years’ worth of death data, focusing on whether obesity was listed as the main cause of death or only a contributing one. The researchers warn that obesity-related death may be more common than believed because it is rarely listed as the main cause of death.
While this study was actually examining the process of recording deaths, it highlights the important association between obesity and health – the researchers say that one possible reason for the increase in obesity-related death recordings is due to a real increase in obesity prevalence. The findings of this research will be important to public health practitioners or researchers who use death records to monitor obesity-related deaths.
Where did the story come from?
The study was carried out by Dr Marie Duncan and colleagues from the University of Oxford’s Department of Public Health and the National Obesity Observatory. The study was funded by the English National Institute for Health Research via its National Coordinating Centre for Research Capacity Development. The research was published in the peer-reviewed European Journal of Public Health.
This time series study has demonstrated increased certification of obesity as a cause of death in England, although it's usually selected as a contributing cause rather than an underlying cause.
What kind of research was this?
A number of health risks are associated with obesity and the condition leads to an increase in overall mortality. This was a time series study analysing the changing trends in obesity-related deaths over time. According to its authors, figures from the 2007 Health Survey for England show that 24% of men and 25% of women are classified as obese.
What did the research involve?
The researchers used two separate datasets to investigate the trends in obesity-related mortality – the Oxford record linkage study (1979-2006) and English national mortality data (1995-2006). The researchers say the Oxford study is considered to be the “longest continuous run of systematic, ready-to-analyse, coding of all mentions on death certificates in a large defined population in England”. The English national mortality dataset also provides all certified causes of an individual’s death, not just the underlying cause of death. Both datasets were searched for mentions of obesity.
Researchers then used each dataset to calculate age-specific mortality rates in five-year age bands to calculate an “age-standardised mortality rate” for each age group. This means that they standardised the death rates in the different datasets against a theoretical population that has the same age structure as England. In this way, death rates from the two datasets were made comparable with each other and the national situation.
The researchers analysed the Oxford data within four time periods corresponding to changes in the regulations about recording deaths – 1979-83, 1984-92, 1993-2000 and 2001 onwards. This dataset was used to see whether changes in directives about coding led to any changes in the way obesity-related death was recorded. The national English dataset was used to assess whether there were any significant increases or decreases in deaths associated with obesity.
Changes in coding
From 1984, the rules governing the selection of the underlying cause of death changed – revision of the International Classification of Diseases (ICD) specified that certain diseases, which can be modes of dying rather than causes of death, should not be recorded as the underlying cause if another ‘primary’ condition is present. There were further changes in 1993, which saw the introduction of automatic coding software by the Office for National Statistics and the use of multiple-cause coding become standard practice in England.
What were the basic results?
Of 656,443 deaths recorded in the Oxford data, obesity was a certified cause of death in 1,002 cases (0.15%) and recorded as the underlying cause of death in 26% (259/1,002) of these.
The researchers then analysed deaths in relation to periods of different coding practice. The proportion of obesity-related deaths with obesity as an underlying cause was 22.2% in 1979-83, 36.4% in 1984-92, 25.8% in 1993-2000 and 17.4% in 2001-06. The researchers say that the increase between the first two periods and the decrease between periods two and three were statistically significant and “coincided with coding rule changes”.
English national mortality data from 1995 to 2006 showed that obesity was a certified cause of death in 8,450 of 6,054,897 deaths (0.14%). It was recorded as the underlying cause of death in 24.8% of these. The percentage of all deaths in England with obesity on the certificate doubled from 0.11% in 1995 to 0.23% in 2006. The researchers estimated that this represented an average annual increase of 7.5% for men and 4.0% for women.
How did the researchers interpret the results?
The researchers concluded: “There is an emerging trend of increased certification of obesity as a cause of death in England.” They also say that relying on the underlying-cause mortality statistics alone “fails to capture the majority of obesity deaths”.
This study highlights important issues surrounding the complex nature of recording of the causes of death. In their work the researchers note that there is an increase in the recording of obesity as a cause of death, but that it is usually noted to be a contributing, rather than underlying, cause. The researchers also say that until recently in England, only one underlying cause of death from each death certificate is used for routine coding and analysis in national systems. There are problems with this approach, including missing data on contributing causes.
The increase in certification of obesity-related deaths also suggests that there is a better way to use these routine statistics to assess mortality – studies that assess mortality based only on obesity as a primary cause of death would have missed this increase. The researchers also make a sensible recommendation over using broader measures when monitoring obesity-related deaths in public health planning, saying that “public health practitioners should consider the importance of using all certified causes of death and not just the underlying cause”.
The researchers say that it seems likely that the increase in obesity-related deaths noted in their study is linked to increasing prevalence of obesity, but that there are other potential reasons for these changes. These include an increase in the prevalence of disease, an increase in severity of the disease (i.e. obesity at levels more likely to kill), increased clinical awareness and certification practice changes such as an increase in willingness to certify obesity.