Thematic analysis.

Author 1 started with the sample from 1993, reading the sample to develop familiarity with the themes and content presented [55]. Axial and selective coding following Ezzy [60] in NVivo then developed distinct coding structures within each sample. These were reviewed by Authors 2 and 3, with findings were recorded as ‘initial’ observations. This process was repeated with 2001, then 2013. At the end of thematic analysis, Author 1 wrote up observations about the ways in which findings differed across the three samples. Authors 2 and 3 reviewed these drafts, providing input for conceptual clarity and rigour.

Framing analysis.

We conceived that medical, behavioural and societal frames would be useful heuristic devices in this research. These three categories are important to public health where health is conceptualised to exist and operate through medicine, behaviour, and broader society. Findings from prior research in the field could also be organised according to these categories, providing further rationale they would be useful frames to examine within the data.

After the thematic analysis, Author 1 approached the 1993 sample with the a-priori framework (i.e. see Table 3 for an example of the societal frame), re-reading all the data while coding data in NVivo to only one of the four frame components under the relevant medical, behavioural, or societal frame (i.e. a total of 12 possible coding locations for each year). Some of the latent meaning embedded in news texts proved difficult to code, and coding concerns were resolved in collaboration with Authors 2 and 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Example of societal coding frame and framing components.

https://doi.org/10.1371/journal.pone.0225794.t003

The findings from 1993 were recorded as initial observations, and the process was then repeated with the sample from 2001, then 2013. At the end of the framing analysis, Author 1 wrote up observations about the ways in which findings differed across the three samples. At this point, having completed both analyses, Author 1 considered the six groups of initial findings, reading between and across them, and explicating the differences derived from the two methods across the different years. These findings were tabulated and then discussed with Authors 2 and 3, which facilitated refinement of the six categories, enhanced the clarity of how each category differed (or overlapped), and provided working reflections drawn on for later interpretation. No major conceptual changes were made to these categories, however the language used to describe the findings emerged and evolved throughout these discussions.

Modifiable and non-modifiable risks.

In 1993, rather than diabetes being a main issue of interest in relation to obesity, information about diabetes tended to be communicated alongside or through discussions about heart disease, at that time cited as a major health issue. Diabetes was asserted to be ‘part of the ageing process’ [61] but dangerous when combined with heart disease and obesity, because it increased the risk of heart attacks.

Obesity was clearly identified as a risk factor for diabetes during 1993.

… obesity, with a BMI of 30 or more… is associated with significant health risks, especially heart disease, stroke, diabetes and cancer.

[62]

Only one account in this sample identified that weight might be a possible symptom of another disease, rather than a risk [63].

In 1993, the only piece to talk about genetics or ethnicity described the evolutionary mismatch between western diets and the genome of Pima Indians of Arizona. Family history or heritability was not linked to diabetes within this sample.

They had survived on crops such as sorghum and often faced long periods of hunger until the white man introduced them to fatty foods. Now almost all the elderly members of the tribe are overweight and diabetic because of a genetic tendency to make excess amylin. This is useful when food is scarce as it concentrates glucose on fatty energy reserves. But it can prove disastrous when plentiful fat-rich food is available, as the amylin pours more and more glucose into fat and triggers diabetes.

[64]

Interpreting proposed responses to the issue.

Pharmaceutical management was most commonly recommended as a response to diabetes in 1993. The focus on diabetes as a medical problem with medical solutions contributed to diabetes being framed as a medical issue during this time period, despite links made with obesity as a risk factor. Some behavioural conceptions were present, in that sedentary lifestyles were noted to be leading to an increase in chronic disease, including diabetes. Exercise and diet were identified as useful in managing diabetes once it was diagnosed.

One article suggested some moral evaluations of people who had diabetes, although it inconsistently delineated ‘dieters’ from ‘diabetics’ throughout the piece, so it wasn’t clear if they were inferring that diabetics were big eaters, or that dieters were lucky that they could use the diabetes pill.

Dieters’ dream pill; ‘CURE’ FOR DIABETES COULD LET THEM EAT CAKE AND STAY SLIM, SAY SCIENTISTS

[64]

Modifiable risks/causes.

By 2001, diabetes and obesity had started to become more topical, evidenced by an increasing number of news articles. Diabetes was often discussed in articles describing the ‘obesity epidemic’, while other articles focused exclusively on diabetes as an emerging health problem, for which obesity was reported as a risk factor and/or a cause.

Discussion about the aetiology of diabetes became more technical, with in-utero risk acknowledged to have some role. During 2001, medical framings emphasised enhanced levels of diabetes risk.

All the same, the symptoms are all there: high blood pressure, raised levels of telltale fats called triglycerides found in the blood, and insulin resistance—an acquired resistance to the body's vital glucose-handling hormone. Diabetes and heart disease are lying in wait for anyone with this group of symptoms, collectively known as Syndrome X.

[65]

Behavioural framings of diabetes were most prominent in 2001 in comparison to medical or societal framings. Notably, obesity started to be explicitly labelled as a cause in addition to a risk factor.

One of the main causes of Type 2 diabetes is obesity.

[66]

In addition, being overweight was also implicated as causal to diabetes, expanding the body sizes that were correlated to ‘result in’ diabetes.

Other consequences of being overweight are more dangerous and include diabetes type 2.

[67]

Sugar was talked about specifically during 2001 as a meaningful part of the weight gain equals diabetes conundrum at this time, extending the equation to: sugar = weight gain = diabetes. It was more common that sugar was described in relation to risk rather than fats, and as such, sugar become enmeshed in the conceptual picture available to individuals regarding the causation of weight gain and diabetes.

SUGAR: A form of carbohydrate energy, so it's fine in moderation. But too much, particularly spread through the day rather than eaten in just one or two sessions, can lead to tooth decay and can increase the amount of insulin you release, leading, over a lifetime, to a greater chance of weight gain (insulin encourages fat deposition) and diabetes.

[68]

In 2001, seven accounts noted that age had a role in developing diabetes. This fact was often quoted in order to situate the earlier onset of the disease, particularly in reference to increasing diabetes rates in the UK and most regularly as a partner-in-crime to obesity—the ‘trigger’ for diabetes.

Reasons for the predicted increase include an ageing population: diabetes is more common in older people and the epidemic of obesity, which can trigger diabetes.

[69]

…in 20 years’ time more of today's children will be dying from obesity-related diseases, such as type 2 diabetes… “We're talking about an epidemic,” he says, “and if we don't do something now, the next time we see these children is when they turn up in their thirties with heart disease and diabetes.”

[70]

Non-modifiable risks/causes.

As with the 1993 sample, only one account provided an alternative account of how weight was related to diabetes, mentioning it as a symptom of Type 1 Diabetes in children [71]. Considering the growth in sample size between these years (i.e. n = 19 in 1993 and n = 119 in 2001), this notion became more marginal. One other 2001 article identified that although better screening may have been driving the increases in Type 2 Diabetes, ‘fatness’ was the real driver of increased diabetes rates.

The researchers, from Cardiff University and the Heart of England NHS Trust in Birmingham, said the rise [in diabetes rates] may be partly due to better screening, since GPs now have performance-related pay that rewards them for diagnosing sufferers. However, they said rising obesity was the main driver. Lead researcher Professor Craig Currie said: It s almost entirely obesity. How fat you are is the top and bottom of it.

[72]

In 2001 four ethnic ‘divisions’ were referenced in risk for diabetes; with African, Carribean and Asian groups identified to be at five times the risk of other groups [66]. The term ‘ethnic minority’ was used once in relation to higher risk of the condition [73]. Insulin resistance, identified as a precursor to diabetes, was discussed as particularly common to various ‘races’ [74].

Despite being a strong indicator for diabetes risk in the extant literature, family history was only associated with diabetes risk in five Daily Mail stories and one from the Guardian. Of those accounts, only one asserted that the risk was correlated with how close the family member was [75]. Another stated personal risk of diabetes as 15% if either one of your parents had it [66], which under-represents the relationship. Individuals were instead tasked with negating diabetes risk by keeping a healthy weight.

Q: DO you have a family history of heart disease, diabetes, prostate or bowel cancer? Obesity is a risk factor for all of these conditions, so keeping to a healthy weight as you age can reduce your risk.

[76]

Interpreting proposed responses to the problem.

Modifiable risks for diabetes were presented as critical for preventing the condition in 2001. Weight loss through exercise and diet were often promoted as ‘solutions’ to ‘protect’ yourself from diabetes risk. Moral evaluations started to surface around people who developed diabetes in light of its preventable nature—neglecting to action ‘simple’ changes in diet and exercise. In the second quote, a link between individual action and the ‘epidemic of Type 2 diabetes’ is made.

This evidence that diabetes is a preventable illness has enormous public health and economic implications… anyone with one or more of these risk factors can take effective steps to protect themselves with specific lifestyle changes.

[66]

Walking for just 30 minutes a day could halve the risk of developing diabetes, researchers have found. Simple changes in diet and exercise may halt the epidemic of Type 2 diabetes… ‘…to go for a walk with your spouse doesn't seem to me to be asking the impossible.’

[77]

During 2001 it was also identified that obesity and poor nutrition logically ‘lead to’ diabetes across the entire life span, including in children. The ‘poor diet time bomb’ is presented as a mix of obesity, sedentariness, poor diet and diabetes in the context of the current generation growing into adulthood.

POOR DIET TIME BOMB: TYPE TWO DIABETES… ‘There's a lot of research to show that obesity, lack of exercise and poor diet lead to type two diabetes in adulthood,’ says Mairi Benson of Diabetes UK.

[78]

The notion of children-at-risk was a prominent topic in 2001 and tended to frame the diabetes problem as an individual responsibility issue that had implications for broader society. This coincided with the increased attention on maternal bodies and mothering practices, where in-utero risk, pre-pregnancy diets and early life practices were portrayed as entwined with diabetes risk for children. The attention on mothers’ biology (i.e. medical frame) was juxtaposed with a responsibility on mothers to prevent diabetes in children (i.e. behavioural frame).

In the United States children and young people in their teens are developing this sort of diabetes, and it is linked with obesity. Copelman says that although diet is important, the key factor is exercise. Children are leading couch-potato lives. They do less exercise at school and are driven everywhere because of increasing concerns about safety. Our children's lifestyles reflect ours as parents.

[79]

The cost of diabetes to society via health services and the national economy, as a result of being overweight, was a dominant topic in the societal frame during 2001. Prevention and early detection were presented as part of the answer to this problem; however this societal issue was often depicted as a result of overweight.

About 1.4m Britons are diagnosed with diabetes, a figure that will probably double by the end of the decade since more people are expected to become overweight.

[73]

Modifiable risks/causes.

The ability to position diabetes and obesity as a societal issue during 2013 relied on many of the important developments from prior years. For example, during 2001, obesity and overweight were both described as discrete causes of diabetes. In 2013, weight is used to explain rising rates of diabetes at a societal level more consistently and confidently, continuing the presentation of weight/obesity as causal to diabetes. This was seen across all three papers. No accounts in this sample challenged the view that being overweight could be a symptom of diabetes, rather than a ‘cause’.

ONE in 20 Britons is suffering from diabetes as a result of soaring obesity rates, a shocking study has found.

[80]

Since 1996 the number of people with diabetes has increased from 1.4 million to 2.9 million, largely linked to weight increase.

[81]

…the academy says doctors are “united in seeing the epidemic of obesity as the greatest public health crisis facing the UK. The consequences of obesity include diabetes, heart disease and cancer and people are dying needlessly from avoidable diseases.”

[82]

The salience of sugar in catalysing obesity, and therefore also playing a causal role in the epidemic of diabetes, was portrayed in many articles during 2013. A moral reading on sugar and sugar consumption was also prevalent at this time, and fostered certain interpretations about people seen to be eating sugar.

Sugar is behind the global explosion in type 2 diabetes, researchers will announce today in a report claiming it plays a uniquely damaging role in a disease that experts fear could overwhelm the NHS. Obesity is usually cited as the main driver of diabetes. But a study by US medical researchers identifies sugar as a predictor of diabetes separate from obesity.

[83]

Still, it seems ridiculous to liken food to hard drugs—is it an overstatement? There is evidence that sweet foods do hit the pleasure and reward centres in the brain rather like hard drugs… leads to insulin resistance (where insulin becomes less effective), high blood pressure, diabetes and unhealthy cholesterol levels—all risk factors for cardiac disease. You don't even have to be overweight for this to occur.

[84]

Further, the danger attributed to diabetes as a medical issue in 2001 was transplanted into ‘alarming’ reports of diabetes as a societal issue in 2013.

The charity's Scottish director, Jane-Claire Judson, said: “We are not scaremongering … if 18,000 to 19,000 people a year were contracting meningitis or bird flu we would be really, really concerned. It is a major problem.”

[85]

Non-modifiable risks/causes.

Only 14 of the 324 articles in 2013 described ageing having a role in diabetes development, or an ageing population influencing diabetes statistics. Unhealthy lifestyles and fatness were at times used to minimise the importance of ageing in the causal profile of diabetes. At other times, they were presented as the two key drivers of diabetes in the UK.

The soaring rate of type 2 " which mostly affects middle-aged people " is being blamed on obesity.

[80]

With the number of diabetics expected to double from the present 2.5 million to 5 million within a generation—a consequence of an ageing population and unhealthy lifestyles—they say that the NHS risks being overwhelmed by the cost of treatment.

[86]

In 2013, 13 articles identified family history in T2DM aetiology: 11 from the Daily Mail and two from the Times. Similarly to 2001, however, the importance of familial risk in disease aetiology was understated in relation to other risk factors, most notably weight and/or ‘lifestyle’, often used loosely to refer to being overweight.

Some are genetically more at risk but lifestyle is the key trigger, he said, adding: It’s still quite rare but we do see them. In the huge majority of cases type 2 diabetes develops as a consequence of being overweight. So as the weight of the nation increases, the incidence goes up too.

[72]

“Certain genes are dictators, like the genes for blue eyes or brown hair. There's nothing you can do about them. The genes for Alzheimer's and other diseases such as diabetes are more like committees: they make suggestions. Lifestyle factors have a profound effect on determining our health, regardless of our family's medical history. What we eat and drink can help determine who thinks clearly decade after decade and who is affected by a terrible tragedy.”

[87]

Only one article in 2013 endorsed the importance of a genetic influence in the development of diabetes, in addition to weight and lifestyle.

Type 2 diabetes is normally associated with obesity and a couch potato lifestyle—not heroic athleticism. Even now no one can tell for sure why he developed it—but it could be that it was unlucky genes… Although the biggest risk factor for type 2 diabetes is weight, genes can also have an important role… If one parent has type 2, then you have a 30 per cent chance of developing it. If both parents do, then that risk is 50 per cent. It is possible to develop type 2 diabetes even if you are of a normal weight—just because of the genetics you inherited.’

[88]

Ethnicity also became an identifier in evaluating diabetes risk during 2013. 11 accounts discussed links between diabetes and ethnicity, tying it in with links to lifestyle and weight. Eight articles talked exclusively about South Asians, two also refer to black, Caribbean or African groups, and one article only uses the term ‘ethnic’. These groups were established to be at higher risk at a younger age and lower level of obesity—maintaining the link between diabetes and obesity [89, 90]. The aerobic fitness of these groups was presented to be lower than other ethnicities [91], and specific gyms were marketed as important for young Asian females at high risk of the disease [92]. A testimonial account unpacked the experience of living at such high risk for the disease:

‘I didn't want to be skinny, I wanted to be fit and strong, to have the energy to play with my kids, look good in my clothes and reduce my risk of problems like heart disease and diabetes. As an Asian woman, my ethnicity is proven to put me at greater risk of developing them.’

[93]

Two stories released on the same day in the Times and the Daily Mail reported on the high levels of diabetes in some boroughs of London. The Daily Mail piece extended the links genetic risk had with ‘race’ to culture and lifestyle, and the interplay of these elements.

Almost one in five people in Brent is from India: diabetes is especially prevalent among people of South Asian origin.

[89]

Scientists believe this is partly down to a genetic difference which means that their muscles do not burn fat well. Traditional cooking methods using oil, creamed coconut milk and high-fat butter called ghee may also play a role.

[94]

Interpreting proposes responses to the problem.

Critiques of the necessary public health response in news articles mirrored the notion that diabetes was caused by unhealthy lifestyles, for which obesity was often a symbol. The recommended response tended to be broad scale prevention of diabetes via awareness raising and education, particularly around the role of lifestyle and weight. While a ‘pandemic of inactivity’ was described to be contributing to society’s diabetes problem, the inferred responsibility for solving this problem circulates around individual action rather than societal change.

The charity's chief executive Barbara Young said the aim is to ‘lay to rest the myth’ that type 2 diabetes is a mild condition. ‘This is a misconception that is wrecking lives and is the reason that as a country we are sleepwalking towards a public health disaster of an almost unimaginable scale,’ she said. Losing weight, eating more fruit and vegetables and becoming more active are thought to cut the risks of diabetes.

[95]

Four out of ten men and five out of ten women are still not active enough to benefit their health. This increases the risk of serious illnesses like type 2 diabetes, heart disease and certain cancers, and makes it more likely that people will be overweight or obese… [this] brings us closer to understanding the kind of societal shift that needs to happen before we truly combat the pandemic of inactivity. The figures are alarming and show that we need to take action now.

[96]

Another excerpt from the same Times piece denotes that this ‘pandemic’ is a result of non-compliance from medical advice, rather than an unintended impact of structural forces such as industrialisation.

Thousands of cases of cancer, heart disease and diabetes would also be prevented if people were as active as doctors recommend… The death and illness toll of inactivity is comparable to that of smoking, they say.

In 2013, the problem/s of diabetes and obesity also started to be portrayed as issues of societal safety. Warnings suggested that together they posed significant threats to societal and economic progress.

Poor lifestyles, increasing sedentary habits, obesity and diabetes threatened to eradicate advances made in Britain in recent years …

[97]

Soaring levels of obesity, which have been predicted to double by 2030, coupled with rising levels of diabetes and heart disease linked to poor diets and a lack of exercise, will overwhelm the dwindling numbers of nurses.

[98]

The children-at-risk narrative which was visible in 2001 was also drawn on to further demarcate diabetes as an issue of public safety by 2013. Although mothers still received specific attention in 2013 news articles, the focus broadened to include ‘parents’ as responsible for diabetes prevention, perhaps in line with reporting that paternal sperm could be attributed to diabetes risk. These accounts are much more definitive in the tone and extent of their accusation than in the previous years, and link obesity or poor nutrition with diabetes.

We're used to hearing words like "obesity" and "diabetes" and "risk". We're also used to switching off. But this, unfortunately, is because we don't think about what it really means… we Brits are the biggest in western Europe. This is not a joke. This is, in fact, the opposite of a joke. We know what makes people fat. It's bad diet and not enough exercise, not cooking properly and not eating enough vegetables, and drinking too many fizzy drinks. And it's growing up with parents who don't care enough about what you eat.

[99]

Are we destined to self-destruct? Not necessarily. It seems particularly terrible to me that small children are developing type 2 diabetes because of what their parents feed them.

[100]