Diet and Nutrition
Article
Investigating the media power of a population health monitoring survey: case study of the NSW Schools Physical Activity and Nutrition Survey (SPANS) Abstract
Paola T. Espinel, Rachel Laws
Objective: To examine the extent and nature of news coverage of a governmentfunded population monitoring survey of children and the potential implications of this coverage for public health advocacy. Methods: Case study of the NSW Schools Physical Activity and Nutrition Survey (SPANS), a population monitoring survey of school-aged children’s weight and weightrelated behaviours, conducted in 1997, 2004 and 2010. Printed news items from all Australian newspapers between January 1997 and December 2011 mentioning the survey findings were identified from the Factiva database and a descriptive analysis of the content conducted. Results: Overall, 144 news items were identified. The news angles focused mainly on physical activity/sedentary behaviour; overweight/obesity and nutrition; however these angles changed between 1997 and 2011, with angles focused on physical activity/sedentary behaviour increasing, compared with overweight/obesity and nutrition angles (p=0.001). Responsibility for obesity and weight-related behaviours was most frequently assigned to parents and food marketing, and the most common solutions were policy strategies and parental/child education and support. Conclusions: Population health surveys are newsworthy and when coupled with strategic dissemination, media can contribute to communicating health issues and interpreting findings in ways that are relevant for consumers, policy makers and stakeholders. Implications: This case study emphasises the news value of government-funded population surveys, while providing a cautionary note about media focus on individual studies rather than a larger body of research evidence. Key words: newspapers, public health, advocacy, obesity, public opinion, children Aust NZ J Public Health. 2013; 37:284-9 doi: 10.1111/1753-6405.12072
284
Prevention Research Collaboration, School of Public Health, University of Sydney, New South Wales
Catriona Bonfiglioli Faculty of Arts and Social Sciences, University of Technology, New South Wales
Louise L. Hardy, Lesley King Prevention Research Collaboration, School of Public Health, University of Sydney, New South Wales
N
ews is an important means of communicating health information to the community and can influence how people understand, view, interpret and act on health issues. 1,2 News media can draw attention to problems, and highlight or neglect specific aspects of an issue1,3 depending on how journalists focus the story. Journalists’ decisions about what stories to tell and which angles to focus on are based on judgements of newsworthiness (‘news values’) rather than on public health concepts of significance. Assessing news value involves making judgments about how important and interesting an event or issue is to the key audience of that media outlet.4 Analyses of news media can identify mismatches between media and expert discourses about health, identify which aspects of an issue attract news media attention and discover news frames that may be undermining health.5 An examination of how news portrays an issue can illuminate how news media sources and journalists contribute to the social construction of issues, events and people in the minds of audiences.6-8 Researchers and funders can use the findings to modify strategies for gaining news coverage, guide the shaping of news angles, and enhance the quality of information offered to news consumers. Submitted: August 2012
Obesity is a high-profile public health issue and in Australia has received substantial television and print media coverage since 1996.9,10 Media coverage of obesity, in terms of causes and responsibilities, has been diverse; and news media studies have identified a substantial increase in media coverage of obesity in Australian newspapers, particularly since 2006.9-11 At that time, the focus of news items was on nutritional explanations more than inactivity and environmental factors;12,13 and an emphasis on individual, rather than structural or environmental solutions.10-12,14 One Australian source of evidence on child obesity is a government-funded population monitoring survey of NSW school aged children: the NSW Schools Physical Activity and Nutrition Survey (SPANS) which has been conducted in 1997, 2004 and 2010.15-17 Active dissemination strategies of the findings through ministerial announcements, media releases and presentations to stakeholders were incorporated in order to achieve high levels of professional and public awareness. Further media attention was gained by key researchers through media releases coinciding with the publication of findings in the scientific literature. To our knowledge there have not been any specific analyses of media coverage of population health surveillance
Revision requested: November 2012
Accepted: February 2013
Correspondence to: Mrs Paola T. Espinel, Prevention Research Collaboration, School of Public Health, Level 2, K25 Medical Foundation Building, The University of Sydney, NSW 2006; e-mail:
[email protected]
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH © 2013 The Authors. ANZJPH © 2013 Public Health Association of Australia
2013 vol. 37 no. 3
Diet and Nutrition
Media coverage of a population health monitoring survey
surveys, such as SPANS, or of the ways in which the findings of surveys are represented in the media. This study aimed to use SPANS as a case study to investigate the extent and nature of print media coverage of its findings and the potential implications of this coverage for public health advocacy. Specifically, this study examined the content of Australian print media stories, including which aspects of the survey attracted attention, how survey findings were reported, how responsibility for children’s nutrition, physical activity and weight were reported by journalists, and what solutions to the problem were presented.
Methods This study was a descriptive analysis of the content of Australian print news items featuring articles on SPANS. Australian print news items associated with SPANS findings between January 1997 and December 2011 were identified through the Factiva online database which provides full-text news, commentary, pictures and media transcripts from Australian and overseas media sources. An initial search was undertaken by two authors (PE, RL) using the search terms: ‘NSW’ and ‘schools’ and ‘survey’ and [(physical activity) or (nutrition) or (overweight) or (obesity)] and [(children) or (teenagers) or (teens) or (kids)]. News items were eligible to be included in the study sample if they cited SPANS (in full or abbreviated) or a non-branded reference to the survey (deduced by coder, and cross-checked with a SPANS investigator if required). Duplicates, same article in the same publication on the same date in the same edition, were excluded from the initial sample, as well as non-press articles (i.e. peer-reviewed publications and company annual reports). Given that key SPANS researchers also coordinated media releases to coincide with academic publications of findings, a second Factiva search was conducted to identify additional print news items reporting SPANS findings associated with these publications one month prior to, and three months following, each academic publication, using the primary authors’ names as the search term. To determine the number of SPANS news items as a proportion of all Australian print news items on child obesity, nutrition and physical activity in NSW, a third Factiva search was conducted for the same time period using the search terms: ‘NSW’ and [(physical activity) or (nutrition) or (overweight) or (obesity)] and [(children) or (teenagers) or (teens) or (kids)].
Coding and analysis Each news item was categorised according to a range of factors associated with the content of each item using criteria from previous research on print media coverage of obesity and physical activity topics.10,18 These were: • Placement of news item: name of news outlet. • Length of item: word count. • Prominence of news item: page position. • Emphasis: topic mentioned in the first paragraph. • Prominence of survey findings: whether findings were reported in the title, or as background. • Branding of SPANS: survey reference.
2013 vol. 37 no. 3
• Funding acknowledged: Implicit or explicit acknowledgement to NSW Health or NSW Government. • News angle: aspect of SPANS which generated news media attention, determined from headline and first paragraph only. • SPANS context: the overall content of the news item and how survey findings were used in the item. • Responsibility assigned: who was responsible for the problem discussed, as reported, • Solutions: solutions offered to combat the problem, as reported. The content of each aspect was categorised, with categories generated by researchers (RL, PE) through preliminary analysis of the full sample of news media items. For example, the content of news angles was categorised and coded as: nutrition (only), physical activity and sedentary behaviour, nutrition and physical activity, overweight and obesity, health risk, and positive health news. The coding categories for each aspect are listed in Appendix 1 in the supplementary material. News angles were identified by analysing the headline and first paragraph which have been identified as the ‘summary’ of news articles.19 The content categorisation was conducted manually by two authors (RL, PE). Inter-rater reliability was calculated on more than 10% of the sample and showed 100% agreement for 18/32 coding categories (mean 94%, range 66.7-100%). Simple descriptive statistical comparisons (between content categories) were conducted using SPSS (Version 20 for Windows, Chicago, IL, USA). To examine temporal changes in media reporting, news items were partitioned into two time frames: items published between 1997 and 2006; and items published between 2007 and 2011. Descriptive statistics were calculated and temporal changes between the two time frames assessed using chi-squared analyses.
Results A total of 462 news items were initially identified. Of these, 48 were duplicates and 52 were non-print media articles. The remaining 362 news items were screened for eligibility and 218 were excluded, resulting in 144 news items in the study sample. Of these, 51 news items were identified in the second Factiva search. The third search, of all Australian news items referring to obesity, diet, physical activity or sedentary behaviour in children in NSW, showed that the number of news items has risen dramatically from 124 in 2001 to 740 in 2011, with a peak of 964 news items in 2006. Figure 1 shows the number of news items per year referring to SPANS and SPANS items as a proportion of the total number of news items on child obesity and related behaviours in NSW that were featured in Australian print news . There were two peaks in the number of SPANS-related news items, in 2006 and in 2011, which corresponded to the dates of ministerial announcements of the survey findings. In 2001, however, news items with a SPANS reference made up 9% of total Australian news items featuring childhood obesity and related behaviours in NSW.
Placement, prominence and branding of the survey findings in news items A summary of the content of news items in terms of their placement, prominence and branding of SPANS is shown in Table 1.
AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH © 2013 The Authors. ANZJPH © 2013 Public Health Association of Australia
285
Espinel et al.
Article
10
10
8
2.3%
2.3% 5 3
6
2%
8
2.3%
6
1.4%
# of SPANS news items
SPANS as % of total obesity news items 10%
30
30
9.0%
30
33
9% 8%
7.4%
25
15
7% 6%
20
15
4.7%
13
4.1%
11
3.3%
3.0%
3
0 00.0% 1997
0 0.0% 1998 1999
3 0.4% 2000
2001
2002
2003
3%
2.3% 5
6 1.4%
2
4%
8
2.3% 5
5%
15
2004
2005
0.9% 2006
2%
3 0.7%
2007
1%
0.5%
2008
2009
2010
Proportion of all print media items
SPANS as % of total obesity news items
# of SPANS news items
10
The three most prevalent types of news angle were: physical activity and sedentary behaviour, overweight and obesity, and nutrition, and these comprised about three-quarters of the news
5 0.9%
1.4%
3
35
Type of news angles: aspect of SPANS which generated news media attention
0% 2011
Year
Table 1: Placement, prominence, branding of SPANS findings. All (n=144)
# of SPANS news items
SPANS as % of total obesity news items
News items n (%)a 1997-2006 (n=85) 2007-2011 (n=59)
P Value for trend
Placement of news item: name of news outlet National
27 (19)
19 (22)
8 (14)
Major metropolitan
78 (54)
52 (61)
26 (44)
Regional/Local
39 (27)
14 (17)
25 (42)
0.003
Prominence of news item: page number First
7 (5)
7 (8)
0 (0)
2-5 page
32 (22)
22 (26)
10 (17)
Other
83 (58)
41 (48)
42 (71)
Not stated
22 (15)
15 (18)
7 (12)
0.02
Length of item: word count Short ( up to 300 words)
44 (31)
26 (31)
18 (30)
100 (69)
59 (69)
41 (70)
Non branded reference to SPANS survey
80 (56)
58 (68)
22 (37)
Full title only
53 (36)
24 (28)
29 (49)
Full title and abbreviation
7 (5)
3 (4)
4 (7)
Abbreviated only
4 (3)
0 (0)
4 (7)
In headline and first paragraph
17 (12)
12 (14)
5 (9)
In first paragraph only
43 (30)
20 (24)
23 (39)
As background
84 (58)
53 (62)
31 (52)
Long (>300 words)
0.9
Branding of SPANS: survey reference 0.001
Prominence of survey findings: position within the item 0.12
a Percentages rounded.
286
2.3% 5
3 Figure 1: Number of SPANS news items and1%2as a 2 0.9% 0.7% 0.7% 0.5% 0.5% 0 0.4% 0 0 0.4% proportion of all news items child obesity, 0% nutrition 00.0% on 0.0% 0.0% 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1997 2008 1998 2009 1999 2010 2000 2011 2001 2002 2003 2004 2005 2006 2007 2008 2009 201 –Figure 1 Number of SPANS news items and as a proportion of all news items on child obesity, nutrition and physical activity between and physical activity between 1997 and 2011. Year Year 1997 and 2011 3
Number of print news items
recognition was more frequent in news items published between 5 2007 and 2011, compared with news items published between 1997 0 00.0% and 2006 (p=0.001). Only 14% of news items linked a research 1997 group to SPANS; and government funding was rarely mentioned (17.4%). In fact, acknowledgement of survey funding decreased in news items published between 2007 and 2011, compared with those between 1997and 2006 (p
