Plenary: Robert Carroll - University of Bristol

Impact of the growing use of narrative verdicts by coroners on geographic variations in suicide: analysis of coroners’ inquest data

Carroll R, Hawton K, Kapur N, Bennewith O, Gunnell D

Background

Death certificates issued by Coroners form the basis of suicide statistics in England and Wales. There have been recent concerns that a change in Coroners’ recording practices – an increased use of narrative verdicts – may affect the reliability of local and national suicide rates.

Aims

To assess the impact that recent changes in Coroners practice may have had on the reliability of statistics regarding the incidence of suicide.

Methods

We used Ministry of Justice (MoJ) data on the outcome of Coroners’ inquests in 2008 and 2009, and suicide mortality data for Local Authorities (2001-2 and 2008-9) to investigate variation in use of narrative verdicts between coroners and the impact of such variations on official suicide rates. MoJ data do not separately identify deaths given narrative verdicts so we used “other” verdicts (79% of which are narratives) as a proxy.

Results

There was wide geographic variation in the use of ‘other’ (mainly narrative) verdicts between Coroners (from 0%-50%, median=9%). Coroners who gave more ‘other’ verdicts gave fewer suicide verdicts (r = -0.41; p<0.001). In the ten Coroners’ jurisdictions where the highest proportion of ‘other’ verdicts were given, the incidence of suicide decreased by 16% between 2001/2 and 2008/9 whereas it did not change in areas served by the 10 coroners who used narratives the least.

Conclusions

Variation in coroners’ use of narrative verdicts influences the validity of reported regional suicide rates. Action is warranted to ensure Coroners who record a narrative verdict also include a likely cause of death in a short form verdict. In the meantime small area suicide rates, and changes in these rates over time in the last decade, should be interpreted with caution.

Plenary: Rona Campbell - University of Bristol

A quantitative exploration of multiple risk behaviours in adolescence using data from the ALSPAC cohort

MacArthur GJ, Heron J, Kipping R, Smith M, Hickman M, Campbell R

Background

Risk behaviours such as smoking, alcohol use and antisocial behaviour cluster in adolescence and can significantly compromise health and wellbeing later in life. Most studies have focused on a limited number of health risk behaviours and little is known about the clustering of a wide range of risk behaviours in young people.

Aims

To examine the prevalence and clustering of 13 risk behaviours in young people participating in a longitudinal birth cohort (the ALSPAC study).

Methods

Analysis of the prevalence and distribution of multiple risk behaviours in young people aged 15-16 years using self-reported data from questionnaires (n=2,695). Data relating to risk behaviours involving substance use, sexual behaviour, offending, antisocial behaviour, self harm, injury risk behaviour, and low levels of physical activity were analysed by gender and socio-economic status. Correlation matrices were constructed to explore clustering.

Results

 At 15-16 years of age, 40% (n=1059) of adolescents engaged in between three and five behaviours (42.0% boys, 38.4% girls) and 6.2% (n=164) engaged in seven or more (6.0% boys; 6.3% girls). Just 5% (n=125) did not engage in any of the risk behaviours examined. The most prevalent risk behaviours were physical inactivity (74.3%), antisocial behaviour and criminal offending (42.1%), and hazardous drinking (33.7%); 28% of adolescents also engaged in car passenger-related risk behaviour. A significantly greater proportion of girls reported physical inactivity, self harm, tobacco smoking, and sexual risk behaviour (all p<0.01), whilst a greater proportion of boys reported antisocial behaviour and offending, cannabis use, cycling without a helmet, and scooter risks. Car passenger-related risk behaviour, scooter risk, tobacco smoking, sex before 16 years of age, self harm, physical inactivity and time spent viewing television were associated with social class (all p<0.05). Substance use behaviours tended to co-occur and cluster with self harm and with criminal and anti-social behaviours. Physical inactivity and time spent viewing television did not cluster with other risk behaviours.

Conclusions

Young people exhibit a number of health risk behaviours and there is evidence of patterning by gender and social position. Some risk behaviours cluster together and such clustering may have implications for the development of preventative interventions.

A1: Mathew Dominey - NHS Somerset

Childhood obesity and deprivation: A quantitative analysis of primary school children

Dominey M, Wyatt K, Henley W

Background

Obesity is a major and growing problem in today’s society and has significant health consequences and societal costs. There is growing acceptance that levels of obesity need tackling although there are limited resources, the evidence with regard to what is most effective is unclear and further research is needed.

Aims

The aim of this study is to analyse the association between deprivation and obesity in primary school children, exploring the hierarchical structure of the data whereby data may be nested within areas or schools.

Methods

A secondary quantitative analysis of data from the National Child Measurement Programme is undertaken to analyse the association between deprivation and obesity. The following variables were obtained academic year, gender, age, school year, body mass index and child and school super output area, giving a total of 64845 records. Standard methods of statistical estimation and inference are used to test for variation in the levels of obesity across different groups and to quantify the magnitude of any differences with more complex multi level modelling techniques being adopted to explore the hierarchical nature of the data.

Results

The initial analyses showed significant differences in obesity across counties, academic years, genders and ages with correlation analyses revealing an association between deprivation and obesity. More complex multi level modelling techniques indicated that a small proportion of the variability in obesity levels, 1% to 4%, could be explained by clustering at a school level with around 10-15% of this variability explained by pupil factors and 3-5% by school factors.

Conclusions

The results from this research have implications for the practitioner, in terms of identifying schools which would benefit from a health promotion intervention, and the academic, in terms of encouraging further explorative and theoretical research to understand the relationship between deprivation, schools and obesity.

B1: Lianne Strauss - NHS North Somerset

Improving the uptake of Childhood Immunisations in North Somerset: A Review

Strauss L, Yeo H, Kipping R

Background

North Somerset is failing to meet national targets for uptake of immunisations for children aged <5.

Aims

To undertake a review to understand and address factors contributing to low vaccination rates.

Methods

The review included a health equity audit, questionnaire to General Practices, and a social marketing analysis. Data for uptake rates were obtained from the Child Health Information System (CHIS) for 2009/10. Differences in uptake were analysed by General Practice, deprivation quintiles, Children’s Centres, locality areas, and maternal age. The GP survey was emailed to 25 practice managers and responses were analysed in Access. The social marketing analysis (MOSAIC) compared 61 identified ‘lifestyle types’.

Results

The Health equity audit found the lowest rates of vaccination were among children of mothers aged ¬¬<25 years old >35 years old. For immunisations due by age one, children of mothers aged 25-29 had a 99% uptake rate whereas mothers aged 20-25 and >40, had rates of 93% - 94%. Children living in the Central Weston Children’s Centre catchment area consistently had the lowest immunisation rates. There was no strong evidence of an association between deprivation and rates of immunisation. 88% of practices responded to the survey. Practice waiting lists, inconsistent access to staff training, and variation in effective appointment systems were identified as possible reasons for variation in immunisation uptake. The social marketing analysis identified that four of the five ‘lifestyle’ groups least likely to be immunised were likely to respond to poster and film/television advertising.

Conclusions

In North Somerset, the Weston area should be prioritised for any interventions to promote uptake. Practices and the PCT need to work together to ensure practices are aware of the processes and systems that can affect immunisation rates. Local campaigns to improve immunisation rates should consider the results of the social marketing analysis.

C1: Kate Glyn-Owen - NHS Wiltshire

Marriage and Mortality: Analysis of the relationship between marital status and mortality with respect to biological, psychosocial, cognitive, lifestyle and relationship factors

Background

Over a century of research has demonstrated that marriage confers a protective effect against mortality. Estimates are that married people may experience 10-15% lower mortality rates than the not married. Identifying the mechanisms by which marriage protects against mortality could help to identify new channels by which public health specialists and policy makers can work to improve health.

Aims

To use data from a prospective cohort study of Welsh men (the Caerphilly Prospective Study) to investigate the relationship between marital status and mortality and to identify significant modifiable variables contributing to this relationship, which might be relevant to public health work.

Methods

The data were already collated and available. Multivariate logistic regression analysis was performed to investigate the relationship between marital status and mortality with respect to a wide range of biological, psychosocial, cognitive, lifestyle, relationship and risk factors.

Results

After adjusting for confounding variables (height, employment status and social class) the odds ratio for mortality for married men compared to unmarried men was 0.60 (95% confidence intervals 0.47, 0.78 p<0.001). For men in the Caerphilly Prospective Study, the relationship between marital status and mortality is at least partly explained by significant differences in dietary intake of fruit, vegetables and vegetable fibre. Body weight is at least one pathway by which all these variables influence mortality.

Conclusions

These findings are consistent with the widely recognised health benefits of eating fruit and vegetables, reflected in both UK-wide and worldwide public health campaigns to increase fruit and vegetable consumption. Our results suggest that it may be beneficial to target public health messages regarding fruit and vegetable consumption at individuals who are not married. These findings are unlikely to be generalisable to women or to men of significantly different racial, cultural or geographical origins to the Caerphilly Prospective Study men.

D1: Jenny Weeks - SWPHO

Self-harm in those aged under 15 years

Background

Children under the age of 15 account for 5% of all hospital admissions for self-harm in the South West (1). The importance of early intervention when mental health problems arise is reflected in national strategy.

Aims

To identify the groups where admissions are greatest to help target preventative services at those of greatest need.

Methods

We analysed data of patients aged under 15 years using the Hospital Episode Statistics extracted for the SW suicide report (1).

Results

Between 2001/02 and 2008/09 there were 3,529 admissions for self-harm for 2,896 individuals aged under 15 years, 14% being admitted more than once. Admissions did not rise over time and were highest in areas of greatest deprivation. Children aged 5-9 accounted for 2% of admissions and those aged 0–4 for 5%. The majority of admissions (93%) were children aged 10 and over and were 5.7 times more likely to be girls than boys. For all ages, 89% of admissions were due to poisoning, and 5% due to a sharp object.

In children aged over 10 years, poisoning was the cause of 92% of admissions (94% for girls). The principle cause of poisoning was over the counter drugs (70%) followed by prescribed drugs (23%), drugs of abuse (4%) and alcohol (2.5%). Children aged over 10 years from the most deprived quintile had more than double the rate of admissions for poisoning compared to the least deprived. They also had a higher proportion admitted for poisoning by prescribed drugs and drugs of abuse.

Conclusions

Children admitted for self-harm in children are predominantly girls aged 10 years or older due to poisoning mostly by over the counter medicines, but with a higher incidence of prescribed medicine in children from more deprived areas. It is unlikely that children under the age of 5 will have committed deliberate self-harm, suggesting a need for a coding review.

(1) Suicide and self-harm in the South West, SWPHO September 2011

A2: Penny Marno - NHS Somerset

A qualitative study to understand communication around obesity and healthy eating between health professionals, young people and families

Marno P, MacArthur G, Dickens F, Chant S, Grant T

Background

PoHeFa (Policy, Health and Family Learning) is a collaborative European Project, involving NHS Swindon and NHS Somerset, which aims to improve the quality, effectiveness and sustainability of health promotion initiatives.

Aims

To understand how information is communicated and received about obesity and healthy eating, whether current approaches are effective at changing behaviour, and how knowledge about healthy lifestyles can be improved.

Methods

Five focus groups were run in both Swindon and Somerset with health professionals, young people and families. Perspectives around nutrition and being overweight or obese, views on communication, and experiences of giving or receiving information were explored. Focus groups were transcribed verbatim, and analysed thematically.

Results

Being overweight was seen by some as a ‘social norm’ often across generations of families. Health professionals felt that people perceived being challenged about their weight as insulting, and there was confusion over whether overweight / obesity was a medical or lifestyle issue. A range of barriers to communicating around overweight and obesity and to leading healthy lifestyles were identified. Young people’s perceptions were influenced by parents, school, and the media. Achieving behaviour change depended on building the knowledge-base as well as improving self-esteem and confidence so people feel empowered to make positive choices around healthy lifestyles. Parents and young people in Swindon focused more on social barriers to lifestyle change and the role of services, whereas Somerset groups gave more detail on the type of information they would find useful.

Conclusions

Communicating information about obesity and healthy eating is complex, and health professionals are just one of many influences on people’s perceptions and lifestyle choices. There was little difference in the issues raised in a rural or urban area but work is needed on improving both the content and method of communication to ensure messages are accessible, appropriate and consistent.

B2: Linda Matthews - UWE

A Review Is MMR delivering? A study of rubella susceptibility and post partum immunisation in pregnant women in South Wales

Matthews L, Lawrance LM, Gray D, Gray S

Background

Rubella, an insignificant childhood illness, can result in Congenital Rubella Syndrome (CRS) in children of women infected in pregnancy. Vaccination of schoolgirls was introduced in 1970, replaced by the MMR in 1988, with a second dose added in 1996. CRS incidence remains low but vaccination policy changes may yet have an impact; the media driven controversy around the MMR has affected uptake.

Aims

To report data, drawn from ante-natal screening samples, for one Welsh Health Board area over a six-year period (2005-10) and include an examination of post partum MMR immunisation uptake.

Methods

Data from screening samples tested by ELISA (Diasorin ETI-RUBEK) was analysed and 114 susceptible women completed questionnaires and also consented to access of their immunisation records. Data was analysed using IBM SPSS and evidence of post partum immunisation was gathered from patients’ notes.

Results

The level of rubella susceptibility (defined by <10 IU/ml) has risen overall. Variations are observed between age-groups and number of pregnancies; for first pregnancy susceptibility has increased from 6.4% to 8.9% over six years. Examination of the immunisation records for women with an antibody level <10 IU/ml showed, 24% of women had no Rubella immunisations, 13.5% had received one dose, 62.5% had received two doses or more. Immune women also showed a marked drop in antibody titre in those receiving combined vaccine. Over 30% of women in the study did not receive post partum MMR.

Conclusions

This increase in Rubella susceptibility, whether due to reduction in uptake of MMR or because of the change in vaccination policy, may result in the re-emergence of rubella and cases of congenital rubella. The poor uptake of post partum MMR immunisation is also an area of concern.

C2: Fiona Spotswood - Bristol Business School, UWE

An ethnographic study of physical inactivity in a deprived community: Implications for social marketing

Spotswood F, Tapp, A

Background

Leisure time physical activity (LTPA) can limit obesity, which creates individual and community cost. UK LTPA is socially patterned, with lower socio-economic groups participating less. Existing research tends to focus on individual predictors such as income or opportunity rather than viewing the pattern as having class-cultural roots. To explore the class-cultural perspective, the authors found theories commonly used by social marketers lacking, as they do not provide adequate insight into the determinants of culturally-driven habitual behaviours; but rather focus on cognitive factors. In contrast, Bourdieu’s habitus, our theoretical framework, views social class holistically as a shared set of cultural dispositions.

Aims

To explore the ‘working class’ cultural habitus and its influence over LTPA participation with a view to making recommendations for social marketers.

Methods

Three months of ethnographic research was undertaken with five families living on a deprived estate in Stockport. Ad hoc and formal interviewing bolstered the participant observation. Fieldnotes were analysed using NVIVO.

Results

A set of cultural pre-dispositions towards physical activity were observed, making it unappealing or dismissed. Observed priorities were ‘family survival’, image management and withdrawal through fantasy. There was a ‘clash’ between the ‘end state goals’ of the people observed and LTPA. For example, the goal of instant pleasure would clash with LTPA, which has longer term benefits. Findings led to the retroductive modelling of three mechanisms underpinning these behaviours; a lack of ‘control’, ‘participation’ and ‘perspective’. These mechanisms will be explored in further depth.

Conclusions

The three differing options suggested for social marketers are to market LTPA as being ‘in line’ with the group’s existing cultural worldview; attempt to change the ‘working class’ habitus through community development so LTPA will be more readily adopted; or to use 'cue-based' marketing techniques based on behavioural economics or ecological theories to increase LTPA and PA levels.

D2: Yvette Morey - UWE

Understanding Self-Harm Online

Morey Y, Warren S, Verne J

Background

Recent years have seen an increase in deliberate self-harm content on the Internet. In addition to message-boards and support sites, discussions and representations of self-harm can be found on a number of social media sites (Tumblr, LiveJournal, Facebook, Bebo, and You Tube). Research has either focused narrowly on specific sites (i.e. message-boards) or more generally on issues surrounding problematic youth generated content. Contrasting arguments are emerging about the advantages and disadvantages of self-harm content online (e.g. providing an outlet for difficult emotions or normalising risky behaviour) however it is still too early to know what effects this content will have (Boyd, Ryan and Leavitt, 2010). A report by the South West Public Health Observatory (2011) points to a nationwide increase (49%) in hospital admissions for self-harm between 2002/3 and 2008/9. However, it has been established that the majority of self-harmers never seek help or medical attention (Adler & Adler, 2007) indicating that these figures could potentially represent the tip of the iceberg.

Aims

To conduct a review of self-harm sites and content online; to explore the ways in which young people engage with these; to map the different functions fulfilled by these.

Methods

The initial part of this two-part study involves: i) a literature review, ii) an online observation or netnography (Kozinets, 2002a) of sites and content related to self-harm.

Results

Initial results reveal a diverse range of content combining popular culture references with personal reflections and perceptions. Sites contain a significant social element including the sharing of content. Where attempts have been made to restrict sites or content these often resurface elsewhere.

Conclusions

Given that many self-harmers remain invisible to health and social organisations, the study of self-harm online offers an innovative and rich way of gaining insight into the individual, social and cultural meanings of this behaviour.

A3: Nicholas Young - University of Bristol

A cross-sectional analysis of the association between diet and the insulin-like growth factor-axis in men in the United Kingdom

Background

The risk of developing prostate cancer is positively associated with serum levels of insulin-like growth factor-I (IGF-I). Evidence for associations between prostate cancer and insulin-like growth factor-II (IGF-II), insulin-like binding protein-2 (IGFBP-2) or insulin-like binding protein-3 (IGFBP-3) is inconclusive.

Aims

To examine the association between dietary variables linked to prostate cancer and serum levels of IGF-I, IGF-II, IGFBP-2 and IGFBP-3, investigating their role as biomarkers and mediators of lifestyle associations with this disease.

Methods

A cross-sectional analysis of self-reported 12-month dietary intake and serum IGF and IGFBP levels was performed using data from 1851 subjects screened for prostate cancer as part of a UK multicentre trial comparing treatments for this condition. Fully-adjusted multivariable linear regression models were used to test associations between dietary intake and IGFs or IGFBPs.

Results

For a one standard deviation (s.d.) increase in dairy product and dairy protein intake there is an increase in IGF-I of 4.66 ng/ml (95% Confidence Interval (CI) 1.97,7.35 ng/ml) and 6.02 ng/ml (95% CI 3.34,8.71 ng/ml) respectively. A 25% increase in calcium and selenium intake increases IGF-I by 5.92 ng/ml (95% CI 3.77,8.07 ng/ml) and 2.61 ng/ml (95% CI 1.10,4.13 ng/ml) respectively. There are no dietary associations with IGF-II. A one s.d. increase in animal protein decreases IGFBP-2 by 6.20% (95% CI -8.91,-3.41%) and there is some evidence of an inverse association of this peptide with dairy protein and calcium. There is no evidence of any consistent dietary associations with IGFBP-3.

Conclusions

IGF-I and IGFBP-2 levels are associated with dietary intake in self-reported food questionnaires in men in the UK, and warrant further investigation as part of randomised trials of dietary interventions to reduce the risk of prostate cancer. There is no evidence that IGF-II or IGFBP-3 are biomarkers or mediators of dietary associations with prostate cancer.

B3: Joanna Jacomelli - HPA

Broad factors affecting the uptake of the H1N1 Influenza A pandemic vaccine in nurses

Jacomelli J, Pollock J

Background

Existing research reports that healthcare workers accepting the H1N1 influenza vaccine tended to be older, male physicians. History of previous seasonal flu vaccine was also strongly associated with vaccine acceptance. Reasons given for acceptance centred on self-protection and for refusal, concern about side effects. However, there is limited research into acceptance of the vaccine by nurses in acute settings who tended to have particularly low uptake of the vaccine.

Aims

Identify differences between those hospital based nurses who received the pandemic influenza vaccine and those who did not and examine the reasons affecting their decision.

Methods

A structured questionnaire was distributed to 200 nurses over two weeks at an Acute Trust with the lowest vaccine uptake in the region. A qualitative interview was undertaken with a member of Occupational Health to better understand the organisational factors that may have influenced uptake of the vaccine. Logistic regression analysis was undertaken to identify those demographic/behavioural factors predicting H1N1 vaccination.

Results

Of the nurses responding (response rate 32.5%), 61.7% accepted the vaccine across six wards. No significant association between uptake of the vaccine and age, specialty, years working in the NHS or type of contact with patients was found. Those who had the seasonal influenza vaccine were significantly more likely to accept the pandemic vaccine than those who did not (odd ratio 13.0, 95% CIs 3.87 – 43.67, p<0.01). Respondents reported most frequently “protect one’s family” as very important in making their decision to have the vaccine. Those rejecting the vaccine most frequently cited “I just didn’t want the vaccine” as very important.

Conclusions

The study supports the finding in the literature that seasonal influenza vaccination is associated with pandemic influenza vaccination in nurses. Uniquely, the main factor related to refusing the vaccine was non-specific and acceptance was greatly influenced by family circumstances.

C3: Jane Powell - UWE

A systematic review of effective behaviour change techniques in community walking and cycling interventions: the iConnect study

Bird E, Baker G, Powell J, Mutrie N on behalf of the iConnect Consortium

Background

iConnect, funded by EPSRC, is an interdisciplinary research programme, aiming to measure and evaluate changes in travel, physical activity and carbon emissions associated with the Sustrans’ Connect2 programme. Until recently, the categorisation of behaviour change techniques (BCTs) has been problematic due a failure to standardise the vocabulary used across intervention programmes. This has restricted the potential for understanding how intervention content is related to the efficacy of an intervention and this is an important consideration for the Connect2 programme. This research attempts to address this gap in the evidence base for the design of interventions to promote walking and cycling in the UK.

Aims

Adopting a recently developed twenty-six item taxonomy of behaviour change techniques, this systematic review categorised the BCTs incorporated into the content of interventions designed to promote walking and cycling; with the aim of identifying which techniques were associated with effectiveness, to help guide the development of future walking and cycling interventions.

Methods

A systematic review of controlled interventions that reported a significant change in walking and cycling were obtained from eight electronic databases. Study design features, context, methodology and outcomes were extracted. Intervention content was independently extracted and categorised according to a validated twenty six item taxonomy of BCTs.

Results

Eighteen studies met the inclusion criteria. Overall study quality was fair. Intervention content varied greatly in terms of vocabulary and level of detail provided. Studies reporting a significant improvement in walking and cycling outcomes incorporated a significantly higher number of BCTs into their design, when compared with studies that failed to produce significant changes, or did not report on significance.

Conclusions

Although the inclusion of multiple BCTs into intervention design is a time consuming, complex process; their inclusion appears to improve the potential of the intervention to achieve the desired outcomes of promoting and increasing walking and cycling behaviour.

D3: Frances Clarke-Stone - NHS Gloucestershire    

      Diane Crone - University of Gloucestershire

From the patient perspective, what is the value of participating in an art and health programme within primary care? Qualitative evaluation of the Gloucestershire Art Lift project.

Clark-Stone FJ, O’Connell EE, Crone DM, James DVB, Tyson PJ

Background

Improving mental well-being is a public health aim, as evidence shows that mental well-being is positively associated with a range of improved health and social outcomes. The use of art to improve well-being is widespread, however there is a shortage of rigorous evaluative literature measuring the impact on participants. Gloucestershire Art Lift is an art intervention where health professionals refer patients for a ten week art programme, usually delivered in a primary care setting

Aims

Building on quantitative findings including a statistically significant improvement in mental well-being, this qualitative study investigated the experiences, opinions and perceptions of the art intervention from the patient perspective.

Methods

Recruitment and informed consent procedures followed an agreed protocol from the NHS Ethics approval application. Patients who completed the intervention were invited to take part in the qualitative part of the evaluation via a telephone call and information letter. One in three patients who volunteered were selected to take part in focus groups, and some patients opted for one to one interviews via telephone or in person. Focus group and individual interviews were recorded and transcribed verbatim. Inductive content analysis was undertaken in the context of the evaluation aims and objectives. Analysis was managed using the qualitative software package NVivo 8 (QSR International Pty Ltd, 1999-2008).

Results

The qualitative results added more depth to the wellbeing changes and suggested some explanation for these improvements, through perceived benefits such as enjoyment, a new interest, improved confidence, distraction, therapeutic value and social support. The importance of continuation and sustainability for participants was also evident. The combined findings highlighted the different aspects of the success of the project and the potential that these programmes have for health improvement in primary care.

Conclusions

Art interventions in primary care can improve the mental well-being of patients because they provide enjoyment, social support and distraction from illness, and improve confidence. As improving mental well-being is positively associated with a range of improved health and social outcomes, art may play a useful role in improving public health.

A4: Lesley Wood - School Food Trust

Evaluation of the Let's Get Cooking programme

Background

There has been growing concern about the decline in cooking skills over the past few decades, which may have contributed to poor diet and the increase in prevalence of obesity. Let's Get Cooking received £20 million from the BIG Lottery fund to establish a programme of 5000 school-based cookery clubs across all English local authorities over a five year period, with an increasing emphasis on targeting areas of high deprivation.

Aims

To evaluate the effectiveness of a school-based cooking club on three outcomes agreed by BIG.

Methods

A random sample (stratified by region and dichotomised eligibility for free school meals) of clubs was selected. Purpose-designed simplified food-frequency questionnaires were used to collect basic dietary data from club participants on two occasions. Data were analysed using SPSS. Ethical approval for the evaluation was granted by King’s College London.

Results

Data were received from 2433 participants attending 243 clubs. The majority (92%) of members replicated a skill learnt at the club when at home. Each club member passed on their new knowledge to an average of 1.22 people and 1,597,529 people have been reached through Let's Get Cooking. There was a significant increase in the mean number of 'healthy' food items eaten after participation in club activities (15.2 items compared with 12.5 items prior to participation; p<0.001). The ratio of less healthy to healthy items was also greater at follow-up compared with baseline (8.6 vs. 6.8; p<0.001), providing evidence of a beneficial shift in eating habits after club attendance. Comments received from the participants suggest that the benefits of attending the club can also improve social interaction and self-esteem.

Conclusions

This evaluation provides evidence that a six to eight week school-based cookery course can improve nutritional knowledge and eating habits which may be of benefit for longer term health.

B4: Maya Gobin - HPA

First Escherichia coli O157 outbreak linked to crab meat, South West England, August 2011

Matulkova P, Gobin M, Oshin F, Thould G, Oliver I, O’Connor K

Background

In August 2011 five adult cases of Vero cytotoxin-producing Escherichia coli (VTEC) O157 phage type (PT) 21/28 verotoxin (VT) 2 infection were reported in a city in South West England. An epidemiological study was conducted after initial investigations identified exposure to crab meat in the city as a possible source.

Aims

To determine the source of the outbreak to prevent further spread, by testing the null hypothesis that infection was not associated with consumption of crab in the city.

Methods

Further cases were identified through VTEC enhanced surveillance. A case-control study was conducted. A case was defined as a person who developed diarrhoea after 19/07/2011 with either laboratory diagnosis of VTEC PT21/28 VT2 or an epidemiological link and who lived in or visited the city within the incubation period. Controls (3:1) were healthy adults living in the city, identified by case-nomination or selection through volunteers in Local Authority and NHS. A bespoke structured questionnaire was developed. Environmental health investigation, supplier tracing and other public health actions were undertaken.

Results

A total of nine cases were identified, eight laboratory confirmed and one with an epidemiological link. The study included eight confirmed cases (mean age 57, female 75%, symptoms onset between 30/07 and 16/09/2011, bloody diarrhoea 75%) and 28 controls (mean age 51, female 61%). The univariate analysis showed a statistically significant association between cases and consumption of crab meat away from home (OR=58.3; 95% CI 4.2-2725.8). Environmental health investigation, which is not yet concluded, revealed a possible link to unapproved crab supplier(s) in the city.

Conclusions

This is the first reported outbreak of E Coli O157 associated with consumption of crab. Cross-contamination of crab meat in an unregulated premise is a possible explanation. The small number of cases in the study may have limited the detection of other possible vehicles.

C4: Elissa Fay Southward - University of Bristol

Mapping the journey to and from school in secondary school age children: the PEACH project

Southward EF, page As, Wheeler BW, Cooper AR

Background

Active travel has been suggested as a possible method to increase overall physical activity in children, but identifying the contribution of the journey itself from other time spent indoors and outdoors is challenging. This study combines accelerometer and GPS data to quantify physical activity on the walk to and from school in relation to overall daily levels.

Methods

Participants were 953 children aged 11-12 from Bristol, England, measured between 2008 and 2009. Accelerometers measured physical activity and Global Positioning System receivers measured outdoors location before and after school every 10 seconds for 3 weekdays. Journey data were time-matched and mapped in a Geographic Information System. Mode of travel was self-reported using a questionnaire. Minutes of moderate to vigorous physical activity (MVPA) were compared for the journeys between 8am-9am and 3pm-5pm and in relation to whole-day levels.

Results

The journey to school accounted for 19% of daily MVPA and the journey home 20%, a total of 22 minutes of MVPA each day. Physical activity on the journey did not differ between boys and girls, but since girls have lower levels of daily physical activity than boys (boys acquired 7.5 more minutes MVPA), the journey contributed a greater proportion to their daily MVPA (40.8% vs. 38.2%).

Conclusions

The journey to and from school is a significant contributor to daily physical activity and MVPA in 11-12 year old children. Combining GPS and accelerometer data within a GIS is a useful approach to quantifying specific journeys.

D4: Emma Bird - UWE

‘Happy Being Me’ in the UK: A controlled evaluation of a school-based body image intervention with pre-adolescent children

Bird E, Halliwell E, Diedrichs P

Background

Body image concerns among children are commonly associated with the onset of severe health-related implications including reduced physical activity, unhealthy eating behaviours, and low self-esteem, among others. Historically, attempts to address these issues have been developed for, and conducted with, adolescent girls alone. However given the increase in evidence for the onset of negative body image in pre-adolescent boys as well as girls, and if these damaging and costly health problems are to be addressed; early intervention that aims to prevent the onset of negative body image in children, is an important public health priority.

Aims

In response to the dearth of research investigating the efficacy of body image interventions with a mixed, pre-adolescent cohort; this novel study aimed to evaluate an adapted version of ‘Happy Being Me’, a school-based body image intervention for pre-adolescent girls and boys.

Methods

Forty-three 10-11 year olds participated in a three-week, school-based intervention programme (45 in control group), completing measures on risk factors for negative body image, eating behaviours and self-esteem at baseline, post-intervention and three month follow-up. The intervention focused on educating children about the negative consequences associated with internalisation of the thin-ideal, appearance-related teasing, appearance-related conversations and body comparison. The programme also targeted self-esteem and unhealthy eating behaviours.

Results

For girls, participants in the intervention condition reported significant improvements in risk factors for negative body image, eating behaviours and intervention topic knowledge at post-intervention, although changes were not all sustained at three month follow-up. For boys, participants in the intervention condition reported significant improvements in risk factors for negative body image at post-intervention, but were not maintained.

Conclusions

The findings indicate that this adapted school-based intervention, facilitated with a mixed sex, pre-adolescent cohort may have the potential to address the ever-increasing body image concerns observed among young people across the UK.

A5: Adam Lester-George - SWPHO

Building the Business Case for Tobacco Control in South West of England

Lester-George A, Andrews F

Background

In light of the economic challenges facing the public sector, funding for tobacco control is coming under increased scrutiny. Healthcare commissioners and smoking cessation professionals could be supported in this environment through access to evidence-based estimates of the financial and health impacts of tobacco and how these could be moderated through local and sub-national policy.

Aims

To use a recently developed user-friendly and accessible toolkit (Trapero-Bertran, Pokhrel & Trueman, 2011)1 to estimate the economic impact of tobacco control and test whether this tool will provide robust, real-world outputs to support continued investment in tobacco control and Stop Smoking Services.

Methods

The toolkit works on a cohort of individuals at a selected locality divided into current smokers, former smokers or never smokers based on current prevalence data. The individuals are then followed over time through a 3-state Markov model to simulate changes in smoking behaviour and the impacts of this. It is assumed that the natural background quit rate can be elevated both in the smoking population as a whole (through the influence of tobacco control programmes) and, independently, in individual smokers (through recruitment to local cessation interventions). The number of smokers in a locality receiving different interventions was manipulated to model the effects of altering dominant treatments and increasing Stop Smoking Services (SSS) throughput.

Results

The toolkit was used to produce case studies for example areas in SW. Results suggest that the presence of effective SSS in North Somerset results in £422k in savings over two years, increasing to £1m through the existence of a sub-national programme. The same service usage across SW could result in short-term savings of approximately £28.1m (compared with no investment in smoking cessation activities).

Conclusions

Case studies presented here estimate substantial savings to public finances through the implementation of sub-national tobacco control programmes and/or increased use of local smoking cessation services. The practicality of using the tool will be discussed.

Reference: Trapero-Bertran, Pokhrel & Trueman. Building the Business Case for Tobacco Control: A toolkit to estimate the economic impact of tobacco, Brunel University 2011

B5: Matt Hickman - University of Bristol

The primary prevention of hepatitis C among injectors: model projections of the impact of opiate substitution therapy, needle exchange and antiviral therapy

Hickman M, Martin NK, Vickerman P, Coleman B, Annett H

Background & Aims

Bristol like many UK cities has a high prevalence of HCV among injectors. In serological surveys in 2006 and 2009 over 50% of injectors were HCV antibody positive. HCV incidence was estimated to be ~40 per 100 person years in 2006 but was estimated to be <10 per 100 person years in 2009. Bristol has high levels of intervention coverage – opiate substitution therapy (OST) and needle and syringe programmes (NSP). We consider the evidence on the effectiveness of OST and NSP on reducing HCV transmission and what impact scaling up OST and NSP or introducing eHCV antiviral treatment may have on prevention of HCV among IDU in the population.

Methods

(I) Meta-analysis and pooled analysis of UK studies measuring intervention exposure and HCV incident infection. (II) Development of a dynamic HCV transmission model amongst IDUs incorporating information on intervention exposure and effect, and introducing HCV treatment. First, we project the likely impact for Bristol of scaling-up the coverage of NSP and OST from 50 to 60, 70 or 80%. Second, we project impact of introducing HCV treatment on HCV prevalence and perform a probabilistic cost-utility analysis, estimating long term costs and outcomes to determine the cost-efficacy of treating IDUs or treating only ex or non-IDU as compared to no treatment.

Results

(I) Both receiving OST and high NSP coverage were associated with a reduction in new HCV infection [adjusted odds ratios 0.41, 95%CI 0.21-0.82 and 0.48, 95%CI 0.25-0.93, respectively]. Full harm reduction (on OST plus high NSP coverage) reduced risk of new HCV infection by nearly 80% (AOR 0.08-0.52). (II) Model projections suggest that current intervention coverage in Bristol has averted HCV infection (without OST and NSP HCV prevalence would be greater than 80%). But increasing OST and NSP coverage in the UK will reduce HCV prevalence by <25% over twenty years unless the coverage of both is ≥80%. In contrast annually treating people with HCV (who are at risk of transmitting infection to others) could generate marked reductions in HCV prevalence, despite risks of re-infection. For example, in Bristol if 70 IDU with HCV were treated annually chronic HCV prevalence may fall by 30% in ten years, and would halve in 20 years. In addition, the incremental cost efficacy ratio (ICER) compared with no treatment is substantially lower for treating IDUs than treating non or ex-IDUs, estimated at less than £4000 for populations with a chronic infection prevalence among IDU of less than 60%.

Conclusions

There is evidence that uptake of OST and NSP can reduce the risk of HCV transmission among injecting drug users. But further scale up of these interventions may not lead to substantial reductions in HCV prevalence in the population. Model projections suggest that HCV treatment could have a primary prevention role, and treating IDUs may be more cost effective than treating ex-IDUs. Evaluating and extending strategies to treat hepatitis C among injectors is warranted.

C5: Toity Deave - UWE

New hazards in the home? An interview study of home fire safety practices of parents of pre-school children

Deave T, Goodenough T, Towner E, Kendrick D, Stewart J

Introduction

Unintentional injury is the major public health challenge facing pre-school children in England. There are steep social gradients in mortality and morbidity. Patterns and types of injury are closely linked with child development. The NIHR funded ,‘Keeping Children Safe at home’, Programme focuses on the prevention of home injuries in pre-school children in England, to develop evidence-based interventions that can be implemented in Children’s Centres, health centres and other settings.

This study explored knowledge and reported safety practices of parents of pre-school children in relation to thermal injuries. The findings have contributed to the development of an Injury Prevention Briefing for fire-related injuries which will be evaluated in a randomised controlled trial.

Methods

Structured interviews were conducted with parents of children under 5 who attended Children’s Centres in four study centres: Nottingham, Norwich, Newcastle, Bristol. The focus was on fire-related safety practices.

Results

200 parents from 20 Children’s Centres in the four study centres completed questionnaires. Working smoke alarms were reported in 191 (96%) households, 163 (85%) were situated at each level. 105 (54%) families had a bedtime routine and 81 (43%) had a fire escape plan. 140 (70%) families owned hair straighteners, 31% were used daily. 36 (22%) families reported keeping matches where under 5’s could find them and 30 (19%) families where under 5’s could reach them.

Conclusions

Primary healthcare professionals are in an ideal position to promote child injury prevention messages. This study highlights key issues not widely addressed. Families lack fire escape plans and fire-prevention bedtime routines, shown to be effective in reducing fire-related deaths. Parents are not always able to anticipate how rapidly children’s abilities change that put them at risk of injury. Finally, new risks emerge, such as those from hair straighteners, and messages change when new evidence on risk emerges.

D5: Daz Harding - NHS Wiltshire

Compliance with Orlistat prescribing guidelines: an assessment among Wiltshire General Practitioners

Harding K, Gray H, Jones M, Cretney N

Background

NICE Obesity Clinical Guideline 43 sets out a recommended framework to prescribe the anti-obesity drug Orlistat. Briefly, it should only be prescribed after dietary, exercise and behavioural approaches have been used and evaluated in patients with a Body Mass Index (BMI) of 30 or more (or 28 with co morbidities). The medication should only be continued for longer than three months if the person has lost 5% of their initial body weight since starting drug treatment.

Aims

To assess the compliance with NICE prescribing guidelines among the ten highest prescribers of Orlistat in Wiltshire between April and June 2010. Feed back the results to GPs and monitor any subsequent changes in prescribing practice.

Methods

Practices were selected using PCT prescribing data. Each was asked to select the first ten patients prescribed Orlistat who appeared on a search of their records. They could elect to collate the requested information themselves or have the collation of information carried out by a member of the Public Health Department.

Results

Nine out of ten practices responded. 46% of patients audited had no period of supervised dietary, exercise and behavioural change evaluated before Orlistat prescription. 15% of patients did achieve a 5% weight loss over 3 months. Of those that did not achieve the 5% weight loss in this period, 70% were however given another prescription (15% of these had diabetes).

Conclusions

Compliance with NICE Obesity Clinical Guideline 43 in relation to Orlistat varied markedly among Wiltshire GPs. Results of this study were fed back to all Wiltshire GPs together with a flowchart detailing the NICE guidelines. Orlistat prescribing costs for individual GP surgeries will be tracked on an ongoing basis.

A6: Nicola Bowtell - SWPHO

An exploratory study: indoor and outdoor work, socio-economic status and skin cancer

Bowtell N, Hollingworth W, Hawkins J, Kipping R

Background

UK studies exploring the risk of skin cancer in occupational settings are limited.

Aims

To investigate the relationship between indoor and outdoor work, socio-economic status and skin cancer.

Methods

Data on persons (3,096), in the South West, where skin cancer (C43 and C44) was recorded as an underlying or contributory cause of death on the Death Certificate (2001-2010) were extracted. Occupation was classified in terms of exposure using an occupational health led New Zealand model (outdoor, intermittent, and indoor), and socio-economic status (NS-SEC).

Results

Most had worked indoors (68.5%), 19.6% outdoors, 11.9% in intermittent occupations. Only 6.2% of outdoor and 24.9% of intermittent workers, were women (p<0.001). The distribution across the three social economic groups, for both indoor and outdoor groups was similar, with the highest proportion of workers in the intermediate and routine occupations. The intermittent group had a greater proportion of professional occupations (p<0.001). Outdoor workers were more likely to die older (75 years and over) with a non melanoma, and indoor workers die younger (74 years and under) with a melanoma (p<0.001). No significant difference was observed for intermittent workers.

Two outdoor professions (agriculture and unskilled outside) appear to have a significant association with non melanoma skin cancers, whereas two indoor groups (medical and allied professions and managerial) appear to be associated with melanoma (p=0.001). No association was shown for clerical workers and outdoor professionals.

Conclusions

There appears to be different induction mechanisms for melanoma and non melanoma, with outdoor workers more likely to die with a non melanoma and indoor workers (especially those from higher socio economic groups) to die with a melanoma. The study highlights the need for protective measures against all types of solar radiation (occupational and recreational: intermittent and continuous).

B6: Petra Matulkova - HPA

Hepatitis C Retesting Exercise at the Health Protection Agency Laboratory, South West England, 2011

Matulkova P, Oliver I, Jacomelli J, Carrington D, Muir P, Pankhania B

Background

A PCR test is used as part of care pathways to identify active hepatitis C infection and may influence patient referral for antiviral treatment. The Health Protection Agency (HPA) Laboratory in Bristol identified an inhibition problem with its in-house HCV RNA PCR assay, resulting in false negatives. The HPA conducted a look back exercise and offered retesting with a commercial assay to patients previously tested HCV PRC negative.

Aims

To ensure that any false negative individuals that may benefit from specialist referral are identified and managed according to local care pathways.

Methods

The retest cohort included 689 patients testing HCV PCR negative from 01/01/2009 to 06/05/2010. Doctors were invited to send patients samples for retesting. To monitor the outcome of this exercise, they were asked to provide demographic, clinical and risk factor information on their patients, and where relevant state reasons for not retesting.

Results

Among 486 responders 202 patients have been retested. Twenty-seven patients were HCV viral load positive. In 12 of these the clinical history suggested possible treatment failures or relapses. Fifteen were identified as false negatives and they had not been offered specialist referral. Doctors returned 283 request forms without a sample. Reasons stated for not retesting included ‘patient did not attend/loss of follow-up’ (n= 76) and ‘already retested/in treatment’ (n=71). We identified that thirty-seven patients had died in this period. Injecting drug (ID) use was reported as a risk factor in only 57% of cases.

Conclusions

Fifteen identified false negatives were managed according to care pathways. The proportion of ID users was lower than expected, this is likely due to lower response among this hard-to-reach group. The performance comparison of the enhanced in-house assay against the commercial assay was considered adequate. High cohort mortality was identified and is being investigated. Lessons learnt will be reported.

C6: Mathew Illes - SWPHO

Using Travel Time Analysis to Inform the Decision Making Process Regarding the Provision of Late Effects Care Clinics for Childhood Cancer Survivors in Peninsula.

Illes M, Stevens M, Beynon P, Verne J

Background

Survivors of childhood cancer face a lifetime of follow-up.

Aims

This analysis was conducted to inform the decision making process as to the optimal number and geographical positioning of late effects care clinics for childhood cancer survivors in the Peninsula Cancer Network area.

Methods

245 eligible childhood cancer survivors were identified from the South West Childhood Cancer Research Registry (SWCCRR). Questionnaires were sent to identify concerns and issues facing this population, including travel for follow-up. The response to the question regarding how long/far a patient would be willing to travel to attend a proposed late effects clinic was studied.

Five existing acute hospital sites were put forward as the location of the late effects clinics. A total of 31 combinations of sites exist; one site, 10 dual sites, 10 triple sites, five quadruple sites or all possible sites exist at once. A Graphical Information System (MapInfo) was used to analyse the postcode of residence for survivors with a valid postcode (209, 85%). Travel times were calculated to each of the possible 31 combinations of sites to assess the optimal configuration/number of sites.

Results

It was found that 77% of respondents would be willing to travel for up to one hour. A clear, incremental reduction in patients travel times was seen when increasing the number of hospital sites. A three site model would provide sufficient coverage to meet the majority, 68%, of respondent’s expectations to reach a clinic within an hour. One of the four site models provided coverage for 100% of the survivor cohort to access a clinic within one hour. It does not appear advantageous to consider a five site model over a four site model.

Conclusions

Based on a purely geographical/travel time standpoint a four site model is most appropriate. Other factors including cost will play an important role in deciding the final configuration of sites.

D6: Georgie MacArthur - NHS Bristol

Evidence for the effectiveness of opioid substitution treatment in relation to HIV transmission in people who inject drugs: a systematic review and meta-analysis

MacArthur G, Minozzi S, Davoli M, Vickerman P, Hickman M

Background

Injecting drug use remains a major risk factor for the acquisition and transmission of HIV between people who inject drugs (PWID) and the wider community. Worldwide it is estimated that 3 million (range 0.8-6.6 million) PWID may be HIV-positive. Opiate substitution treatment (OST) reduces drug use, injecting risk behaviour and HIV transmission but implementation of OST is variable both around Europe and worldwide. Although evidence indicates that OST is effective in reducing HIV transmission, no meta-analysis has been carried out to obtain a summary estimate of effect and this is required to model the impact of scaling up OST coverage.

Aims

To identify and obtain published and unpublished data to examine the impact of OST in relation to HIV transmission in PWID.

Methods

Medline, EMBASE and PsychINFO were searched up to June 2011 to identify primary studies that have examined the effectiveness of OST in relation to HIV transmission, and prospective studies that have assessed HIV incidence in PWID.

Results

10 primary research studies were identified that investigated the impact of OST on HIV transmission, including a total of 21,859 person-years of follow-up and 625 incident HIV infections. Preliminary random effects meta-analysis provides strong evidence that OST reduces risk of HIV transmission among PWID (OR 0.46, 95% CI 0.30-0.72; p=0.001) although there was significant heterogeneity between studies (I2 68.6%, χ2 19.13, p=0.004). A further 15 cohort studies were identified that examined HIV incidence in PWID; unpublished data relating to OST will be sought and included in further meta-analyses and heterogeneity explored.

Conclusions

Opiate substitution treatment can significantly reduce HIV transmission between PWID. Ensuring sufficient coverage of OST as part of the full range of interventions and services is crucial to reduce the burden of HIV among PWID.