International Festival of Public Health

LT2 - Health Services Research Abstracts

Estimating population prevalence of potential airflow obstruction using different spirometric criteria: a pooled cross-sectional analysis of adults aged 40-95 years in England and Wales

Shaun Scholes, Alison Moody, Jennifer S Mindell

Objectives: Estimation of the burden of chronic obstructive pulmonary disease (COPD) has been hindered by differences in methods, including different spirometric cut-points for impaired lung function. The impact of different definitions on the prevalence of potential airflow obstruction, and its associations with key risk factors, is investigated.

Design: Pooled cross-sectional analysis of the UK Household Longitudinal Survey (Wave 2) and Health Survey for England 2010, including 7879 participants, aged 40-95 years in England and Wales, without diagnosed asthma, and with good-quality spirometry data. Potential airflow obstruction was defined in three different ways: using reported diagnosed COPD; a fixed threshold (FT) forced expiratory volume in 1 second/forced vital capacity (FEV1/FVC) ratio <0.70; and an age-, sex-, height- and ethnic-specific lower limit of normal (LLN).

Results: Across definitions, participants classed with obstructed airflow were more likely to be older, currently smoke, have higher pack-years of smoking, and be in lower socioeconomic groups. The prevalence of airflow obstruction was 2.8% (95% CI 2.3-3.2), 22.2% (21.2-23.2), and 13.1% (12.2-13.9) using diagnosed COPD, FT, and LLN respectively. Differences in prevalence between FT and LLN increased in older age-groups. After adjustment for key risk factors, sex differences in the risk of obstruction was sensitive to the choice of spirometric cut-point, being significantly higher in men than women when using FT but no significant difference using LLN. Strong associations with pack-years of cigarette smoking were found in each definition.

Conclusions: A different distribution of the disease burden exists when applying FT or LLN spirometric cut-points.

Helping Families programme in Salford 2014

Eejay Whitehead

The burgeoning social and economic cost of 'troubled families' to the public sector has led to the development of a nationally led programme to turn around the lives of 120,000 families across the country by 2015. In 2012/13, 479 families were defined in Salford using two or more of the national criteria around crime, absence from school and unemployment. A piece of Joint Strategic Needs Assessment work was designed to integrate intelligence on health and social factors for these families, and explore the impact of health on the problems they faced.

The first phase investigated the health needs using anonymised data from the Salford Integrated Record (a unique record linking primary and secondary care health records for the local population) and social marketing profiling. The profiling showed on average that the families lived in areas with high crime rates and drug/alcohol problems and low levels of qualification and employment. The health data showed levels of obesity and smoking that were significantly higher, and levels of alcohol consumption that were similar, to the Salford average. The estimated life expectancy by deprivation showed levels well below other areas in Salford, and consequently well below England.

This work highlights the challenges of integrating health and social intelligence, but also identifies the opportunities it could bring to support preventative and early intervention approaches. Further work is planned to corroborate health data and identify trigger points at which family problems become significant.

Burden of recurrent unplanned hospital admissions in children and young people in England

Linda Wijlaars, Pia Hardelid, Jenny Woodman, Janice Allister, Ruth Gilbert

Background: Since 1999, unplanned admissions in children and young people have increased by 28%. We aim to describe to what extend recurrent admissions contribute to the burden of admissions in the National Health Service (NHS) in England.

Methods: We analysed all hospital admissions to the NHS in England using Hospital Episode Statistics (HES) from 2009-2011 for children and young people aged 0-24 years. We followed children and young people for two years from their first unplanned admission in 2009 and counted subsequent admissions, time between admissions, and length of stay. We determined the proportion of admissions for injury and in children affected by chronic conditions.

Results: 869,885 children had an unplanned admission in 2009, resulting in a further 939,710 admissions during the 2-year follow-up period. Of the children with an index admission, 32% had recurrent unplanned admissions, accounting for 41% of all admissions in the cohort. Five percent of children and young people had unplanned admissions. 25% of first recurrent admissions occurred within 30 days after discharge from the index admission. 78% of recurrent unplanned admissions were in patients affected by chronic conditions.

Conclusions: This snapshot of hospital flow over a 2-year period shows that recurrent unplanned admissions account for a substantial minority of all unplanned admissions. They recur soon after a prior discharge and occur predominantly in children and young people with chronic conditions. Improved planning, support at hospital discharge for parents caring for children with chronic conditions, and liaison with community services may reduce recurrent admissions.

Rising to the challenge - delivering whole system HIV prevention, treatment and care in a fragmented commissioning world

Sarah Doran, Sarah Stephenson, Jon Dunn, Roz Jones

Background: Recently, there has been a significant change in the commissioning of HIV prevention, treatment and care services. Local authorities now commission HIV prevention services, including HIV testing, and services to care for people living with HIV. NHS England commissions HIV treatment services.

Aims: The Greater Manchester Sexual Health Network needs to ensure that the split responsibility in commissioning HIV services does not impact on patient outcomes.

Participants: The Network covers the 10 local authorities and 12 clinical commissioning groups in Greater Manchester.

Design: A whole system approach to commissioning sexual health and HIV services through co-ordinating the: Commissioning leads meeting - commissioners from local authorities, NHS England and CCGs meet to ensure there is a robust whole system approach to commissioning HIV services HIV prevention group; representatives from local authorities, third sector organisations and Public Health England meet to ensure there is a co-ordinated approach to HIV prevention activitiesHIV clinicians group (in development); HIV clinicians and NHS England local area team commissioner meet to develop clear HIV treatment pathways and discuss changes to HIV services.To link the parts of the system together, the Network hosts a priority action group on HIV with representatives from the Commissioning Leads Group, HIV Clinicians Group, Prevention Group, the Network core team, Public Health England and GPs.

Results: The Network supports the whole HIV pathway and ensures that, through partnership working, high quality services are delivered through seamless integrated pathways. ConclusionsAlthough fragmentation of commissioning poses challenges, these can be managed safely through a Network's collaborative working model.

Economic evaluation of single-fraction versus multiple-fraction palliative radiotherapy for painful bone metastases in breast, lung and prostate cancer

Lucie Collinson, Giorgi Kvizhinadze, Tony Blakely

Objectives: There is little difference in effectiveness between single- and multiple-fraction external beam radiotherapy (SFX-EBRT and MFX-EBRT) as palliative treatment for localised metastatic bone pain. MFX-EBRT is more time consuming and expensive, and SFX-EBRT is associated with a higher retreatment rate and pathological fracture rate. This study estimates the cost-effectiveness of SFX-EBRT- and MFX-EBRT for metastatic bone pain in breast, prostate and lung cancer, for treatments compared to analgesia only and each other, and allowing for varying retreatment and pathological fracture rates.

Methods: A Markov micro-simulation model was used to estimate quality adjusted life years and health system costs. Routine cancer epidemiological and costing data, and effect estimates from RCTs and other studies, were used to parameterize the model.

Results: QALY gains were slightly greater for SFX-EBRT compared to MFX-EBRT across the three cancers, but per patient costs were always much less for SFX-EBRT (NZ$1469 (95% uncertainty interval $1112 to $1886) for lung cancer, $1316 ($810 to $1854) for prostate cancer, and $1344 ($855 to $1846) for breast cancer). Accordingly, SFX-EBRT always dominated MFX-EBRT and was the preferable treatment option. A range of sensitivity analyses about input parameter values (e.g. differences in transition rates to pain relief) and model structure (e.g. not allowing for radiotherapy retreatment in either radiotherapy arms) did not overturn the preference for SFX-EBRT.

Conclusion: For all three cancers, SFX-EBRT was clearly more cost-effective than MFX-EBRT, and equally effective. This study adds to the case for desisting from offering MFX-EBRT to patients with localised metastatic bone pain.

The troubled interface between Clinical Commissioning Groups and Health and Wellbeing Boards, as seen through 'windows'

Lynsey Warwick-Giles, Surindar Dhesi

Background: Since 2010, the NHS and public health systems in England have undergone significant reorganisation. GP-led commissioning organisations (CCGs) have replaced Primary Care Trusts, and Health and Wellbeing Boards (HWBs) were created to set the local strategic direction for health and wellbeing.

Aim: The paper brings together two doctoral research projects; one focussing on CCGs and one on HWBs. Both explore national policy implementation at a local level, and describe the challenges and tensions between the organisations during their development, using Exworthy and Powell's (2004) 'big windows' and 'little windows' framework.

Methods: Fieldwork was conducted for eighteen months from early 2012 in seven longitudinal case study sites comprised of three shadow CCGs in the North of England and four shadow HWBs in the Midlands and North. Data included around 250 hours of CCG / HWB meeting observations; 72 semi-structured interviews with CCG governing body members, HWB members, and other key individuals, with additional environmental health professionals from all English regions. Documents produced by CCGs and HWBs were also read. Data was analysed using Atlas ti.

Results: Implementation of policy focusing on bringing new organisations together is complex. For success, three policy streams need to be aligned (central-local, central-central and local-local). The research identified governance, authorisation requirements, structures, and ways of working as significant issues. These dynamics were influential in the enactment of policy at local levels.

Conclusion: The 'big windows' and 'little windows' concept provides a useful lens to understand the complexities of policy implementation aimed at tackling 'wicked' health and public health issues.

Dying in prison: the public health challenges of growing numbers of prisoners requiring end of life care

Marian Peacock, Mary Turner, Sheila Payne, Katherine Froggatt, Andrew Fletcher, Gill Scott, Bob Gibson

In the last decade the number of older men in UK prisons has doubled, with the sharpest increase (226%) in prisoners aged over 70, resulting in a rise in anticipated deaths in prison. Although such deaths are small in number, the trajectory is sharply upwards. Current end of life care provision in prisons is uneven, and is further complicated by over 40% of older prisoners being sex offenders, many of whom are in prison for the first time in old age due to historic abuse. These shifts in the prison population raise considerable public health challenges. The Parliamentary Justice Committee 2013-2014 recommended increasing the capacity of prisons to provide palliative care, in order to maintain dying prisoners within the prison environment, rather than transfer them to other care settings or grant compassionate release.

This paper will draw on preliminary findings from the 'Both sides of the fence' study, an action research project aimed at developing high quality palliative care in prisons, currently underway in North West England. Focus groups, interviews and informal meetings have been conducted with over 60 participants (including prison healthcare and discipline staff, chaplains, prisoners, community based palliative care staff and others) focused around a single prison. Currently, there is a lack of adequately resourced provision and there are questions as to how appropriate services can best be delivered. We have also identified a range of complex practical and emotional challenges for the differing groups of staff whose work involves this population.

Live Dictation: Seize the opportunity

Omer Ali, Sarah Collins, Afnan Chaudhry, Navneet K Ahluwalia

Background and aims: Clinicians often cite a lack of time and skills as a barrier to effectively communicate health messages to patients. It is common practice to dictate a letter about the consultation that is sent to the GP only. We believe that this unique opportunity can be used to deliver health messages to patients by dictating the letter in the patient's presence. We conducted this study to determine the impact of live dictation on patient experience.

Design: 40 consecutive patients attending a gastroenterology clinic completed an anonymised questionnaire. All patients had experience of both dictation in their presence, live dictation and dictation after they had left, delayed dictation.

Results: 95% of patients felt that live dictation encourages clinicians to use patient-friendly language. All patients in our sample, felt that they had a greater opportunity to ask questions. Patients particularly commented on the transparency and improvement in recall as a result of live dictation.

Conclusion: Live dictation in the presence of the patient is an excellent opportunity to convey and emphasise their message to patients, resulting in better rapport and compliance. Live Dictation complements communication with patients, hence resulting in greater retention by the patient since the dictation gets their undivided attention. We propose piloting live dictation in different clinics with different cohorts of patients.

Data linkage errors and the Hospital Episode Statistics identification (HESID) pseudoanonymisation algorithm: Insights from the Paediatric Intensive Care Audit

Gareth Hagger-Johnson, Tom Fleming, Katie Harron, Rebecca Landy, Ruth Gilbert, Harvey Goldstein, Roger Parslow

Aims: To evaluate data linkage errors in hospital records, following application of the Hospital Episode Statistics pseudoanonymisation (HESID) algorithm: false matches (same HESID, more than one patient) and missed matches (different HESIDs, same patient).

Design: We applied a replication of the HESID algorithm to identifiable patient records, in order to evaluate the extent of data linkage error and its causes. Patient identifiers available included names, date of birth, sex, NHS number, local ID number and postcode.Setting. Paediatric Intensive Care Audit Network (PICANet) database at University of Leeds, covering 33 units (England, Scotland and Wales).

Participants: Patient records from 166,406 admissions (2004 to 2014) with PICANet ID treated as a gold standard.Results. After assigning a HESID, 77,940 (46.8%) of the records were classified as links (readmissions). Provisional results show that false links (N = 169; 0.2% of linked records) are significantly more likely to occur for males, Asian and Black ethnic groups, patients with no ethnic group recorded, and those in areas of high socio-economic deprivation. Missed links (N = 3,380 3.8% of non-linked records) are more likely to occur for younger infants, Black/Other ethnic groups, those with no ethnic group recorded, and those living in deprived areas.

Conclusions: Pseudoanonymisation of patient records always produces a proportion of linkage errors. False and missed matches disproportionately affect certain patient groups, particularly ethnic minorities those in deprived areas. Errors in patient identifiers need to be evaluated and resolved prior to pseudoanonymisation, in order to reduce linkage error.

IRIS (Identification and Referral to Improve Safety) programme in Primary Care in Manchester

Clare McCann, Clare Ronalds, Catherine Cutt

IRIS AbstractAimsIRIS (Identification and Referral to Improve Safety) is a General Practice based Domestic Violence and Abuse (DVA) training, support and referral programme for primary care staff.

Design: The training involves 2 two hour sessions for clinical staff and a separate session for practice administrative and reception staff, in order to achieve IRIS accreditation.The training enables General Practice staff to ask questions about abuse in a safe way and make appropriate referrals. Training is delivered by the Advocate Educator (AE) and a GP clinical lead, referrals are then made from General Practice to the AE for support and case management.

Setting: General Practice in Manchester

Participants: There are currently 16 practices that are IRIS accredited in ManchesterResultsIn the first two years of operation, 116 clinical staff and 136 administrative and reception staff have been trained and the Advocate Educator has received 169 referrals.In a recent National IRIS report Manchester had the lowest number of IRIS accredited practices, but had generated the highest number of referrals, accounting for 22% of the total number of IRIS referrals.Manchester is the only site where IRIS is operating in the North of the UK.

Conclusions: The IRIS project is innovative and provides quality training for all practice staff to raise awareness of and response to DVA, with specific communication skills training for clinical staff and an associated AE service who takes referrals, sees patients in their own practice, carries out risk assessment and case management.