National Health Promotion in Hospitals Audit

Transcription

1 National Health Promotion in Hospitals Audit Acute & Specialist Trusts Final Report

2 Report Structure This report has been divided into three sections: Section 1 Section 1 provides introductory information, an overview of the findings from the audit and organisational survey and recommendations on how acute trusts might improve health promotion practice. Section 2 Section 2 presents the data for all participating sites, enabling benchmarking between organisations. This is followed by a detailed statistical analysis of the NHPHA data. Section 3 Section 3 focuses on the findings from the organisational survey; and where possible relates these findings to the audit data. 2 National Health Promotion in Hospitals Audit

3 Contents Section Introduction Background Standards Development of the NHPHA Participation Pilot Main audit Data collection Figure 1: Geographical Distribution of Acute/ Specialist Trusts Participating in the NHPHA Participating acute/specialist trusts Additional pilot acute sites Main Findings - Overview Smoking Alcohol Obesity Physical activity Demographics Overview of statistical analysis of NHPHA data: Length of stay Age Treatment Specialty Primary Diagnosis Type of Health Promotion Summary of findings from the Organisational Survey: Recommendations Changes to the Audit Structure & Process Steering Group Acknowledgements Section 2 NHPHA Data Presentation & Analysis Case Mix Sex Age Figure 1: Age distribution for whole NHPHA sample Table 1: Descriptive statistics for age Length of stay Figure 2: Length of stay distribution for whole NHPHA sample Table 2: Descriptive statistics for length of stay Treatment Specialty Figure 3: Treatment Specialty distribution per participating site Primary diagnoses Dementia/delirium diagnoses Unconscious for whole spell, delirium, dementia Interpretation of summary main findings Figure 4: Relative percentages and standards for total audit sample Figure 5: Absolute percentages and standards based on total audit sample size Table 3: Summary of standards met for each participating Trust Interpretation of figures and tables on assessment and prevalence of risk factors and health promotion delivered 25 Figures Tables, Health Promotion Figures 6 to 9: Assessment of risk factors Figures 10 to 13: Prevalence of risk factors National Health Promotion in Hospitals Audit 3

4 Health Promotion for risk factors: Figures and Tables Figure 14: Health Promotion Delivered for Smoking Table 4: Forms of health promotion delivered to smokers Table 5: Percentage of total sample of smokers receiving different forms of heath promotion Figure 15: Health Promotion Delivered for Alcohol Misuse Table 6: Health promotion delivered to patients identified as misusing alcohol Table 7: Percentage of total sample of patients misusing alcohol receiving heath promotion Figure 16: Health Promotion Delivered for Obesity 40 Table 8: Health promotion delivered to obese and morbidly obese patients Table 9: Percentage of total sample of obese & morbidly obese patients receiving heath promotion.. 42 Malnutrition At risk of malnutrition Table 10: Health promotion delivered to patients identified as physically inactive Table 11: Percentage of total sample of physically inactive patients receiving heath promotion Detailed Analysis Cluster analysis findings Reliability of Audit Data Section 3 - Organisational Survey Introduction Results Population Health promotion policy and people Figure 1: Health Promotion Policy & People Table 1: Cross tabulation of Health Promotion Policy & People Training Table 2: Number of Trusts reporting that staff have access to training on assessing and delivering health promotion to patients for risk factors Integrated Care Pathways (ICPs) Table 3: Ranking of risk factors for ICP elements.. 61 Figure 2: Do care pathways have elements of assessment for Figure 3: Do care pathways have health promotion/education for Figure 4: Do care pathways have referral to health promotion services for Use of validated assessment tools Table 4: Identification of Trusts using validated alcohol assessment tools Leaflets/written advice Table 5: Number of Trusts with health promotion leaflets available Table 6: Combination of health promotion leaflets available within Trusts Figure 5: How leaflets are accessed Specialists & specialist services Table 7: Healthcare professionals that prescribe NRT Alcohol - Alcohol liaison nurses Community alcohol teams Table 8: Alcohol liaison nurses/workers and referral process to Community Alcohol Teams within Trusts Nutrition/diet and physical activity Health trainers Commissioning Concluding Recommendations References Appendix Appendix 1 - Health Promotion for Malnourishment.. 78 Appendix 2 Analysis Appendix 3 - Inter-rater reliability Appendix 4 Health Promoting Hospitals Standards. 92 Appendix 5 Integrated Care Pathways Appendix 6 Health Promotion Leaflets Photography Credits National Health Promotion in Hospitals Audit

5 Section 1 NHPHA Overview National Health Promotion in Hospitals Audit 5

6

7 Introduction Background The National Health Promotion in Hospitals Audit (NHPHA) is a web-based audit designed to measure the delivery of health promotion to hospitalised patients within all English hospitals. The development of this audit was funded by the National Clinical Audit Programme, Department of Health 1. The purpose of the audit is to provide participating hospital with details of the proportion of their adult hospitalised patients who were assessed for a risk factor (smoking, alcohol, obesity and physical inactivity), had a risk factor, and were delivered health promotion (form: verbal advice, written advice, referral to a specialist or service). Each hospital has already received a summary report of their data which they were able to download immediately following completion of the audit data collection. In this final report we have presented the data from all hospitals in order to enable benchmarking across organisations; and further statistical analysis has been undertaken to explore whether differences between hospitals in assessment of risk factors and delivery of health promotion are related to differences in length of stay/age/gender/ primary diagnosis/treatment specialty of patients. Participating Trusts were also asked to complete an organisational survey to provide baseline information on health promotion services within and available to them. Following on from the findings from the audit and organisational surveys several recommendations have been made on how to develop health promotion services within hospitals. The report is in three sections, this first section provides an introduction and overview of the main findings from the audit and organisational survey. The second section provides a detailed analysis and presentation of the NHPHA data; and the final section discusses the findings from the organisational survey. Standards The standards for 2009 were developed in reference to findings from a Greater Manchester Health Promotion in Hospitals audit (which was the basis for the NHPHA) and the pilot results for the NHPHA. Standards represent what we found was realistically achievable for hospitals to deliver. Standards will be reviewed following the outcomes of the NHPHA and in line with emerging public health policy. Risk Factor Smoking Alcohol Obesity Physical activity Assessment 100% of patients 95% of patients 45% of patients 35% of patients Health promotion delivered 35% of smokers 50% of hazardous/ harmful drinkers 45% of obese patients 45% of physically inactive Development of the NHPHA The NHPHA was developed by a team at Stockport NHS Foundation Trust and an IT company specialising in developing dataentry and reporting solutions for healthcare: Advent IT 2, was commissioned in April 2008 to develop a public website providing information about the project and a secure login area for NHPHA participants to enter audit data and access their summary reports ( The public side of the website was launched in summer 2008; and the securely accessed online audit tool was available in October 2008, at which point the pilot started. The main audit was rolled-out in spring NCAAG/DH_ co.uk/ National Health Promotion in Hospitals Audit 7

8 Figure 1: Geographical Distribution of Acute/ Specialist Trusts Participating in the NHPHA Strategic Health Authority 1 North West (n = 16) 2 West Midlands (n = 6) 3 South West (n = 11) 4 South Central (n = 3) 5 South East Coast (n = 1) 6 London (n = 9) 7 East of England (n = 2) 8 East Midlands (n = 1) Participation Pilot Sixteen acute/specialist hospitals and five mental health trusts participated in the pilot. Twelve of the acute/specialist pilot sites also participated in the main audit. Some of the pilot sites that did not participate in the main audit reported that the pilot had instigated changes to health promotion which would not be in practice in time for the audit, but that they were interested in participating in following years. Main audit Following a first mail out to Chief Executives, Directors of Nursing and Directors of Public Health, we received positive responses from 86 trusts concerning participation in the pilot and/or main audit. We had confirmation of participating in the audit from fifty six acute/specialist hospitals and fifty three successfully completed the audit. Trusts that decided not to participate identified that this was either due to a lack of resources/capacity to undertake the audit work, that the timing meant it wasn t possible to participate in 2009, but they were interested in participating in the future, or that they were undertaking their own data collection and analysis for public health issues. Details of the geographical distribution of participating Trusts are shown in figure 1. Data collection Each Trust collected data from a random sample of 100 hospitalised adult patients discharged alive between January 5th and January 30th 2009 inclusive. On-line data input was available from 1st June 2009 and the original deadline for completing data input was 31st August, but this was extended to 30th September. The organisational Survey was sent out in November with a deadline of mid- December, which was extended to mid-january Yorkshire and The Humber (n = 3) 10 North East (n = 1) 8 National Health Promotion in Hospitals Audit

9 Participating acute/ specialist trusts Barts and The London NHS Trust* Basildon and Thurrock NHS Foundation Trust Basingstoke and North Hampshire NHS Foundation TrustP Blackpool, Fylde and Wyre Hospitals NHS Foundation Trust Bradford Teaching Hospital, St Luke s* Buckinghamshire Hospitals NHS Trust*P ūū Stoke Mandeville Hospital ūū Wycombe Hospital County Durham and Darlington NHS Foundation TrustP Doncaster and Bassetlaw Hospitals NHS Foundation Trust East Lancashire Hospitals Trust*P Epsom and St Helier University Hospitals Foundation Trust,* Gloucestershire Hospitals NHS Foundation Trust ūū Gloucestershire Royal Hospital ūū Cheltenham General HospitalP Great Western Hospitals NHS Foundation Trust,* The Hillingdon Hospital NHS Trust Lancashire Teaching Hospitals NHS Foundation Trust* Liverpool Heart and Chest NHS Trust*P Mid Cheshire Hospitals NHS Trust, Leighton Hospital Mid Yorkshire Hospitals NHS Trust Northamptonshire General Hospital NHS Trust* Northamptonshire Healthcare NHS TrustP Pennine Acute Hospitals NHS Trust* Princess Alexandra Hospital Royal Bolton Hospitals NHS Trust* Royal Brompton and Harefield NHS Trust Royal Devon and Exeter NHS Foundation Trust* Royal Liverpool and Broadgreen University Hospitals NHS Trust Royal Surrey County Hospitals NHS Trust Royal United Hospital Bath NHS Trust* Salford Royal NHS Foundation Trust Salisbury NHS Foundation Trust* Sandwell and West Birmingham Hospitals NHS Trust* ūū City Hospital ūū Sandwell General Hospital Shrewsbury and Telford Hospitals NHS Trust ūū Princess Royal Hospital ūū Royal Shrewsbury Hospital South Devon Healthcare NHS Foundation Trust* St George s Healthcare NHS TrustP Stockport NHS Foundation Trust* Tameside Hospitals NHS Foundation Trust* Taunton and Somerset NHS Foundation Trust The North West London Hospitals NHS Trust* ūū Central Middlesex Hospital ūū Northwick Park Hospital The Queen Elizabeth Hospital King s Lynn NHS Trust The Royal Marsden NHS Foundation TrustP University Hospital of North Staffordshire NHS Trust* University Hospitals Bristol NHS Foundation Trust University Hospitals of Morecambe Bay* ūū Furness General Hospital ūū Royal Lancaster Infirmary ūū Westmorland General Hospital Walsall Hospitals NHS Trust* Weston Area Health Trust, Weston General HospitalP Wirral University Teaching Hospital NHS Foundation TrustP Wrightington, Wigan and Leigh NHS TrustP Yeovil District Hospitals NHS Foundation Trust* Additional pilot acute sites East Kent Hospitals University NHS Foundation Trust Medway NHS Foundation Trust The Newcastle upon Tyne Hospitals NHS Foundation Trust The Whittington Hospital NHS Trust A * indicates that these Trusts responded to the service evaluation survey. P: also participated in the pilot. Some Trusts had more than one hospital site participating separately in the audit, these are indicated by -. National Health Promotion in Hospitals Audit 9

10 Main Findings - Overview Only 1 trust met all the standards for smoking, alcohol misuse, obesity, and physical inactivity; and there was considerable variation between trusts in meeting the standards. The lower standards for obesity and physical inactivity means that more trusts appear to do well in assessing and delivering health promotion for these risk factors in comparison to smoking and alcohol, where in actuality hospitals need to make considerable improvements in the assessment and delivery of health promotion for all risk factors. A summary of the findings for each risk factor in relation to the standards, supported by relevant findings from the organisational survey 3 is provided below. Alcohol 4 Trusts met the standard that 95% of patients should be assessed for alcohol use. Overall 69% of patients in the NHPHA were assessed for alcohol use. On average only 11% of those assessed for alcohol use were found to be hazardous/ harmful/dependent drinkers. This is far lower than the expected prevalence of 20%, which was evident in 27 Trusts 4. One reason for a low identification of alcohol misuse may be the lack of validated alcohol tools readily available for healthcare professionals to use only half of the respondents to the organisational survey reported that their hospitals assessed patients for alcohol use with a validated alcohol tool. Of these 12 hospitals, 7 had the expected prevalence of alcohol misuse. 3 The organisational survey was completed by just over half of the participating Trusts. 4 For 4 Trusts this was on the basis of percentage of assessed patients identified as misusing alcohol, and in a further 23 Trusts on the basis of 95% upper confidence interval (CI) values. Smoking Only 1 Trust met the standard that all patients should be assessed for smoking. On average 81% of patients were assessed for smoking. 25% of those assessed for smoking were found to be smokers; and on average 20% of identified smokers received health promotion. In total 21 Trusts met the standard that 35% of smokers were delivered health promotion. Verbal advice was by far the most common form of health promotion delivered to patients who smoke. On average 45% of patients misusing alcohol received health promotion and 24 Trusts met the standard that 50% were delivered health promotion. The finding that 85% of patients who were clearly dependent drinkers (i.e. prescribed chlordiazepoxide) received at least one form of health promotion for alcohol was reassuring as one would hope that all such patients were provided with assistance to enable them to stop drinking. As with smoking, verbal advice was the most common form of health promotion delivered to all patients misusing alcohol. 10 National Health Promotion in Hospitals Audit

11 Obesity 25 Trusts met the standard that 45% of patients should be assessed for obesity. 40% of the whole NHPHA sample was assessed for obesity and 21% of those assessed were identified as obese, which is in line with the prevalence of obesity in the general population. Overall 22% of obese patients received health promotion; and only 5 Trusts met the standard that 45% of obese patients were delivered health promotion. Verbal advice and referral to a specialist in nutrition/diet were the most common forms of health promotion delivered to patients, followed by the provision of written advice. Physical activity The standard that 35% of patients should be assessed for physical activity was met by 32 Trusts; and it was encouraging to see that 39% of the whole NHPHA sample was assessed for physical activity. However only 17% of those assessed were identified as physically inactive, which is far lower than the prevalence of physical activity in the general population, even when taking into account the number of patients who were not independently mobile; and given that physical activity declines with age and that the NHPHA sample predominantly consists of patients who are middle aged to elderly, one would expect a far higher percentage of physically inactive patients to be identified. It was clear from the organisational survey that hospitals are the least resourced/prepared for assessing physical activity in comparison to the other risk factors. Prompts and tools for the assessment, health promotion/ education, and referral to health promotion services for physical activity were significantly less likely to be incorporated into integrated care pathways in comparison to all other risk factors; and staff were less likely to be able to access training on physical activity assessment and health promotion for physical activity compared to other risk factors. However over 80% of respondents to the survey reported that there were physical activity programmes that they could refer patients to within the community and/or hospital setting. This may explain the relatively good performance on delivering health promotion to physically active patients: On average 46% of patients who were identified as physically inactive did receive health promotion and 26 Trusts did meet the standard that 45% of physically inactive patients were delivered health promotion. However, given the strong likelihood that physical inactivity was not identified in many patients, one can not conclude that health promotion is being adequately delivered to those patients that require it. Verbal advice was the most com- National Health Promotion in Hospitals Audit 11

12 5 P value < accepted as indicating significance. mon form of health promotion delivered to physically inactive patients; and many were also referred to a physiotherapist. Demographics There were slightly more female than male patients in the whole NHPHA sample; and females tended to have slightly longer lengths of stay. This may reflect the fact that females were significantly older than males and that there was a significant correlation between age and length of stay. The age of the NHPHA sample was skewed towards the older age group, with the median age being 65 years. 60% of patients were general surgery patients. Differences in age, length of stay, treatment specialty and diagnosis were taken into account in data analysis. Overview of statistical analysis of NHPHA data: While some of the statistical analysis was exploratory in nature, it was also undertaken to assess several hypotheses which are detailed below, followed by relevant findings (an indication of whether the findings supported the hypotheses is denoted by a and a indicates that the findings did not support the hypothesis): Length of stay Hypothesis: Risk factor assessment is NOT related to length of stay. Because assessments of risk factors are expected to be undertaken on/near admission we do not expect length of stay to have a significant impact on assessments. Significantly 5 shorter length of stay in patients assessed for smoking and alcohol. Assessment of obesity was not related to length of stay. Patients assessed for physical activity had significantly longer lengths of stay than patients not assessed. Hypothesis: Health promotion delivery is positively related to length of stay. Delivery of health promotion for smoking, alcohol and obesity not clearly related to length of stay. but some indication from cluster analysis that obesity health promotion was not delivered to physically inactive obese patients who had short lengths of stay. Delivery of health promotion was positively related to length of stay for physical activity. Age Hypothesis: There will be a normal distribution for age and health promotion delivered - with the young and very old receiving relatively little health promotion. no significant differences overall in age between patients who did and did not receive health promotion for smoking, alcohol and obesity; but, contrary to predictions, cluster analysis indicated that, overall, it was middle-aged patients that were least likely to receive health promotion for all risk factors. Older inactive patients were most likely to receive health promotion for physical activity. Indication that elderly patients with alcohol problems were least likely to receive health promotion for alcohol misuse. Treatment Specialty Hypothesis: General surgery patients are more likely to be assessed for risk factors and delivered health promotion than general medicine patients. There does not appear to be a clear cut relationship between treatment specialty and assessment or health promotion delivery for smoking, alcohol or physical activity; 12 National Health Promotion in Hospitals Audit

13 however, general surgery patients were more likely to be assessed for obesity than general medicine patients. Primary Diagnosis Hypothesis: Patients with primary diagnosis ICD10 codes identified as related to risk factors(6) will be more likely to be assessed for risk factors AND delivered health promotion. Evidence that respiratory patients were highly likely to be assessed for smoking and delivered health promotion for smoking. / Patients with cancer diagnoses had relatively high levels of assessment for all risk factors and were highly likely to be delivered health promotion for smoking, obesity and physical inactivity, but not for alcohol. Except for a group of young cancer patients who smoked and received very little health promotion for smoking and all other risk factors. In addition, cancer patients were most likely to be surgical patients, so we do not know whether high levels of assessment and health promotion are related to having a cancer diagnosis or being surgical patients. In the cluster of patients with the highest level of alcohol related diagnoses nearly all received health promotion for alcohol. Hypothesis: Predict that patients with diagnoses of dementia/delirium were less likely to be assessed risk factors AND delivered health promotion. / Cluster analysis indicated that patients with dementia were less likely to be assessed for smoking and alcohol, but highly likely to be assessed for obesity and physical inactivity. When identified as having risk factors these patients actually received relatively high levels of health promotion for obesity and physical activity; and in line with average for the whole NHPHA sample for smoking and alcohol health promotion. Type of Health Promotion Hypothesis: Health promotion is most likely to be provided as verbal advice. It was evident that health promotion for all risk factors was most likely to be in the form of verbal advice. This is of some concern as medical information has been shown to be least likely to be recalled if it is delivered verbally; and most likely to be recalled if it is provided in written/pictorial material alone or in combination with verbal information yet written advice is provided to relatively few patients, with less than 20% of patients requiring health promotion receiving it in the form of written materials. Summary of findings from the Organisational Survey: Policy & People: Half of responding Trusts had health promotion as part of their stated aims and mission, but this was rarely backed up with a health promotion strategy and personnel. Only 2 Trusts had a mission statement, written strategy and health promotion roles within job descriptions and an additional 5 Trusts had a board champion, health promotion group and coordinator. It therefore appears that the majority of trusts do not have in place a comprehensive management policy and mechanisms that are supportive of health promotion. While only 1 Trust had an identifiable budget for health promotion services and materials, 78% of respondents were in discussions with local commissioners on providing services within the hospital to encourage healthy behaviours. Only 3 Trusts had personnel directly employed by the PCT to assist in delivering health promotion to hospital patients. Training: Over 70% of Trusts have training available to staff on the assessment and delivery of health promotion for smoking National Health Promotion in Hospitals Audit 13

14 and (healthy) diet. Training for alcohol and weight is available to staff in approximately 65% of Trusts; but for physical activity training is only available in half of the Trusts. When asked specifically about the availability of staff training on behaviour change, this was only available in 40% of the Trusts. Integrated care pathways (ICPs): A picture emerged of hospital ICPs incorporating prompts or tools for the assessment of risk factors, but a lack of tools for the next step of delivering health promotion the implications of which are clearly seen in the audit results. Physical activity assessment prompts/ tools and health promotion guidance were far less likely to be incorporated into ICPs in comparison to smoking, alcohol and obesity assessment and health promotion tools. Leaflets: Most Trusts had a good availability of leaflets, in particular for local services, yet we have seen from the audit results that written health promotion materials are infrequently provided to patients that would benefit from receiving them. Smoking cessation: Hospital nurses trained in smoking cessation can help increase smoking cessation rates through the provision of advice, behavioural therapy and/or nicotine replacement therapy (NRT). It was therefore good to observe that 17 Trusts (74%) reported having smoking cessation nurses. However smoking cessation nurses can prescribe NRT at only 6 of these Trusts. At the majority of Trusts it was doctors/medical staff who were able to prescribe NRT (n = 13; 57%); and at 4 Trusts (17%) there were no policies in place for healthcare professionals to prescribe nicotine replacement therapy. Specialist alcohol service: One of the best means of ensuring an optimal alcohol service is delivered is through the employment of alcohol liaison nurses/workers. Alcohol liaison nurses/workers were employed at 14 Trusts (61%). While the majority of Trusts (21 Trusts: 91%) had a local community alcohol team (CAT), only 11 had a standardised system for referring hospital patients to their local CAT. Of concern is the finding that at least 4 Trusts had neither alcohol liaison nurses nor referral processes in place to CATs. Specialist weight, diet, and/or physical activity services: Nutrition specialists/ dietitians were employed in all but one of the Trusts; and in 16 Trusts patients could be referred directly to these specialists solely for the treatment of obesity (i.e. obese patients did not have to have nutrition related comorbidities). 15 Trusts also reported that they have a system in place for referring patients to community weight loss programmes. 19 Trusts reported that there were physical activity programmes that they could refer patients to within the community and/ or hospital setting. 14 National Health Promotion in Hospitals Audit

15 Recommendations We recommend that: All Trusts have at least one nurse per ward trained in smoking cessation techniques. All Trusts undertake the necessary measures to ensure that smoking cessation nurses can prescribe NRT. Mechanisms are put in place to enable patients to receive follow-up smoking cessation support for at least 1 month following discharge. All Trusts incorporate a validated alcohol assessment tool such as AUDIT or CAGE into their ICPs for in-patient care. All Trusts ensure that some basic training is available to ensure healthcare professionals feel confident in using alcohol assessment tools. All Trusts ensure that their staff are aware of community weight loss and physical activity programmes All Trusts establish referral processes to community weight loss and physical activity programmes for suitable patients. Further research is required to identify an appropriate tool for identifying physical inactivity in hospitalised patients. Once validated for use in hospitalised patients the tool should be incorporated into a health promotion ICP. All Trusts should review their ICPs and consider implementing a specific health promotion ICP which has appropriate assessment tools AND guidance on health promotion. Trusts ensure that healthcare professionals are aware of the importance of ensuring verbal advice is supported by written health promotion materials given to patients (either leaflets or healthcare professional providing written tailored information). All Trusts should ensure that appropriate health promotion leaflets are always available in patient areas. A system of having individuals (e.g. health promotion coordinator, alcohol liaison worker, etc) and/or a department (e.g. occupational health) ordering leaflets and volunteers distributing the leaflets to patient areas has been shown to work well in several Trusts. All Trusts ensure that a commitment to delivering health promotion to patients, staff and visitors is explicitly incorporated into their stated aims and mission. Healthcare professionals can not be expected to deliver health promotion to patients if it is not part of the organisation s vision. The strategic and operational development of health promotion should be informed by a specific group which has representatives from both the Acute and Primary Care Trusts and other partner organisations as appropriate. More detailed recommendations are provided within the body of this report in the review of the organisational survey findings. National Health Promotion in Hospitals Audit 15

16 Changes to the Audit Structure & Process In order to assist the interpretation of data from the NHPHA, in the future all participants will be asked to complete the organisational survey prior to the start of data collection for the NHPHA. Because the NHPHA health promotion questions for physical activity have only moderate inter-rater reliability the NHPHA team will investigate whether the wording of the audit pro forma questions need changing; and detailed clarification regarding physical activity data items will be provided in the user help notes. Unfortunately the postcode data (first half only) was insufficient to provide a single Index of Multiple Deprivation Score. We will investigate whether Trusts are willing to provide the first digit of the second half of the postcode to enable analysis into the provision of health promotion to patients from different socioeconomic backgrounds. The paper-based audit pro forma will be redesigned to reflect the logic that the IT system implements so that redundant data are not collected by data collectors. NHPHA standards will be reviewed. Due to funding constraints the audit will be repeated bi-annually, with the next audit data input period to start summer 2011 using data from patients discharged in January This will also have the benefit of providing hospitals with a good time period to make changes to health promotion practice and policy and for these changes to have a measurable impact. Steering Group The steering group contributed greatly to the development and running of the NHPHA. Members of this group are: Angela Bartley, Public Health Lead, Royal Free Hampstead NHS Trust Toni Doyle, Head of Nursing and Patient Experience, Salford PCT and Chair, Greater Manchester Essence of Care (resigned, post changed) Dr Gary A Cook, Consultant in Public Health, Stockport NHS Foundation Trust Dr Ann Hoskins, Deputy Regional Director of Public Health/Acting Director of Children and Young People, NHS North West (resigned, post changed) Dr Melanie J Maxwell, Head of CPRU/Consultant in Public Health Medicine, Wirral University Teaching Hospital NHS Foundation Trust Prof Mike Pearson, Professor of Clinical Evaluation, University of Liverpool Alison Reavy, Clinical Audit Facilitator, Cheshire & Wirral Partnership NHS Foundation Trust (Mental Health Trust) Acknowledgements We would like to thank the Department of Health for their financial support in the development of the NHPHA, Advent IT for their work on developing the on-line tools, Deborah Kenyon for her administrative support and all of the individuals who collected and inputted data for the NHPHA in the first year of this audit. 16 National Health Promotion in Hospitals Audit

17 Section 2 NHPHA Data Presentation & Analysis National Health Promotion in Hospitals Audit 17

18 Case Mix Sex There was a very similar distribution of females to males with 2789 female patients in the whole audit sample (52.62%) and 2511 males (47.38%). Age Figure 1: Age distribution for whole NHPHA sample 35% 30% 25% 20% 15% 10% 5% 0% yrs yrs yrs yrs yrs yrs yrs The age distribution is skewed towards the older population. Female Male Table 1: Descriptive statistics for age Range Mean ± SE Median Males 17 to ± Females 17 to ± Total 17 to ± All data above are in years. There were significant differences in age between the sexes, with females significantly older than males (P < ). 18 National Health Promotion in Hospitals Audit

19 Length of stay Figure 2: Length of stay distribution for whole NHPHA sample 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Up to 2 Weeks 2 to 4 Weeks 4 to 8 Weeks 8 Weeks or More Female Male Table 2: Descriptive statistics for length of stay Range Mean ± SE Median Males 1 to ± Females 1 to ± Total 1 to ± All data above are in days. There was a small significant difference in length of stay between the sexes, with females having longer lengths of stay than males (P < 0.02). National Health Promotion in Hospitals Audit 19

20 Figure 3: Treatment Specialty distribution per participating site General Medicine General Surgery Hospital H53 H40 H44 H42 H50 H5 H32 H43 H22 H23 H24 H4 H49 H15 H29 H12 H16 H9 H17 H30 H34 H8 ALL H2 H31 H33 H48 H10 H11 H7 H26 H14 H21 H25 H28 H35 H47 H36 H52 H37 H6 H18 H45 H46 H13 H20 H38 H41 H39 H19 H51 H27 H1 H3 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 20 National Health Promotion in Hospitals Audit

21 Treatment Specialty As can be seen from Figure 3 the distribution of general medicine and general surgery is not even within each site s audit sample. 60% of the patients in the whole audit were from general medicine and 40% were from general surgery. Primary diagnoses The variation in diagnosis codes is too wide to describe or present graphically. For the purposes of analyses investigating the relationship between assessment and delivery of health promotion and evidence of risk factors, diagnoses have been categorised into codes representing conditions that may be related to risk factors 6, dementia and delirium codes (while these patients should not be excluded from assessment and health promotion, patients condition may make it impractical to undertake these actions), and codes not obviously related to risk factors. Dementia/delirium diagnoses A total of 74 patients (1.4%) in the audit had a primary diagnosis of dementia or delirium. Of these patients 33 (i.e. 44.6%) were not assessed for smoking, alcohol, and/or physical activity due 7 to their dementia/delirium. Unconscious for whole spell, delirium, dementia A further 34 patients with delirium and 162 patients with dementia were identified as not being assessed for smoking/alcohol use/ physical activity because of these conditions, and 12 patients were not assessed due to being unconscious for the whole spell. 6 ICD10 codes for Alcohol, Cancer, Cardiovascular, Dementia, Gastrointestinal, Liver, Spontaneous abortion, Obesity, Pancreas, Respiratory, Vascular, and a grouping termed smoking which includes all ICD10 diagnosis codes identified by the Surgeon General as related to smoking; reference: Diseases and Adverse Health Effects Related to Smoking: 2004 Surgeon General s Report on the Health Consequences of Smoking and July 1, 2005 MMWR on Annual Smoking-Attributable Mortality, Years of Potential Life Lost, and Productivity Losses in the US, gov/library/ smokingconsequences/ 7 an option for not assessing obesity due to dementia or delirium was not provided in the audit as it does not require verbal communication to be determined National Health Promotion in Hospitals Audit 21

22 8 The 95% CI indicates the range of values within which one can be 95% confident that the true value for the whole adult in-patient population lies for each Trust. 9 Please contact support@ nhphaudit.org if you require the details of your Trust s code Interpretation of summary main findings Figure 4 provides details of the overall percentage ± 95% confidence intervals (CI 8 ) of patients in the whole audit sample assessed and delivered health promotion for each risk factor, alongside the corresponding standard set for Information for each risk factor is grouped together, starting with assessed for, prevalence of risk factor ( evidence of. ) and health promotion delivered for.. Data in assessed for is always out of the whole sample size of 5,300 patients, but the sample size becomes smaller depending on how many patients were assessed (the total assessed provides the sample size value for evidence of. ) and how many patients were found to have the risk factor that was being assessed (the total with evidence of a risk factor provides the sample size value for health promotion delivered for. ). Each risk factor has its own colour throughout the figures in the report: smoking is represented by orange, alcohol by green, obesity by red and physical activity by blue. The standards are represented as grey columns. In figure 5 data have been presented along the same lines as in figure 4 but the data is in absolute terms, with the denominator for assessed for, evidence of. and health promotion delivered for. always remaining at 5,300 patients, and the standards adjusted accordingly. Finally, in Table 3 each participating audit site has been listed in chronological order of its code (staring from H1 to H53 9 ) with details of whether or not the audit standard was met for assessing and delivering health promotion for each risk factor. A red tick ( ) indicates that the standard was met on the basis of percentage, a black tick ( ) indicates that the standard was met on the basis that the standard value fell within the 95% CI; and an N means that the standard was not met. For example, if we look at the standard of 100% patients assessed for smoking, a would be given if all 100 of the patients assessed were asked about their smoking, but a would be given if, for instance, 99 of 100 patients had been assessed because the 95% CI would be between 94.6% and 100.0%. The final row in Table 1 provides details of the total number of Trusts that met the standard on the basis of percentage assessed or delivered health promotion and in parenthesis the number meeting the standard on the basis of their 95% CI value. 22 National Health Promotion in Hospitals Audit

23 Figure 4: Relative percentages and standards for total audit sample NHPHA Standard Figure 5: Absolute percentages and standards based on total audit sample size Proportional Standard National Health Promotion in Hospitals Audit 23

24 Table 3: Summary of standards met for each participating Trust Assess smoking HP smoking Assess alcohol HP Alcohol Assess physical HP physical Assess obesity HP obesity misuse misuse activity inactivity All N N N N N N H1 N N N N N N N H2 N N N N N N H3 N N N N N N H4 N N N N N H5 N N N N N N H6 N N N N N N H7 N N N H8 N N N N N N N H9 N N N N N N N H10 N N N N N H11 N N N N H12 N N N N N N N H13 N N N N N N N N H14 N N N N N N H15 N N N N N H16 N N N N N N H17 N N N N H18 N N N N N H19 N N N N N N N H20 N N N N N N N N H21 N N H22 N N H23 N N N N N H24 N N N N N N N N H25 N N N N N N H26 N N N H27 N N N N N N N N H28 N N N N N H29 N N N N N N H30 N N N H31 N N N N N N H32 N N N N N N N H33 N N N H34 N N N N N N H35 N N N N N N N H36 N N N N N N N N H37 N N N N N H38 N N N N N N H39 N N N N N N H40 N N N N N N N H41 N N N N N H42 N N N N H43 N N N N H44 N N H45 N N N N N N H46 N N N N N N N H47 N N N N N H48 N N N N N H49 N N N N N N N H50 N N N N N N H51 N N N N N H52 N N N N N H53 Total 1 12 (9) 1 (3) (4) 4 (1) 26 (6) 24 (2) 24 National Health Promotion in Hospitals Audit

25 Interpretation of figures and tables on assessment and prevalence of risk factors and health promotion delivered Figures Data in all figures are presented in rank order from top to bottom indicating the highest to lowest performing Trusts respectively. All data shows the percentage (solid bar) and 95% CI (line dissecting the top of the bar). Figures depicting the assessment (figures 6 to 9) and delivery of health promotion (fig ures 14 to 16, and19) for smoking, alcohol, obesity and physical activity all have a solid line illustrating the NHPHA standard. For the purposes of benchmarking, information on the upper and lower quartiles for assessment and delivery of health promotion for each risk factor is represented by a dashed and long dashed line respectively. Those hospitals with data on or above the upper quartile line are those with values in the highest 25% of data and those with data on or below the lower quartile line are within the lowest 25% of data. In figures 16 and 19 the lower quartiles were 0% and as such the long dashed line has not been drawn as this would interfere with the legibility of the figure. Tables, Health Promotion Each table depicts the percentage of patients identified as having a risk factor who were delivered each possible form of health promotion. Data are presented in order of hospital code. A traffic light system has been used to indicate those Trusts whose data were in the upper quartile: green, interquartile range: amber, and lower quartile: red for each type of health promotion delivered. Please note that each patient could receive more than one type of health promotion. Tables detailing the descriptive statistics for each form of health promotion for all NHPH audit data combined are also provided. National Health Promotion in Hospitals Audit 25

26 Figures 6 to 9: Assessment of risk factors Figure 6: Assessed for Smoking Standard Upper Quartile Lower Quartile Hospitals 26 National Health Promotion in Hospitals Audit Percent of patients in audit assessed for Smoking

27 Figure 7: Assessed for Alcohol Misuse Standard Upper Quartile Lower Quartile Hospitals Percent of patients in audit assessed for Alcohol Misuse National Health Promotion in Hospitals Audit 27

28 Figure 8: Assessed for Obesity Standard Upper Quartile Lower Quartile Hospitals 28 National Health Promotion in Hospitals Audit Percent of patients in audit assessed for Obesity

29 Figure 9: Assessed for Physical Activity Standard Upper Quartile Lower Quartile Hospitals Percent of patients in audit assessed for Physical Activity National Health Promotion in Hospitals Audit 29

30 Figures 10 to 13: Prevalence of risk factors Figure 10: Evidence of Smoking Expected Prevalence Hospitals 30 National Health Promotion in Hospitals Audit Percent based on sample assessed for Smoking

31 Figure 11: Evidence of Alcohol Misuse Expected Prevalence Hospitals Percent based on sample assessed for Alcohol Misuse National Health Promotion in Hospitals Audit 31

32 Figure 12: Evidence of Obesity Expected Prevalence Hospitals 32 National Health Promotion in Hospitals Audit Percent based on sample assessed for Obesity

33 Figure 13: Evidence of Physial Inactivity Expected Prevalence Hospitals Percent based on sample assessed for Physical Activity National Health Promotion in Hospitals Audit 33

34 Health Promotion for risk factors: Figures and Tables Figure 14: Health Promotion Delivered for Smoking Standard Upper Quartile Lower Quartile Hospitals 34 National Health Promotion in Hospitals Audit Percent based on sample with evidence of Smoking

35 Table 4: Forms of health promotion delivered to smokers Hospital verbal advice written advice NRT prescribed Referred to smoking Referred to specialist Advised to contact cessation nurse respiratory nurse GP/practice nurse H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H National Health Promotion in Hospitals Audit 35

36 Only one Trust: H26 is within the upper quartile for all forms of health promotion delivered to smokers. Table 5: Percentage of total sample of smokers receiving different forms of heath promotion Health Promotion Mean ± SE 95% CI Lower quartile Upper quartile Verbal advice ± to Written advice 6.00 ± to NRT prescribed 5.98 ± to Referred to smoking cessation nurse 6.38 ± to Referred to specialist respiratory nurse 2.84 ± to Advised to contact GP 4.31 ± to All values in the table above are in percentages A total of 1071 people were identified as smokers, of whom only 215 received health promotion. Verbal advice was the most common form of health promotion delivered to patients who smoke. Review of the 95%CIs indicates that there was no significant difference in the use of any of the other forms of health promotion over another (written advice, NRT, referrals and advice to contact GP or practice nurse). 36 National Health Promotion in Hospitals Audit

37 Figure 15: Health Promotion Delivered for Alcohol Misuse Standard Upper Quartile Lower Quartile Hospitals Percent based on sample with evidence of Alcohol Misuse National Health Promotion in Hospitals Audit 37

38 Table 6: Health promotion delivered to patients identified as misusing alcohol Hospital Verbal advice Written advice Advised Advised contact Referred to Given CAT Contact contact GP alcohol service alcohol service referred to HALW H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H National Health Promotion in Hospitals Audit

39 Data for H33, H37, and H42 are within the upper quartile for all forms of health promotion delivered to smokers. Table 7: Percentage of total sample of patients misusing alcohol receiving heath promotion Health Promotion Mean ± SE 95% CI Lower quartile Upper quartile Verbal advice ± to (59.71 to 77.23)a Written advice 9.97 ± to14.79 (14.46 to 30.15)a Advised to contact GP/ practice nurse 9.29 ± to13.69 (10.86 to 25.36)a Advised to contact alcohol service ± to (27.49 to 45.65)a Given Community Alcohol Team (CAT) Contact 8.85 ± to (19.70 to 36.66)a Referred to alcohol service ± to (33.94 to 52.63)a Referred to Hospital Alcohol Liaison Worker 8.70 ± to (18.94 to 35.74)a All values in the table above are in percentages. a: 95% CIs for percentage of patients delivered health promotion who were also prescribed chlordiazepoxide (i.e. dependent drinkers). A total of 406 people were identified as hazardous or harmful drinkers, and of these, 116 (29%) were prescribed chlordiazepoxide, a clear indicator of alcohol dependency (although some of these patients were identified as hazardous, which is very likely to be an underestimation of an individual s alcohol intake). In total, 182 patients identified as misusing alcohol were delivered health promotion. As with smoking, verbal advice was the most common form of health promotion delivered to all patients identified as either hazardous or harmful drinkers (the latter category was also used to include dependent drinkers). Review of the 95% CIs indicates that there were no significant differences in the choice of any of the other forms of health promotion over another. It is reassuring to observe that dependent drinkers had relatively high levels of health promotion delivered, for example, between 60% and 77% were given verbal advice and between 34% and 53% were referred to an alcohol service. While a separate standard was not set for these drinkers, given the seriousness of alcohol dependency, all such patients (i.e. 100%) should receive referrals to a specialist alcohol service/worker; and a separate analysis revealed that 98 (i.e. 84%, 95% CI = 77% to 91%) did receive at least one form of health promotion for alcohol. National Health Promotion in Hospitals Audit 39

40 Figure 16: Health Promotion Delivered for Obesity Standard Upper Quartile Hospitals Percent based on sample with evidence of Obesity 40 National Health Promotion in Hospitals Audit

41 Table 8: Health promotion delivered to obese and morbidly obese patients Hospital Verbal advice Written advice Advised to contact GP Advised to join weight loss programme referred to GP Referred to specialist in nutrition/diet Referred to weight loss programme Currently on weight loss programme Referred to hospital gym H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H H Grey cells have been used as a visual aid to identify when patients have received some health promotion; given the very small percentage of patients receiving health promotion in these columns, both the upper and lower quartiles were at zero, and hence the traffic light system could not be used. Referred to community organisation National Health Promotion in Hospitals Audit 41

42 Table 9: Percentage of total sample of obese and morbidly obese patients receiving heath promotion Health Promotion Mean ± SE 95% CI Lower quartile Upper quartile Verbal advice ± to Written advice 5.78 ± to Advised to contact GP / practice nurse 2.63 ± to Advised to join weight loss programme 2.53 ± to Referred to GP/ practice nurse 0.89 ± to Referred to specialist in nutrition/diet 7.83 ± to Referred to weight loss programme 2.01 ± to Referred to hospital gym 0.29 ± to Referred to community organisation 0.57 ± to All values in the table above are in percentages 437 patients were identified as obese and 98 received health promotion. Yet again, verbal advice was the most common form of health promotion delivered to patients, provided to significantly more patients than any other form of health promotion except for referral to a specialist in nutrition/diet, which was the second most delivered form of health promotion (the latter category is a combination of referral to a nutritionist, other nutritionist and/or dietitian), followed by the provision of written advice. 42 National Health Promotion in Hospitals Audit

43 Malnutrition Out of a sample size of 5300, only 94 patients were identified as malnourished (1.8%); and 19 acute/specialist trusts had no patients with malnourishment in their samples. In the 34 acute/specialist trusts that had malnourished patients within their audit samples, 64% received nutritional support, and 53% were referred to a dietician or nutrition specialist. Ideally we would hope all malnourished patients have contact with a dietitian/nutrition specialist; however, with an estimated 25% prevalence of malnutrition in hospital in-patient populations 10, there may not be an appropriate number of dietitians/nutrition specialists within hospitals to be able to see all malnourished patients; and hospitals may have in place protocols/ pathways that ensure other ward staff can appropriately address the needs of malnourished patients - NICE guidance stipulates that all hospitals should ensure that patients who are either at risk of or have malnutrition should have access to a dietitian if necessary. The relatively small number of dietitians in most hospitals, means that some of their roles must be delegated to other ward staff. The dietitians therefore need to develop hospital protocols and care pathways on nutrition support, and to participate in the nutritional education of the entire clinical workforce. The aim should be that all hospital healthcare professionals should understand the importance of nutrition in patient care and the means available to provide it safely and effectively. cg032fullguideline.pdf P.57, Section3.3.2 The organisational survey, requested details on the number of dietitians/nutrition specialists available within each Trust. Given that the median value is 6.42 WTE for individual sites within Trusts, and the range is from zero to a maximum of 36, it would appear that the majority of Trusts would find it difficult to ensure that all malnourished patients are seen by a dietitian/nutritionist. Figure 17 provides details of the number of patients with malnourishment who were provided with nutritional support and/or delivered dietary health promotion. Data are ordered from left to right in order of the highest number of malnourished patients receiving nutrition support and health promotion. A measure of health promotion is derived from data collected under the weight and nutrition section of the audit for verbal advice, written advice, referred to dietician/ nutrition specialist 11, referred to GP, referred to community organisation, and/or advised to contact GP/Practice nurse. It is not necessarily always appropriate for patients with malnourishment to receive this health promotion and so data should not be used as an indicator of whether or not patients received appropriate health promotion care. Table 1, Appendix 1 provides a breakdown of the number of malnourished patients provided with each potential form of health promotion. At risk of malnutrition 284 patients were identified as at risk of malnourishment (5.4% of the whole sample) within 41 of the 53 acute/specialist trusts (see Figure 18). 54% of these patients received nutritional support, and 39% were referred to a dietician or nutrition specialist. Health Promotion was delivered to 43% of patients. A breakdown of the different forms of health promotion is provided in Table 2, Appendix Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M et al. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the malnutrition universal screening tool ( MUST ) for adults. British Journal of Nutrition 2004, 92(5): While we would expect this to be part of normal medical care for malnourished patients, we have also included referral to dieticians or nutrition specialists as health promotion as patients may receive general dietary health promotion from these specialists. National Health Promotion in Hospitals Audit 43

44 Figure 17: Number of malnourished patients receiving nutritional support and/or health promotion for nutrition Hospital Yes HP - Yes NS Yes HP - NS Inappropriate Yes HP - No NS No HP - Yes NS No HP - NS Inappropriate No HP - No NS HP: Health Promotion, NS: Nutritional Support Number 44 National Health Promotion in Hospitals Audit

45 Figure 18: Number of patients at risk of malnutrition who received nutritional support and/or health promotion for nutrition Hospital Yes HP - Yes NS Yes HP - NS Inappropriate Yes HP - No NS No HP - Yes NS No HP - NS Inappropriate No HP - No NS HP: Health Promotion, NS: Nutritional Support Number National Health Promotion in Hospitals Audit 45

46 Table 10: Health promotion delivered to patients identified as physically inactive Hospital Verbal advice Written advice Referred to Referred to Referred to physical Physiotherapist rehabilitation Specialist activity programme H H H H H H H H H9 H H H H H H H16 H H H H H H H H H25 H H H H H H H H H34 H35 H H H H H H H H H H H H H H H H H H Blank rows in the table above indicate that there were no patients identified as physically inactive in these Trusts and as such no health promotion was expected to be delivered to patients. 46 National Health Promotion in Hospitals Audit

47 One Trust: (H53) is within the upper quartile for all forms of health promotion delivered to physically inactive patients. Table 11: Percentage of total sample of physically inactive patients receiving health promotion Health Promotion Mean ± SE 95% CI Lower quartile Upper quartile Verbal advice ± to Written advice ± to Referred to physiotherapist ± to Referred to rehabilitation specialist Referred to physical activity programme All values in the table above are in percentages ± to ± to patients were identified as physically inactive and 162 were delivered health promotion for physical activity. Verbal advice and referral to a physiotherapist were the most common forms of health promotion delivered to physically inactive patients. While referral to a physiotherapist has been included under health promotion, physiotherapists may only provide treatment for the presenting physical condition, and not provide patients with additional information/support to enable them to lead physically active lives. There were no significant differences in the provision of written advice, referral to a rehabilitation specialist or physical activity programme to physically inactive patients. National Health Promotion in Hospitals Audit 47

48 Figure 19: Health promotion delivered for Physical Activity Standard Upper Quartile Hospitals Percent based on sample with evidence of Physical Inactivity 48 National Health Promotion in Hospitals Audit

49 Detailed Analysis Cluster 5 Our initial plan was to undertake regression analysis in order to predict membership in the following groups: assessed, evidence, and health promotion delivered for each risk factor on the basis of patients length of stay, age, primary diagnosis and treatment specialty; however these variables were not sufficient enough to significantly and meaningfully predict membership in those groups. Instead, cluster analysis was undertaken as this is a powerful exploratory tool to examine differences in populations by sorting data into associated groups (clusters) that use meaningful variables such as assessed for smoking or length of stay to describe variance within populations. Multiple analyses were undertaken and are described in Appendix 2. Cluster analysis findings The main findings of interest from the cluster analysis are provided below and a detailed account of the findings from the main cluster analysis is provided in Appendix 2. A cluster analysis utilising all the data from the audit resulted in six clusters. Some hospitals contributed more data (cases) to a specific cluster than others (see Table 2, Appendix 2 for details). A review of the variables which contribute to these clusters revealed that: Elderly patients (mean 77 years) who also had long lengths of stay (mean 38 days) and high levels of dementia (cluster 5) had relatively 12 low levels of assessment for smoking and alcohol, but high levels of assessment for obesity and physical activity. Of those that were assessed for risk factors, they all had relatively low levels of evidence of any of the risk factors; but of those that were identified as requiring health promotion, a relatively high proportion received health promotion for obesity and physical activity; and health promotion for smoking and alcohol was in line with the rest of the NHPHA population. The hospitals that contributed the most data to cluster 5 were, in order of the most number of patients contributing data H1, H46, H14, H18, H6, H35, H15, H51, H22, H34, H37, and H10. Another cluster with elderly patients (mean 71 years) had lengths of stay on average just over a week, with a high prevalence of respiratory conditions and consisted almost exclusively of patients within general medicine. This cluster had relatively high levels of assessment for smoking, alcohol use and physical activity, but low levels of obesity assessment (cluster 3). These patients had a relatively low prevalence of smoking, alcohol misuse and obesity; and relatively high levels of health promotion for smoking, alcohol misuse and physical inactivity, but low levels of health promotion for obesity. The hospitals that contributed the most data to cluster 3 were H45, H33, H39, H18, H26, H47, H52, H1, H7, H41, H3, and H36. One of the clusters which consists of younger, middle-aged patients (mean age of 54 years) with short lengths of stay (mean 3 days) had very high levels of assessment for smoking and alcohol misuse (cluster 1); relatively high levels of evidence of smoking but low levels of health promotion for smoking; relatively high levels of alcohol misuse and obesity and high levels of health promotion for these two risk factors. The hospitals that contributed the most data to cluster 1 were H26, H32, H50, H41, H21, H4, H33, H35, H8, H18, H5, H9, and H19. Another cluster (cluster 2) has middle-aged patients (mean age of 55 years) with very short lengths of stay (mean 1 day). These patients have low levels of assessment for physical activity but relatively high levels of physical inactivity. Patients in this cluster also had high levels of smoking and obesity; but health promotion for all risk factors was delivered infrequently to these patients. The hospitals that contributed the most data to cluster 2 were H42, H25, H17, H44, H11, H19, H3, H9, H10, H13, H43, and H30. Cluster 3 Cluster 1 Cluster 2 12 In reporting of the cluster analysis relatively is always in relationship to the values for the total audit data unless otherwise stated. National Health Promotion in Hospitals Audit 49

50 Cluster 6 Cluster 4 Cluster 6 is mainly differentiated by the finding that patients in this cluster received no assessments for smoking and alcohol misuse and below-average assessments for obesity and physical inactivity. These patients were on the cusp of elderly (mean age of 65 years) and their mean length of stay was 6 days. Not surprisingly, this group had low levels of evidence for alcohol misuse (0%) and obesity (16%) but, relative to the whole audit population, average levels of physical inactivity (15%). Health promotion for obesity was very poor (5%) but relatively reasonable for physical activity (44%). The hospitals that contributed the most data to cluster 6 were H40, H37, H46, H48, H12, H15, H36, H45, H50, H43, H5, H24, and H34. Cluster 4 consisted of surgery patients with a high prevalence of cancer diagnosis (63%), unremarkable lengths of stay (mean 9 days) and mainly from the upper end of middle age (mean 64 years). This cluster was characterised as having relatively high levels of assessment for all risk factors, relatively low prevalence of smoking, alcohol issues and physical inactivity but 23% were obese (i.e. slightly higher than NHPHA population overall). Relatively high levels of health promotion were delivered for smoking, obesity and physical activity but not for alcohol misuse. The hospitals that contributed the most data to cluster 4 were H53, H22, H30, H44, H15, H16, H50, H14, H33, H8, H45, and H47. The cluster analysis for the whole NHPHA population indicates that Evidence of smoking, alcohol misuse, obesity, and physical inactivity was found more often in middle-aged than elderly patients. Despite middle-aged patients being more likely to need health promotion for all risk factors, they were less likely to receive it. Cluster analysis of all patients identified as smokers indicates that: Smokers tended to be younger, with shorter lengths of stay than non-smokers (all those assessed for smoking and found to be nonsmokers). Evidence of alcohol misuse in smokers was double that compared to the entire NHPHA population average (24.7% versus 11.2% respectively) and quadrupled compared to the population of nonsmokers (6.1%). Smokers were less likely to be obese than nonsmokers (16.0% versus 22.5% respectively). There were two groups of younger smokers those with a relatively high level of cancer diagnosis but a very short average length of stay (1 day) who had far less health promotion for smoking, alcohol misuse, obesity and especially physical inactivity compared to other smokers; and those with an average length of stay of 10 days, who received health promotion for smoking in line with the average for all smokers, and very high levels of targeted health promotion for their alcohol misuse (92.5%) which was a particular issue within this cluster. Overall, smokers were more likely to receive health promotion for their alcohol misuse (51.0%) than non-smokers (32.1%), but this primarily reflects the successful identification of alcohol misuse in one particular cluster of male smokers. 50 National Health Promotion in Hospitals Audit

51 Cluster analysis of all patients identified as obese indicates that: Cluster analysis of all patients identified as misusing alcohol indicates that: Most patients with evidence of alcohol misuse are male (76%). Alcohol misusers were younger than non-misusers overall. Alcohol-related primary diagnoses were found solely in two Clusters, which also contained the youngest patients (mean age of 44 years and 42 years in clusters 1 and 5 respectively). Only 41% of patients in cluster 1 received health promotion for alcohol misuse but nearly all in cluster 5 received health promotion (97%) More alcohol misusers were found in general medicine than general surgery. Assessment of smoking in patients with alcohol misuse was slightly lower overall compared to non-misusers despite the finding that alcohol misusers were more likely to be smokers (56% compared to 20% in patients who do not misuse alcohol). Younger patients who misused alcohol also had a higher occurrence of physical inactivity (27-33% compared to 17% of non-misusers). Younger patients who misused alcohol were less likely to receive health promotion for physical inactivity and obesity than older patients who misuse alcohol. Clusters with the two oldest populations were least likely to receive health promotion for misusing alcohol - Hospitals need to improve the delivery of health promotion for alcohol misuse to older patients who misuse alcohol. Obese cardiovascular patients received fewer assessments for physical activity than cardiovascular patients who smoke; yet evidence of physical inactivity was higher within this cluster of obese cardiovascular patients than within most other obese patients and especially non-obese patients. Surprisingly, obese patients were slightly less likely to be assessed for physical inactivity (38%) than non-obese patients (44%); but (not surprisingly) obese patients were far more likely to be physically inactive than non-obese patients. Overall, evidence of obesity was found within surgery patients more than general medicine, which reflected the higher number of assessments of obesity in surgery patients. Obese patients had a lower prevalence of smoking and alcohol misuse in comparison to the population average. Obese patients who were assessed and identified as being physically inactive were more likely to receive health promotion for physical inactivity in all clusters than obese patients were likely to receive health promotion for obesity per se. While health promotion for physical activity directly relates to weight and is therefore very appropriate for obese patients, the issue is the content and quality of National Health Promotion in Hospitals Audit 51

52 the health promotion if obese physically inactive patients are only receiving recommendations to undertake the minimum recommended levels of physical activity, this will have little impact on obesity as recommendations for weight loss and obesity prevention in adults is daily participation in minutes of at least moderate intensity physical activity 13. Within the audit, user help notes identified that health promotion for obesity included advice on how to lose weight (which should include details on physical activity and/or diet). Further research is needed to understand whether the physical activity health promotion adequately addresses obesity issues; and to determine whether health promotion options for obesity and physical activity within the audit should (therefore) be combined into one group when looking at health promotion for obesity. Cluster analysis of all patients identified as physically inactive indicates that: No gender differences were evident overall between physically inactive and physically active patients. There was an even distribution of specific diagnoses across the clusters in inactive patients except for one cluster (cluster 4) which was almost entirely composed of cardiovascular patients. Assessment of obesity was relatively high within this cluster; and health promotion for all risk factors was delivered to a high proportion of these patients. Physically inactive patients were more likely overall to receive assessments for obesity (52% assessed compared to 42% of physically inactive); but this was primarily due to high levels of assessment within three clusters (clusters 1, 3 & 4) the common factor between these clusters was relatively short lengths of stay in hospital. Overall, physically inactive patients had a higher occurrence of obesity compared to physically active patients. The relationship between obesity and physical activity assessments is complicated if identified as obese, patients appear less likely to be assessed for physical activity; but if patients were identified as physically inactive assessments of weight status were made clearly further investigation is required to understand why this is the case. The three clusters with the highest evidence of obesity (clusters 1, 3 and 5) had the lowest levels of health promotion delivered for obesity within the physically inactive clusters. These clusters were characterised by having the shortest lengths of stay (mean 1, 3 and 4 days respectively). Whether or not the short length of stay or some other reason(s) is an explanation for the lack of health promotion for obesity requires further research. Reliability of Audit Data Data for 10 cases at each audit site were double data collected for the purposes of assessing inter-rater reliability. This provides an indication of the precision of audit data, indicating how consistently data items are being collected. Low levels of inter-rater reliability may be indicative of poor data collection by one or both of the data collectors and/or that question were not understood by data collectors (and may therefore require further clarification from the NHPHA team). A detailed analysis and interpretation of inter-rater reliability is provided in Appendix 3. Inter-rater reliability was very good for all smoking and alcohol data items, good for most weight and nutrition data items and all physical activity assessment items. 13 Saris, W.H.M., Blair, S.N., Van Baak, M.A., et al. How much physical activity is enough to prevent unhealthy weight gain? Outcome of the IASO 1st stock conference consensus statement. Obesity Reviews ; 4: National Health Promotion in Hospitals Audit

53 Reliability was poor for some weight items: waist measurement taken, referred to nutritionist, and referred to other nutritionist specialist. When the latter two data items are collapsed into one category along with referred to dietitian, inter-rater reliability is much improved; accordingly data are presented for all 3 data items combined into a new category of referral to specialist in diet/nutrition ; and in future there will be only one question relating to referral to specialists in diet/nutrition. There is little impact on the overall meaning of the audit results from a low inter-rater reliability on waist measurement taken, and should not be interpreted as casting doubt on the audit findings concerning weight. For physical activity all health promotion data items showed only a moderate level of reliability (see Table 4, Appendix 3 for further details). A detailed exploration of where data differences between raters exist revealed that discrepancies occurred in only fourteen of the participating sites. This raises some concerns about the precision of the health promotion data and the identification of individuals requiring health promotion in these sites, and in particular within one site (H51) which had differences in seven out of ten data items collected on physical activity health promotion. These fourteen audit sites should view their results on physical activity with some caution (see Appendix 3 for further details). The NHPHA team will investigate further why health promotion questions for physical activity have only moderate inter-rater reliability and, if appropriate, will make changes to the wording of the questions and/or help notes associated with each question to ensure a higher level of reliability. National Health Promotion in Hospitals Audit 53

54

55 Section 3 Organisational Survey National Health Promotion in Hospitals Audit 55

56 14 Please contact nhphaudit.org if you do not have a copy of your survey and we will send you an electronic copy of the data file for your Trust. 15 Survey questions were based on face validity and informed by the Greater Manchester Acute Trusts Survey on Promoting Health and Tackling Health Inequalities nwph.net/agmpcts/ Resources/Health%20 Promoting%20Hospitals/ Reports/Acute%20 Trusts%20Baseline%20 Survey%20Report%20 July% pdf and standards set by the International Health Promoting Hospitals project/network www. healthpromotinghospitals. org and (see Appendix 4 for standards). 16 Several Trusts have reported that they made changes to health promotion policy/people/ tools following a review of their automated NHPHA summary results. Introduction The purpose of the organisational survey was to provide a baseline measure of health promotion services within and available to acute/specialist Trusts (for ease of reading these will be referred to as Trusts from here on). This information can be used to inform guidance on developing health promotion services within individual Trusts. It is recommended that if you completed the organisational survey that you read the report with reference to you own Trust s survey data 14. Questions 15 were asked on The population that each Trust served Health promotion policy and people Training for hospital staff on health promotion For each risk factor (smoking, alcohol, weight, diet and nutrition, and physical activity) information was gathered on ūū Integrated care pathways for patient assessment & health promotion ūū Access to specialists/specialist services ūū Availability and type of leaflets ūū Additional questions specific to enabling the assessment or delivery of health promotion for each individual risk factor Commissioning arrangements The organisational survey was distributed to all Trusts that participated in the NHPHA following completion of the audit. Because the survey was completed in some cases up to 12 months after the data period for discharged cases used in the audit (which was January 2009), direct comparison of survey findings with audit results is not appropriate as changes to health promotion policy, people and tools may well have occurred in the time between the audit data and completion of the survey 16. We would however hope that it is unlikely that health promotion policy/people/tools have worsened since the NHPHA data period started; and so where there is a clear absence of health promotion policy/people/tools we will relate these findings to the audit data. In future the organisational survey will be sent to Trusts for completion at least 3 months prior to the start of data collection for the NHPHA4. Twenty three acute/specialist Trusts (representing twenty eight 53% - of the participating sites) responded to the organisational survey. Because the sample size is small inferences made from the data in the surveys may not be generalisable, but do provide an interesting insight into the responding Trusts health promotion policies, people, tools and services. In future, completion of the organisational survey will be a prerequisite for participating in the NHPHA 17. There follows a predominantly descriptive account of the data from the organisational survey. It is assumed that data provided by Trusts is correct, if you believe this not to be the case, please contact the NHPHA administration team. 56 National Health Promotion in Hospitals Audit

57 Results Population Respondents to the survey were from Trusts serving large, medium and small geographical areas. The populations served by the Trusts ranged from large (maximum: 2,800,000 population) to small population sizes (minimum: 180,000 people), with the average population size approximately 510,000 ± 120,000 (SE). Each trust was asked to provide details of the number of in-patients treated between April 2008 and April 2009 (whole Trust). Because further additional information was not provided on how to answer this question, it appears to have resulted in different interpretations of the question, and so to ensure data collected are the same for each Trust we have used HES Data for admissions between 2008 and Admissions ranged from 12,337 to 210,573 with an average of 89,300 ± 8,145 (SE) and median of 87,175 admissions. Health promotion policy and people In order for hospitals to ensure that health promotion is an integral part of the organi- sation s quality management system 19 it is paramount that there is a health promotion policy that senior management endorse and which is supported by resources, staff training and awareness of the policy. Figure 1 provides details of the percentage and number of Trusts that have health promotion policies and people dedicated to delivering health promotion within hospitals. Further details are provided in Table 1. Data in Figure 1 reveal that approximately half of all respondents have health promotion as part of their Trust s stated aims and mission ( mission statement ), a quarter have a written health promotion strategy ( Written HP strategy ) and 30% of Trusts healthcare professionals with patient contact have a health promotion role written into their job descriptions/contract ( HP job role description ). The majority (70%) of Trusts have a champion for health promotion at board level, 40% have a group responsible for health promotion within their Trust and 30% have specific personnel who co-ordinate health promotion. 17 Case notes for review in the NHPHA are always from patients discharged at least 3 months prior to the start of the audit data collection period. 18 The NHS Information Centre, Hospital Episode Statistics for England. Inpatient statistics, document/e82490.pdf Figure 1: Health Promotion Policy & People HP: Health Promotion. Numbers in columns indicate the number or responses Missing Data No Yes Percentage Description National Health Promotion in Hospitals Audit 57

58 Table 1: Cross tabulation of Health Promotion Policy & People Policy No Mission statement (n = 9) Yes, Mission statement (n =12) Written HP strategy Written HP strategy HP role job description No (n = 8) Yes (n = 1) HP role job description No (n = 7) Yes (n = 5) No (n = 13) 5 1 No 4 3 Yes (n = 7) 3 0 Yes 2 2 Yes, some (n = 1) 0 0 Yes, some 1 0 People No Board Champion (n = 6) Yes, Board Champion (n =16) HP Group HP Group HP Coordinator No (n = 5) Yes (n = 1) HP Coordinator No (n = 8) Yes (n = 8) No (n = 8) 4 0 No 3 1 Yes (n = 7) 1 1 Yes Nicholson, D. May The Year: NHS Chief Executive s annual report 2008/09 dh.gov.uk/dr_consum_dh/ groups/dh_digitalassets/ documents/digitalasset/ dh_ pdf And for an overview see telegraph.co.uk/health/ healthnews/ /nhs- chief-tells-trusts-to-make- 20bn-savings.html 21 Royal Society for Public Health Guide for World Class Commissioners. Promoting Health and well-being: Reducing Inequalities. org.uk 22 NICE (2009) Using NICE guidance to cut costs in the down turn nice.org.uk/aboutnice/ whatwedo/niceandthenhs/ UsingNICEGuidanceTo CutCostsInTheDownturn.jsp 23 Chief Medical Officer (2005) Annual Report. London: Department of Health. uk/prod_consum_dh/ groups/dh_digitalassets/@ dh/@en/documents/ digitalasset/dh_ pdf Only two Trusts (9% of respondents) had a mission statement, written strategy and health promotion roles in job descriptions (see Table 1). These two Trusts (H2 and H30) also had a board champion, health promotion group and coordinator (but no identifiable budget for health promotion services/ materials); and an additional five Trusts (representing seven hospitals: H3, H13, H20, H27, H37, H43, H46) had a board champion, health promotion group and coordinator. While it is a positive sign that half of Trusts have health promotion as part of their stated aims and mission, this is rarely backed up with a health promotion strategy and personnel; i.e. it is clear that the majority of trusts do not have in place a comprehensive management policy and mechanisms that are supportive of health promotion. While health promotion within hospital settings does not necessarily need to be a high cost activity, the finding that Only one Trust (H22) had an identifiable budget for health promotion services and materials is disappointing; yet not surprising given the current economic situation, with NHS Trusts required to make up to 20 billion in efficiency savings by the end of 2013/2014, which translates into efficiency savings of millions of GBP for each hospital 20. However several recent publications indicate that it is now more important than ever before to invest in health promotion Health promotion is the only way of ensuring a financially sustainable health service unless there is substantial investment to support the public in active pursuit of their own health and well-being, the cost of treatment of chronic conditions by the NHS will become unsupportable. 21 Promoting good health and preventing ill health saves money... Increased investment in public health is key to increasing efficiency in the health service. A small shift in resource towards public health prevention activity would offer significant short, medium and long term savings to the service and to the taxpayer nevertheless evidence indicates that public health budgets are often raided to support other activities (prior to the economic crisis ) 23,24 ; and there appears to be reticence about recommending ring-fencing 58 National Health Promotion in Hospitals Audit

59 24 Strategic Review of Health Inequalities Fair Society, Healthy Lives - The Marmot Review Final Report. ac.uk/gheg/marmotreview/ FairSocietyHealthyLives 25 Bernstein H, Cosford P, & Williams A. February Enabling Effective Delivery of Health and Wellbeing. An independent report. dr_consum_dh/groups/dh_ digitalassets/@dh/@en/@ ps/documents/digitalasset/ dh_ pdf funds supposedly allocated to public health activities, even from those supporting health promotion within the NHS - for example, an independent report focussing on improving the delivery of health and wellbeing services in England states We make no recommendations in this report on ringfenced budgets. These can clearly be helpful in protecting investment, but they also tend to identify health and wellbeing as separate from the mainstream function of the NHS and other public sector bodies. Whether or not budgets for health and wellbeing programmes are ringfenced, we consider it vital that the full corporate endeavours of the NHS and the wider public sector are applied to this purpose. 25 while the International Health Promoting Hospitals project/network highlights the importance that hospitals allocate resources to the processes of implementation, evaluation and regular review of the [health promotion] policy [and ensure] the availability of the necessary infrastructure including resources, space, equipment, etc in order to implement health promotion activities 19 (standard 1), it will clearly be a challenge for hospitals to achieve this, thus making it imperative that hospitals work in partnership with other health services, local authorities, voluntary organisations, etc; and this report ends with a discussion of the importance of working with local commissioners. National Health Promotion in Hospitals Audit 59

60 Training Health promotion policy can only be supported by staff that have the tools, knowledge and confidence to deliver health promotion to patients. The importance of staff training has been highlighted by the International Health Promoting Hospitals project/network (standard 4): that hospital management should establish conditions for the development of the hospital as a healthy workplace, and in particular ensure there is a resource strategy that includes the development and training of staff in health promotion skills 19. In relation to staff s training needs only one Trust has undertaken a health promotion training needs analysis (H37). Table 2 provides details of the number of Trusts whose staff have access to training on assessing and delivering health promotion for risk factors. The majority of Trusts (77.3%) have training available to staff for smoking assessment/health promotion, and this is predominantly delivered by PCTs. The majority (72.7%) also have access to training on diet, but this is mainly delivered within hospitals. Training for alcohol and weight is available to staff in approximately 65% of Trusts responding to the question and is usually delivered by the hospital rather than PCT. While there clearly appears to be the least amount of training available to staff for physical activity, with only 50% of Trusts staff having access to training, a Chi- Square analysis did not reveal any significant difference in the provision of training between the risk factors. Table 2: Number of Trusts reporting that staff have access to training on assessing and delivering health promotion to patients for risk factors Risk Factor Delivered by Hospital only PCT only Hospital and PCT Other Organisation Total with access to training No training Smoking Alcohol Diet Weight Physical activity Chi-Square analysis comparing total with access to training and those with no training between the risk factors was χ2 = , DF = 4, P = emphasisnetwork.org. uk/tphn/downloads/ BIsummary pdf Ideally health promotion training should include training on behaviour change interventions (for example brief advice, brief interventions, motivational interviewing and social marketing 26 ); but staff at only nine Trusts have access to such training: staff at three Trusts are able to access behaviour change training in-house and from external organisations (H1, H30, H36), at one Trust training is available in-house only (H37) and at five Trusts training is available from external organisations only (H5, H14, H17, H20, H22). 60 National Health Promotion in Hospitals Audit

61 Integrated Care Pathways (ICPs) Hospital healthcare professionals often report that a lack of time is a major obstacle in delivering health promotion to patients, so any mechanisms that can reduce the time it takes for staff to assess patients for risk factors and that provide clear guidance on how to deliver health promotion (while maintaining an individualised approach to its delivery) and/or whom to refer patients to should improve the delivery of health promotion. One such potential mechanism is the use of ICPs. Figures 2 to 4 show the percentage and number of Trusts who report that all, some or no care pathways have elements of Assessment of risk factors Health promotion and education delivered by Trust s healthcare professionals Referral to (internal and/or external) health promotion services for each risk factor. Analysis of whether there are differences between the degree to which ICPs have elements of assessment, health promotion from hospital healthcare professionals and referral pathways between the risk factors was undertaken through the use of a Friedman Test followed by all pairwise comparisons. This enables ranking of risk factors, which is shown in Table 3. The pairwise comparisons revealed that prompts and tools for the assessment, health promotion/education, and referral to health promotion services for physical activity are significantly less likely to be incorporated into ICPs in comparison to all other risk factors. There were no significant differences in the number of Trusts with ICPs having all, some, or no assessment and health promotion/referral elements between smoking, alcohol, weight and diet/nutrition. Friedman analysis was also undertaken to determine whether there were differences in ICPs for assessment, health promotion by healthcare professionals and referrals for health promotion within each risk factor. Elements of assessment within ICPs are consistently and significantly higher ranked (i.e. more frequently reported as being in some or all ICPs) than both health promotion by healthcare professionals and referrals within each risk factor; and no significant difference in ranking between health promotion by healthcare professionals and health promotion referrals within each risk factor were evident. This reveals a picture of hospital ICPs incorporating prompts or tools for the assessment of risk factors but a lack of tools for the next step of delivering health promotion to patients for risk factors Table 3: Ranking of risk factors for ICP elements Ranking Assessment HP from HCPs HP referral 1 Weight Smoking Smoking 2 Smoking and alcohol Diet Weight 3 Alcohol and weight Diet 4 Diet Alcohol 5 Physical activity Physical activity Physical activity HCP: healthcare Professional National Health Promotion in Hospitals Audit 61

62 27 This reason was reported by hospital nursing staff in interviews concerning attitudes towards health promotion for patients, unpublished report, Stockport NHS FT, (whether these are delivered directly by healthcare professionals within hospitals, or through referral mechanisms to health promotion services). When we looked at the composition of ICPs on the basis of each risk factor (see Appendix 5 for further details) we found that only three Trusts (H31, H35, H38, H49) reported that all their ICPs incorporate assessment and health promotion (including referrals) for smoking; for weight and diet this was also evident in three Trusts (H1, H38, H43, H49 for both risk factors); for alcohol this was found in two Trusts (H22, H38, H49); and for physical activity no Trusts had all ICPs incorporating assessment and health promotion. One Trust had assessment and heath promotion within all ICPs for smoking, alcohol, weight and diet (H38, H49). Given the small proportion of patients assessed in the audit for physical activity (38.8%) in comparison to smoking (81.0%) and alcohol assessment (68.5%), the finding that ICPs are significantly less likely to have elements of assessment for physical activity is not surprising and may to some extent explain why these differences in assessment exist if staff aren t provided with explicit guidelines and prompts, it may be more difficult to undertake assessment, especially as physical activity is relatively harder than other risk factors to assess as a simple question or easily applied objective measure of activity does not exist. This reasoning does not however apply to obesity which is assessed in only 40.1% of patients in the audit and yet all Trusts that responded to the question on ICP assessment indicated that all or some ICPs have elements of assessment for weight status; and the presence of assessment of weight documentation in ICPs is more highly ranked than assessment documentation for any other risk factors. Have assessment prompts/tools for identifying obesity been introduced since the audit was completed; or are there other reasons for why obesity is not being assessed in more patients? Looking at the extent to which ICPs have health promotion and referral pathways in comparison to the audit findings on health promotion delivered, the relative proportion data does not appear to have any relationship to the profile of ICPs (see Figure 4, section 2): health promotion for physical activity and alcohol were delivered most often, followed by obesity and smoking health promotion; but if we look at the absolute proportion of patients receiving health promotion (see Figure 5, section 2), the largest total number of patients received health promotion for smoking (4.1%), then alcohol (3.4%), physical activity (3.1%) and lastly for obesity (1.8%) again it appears that ICPs for obesity health promotion are not translating into expected levels of health promotion for obese patients. Further exploration into whether health promotion documentation in ICPs is a relatively recent development within Trusts and/ or whether healthcare professionals find obesity a particularly difficult issue to address, for example due to a lack of training and/or not feeling comfortable discussing obesity with patients (perhaps due to own observable issues with weight themselves 27 ) is required. 62 National Health Promotion in Hospitals Audit

63 Figure 2: Do care pathways have elements of assessment for.. Smoking Alcohol Weight Diet Physical activity % 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Figure 3: Do care pathways have health promotion/education for... Yes All Yes Some No Missing Data Smoking Alcohol Weight Diet Physical activity % 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Figure 4: Do care pathways have referral to health promotion services for.. Yes All Yes Some No Missing Data Smoking Alcohol Weight Diet Physical activity % 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% National Health Promotion in Hospitals Audit 63

64 28 Mackenzie DM, Langa A, Brown TM: Identifying hazardous or harmful alcohol use in medical admissions: a comparison of AUDIT, CAGE and brief MAST. Alcohol & Alcoholism 1996, 31: See drugslibrary.stir.ac.uk/ documents/alccontools. pdf for information on various validated alcohol tools. 30 J.B. Saunders, O.G. Aasland, T.F. Babor, J.R. de la Fuente and M. Grant. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consumption-II, Addiction 88 (1993), pp ac.uk/pubs/contents/ ea90ff6a-fcd b958- d98f0cc2246a.pdf 32 J.A. Ewing. Detecting Alcoholism. The CAGE Questionnaire. JAMA. 1984; 252(14): jama.ama-assn.org/cgi/ data/300/17/2054/dc1/1 Table 4: Identification of Trusts using validated alcohol assessment tools Use of validated assessment tools The assessment of smoking is relatively straight forward and can be achieved through simple questioning of a patient as to whether or not they currently smoke or have ever smoked. The assessment of all other risk factors is not as simple and is ideally achieved through the use of specific validated tools. For alcohol, many hospitals will have ICPs with question which merely prompt the healthcare professional to ask a patient how much alcohol they drink, relying both on the healthcare professional to have a good understanding of alcohol units, and the truthfulness of the patient. This method has been shown to result in missing a significant number of patients with alcohol problems. It is far more appropriate to use a validated alcohol tool to understand how much someone drinks and their relationship with alcohol 28. There are numerous validated tools available 29, with the WHO recommending the use of AUDIT (Alcohol Use Disorders Identification Test 30 ) within primary care and the Royal College of Physicians recommending its use within hospitals 31. But studies assessing the use of alcohol tools within hospitals are equivocal with some favouring AU- DIT 28 and others suggesting that alternatives such as CAGE 32 may be more appropriate for use with hospital patients 33. A recent pilot study has revealed that nurses do not feel Alcohol Assessment Tool Trust/Hospital codes AUDIT H2, H4, H9, H15, H23, H37, H45 AUDIT-C H2 AUDIT-PC H17 CAGE H4, H15, H23, H35, H36 FAST H2, H15, H20, FRAMES H15 MAST H22 PAT H2, H4, H15, H23, H45 SADQ H45 comfortable asking patients the questions in AUDIT 34. This has had a negative impact on its practical application within the hospital. To address this issue alcohol liaison workers have provided training for nursing staff on how to ask the questions indirectly as part of a conversation rather than to ask the questions as they are written. It is possible that all staff expected to use AUDIT with hospital patients will require such training to gain confidence in using the tool. Regardless of which validated alcohol tool is most appropriate to use within hospital settings, they all show better sensitivity and specificity for identifying alcohol misuse than questions asking how much a patient drink. The survey revealed that nearly half of all Trusts (n = 11; 48%) use one or more validated alcohol tools (see Table 4 for details), with the most frequently used tool AUDIT (n = 6), followed by the Paddington Alcohol Test (n = 4; which was designed to be used within emergency departments) and GAGE (n = 4), FAST (n = 3), AUDIT-PC (n = 2), AUDIT-C (n = 1), MAST (n = 1), SADQ (n = 1) and FRAMES (n = 1). There is a clear preference for the use of AUDIT or one of its shorter versions (AU- DIT-PC or AUDIT-C) to assess alcohol use within hospital patients. We would recommend that those Trusts not already using a validated alcohol tool look into incorporating either AUDIT or CAGE into ICPs for in-patient care; and ensure that some basic training is available to ensure healthcare professionals feel confident in using the tools. The classification of an individual as obese is not complex, but does require that a patient is weighed, their height measured and their body mass index (BMI) determined from these measures (usually through the use of pre-printed BMI charts). Clearly, weighing 64 National Health Promotion in Hospitals Audit

65 scales are required to accurately calculate weight; and nineteen Trusts did respond that they had weighing scales on every ward. Of the three Trusts (H5, H15, H37) that reported they did not have weighing scales on all wards (nor any scales that automatically calculate BMI) none met the standard for assessing obesity. One Trust (H 37) assessed only 2% of its patients for obesity and the other two Trusts assessed less than a third of their patients for obesity. In the absence of weighing scales on all wards, healthcare professionals can not be expected to assess all patients for weight with the identification of both obesity and malnutrition in patients at risk. Scales which calculate BMI are an additional cost, but should ensure that patients have their BMIs calculated accurately and easily (hopefully assisting in the assessment and recording of patients BMI). One Trust reported having BMI scales on all its adult wards (H35), one had BMI scales on all patient areas (H30), four Trusts had BMI scales on some adult wards (H4, H9, H20, H23, H36) and one Trust had BMI scales in the gym and occupational health area (H31). The audit also gathered data on whether patients were assessed for malnourishment. Twenty two Trusts report using a validated screening tool for malnutrition (one Trust did not respond to the question). Sixteen used the malnutrition universal screening tool (MUST 35 ) and nine used another malnutrition tool, three of whom also used MUST. Trusts clearly have appropriate tools in place for the assessment of malnutrition. Leaflets/written advice While information alone will not lead to behaviour change 36, the provision of information on the risks of unhealthy behaviours, Table 5: Number of Trusts with health promotion leaflets available Local National Health benefits/ Tips for Number of Trusts risks change with no leaflets Smoking Alcohol (H5, H30) Obesity (H19, H30, H31, H37) Physical 5 (H5, H15, H20, 14 NA activity H30, H31) NA: not applicable, as there are no national schemes for physical activity Trusts were not asked whether they had leaflets for national services. benefits of changing behaviour (on physical and mental health, finances, etc) and advice on how to change behaviour is one component of health promotion. It is vital that such information is provided in written form to patients as medical information is most likely to be recalled if it is provided in written/pictorial material alone or in combination with verbal information 37,38. The responses in table 5 concerning the availability of leaflets with information on local health promotion services ( Local ), national services ( National ), risks of a risk factor and health benefits of changing lifestyle ( health benefits/risks ) and tips on how to change risk behaviours ( Tips for change ) indicate that most Trusts have a good availability of leaflets, in particular for local services. It is a very positive finding that all Trusts report having leaflets available for smoking. Those Trusts that do not have leaflets available for the other risk factors in particularly H30 which has no leaflets available for alcohol, obesity or physical activity - should look into identifying practical sustainable systems for ensuring health promotion leaflets are available to patients and visitors. 33 R. Hearne, A. Connolly, and J. Sheehan. Alcohol abuse: prevalence and detection in a general hospital. J R Soc Med February; 95(2): Developing a Lifestyle Service in Secondary Care Stockport NHS Foundation Trust. Funding body: Public Health Leadership and Workforce, Department of Health. Report is due May 2010 and will be available to download from www. nhphaudit.org 35 must_tool.html 36 The King s Fund Commissioning and Behaviour Change. Kicking Bad Habits final report. document.rm?id= Ley, P. (1979) Memory for Medical Information. British Journal of Social and Clinical Psychology, 18, Kessels RPC: Patients memory for medical information. Journal of the Royal Society of Medicine, 2003, 96: National Health Promotion in Hospitals Audit 65

66 Table 6: Combination of health promotion leaflets available within Trusts Smoking Alcohol Weight Physical Activity local + national + health benefits + tips NA local + national + health benefits NA local + health benefits + tips local + national 1 1 NA local + tips local 1 national + health benefits + tips NA health benefits + tips tips Total (hospitals) Unpublished report in 2007 on health promotion services healthcare professionals at Stockport NSH Foundation Trust would value included the regular delivery of leaflets to wards. This has now been achieved through the use of hospital volunteers. Table 6 shows the different combination of leaflets available at all Trusts. The majority of Trusts have a wide variety of leaflets for smoking and alcohol, covering local and national services, tips for stopping smoking/ reducing alcohol intake and information on the health benefits of stopping smoking/ sensible drinking and risks of heavy drinking. And for physical activity, the majority of Trusts have leaflets providing details of local physical activity programmes, health benefits of physical activity and tips to increase activity. Types of leaflets available for weight are more varied, with six Trusts having no leaflets available for local weight loss programmes although five of these Trusts have systems for referring patients to weight loss programmes in the community. It is however possible that the community programmes do not have promotional written materials available. Availability of leaflets does not necessarily translate into the provision of these leaflets to patients as we have seen from the audit results, relatively few patients receive written health promotion the ease at which leaflets can be accessed by healthcare professionals, in particular nursing staff, is one vital factor 39. Trusts were asked whether leaflets could be downloaded from the intranet/internet and/or provided by PCT public health units (see figure 5 for results). The majority of Trusts can access at least some leaflets through the intranet, internet and/or their PCT public health unit; but the finding that few of the Trusts can access all leaflets through these mechanisms implies that accessibility of ALL appropriate written information may be patchy. Trusts were also asked to provide details of how hard copies of leaflets are distributed throughout wards. The free text information provided in response to this question was subject to content analysis (see Appendix 6 for details). This revealed that leaflets are distributed by specialists such as nutrition link nurses, alcohol liaison nurses, etc (n = 6) and by volunteers (n = 5). Details also indicated that leaflets are ordered directly by ward staff (n = 4) and that leaflets are displayed on the wards, for example in racks, although the details of how they get there is not always clear (n = 6). A previous hospital staff survey uncovered that the ready availability of health promotion leaflets is very important to nursing staff37. We would suggest that nurses time is not best spent ordering and arranging leaflets. The use of volunteers is perhaps the best method for ensuring regular distribution of leaflets. 66 National Health Promotion in Hospitals Audit

67 Figure 5: How leaflets are accessed 100% 90% 4 80% 1 70% Missing NA None Some All 60% 50% 40% % 20% 10% 0% 1 Intranet Internet PCT 2 Specialists & specialist services Smoking Twenty Trusts reported that they had specialist respiratory nurses. The median number of full time specialist respiratory nurses across the Trusts was 3.0 WTE (whole time equivalent), average was 5.2 ± 1.1 (SE), and range was 1.0 to 23.1 WTE. Seventeen Trusts (74%) reported having smoking cessation nurses (those that have received accredited training to deliver smoking cessation to patients). While the median number of full time smoking cessation nurses across the Trusts was 2.0 WTE, the average was 17.0 ± 12.3 (SE), indicating a large amount of variability in the number of smoking cessation nurses between different Trusts, which is also indicated by the wide range: 0.4 to WTE. One Trust reported employing two smoking cessation advisors who were not nurses. Smoking cessation nurses/advisors should be assisting patients to quit smoking through advice/behavioural therapy/provision of NRT or bupropion/follow-up after discharge. A Cochrane review of interventions for smoking cessation in hospitalised patients found that counselling interventions significantly increased smoking cessation rates only if the counselling was intensive: intervention started in hospital followed by patient follow-up of at least 1 month following discharge. Less intensive interventions those only delivered during hospitalisation or with a follow-up of less than 1 month after discharge were not effective; hence it is important that any patients provided with smoking cessation interventions within hospital are also provided with the option of follow-up after discharge 40 from a smoking cessation nurse/service. The addition of NRT to Intensive counselling did not result in a statistically significant increase in quit rate but there is considerable evidence that all 40 wiley.com/cochrane/ clsysrev/articles/ CD001837/frame.html Rigotti N, Munafo MR, Stead LF. Interventions for smoking cessation in hospitalised patients. Cochrane Database of Systematic Reviews 2007, Issue 3. Art. No.: CD DOI: / CD pub2. National Health Promotion in Hospitals Audit 67

68 41 wiley.com/cochrane/ clsysrev/articles/cd000146/ frame.html Stead LF, Perera R, Bullen C,Mant D, Lancaster T. Nicotine replacement therapy for smoking cessation. Cochrane Database of Systematic Reviews 2008, Issue 1. Art. No.: CD DOI: / CD pub3. 42 The NHS confederation. January 2010.Too much of the hard stuff: what alcohol costs the NHS. Publications/Documents/ Briefing_193_Alcohol_ costs_the_nhs.pdf 43 DH/National Treatment Agency for Substance Misuse. June Models of care for alcohol misusers (MoCAM). prod_consum_dh/groups/ documents/digitalasset/ dh_ pdf forms of nicotine replacement therapy (NRT) - transdermal patches, chewing gum, nasal spray, tablets and inhalers - significantly increase the likelihood that an attempt to quit smoking cigarettes will be successful in comparison to trying to quit without NRT/ placebo 41. Smoking cessation nurses can prescribe NRT at only six of the seventeen Trusts with smoking cessation nurses (see Table 7 for details of the Trusts/hospitals). At four of all responding Trusts (17%) there were no policies in place for healthcare professionals to prescribe nicotine replacement therapy. At the majority of Trusts it is doctors/medical staff who prescribe NRT (n = 13; 57%). Only one Trust reported that all nurses are able to prescribe NRT and at four Trusts some nurses are able to prescribe. Pharmacists can prescribe NRT at seven Trusts; and four other Trusts reported that other healthcare professionals can prescribe NRT within their Trusts. We recommend that All Trusts should have nurses who have been trained to deliver smoking cessation, All Trusts undertake the necessary measures to ensure that smoking cessation nurses can prescribe NRT, Mechanisms are put in place to enable patients to receive follow-up smoking cessation support for at least 1 month following discharge. Alcohol - Alcohol liaison nurses Given the large increases in alcohol consumption within England over the past thirty years, the concomitant increases in hospital attendance and admission rates for alcoholrelated health problems, and the finding that most of the cost of treating alcohol-related acute and chronic conditions is spent in hospitals 42 it is vital that hospitals have appropriate services to address the treatment and health promotion needs of those who misuse alcohol and are dependent drinkers. One of the best means of ensuring an optimal alcohol service is delivered is through the employment of alcohol liaison nurses/ workers - The Royal college of Physicians 31 have highlighted the need for each trust to have one or more dedicated alcohol health workers employed by and answerable to the acute trust. The roles will include: implementation of screening strategies detoxification of dependent drinkers brief interventions in hazardous drinkers referral of patients for on-going support and with access/knowledge about locally available non-statutory and voluntary agencies provision of links with liaison/specialist alcohol psychiatry an educational resource and support focus for other health care workers in the Trust. (Recommendation 7 of 12) The Department of Health and National Treatment Agency for Substance Misuse have also stated that Alcohol liaison posts would help promote alcohol interventions and treatment within hospital settings. 43 Alcohol liaison nurses/workers are employed at fourteen Trusts (61%: H2, H4, H9, H14, H15, H17, H19, H20, H22, H23, H35, H36, H37, H45, H49, H38). One Trust (H45) did not 68 National Health Promotion in Hospitals Audit

69 supply details of the number of alcohol liaison nurses/workers employed at their Trust, but based on data from the thirteen that provided employment details, the median number of full time workers across the Trusts was 1.5 WTE, average was 2.6 ± 1.0 (SE), and the range was 0.24 to 14.0 WTE. Clearly those Trusts who do not have an alcohol liaison nurse/worker should look into either creating the post or working in partnership with other organisations to ensure that hospital staff have direct access to such individuals to ensure that the needs of patients with alcohol problems, especially dependent drinkers, can be appropriately met. While there may be many more community alcohol services available to patients, of importance is whether acute Trusts are aware of these services and can signpost or refer patients to them. Of the five Trusts that potentially do not have community alcohol services their patients can access, one - H31 - also reports having no alcohol liaison nurses, does not use a validated alcohol tool and only has leaflets available with information on tips for reducing alcohol intake. This Trust was one of six Trusts that provided 0% of patients with health promotion for alcohol. Table 7: Healthcare professionals that prescribe NR - Community alcohol teams Nearly all Trusts - Twenty one (91%) - reported that there was a community alcohol team (CAT) in the geographical area that their patient population is from; but only eleven of these Trusts had a standardised system for referring patients to their local CAT (and one additional Trust reported that a new simplified referral form has been developed to enable effective referral ). Of concern is the finding that at least four Trusts (H3, H5, H11, H27, H30, H42, H46) have neither alcohol liaison nurses nor referral processes in place to CATs (see Table 7). Healthcare Professional Doctors/ medical staff Smoking cessation nurses Some nurses All nurses prescribe Pharmacists Other HCPs No one able to prescribe HCP: healthcare professional Hospital Code H1, H2, H3, H4, H13, H14, H15, H23, H24, H27, H30, H35, H36, H43, H46, H49, H38 H1, H3, H13, H27, H31, H37, H43, H46 H11, H24, H36, H37, H42 H1 H1, H13, H17, H24, H31, H37, H38, H49 H1, H2, H5, H20, H30 H9, H19, H22, H45 Trusts were also asked what community and hospital based alcohol services were available to their adult hospitalised patients. Twenty one Trusts reported that services were available to their patients with eighteen providing details of community services and ten Trusts having hospital based services available to patients (eight Trusts had both community and hospital alcohol services). Space was provided for details of up to eight hospital and/or community alcohol services; an average of 3 community services and 1 hospital service was identified. Table 8: Alcohol liaison nurses/workers and referral process to Community Alcohol Teams within Trusts Alcohol Liaison Nurses/Workers? Referral Process to CAT? No Yes No 4 (2) 3 Yes 3 (3) 8 Numbers in parentheses are those Trusts that did not respond to the question concerning whether they had a referral process to their local CAT. National Health Promotion in Hospitals Audit 69

70 Previous research 45 has found that adult hospitalised patients are keen to improve their levels of physical activity and have reported a desire to be able to access hospital gyms. Seven Trusts report that they have a gym within their hospital(s) that patients can use (H3, H4, H11, H17, H22, H23, H27, H30, H31, H42, H46). 44 Healthcare Commission Obesity: identification and management in secondary care. healthcarecommission.org. uk/indicators_2006nat/ Trust/Indicator/ indicatordescriptionshort. asp?indicatorid= Haynes, CL Health promotion services for lifestyle development within a UK hospital - Patients experiences and views. BMC Public Health, 2008, 8:284 com/ /8/ uk/en/publichealth/ Healthinequalities/ HealthTrainersusefullinks/ DH_6590 Nutrition/diet and physical activity Twenty two Trusts have nutrition specialists/dietitians; and sixteen of these Trusts can refer patients directly to these specialists solely for the treatment of obesity (i.e. the patient does not have to have co-morbidities associated with diet). Nineteen Trusts provided details of the number of nutrition specialists/dietitians they employed - the median number working across these Trusts is 11.0 WTE, average of 14.0 ± 2.5 (SE), and the range is 2 to 43.0 WTE. Fifteen Trusts (65%) reported that they have a system in place for referring patients to community weight loss programmes (H1, H2, H4, H11, H13, H14, H15, H17, H22, H23, H24, H30, H35, H37, H42, H43, H45). At ten of these Trusts only patients over a certain BMI can be referred to these programmes, for the majority of Trusts the BMI cut-offs for referral are in line with previous healthcare commission guidance 44 - a BMI over 30 or over 27 with co-morbidities; for one Trust patients must have a BMI over 40 or over 35 with co-morbidities (H22). Three Trusts reported that all patients that perceive they have a weight problem and want to lose weight can be referred to community weight loss programmes (H13, H30, H37). Nineteen Trusts reported that there were physical activity programmes that they could refer patients to within the community and/or hospital setting (the Trusts that did not provide details of community/ hospital physical activity programmes they could to refer patients to: H3, H20, H24, H27, H38, H46, H49). Eighteen Trusts provided details of community services and six Trusts had hospital based services available to patients (four Trusts had both community and hospital physical activity programmes). An average of 3 community programmes and 1 hospital programme was identified (median of 1 and 0 programmes respectively). It is recommended that all Trusts ensure they are aware of community weight loss and physical activity programmes and that they establish referral processes to the programmes for suitable patients. Health trainers Health Trainers provide individual support and advice to help people to identify and achieve their own health goals and to make healthier lifestyle choices, most often in the areas of healthy eating, physical activity, smoking cessation and alcohol. 46 Health trainers clients are generally people living within areas of high deprivation. Health trainers are usually commissioned and managed by PCTs or local authorities and while they work in a wide range of settings 70 National Health Promotion in Hospitals Audit

71 (often within the geographical areas that their clients come from), they are not usually found within hospital settings. However hospital admission rates are positively correlated with measures of deprivation, with a higher proportion of adults from lower socioeconomic groups 47 admitted as emergencies compared with higher socio-economic groups; and hospitals may be the first point of contact with a health care professional for individuals from lower socio-economic backgrounds. Hence, having health trainers within hospitals may help narrow health inequalities if a system is established by which patients from more deprived background can be identified and given the option to access the services of a health trainer. This would also work if links are made with health trainers in the community, and while this was not asked about in the survey, one Trust reported that they could access health trainers within the community (H22). Three of the responding Trusts do have health trainers within their hospitals - at one Trust (H37) they reported that they will have three mainstream health trainer services from March 2010, at another Trust (H45) they have three cardiac lifestyle trainers, one respiratory lifestyle trainer, one fitness trainer, and one psychologist and one Trust reported being able to access health trainers through their PCT (H5). Commissioning As we have seen earlier, hardly any Trusts have an identifiable budget for health promotion and while public health policy indicates that hospitals have an important role in preventing ill health, the lead for commissioning health promotion activities is expected to be taken by PCTs. The King s Fund has recommended that PCTs work with hospitals to ensure hospital patients receive health promotion: PCTs also need to see behaviour change interventions as integral to their full range of commissioning activities, including those in the acute sector. PCTs should ensure that patients in the hospital setting, as well as those accessing primary care, receive appropriate advice and information on behaviour change. PCTs should use contracts and locally agreed care pathways to ensure that patients in the hospital setting are referred to appropriate local support services. 48 With the difficulties inherent in the funding of public health combined with the pressure on PCTs to be supportive of health promotion initiatives, it is crucial that hospitals are in communications with local commissioners in order to facilitate hospital-based health promotion services. The survey asked Have there been discussions with the local commissioners on providing services within the hospital to encourage healthy behaviours? Eighteen Trusts responded positively, while three said there had been no discussions (H9, H17, H31) and two Trusts did not provide a response (H13, H36). Content analysis of the information provided indicates that smoking is the main focus area for discussion with commissioners as this was reported by eight Trusts, 47 Reid, F. D. A., Cook, D. G. & Majeed, A. (1999) Explaining variation in hospital admission rates between general practices: cross sectional study. BMJ, 319, The King s Fund Commissioning and behaviour change. Kicking Bad Habits final report. National Health Promotion in Hospitals Audit 71