Implementing personal health budgets within substance misuse services [final report]

<p>Executive summary1. The personal health budget initiative is a key aspect of personalisation across health care services inEngland. Its aim is to improve patient outcomes, by placing patients at the centre of decisions abouttheir care.2. In 2009 the Department of Health invited PCTs to become pilot sites to join a programme which wouldexplore the opportunities offered by personal health budgets. The Department of Healthcommissioned an independent evaluation to run alongside the pilot programme to provideinformation on how personal health budgets are best implemented, where and when they are mostappropriate, and what support is required for individuals.3. Two pilot sites within the pilot programme explored whether personal health budgets had an impacton outcomes and experiences compared to conventional service delivery among individuals withsubstance misuse problems.Study design and methodology4. The evaluation adopted a longitudinal approach, and included people with drug and/or alcoholaddiction.5. The study used a controlled trial with a pragmatic design to compare the experiences of peoplereceiving a personal health budget with the experiences of people continuing under the currentsubstance misuse treatment support arrangements. After applying initial selection criteria, in one pilotsite people were randomised into the personal health budget group or a control group. In the secondpilot site, the personal health budget group was recruited from patients of those health careprofessionals in the pilot offering budgets, and a control group was recruited from patients of nonparticipatinghealth care professionals.6. A mixed design was followed where both quantitative and qualitative methodologies were used toexplore patient outcomes and experiences, service use and costs, as well as the experiences of thoseimplementing the initiative. In total, an active sample of 166 participants was recruited: 119 in thepersonal health budget group and 47 in the control group. Within the active study sample, 55participants had drug and alcohol addictions and 111 participants had an alcohol addiction only.7. The qualitative analysis involved interviews with personal health budget holders and organisationalrepresentatives. Data were analysed using the framework approach, with the data organised bythemes according to the topic guides used in the interviews.8. The difference-in-difference approach was used to explore whether personal health budgets had animpact on an individual’s quality of life and relapse rates. The analysis subtracted an individual’sfollow-up outcome scores from their baseline score. Due to the small sample size, the analysis did notinclude exploring difference-in-difference multivariate models and therefore we were unable tocontrol for confounding baseline differences.The content of support plans9. Among the personal health budget group, 103 support plans were returned from the two pilot sites.In terms of the size of the budget, 41 budgets were worth between £1,000 and £5,000 per year, while4 budgets were worth more than £10,000. 210. The majority of care/support plans were managed notionally. While one of the pilot sites did haveapproval to offer direct payments, we did not find evidence this deployment was offered during thepilot programme.11. Residential detox was the largest single cost category. The more innovative uses of the personalhealth budget included driving lessons, alternative therapies, leisure activities and educationalcourses. Enabling people to access community detox rather than residential detox could also beregarded as an innovative use of their budget.The impact of personal health budgets on relapse rates, quality of life and service quality12. The shortened version of the Alcohol Use Disorders Identification Test (AUDIT-C) was used to detectsigns of hazardous and harmful drinking. Difference-in-difference analysis indicated that individualsin the personal health budget group had reduced their excessive drinking at follow-up compared tothose in the control group. Similar results were found with the change in drug consumption at followup.13. Difference-in-difference analysis indicated that there were greater improvements in care-relatedquality of life (ASCOT) and psychological well-being (GHQ12) for individuals in the personal healthbudget group compared to those in the control group, although the difference was not statisticallysignificant.14. Individuals in the personal health budget group were more satisfied with the help paid for by thebudget and the care/support planning process than those receiving conventional services.15. While the quantitative results highlighted the positive impact of receiving a personal health budget,firm conclusions around the impact of personal health budgets compared to conventional servicedelivery could not be made, due to the small sample size.Views from patients16. Qualitative in-depth interviews indicated that personal health budgets had a positive impact onservice quality, relationships with health professionals and views on what could be achievedcompared with conventional service detox delivery.17. The importance of effective implementation was highlighted, both in terms of providing the necessaryinformation to enable budget holders to make an informed choice and also to minimise any delays inthe process of obtaining and using a budget. Individuals reported that delays could potentially lead toanxiety and distress.18. A list of suggestions of possible uses of personal health budgets would have been useful during thesupport/care planning stage.19. Personal budget holders reported a lack of after-care services available with this treatment routewhich could potentially have a longer-term impact on relapse rates. This desire for post-detox care toprevent relapse was especially prevalent at follow-up, when patients had completed theirdetoxification and required relapse prevention services.20. Individuals receiving conventional detox services expressed more negative views of the relationshipthey had with health professionals and their experiences of services.Views from the system21. Organisational representatives believed that personal health budgets had a positive impact onoutcomes for budget holders: the way they accessed services, and to a certain extent the content or 3quality of those services. Organisational representatives attributed these impacts to the personalhealth budgets enabling: increased choice and control for budget holders; increased flexibility;encouraging innovation and creativity; greater ‘person-centred’ care/support planning; and theopportunity to reduce costs by accessing alternative services or providers of services.22. A number of challenges within the implementation process were mentioned by organisationalrepresentatives. These included: the length of time required to conduct the care/support planningprocess; the time point at which a personal health budget should be introduced; deciding what canand cannot be included, in particular considering whether the budget should be used for relapseprevention; managing attitudes to risk and the cultural change required for patients in the system; thelogistics of managing multi-agencies involved in a person’s care; and establishing integration betweenservices and creating a jointly-funded budget.Recommendations for policy and practice23. A number of recommendations can be made regarding a possible roll-out of personal health budgetswithin the area of substance misuse from the results of this study:? Personal health budgets increased service satisfaction, facilitated a positive relationship withhealth professionals and improved quality of life supporting a wider roll-out.? The budget-holders we interviewed emphasised the value of information and guidance fromoperational representatives about the size and operation of their budgets, including what serviceswere covered.? Direct payments were viewed as playing a critical role in the success of personal health budgetsfor people with substance misuse problems. However, managing the anxiety and practicalchallenges around offering this deployment option may need consideration.</p>