Childhood adversity and mental health admission patterns prior to young person suicide (CHASE): a case-control 36 year linked hospital data study, Scotland UK 1981-2017.

BACKGROUND: Childhood adversity is associated with increased later mental health problems and suicidal behaviour. Opportunities for earlier healthcare identification and intervention are needed. AIM: To determine associations between hospital admissions for childhood adversity and mental health in children who later die by suicide. METHOD: Population-based longitudinal case-control study. Scottish in-patient general and psychiatric records were summarised for individuals born 1981 or later who died by suicide between 1991 and 2017 (cases), and matched controls (1:10), for childhood adversity and mental health (broadly defined as psychiatric diagnoses and general hospital admissions for self-harm and substance use). RESULTS: Records were extracted for 2477 'cases' and 24 777 'controls'; 2106 cases (85%) and 13 589 controls (55%) had lifespan hospitalisations. Mean age at death was 23.7; 75.9% were male. Maltreatment or violence-related childhood adversity codes were recorded for 7.6% cases aged 10-17 (160/2106) versus 2.7% controls (371/13 589), odds ratio = 2.9 (95% CI, 2.4-3.6); mental health-related admissions were recorded for 21.7% cases (458/2106), versus 4.1% controls (560/13 589), odds ratio = 6.5 (95% CI, 5.7-7.4); 80% of mental health admissions were in general hospitals. Using conditional logistic models, we found a dose-response effect of mental health admissions <18y, with highest adjusted odds ratio (aOR) for three or more mental health admissions: aORmale = 8.17 (95% CI, 5.02-13.29), aORfemale = 15.08 (95% CI, 8.07-28.17). We estimated that each type of childhood adversity multiplied odds of suicide by aORmale = 1.90 (95% CI, 1.64-2.21), aORfemale = 2.65 (95% CI, 1.94-3.62), and each mental health admission by aORmale = 2.06 (95% CI, 1.81-2.34), aORfemale = 1.78 (95% CI, 1.50-2.10). CONCLUSIONS: Our lifespan study found that experiencing childhood adversity (primarily maltreatment or violence-related admissions) or mental health admissions increased odds of young person suicide, with highest odds for those experiencing both. Healthcare practitioners should identify and flag potential 'at-risk' adolescents to prevent future suicidal acts, especially those in general hospitals.

Can proactive support prevent unscheduled care? A controlled observational retrospective cohort study in cancer patients in Scotland.

INTRODUCTION: Preventative spend is a global health and social care strategy. Improving Cancer Journeys (ICJ) is a proactive, holistic, multidisciplinary project consistent with this agenda, currently being rolled out across Scotland and parts of UK. ICJ helps people with cancer access whatever support they need to mitigate their most pressing concerns. This study hypothesised that ICJ service users should subsequently use less unscheduled care than matched cohorts not using ICJ. METHODS: Retrospective observational cohort study using linked national datasets. N = 1,214 ICJ users in Glasgow were matched for age, sex, deprivation, cancer type, stage, and diagnosis year to two control groups: 1. Cancer patients from Glasgow before ICJ (pre-2014), 2. Cancer patients from rest of Scotland during study period (2014-2018). Cancer registrations were linked for 12-month baseline and study periods to: NHS24 calls, A&E admissions, inpatient hospital admissions, unscheduled care, number & cost of psychotropic prescriptions. Per-person mean service uses were compared between groups. RESULTS: There was a significant increase in NHS24 calls in the ICJ group (0.36 per person vs. -0.03 or 0.35), more and longer A&E attendances in ICJ (0.37 per person vs. 0.19 or 0.26; 2.19 h per person vs. 0.81-0.92 h), more and longer hospital admissions in ICJ (4.25 vs. 2.59 or 2.53; 12.05 days vs. 8.37 or 8.64), more care pathways involving more steps in ICJ (0.77 spells vs. 0.39 or 0.57; 1.88 steps vs. 1.56 or 1.21), more psychotropic drug prescriptions and higher costs in ICJ (1.88 prescription vs. 1.56 or 1.21; £9.51 vs. £9.57 or £6.95) in comparison to both control groups. DISCUSSION: ICJ users sourced significantly more unscheduled care than matched cohorts. These findings were consistent with much of the comparable literature examining the impact of non-health interventions on subsequent health spend. They also add to the growing evidence showing that ICJ reached its intended target, those with the greatest need. Together these findings raise the possibility that those choosing to use ICJ may also be self-identifying as a cohort of people more likely to use unscheduled care in future. This needs to be tested prospectively, because this understanding would be very helpful for health and social care planners in all countries where proactive holistic services exist.

Reducing patient delay in acute coronary syndrome: Randomized controlled trial testing effect of behaviour change intervention on intentions to seek help.

OBJECTIVES: The aim of this study was to evaluate the efficacy of a behaviour change intervention to reduce patient delay with symptoms of acute coronary syndrome. DESIGN: A 3-arm web-based, parallel randomized controlled trial. METHODS: The intervention comprised 12 behaviour change techniques (BCTs) embedded in a text-only or text+visual narrative (the techniques were systematically identified through systematic review and a consensus exercise). Between February and November 2017, n = 145 people who had recently experienced acute coronary syndrome were randomly allocated to intervention ('text+visual' or 'text-only') or control. Intentions to phone an ambulance immediately for acute coronary syndrome symptoms were assessed before and after the intervention using symptom scenarios, and the change in intention was compared across the three groups. RESULTS: Significant increases in intention to phone an ambulance immediately for ACS symptoms were seen following the 'text+visual' intervention but not following 'text-only' or control. However, the study was underpowered to detect any significant changes in intention between the 3 groups. There were no unintended effects on intentions for non-urgent symptoms. CONCLUSIONS: A 'text+visual' BCT-based intervention may significantly increase intention to phone an ambulance with symptoms of ACS. Further testing of the effect of the intervention on actual behaviour is required.

Meeting psychosocial needs to improve health: a prospective cohort study.

BACKGROUND: Cancer impacts on patients and their families across a range of different domains. For that reason, optimal cancer care has moved away from a disease-centric focus to a more holistic approach in order to proactively support people with their individual needs and concerns. While international policy clearly advocates this agenda, implementation into routine care is limited. Therefore, relevant interventions that measurably improve patient outcomes are essential to understand if this ideal is to become routine multidisciplinary practice. The aim of this study was to analyse the impact of a proactive, holistic, community-based intervention on health-related quality of life in a cohort of people diagnosed with cancer. Secondary aim was to explore the relationship between changes in health status and: cancer type, cancer stage, number of concerns expressed and change in severity of concerns pre and post intervention. METHOD: Prospective observational cohort study. A convenience sample of 437 individuals were referred to the service 'Improving the Cancer Journey (ICJ) in the UK. Each completed the Euroqol EQ-5D-3 L and visual analogue scale (VAS) and a Holistic Needs Assessment (HNA) during initial visit to the service and again at follow-up review, median 84 days later. Change between scores was tested with paired t-tests and relationships between variables with multiple regression models with heteroscedasticity-consistent standard errors. RESULTS: Participants were White British with median age between 50 and 64 years. Cancer type and stage were varied. EQ-5D utility scores improved at follow-up by 0.121 [0.0891-0.153], p < .001, and VAS scores improved by 7.81 [5.88-9.74], p < .001. The strongest predictor of change was a decrease in severity of concerns. Cancer stage 'palliative care' contributed to a reduction in health status. CONCLUSION: This study is the first to show that a holistic community intervention dedicated to supporting the individual concerns of participants had both a statistically significant and clinically meaningful impact on participants' health-related quality of life. The mean change in EQ-5D scores was more than the 'minimally important clinical difference' described in the literature. This is important because while quality of life has multiple determinants, this study has shown for the first time that it is possible to capture a clinically meaningful improvement as a function of reducing someone's personally identified concerns.

Proactive community support tailored to holistic needs: A cohort study.

BACKGROUND: It is increasingly internationally recognized that a cancer diagnosis impacts on people practically and financially as well as physically and psychologically. It is less clear what to do about this. This study introduces an original community service designed to mitigate this wider impact. Nonclinical "link officers" use holistic needs assessment (HNA) to help newly diagnosed people identify and quantify the severity of their physical, psychological, practical, financial, and social concerns. A care plan is then agreed, usually involving community interventions from partner agencies. Following intervention, assessment is repeated. The primary aim of this study was to establish whether there was a significant difference between initial assessment and follow-up, postintervention. Secondary aim was to identify potential predictors of increased levels of concern at baseline and follow-up. METHOD: Pre- and postintervention observational cohort study. Paired t test examined the difference in mean (SD) concern severity between baseline and follow-up. Multiple linear regression models were computed to hypothesize potential predictors of initial concern severity and severity change. RESULTS: The service saw 2413 people 2014-2017. Participants identified average 5.5 (4.7) concerns, financial concerns being most frequent. Mean severity at baseline was 7.12 (out of 10) (2.50), reducing to 3.83 (3.49) post-treatment, paired t(4454) = 64.68, P < 0.0001, reduction of 3.31 (95% CI 3.21-3.41). Factors associated with higher initial concern included unemployment and caring responsibilities. Unemployment was also associated with a smaller reduction of concern severity at follow-up. CONCLUSION: Patient level of concern went from a level associated with specialist referral to a much more manageable level. This original finding is internationally significant because it extends Khera et al's (2017) "provocative idea" that all patients should be screened for financial problems to show that they can be helped with all their concerns. This article describes a successful, transferable model of community care.