Backgroundfalls can negatively affect patients, resulting in loss of independence and functional decline and have substantial healthcare costs. Hospitals are a high-risk falls environment and regularly introduce, but seldom evaluate, policies to reduce inpatient falls. This study evaluated whether introducing portable nursing stations in ward bays to maximise nurse–patient contact time reduced inpatient falls.Methodsinpatient falls data from local hospital incident reporting software (Datix) were collected monthly (April 2014–December 2017) from 17 wards in Stoke Mandeville and Wycombe General Hospitals, the UK. Portable nursing stations were introduced in bays on these wards from April 2016. We used a natural experimental study design and interrupted time series analysis to evaluate changes in fall rates, measured by the monthly rate of falls per 1000 occupied bed days (OBDs).Resultsthe wards reported 2875 falls (April 2014–December 2017). The fallers’ mean age was 78 (SD = 13) and 58% (1624/2817) were men. Most falls, 99.41% (2858/2875), resulted in none, low or moderate harm, 0.45% (13/2875) in severe harm and 0.14% (4/2875) in death. The monthly falls rate increased by 0.119 per 1000 OBDs (95% CI: 0.045, 0.194; P = 0.002) before April 2016, then decreased by 0.222 per 1000 OBDs (95% CI: −0.350, −0.093; P = 0.001) until December 2017. At 12 months post-intervention, the absolute difference between the estimated post-intervention trend and pre-intervention projected estimate was 2.84 falls per 1000 OBDs, a relative reduction of 26.71%.Conclusionportable nursing stations were associated with lower monthly falls rates and could reduce inpatient falls across the NHS.

BackgroundSleepio is an automated digital programme that delivers cognitive behavioural therapy for insomnia (dCBT-I). Sleepio has been proven effective in improving sleep difficulties. However, evidence for the possible impact of Sleepio use on health care costs in the United Kingdom has not previously been developed.AimWe sought to identify the effect of a population-wide rollout of Sleepio in terms of primary care costs in the National Health Service (NHS) in England.Design & settingThe study was conducted in the Thames Valley region of England, where access to Sleepio was made freely available to all residents between October 2018 and January 2020. The study relies on a quasi-experimental design, using an interrupted time series to compare the trend in primary care costs before and after the rollout of Sleepio.MethodWe use primary care data for people with relevant characteristics from nine general practices in Buckinghamshire. Primary care costs include general practice contacts and prescriptions. Segmented regression analysis was used to estimate primary and secondary outcomes.ResultsFor the 10,704 patients included in our sample, the total saving over the 65-week follow-up period was £71,027. This corresponds to £6.64 per person in our sample or around £70.44 per Sleepio user. Secondary analyses suggest that savings may be driven primarily by reductions in prescribing.ConclusionSleepio rollout reduced primary care costs. National adoption of Sleepio may reduce primary care costs by £20 million in the first year. The expected impact on primary care costs in any particular setting will depend on the uptake of Sleepio.

BackgroundOpioids are strong pain medications that can be essential for acute pain. However, opioids are also commonly used for chronic conditions and illicitly where there are well-recognised concerns about the balance of their benefits and harms. Technologies using artificial intelligence (AI) are being developed to examine and optimise the use of opioids. Yet, this research has not been synthesised to determine the types of AI models being developed and the application of these models.MethodsWe aimed to synthesise studies exploring the use of AI in people taking opioids. We searched three databases: the Cochrane Database of Systematic Reviews, Embase and Medline on 4 January 2021. Studies were included if they were published after 2010, conducted in a real-life community setting involving humans and used AI to understand opioid use. Data on the types and applications of AI models were extracted and descriptively analysed.ResultsEighty-one articles were included in our review, representing over 5.3 million participants and 14.6 million social media posts. Most (93%) studies were conducted in the USA. The types of AI technologies included natural language processing (46%) and a range of machine learning algorithms, the most common being random forest algorithms (36%). AI was predominately applied for the surveillance and monitoring of opioids (46%), followed by risk prediction (42%), pain management (10%) and patient support (2%). Few of the AI models were ready for adoption, with most (62%) being in preliminary stages.ConclusionsMany AI models are being developed and applied to understand opioid use. However, there is a need for these AI technologies to be externally validated and robustly evaluated to determine whether they can improve the use and safety of opioids.

, conducted evaluations with the highest and lowest rates of positive reimbursement decisions: 100% (n = 29) and 57% (n = 30), respectively. No significant association was found between ERG group and reimbursement decision (p = 0.0930). Review year, oncology status, and review type were all tested as possible confounders. Of the models tested, the only significant relationship found was between review type and reimbursement decision, with indications reviewed through the MTA process being 60% less likely to receive reimbursement than those reviewed through the STA process (HSTs omitted from the model because of over-fitting). While reviews evaluated by different ERG groups had different rates of reimbursement, the difference was not statistically significant. The relationship between review type and reimbursement decision should be explored in future research.