Approximately 20% of children with attention-deficit hyperactivity disorder (ADHD) experience clinical levels of impairment into adulthood. In the UK, there is a sharp reduction in ADHD drug prescribing over the period of transition from child to adult services, which is higher than expected given estimates of ADHD persistence, and may be linked to difficulties in accessing adult services. Little is currently known about geographical variations in prescribing and how this may relate to service access.

To analyse geographic variations in primary care prescribing of ADHD medications over the transition period (age 16–19 years) and adult mental health service (AMHS) referrals, and illustrate their relationship with UK adult ADHD service locations.

Using a Clinical Practice Research Datalink cohort of people with an ADHD diagnosis aged 10–20 in 2005 (study period 2005–2013; n = 9390, 99% diagnosed <18 years), regional data on ADHD prescribing over the transition period and AMHS referrals, were mapped against adult ADHD services identified in a linked mapping study.

Differences were found by region in the mean age at cessation of ADHD prescribing, range 15.8–17.4 years (P<0.001), as well as in referral rates to AMHSs, range 4–21% (P<0.001). There was no obvious relationship between service provision and prescribing variation.

Clear regional differences were found in primary care prescribing over the transition period and in referrals to AMHSs. Taken together with service mapping, this suggests inequitable provision and is important information for those who commission and deliver services for adults with ADHD.

Lewy body dementia, consisting of both dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD), is considerably under-recognised clinically compared with its frequency in autopsy series.

This study investigated the clinical diagnostic pathways of patients with Lewy body dementia to assess if difficulties in diagnosis may be contributing to these differences.

We reviewed the medical notes of 74 people with DLB and 72 with non-DLB dementia matched for age, gender and cognitive performance, together with 38 people with PDD and 35 with Parkinson's disease, matched for age and gender, from two geographically distinct UK regions.

The cases of individuals with DLB took longer to reach a final diagnosis (1.2 v. 0.6 years, P = 0.017), underwent more scans (1.7 v. 1.2, P = 0.002) and had more alternative prior diagnoses (0.8 v. 0.4, P = 0.002), than the cases of those with non-DLB dementia. Individuals diagnosed in one region of the UK had significantly more core features (2.1 v. 1.5, P = 0.007) than those in the other region, and were less likely to have dopamine transporter imaging (P < 0.001). For patients with PDD, more than 1.4 years prior to receiving a dementia diagnosis: 46% (12 of 26) had documented impaired activities of daily living because of cognitive impairment, 57% (16 of 28) had cognitive impairment in multiple domains, with 38% (6 of 16) having both, and 39% (9 of 23) already receiving anti-dementia drugs.

Our results show the pathway to diagnosis of DLB is longer and more complex than for non-DLB dementia. There were also marked differences between regions in the thresholds clinicians adopt for diagnosing DLB and also in the use of dopamine transporter imaging. For PDD, a diagnosis of dementia was delayed well beyond symptom onset and even treatment.

Homeless and precariously housed individuals experience a high burden of comorbid illnesses, and excess mortality. Cross-sectional studies report a high rate of cognitive impairment. Long-term trajectories have not been well investigated in this group.

To longitudinally assess risks for premature and/or accelerated cognitive ageing, and the relationship with early mortality in homeless and precariously housed people.

This is a 9-year community-based study of 375 homeless and precariously housed individuals from Vancouver, Canada. Annual cognitive testing assessed verbal learning and memory, and inhibitory control. Linear mixed-effects models examined associations between clinical risk factors (traumatic brain injury, psychotic disorders, viral exposure, alcohol dependence) and cognitive change over 9 years. Cox regression models examined the association between cognition and mortality.

Traumatic brain injury and alcohol dependence were associated with decline in verbal memory. Inhibitory control declined, independent of risk factors and to a greater extent in those who died during the study. Better inhibitory control was associated with a 6.6% lower risk of mortality at study entry, with a 0.3% greater effect for each year of life. For each one-point increase in the Charlson Comorbidity Index score at study entry, the risk of mortality was 9.9% higher, and was consistent across age. Adjusting for comorbidities, inhibitory control remained a significant predictor of mortality.

Findings raise the possibility of a premature onset, and accelerated trajectory, of cognitive ageing in this group of homeless and precariously housed people. Traumatic brain injury, alcohol dependence and cognition could be treatment priorities.

In pharmacological doses, lithium successfully treats bipolar disorder and it can reduce violent crimes committed by individuals with this disorder.

To investigate whether naturally occurring lithium in drinking water lowers rates of violent crime in the general population.

We examined lithium levels in the drinking water of the 274 municipalities of Kyushu Island in Japan and compared these with the crime rates in each municipality.

We found that lithium levels were significantly and inversely associated with crime rates in 2009.

Our findings suggest that even very low levels of lithium in drinking water may play a role in reducing crime rates in the general population.

Patients receiving treatment for opioid-use disorder (OUD) may experience psychological symptoms without meeting full criteria for psychiatric disorders. The impact of these symptoms on treatment outcomes is unclear.

To determine the prevalence of psychological symptoms in a cohort of individuals receiving medication-assisted treatment for OUD and explore their association with patient characteristics and outcomes in treatment.

Data were collected from 2788 participants receiving ongoing treatment for OUD recruited in two Canadian prospective cohort studies. The Maudsley Addiction Profile psychological symptoms subscale was administered to all participants via face-to-face interviews. A subset of participants (n = 666) also received assessment for psychiatric disorders with the Mini International Neuropsychiatric Interview. We used linear regression analysis to explore factors associated with psychological symptom score.

The mean psychological symptom score was 12.6/40 (s.d. = 9.2). Participants with psychiatric comorbidity had higher scores than those without (mean 16.8 v. 8.6, P<0.001) and 31% of those with psychiatric comorbidity reported suicidal ideation. Higher psychological symptom score was associated with female gender (B = 1.59, 95% CI 0.92–2.25, P<0.001), antidepressant prescription (B = 4.35, 95% CI 3.61–5.09, P<0.001), percentage of opioid-positive urine screens (B = 0.02, 95% CI 0.01–0.03, P<0.001), and use of non-opioid substances (B = 1.92, 95% CI 0.89–2.95, P<0.001). Marriage and employment were associated with lower psychological symptoms.

Psychological symptoms are associated with treatment outcomes in this population and the prevalence of suicidal ideation is an area of concern. Our findings highlight the ongoing need to optimise integrated mental health and addictions services for patients with OUD.

Electroconvulsive therapy (ECT) is recommended in treatment guidelines as an efficacious therapy for treatment-resistant depression. However, it has been associated with loss of autobiographical memory and short-term reduction in new learning.

To provide clinically useful guidelines to aid clinicians in informing patients regarding the cognitive side-effects of ECT and in monitoring these during a course of ECT, using complex data.

A Committee of clinical and academic experts from Australia and New Zealand met to the discuss the key issues pertaining to ECT and cognitive side-effects. Evidence regarding cognitive side-effects was reviewed, as was the limited evidence regarding how to monitor them. Both issues were supplemented by the clinical experience of the authors.

Meta-analyses suggest that new learning is impaired immediately following ECT but that group mean scores return at least to baseline by 14 days after ECT. Other cognitive functions are generally unaffected. However, the finding of a mean score that is not reduced from baseline cannot be taken to indicate that impairment, particularly of new learning, cannot occur in individuals, particularly those who are at greater risk. Therefore, monitoring is still important. Evidence suggests that ECT does cause deficits in autobiographical memory. The evidence for schedules of testing to monitor cognitive side-effects is currently limited. We therefore make practical recommendations based on clinical experience.

Despite modern ECT techniques, cognitive side-effects remain an important issue, although their nature and degree remains to be clarified fully. In these circumstances it is useful for clinicians to have guidance regarding what to tell patients and how to monitor these side-effects clinically.

Many psychiatrists are worried their patients, at increased risk for COVID-19 complications, are precluded from receiving appropriate testing. There is a lack of epidemiological data on the associations between psychiatric disorders and COVID-19 testing rates and testing outcomes.

To compare COVID-19 testing probability and results among individuals with psychiatric disorders with those without such diagnoses, and to examine the associations between testing probability and results and psychiatric diagnoses.

This is a population-based study to perform association analyses of psychiatric disorder diagnoses with COVID-19 testing probability and such test results, by using two-sided Fisher exact tests and logistic regression. The population were UK Biobank participants who had undergone COVID-19 testing. The main outcomes were COVID-19 testing probability and COVID-19 test results.

Individuals with psychiatric disorders were overrepresented among the 1474 UK Biobank participants with test data: 23% of the COVID-19 test sample had a psychiatric diagnosis compared with 10% in the full cohort (P < 0.0001). This overrepresentation persisted for each of the specific psychiatric disorders tested. Furthermore, individuals with a psychiatric disorder (P = 0.01), particularly substance use disorder (P < 0.005), had negative test results significantly more often than individuals without psychiatric disorders. Sensitivity analyses confirmed our results.

In contrast with our hypotheses, UK Biobank participants with psychiatric disorders have been tested for COVID-19 more frequently than individuals without a psychiatric history. Among those tested, test outcomes were more frequently negative for registry participants with psychiatric disorders than in others, countering arguments that people with psychiatric disorders are particularly prone to contract the virus.

Premature mortality among patients experiencing forensic care is high. This paper examines the morbidity and mortality of all Scottish high secure patients in 1992/1993 and followed up 20 years later through the context of recovery.

To explore morbidity and delineate which patients are at greatest risk of premature mortality. To assess the extent of suicide and unnatural deaths. To establish which factors, if any, appear protective.

Health and mortality data were extracted from national data-sets and death categorised as premature or post-expected age. Standardised mortality ratios were calculated to explore natural, unnatural and suicide deaths with Cox regression conducted to explore baseline demographics and premature death.

During a mean follow-up of 21.1 years, 36.9% (n = 89) died, at an average age of 55.6 years. Of these, 70.8% (n = 63) died prematurely. Men lost on average 14.9 years and women 24.1 years of potential life. Five lives (5.6%) were lost by suicide and three (3.4%) by unnatural means.

In contrast to other mainstream and forensic cohorts, high rates of suicide and accidental deaths were not apparent. Risk of premature mortality is high. A greater focus upon physical health by community and in-patient services is essential.

Rapid spread of coronavirus disease 2019 (COVID-19) has affected people with intellectual disability disproportionately. Existing data does not provide enough information to understand factors associated with increased deaths in those with intellectual disability. Establishing who is at high risk is important in developing prevention strategies, given risk factors or comorbidities in people with intellectual disability may be different to those in the general population.

To identify comorbidities, demographic and clinical factors of those individuals with intellectual disability who have died from COVID-19.

An observational descriptive case series looking at deaths because of COVID-19 in people with intellectual disability was conducted. Along with established risk factors observed in the general population, possible specific risk factors and comorbidities in people with intellectual disability for deaths related to COVID-19 were examined. Comparisons between mild and moderate-to-profound intellectual disability subcohorts were undertaken.

Data on 66 deaths in individuals with intellectual disability were analysed. This group was younger (mean age 64 years) compared with the age of death in the general population because of COVID-19. High rates of moderate-to-profound intellectual disability (n = 43), epilepsy (n = 29), mental illness (n = 29), dysphagia (n = 23), Down syndrome (n = 20) and dementia (n = 15) were observed.

This is the first study exploring associations between possible risk factors and comorbidities found in COVID-19 deaths in people with intellectual disability. Our data provides insight into possible factors for deaths in people with intellectual disability. Some of the factors varied between the mild and moderate-to-profound intellectual disability groups. This highlights an urgent need for further systemic inquiry and study of the possible cumulative impact of these factors and comorbidities given the possibility of COVID-19 resurgence.

Patients often have very different ideas from clinicians about what they want treatments to achieve. Their views on what outcomes are important are not always reflected in trials.

To elicit the views of people who self-harm on the most commonly used outcome measures and to identify the outcomes that matter to them.

We conducted in-depth interviews with 18 people with histories of self-harm, recruited from hospital and community settings. We conducted thematic analysis using a framework approach and used visual mapping to arrive at our final analysis and interpretation.

Participants' accounts contained a number of challenges to the validity and meaningfulness of current trial outcome measures. Five broad issues emerged: (a) relationship between frequency and severity of self-harm; (b) behavioural substitution; (b) self-management skills; (d) the role of self-harm as survival tool and affect regulator, and (e) strategic self-presentation. We show how these affect the visibility and measurability of commonly used outcomes. The outcomes that mattered to participants focused on positive achievements in three domains: (a) general functioning and activities of everyday living; (b) social participation, and (c) engagement with services. Participants conceptualised these as both measures and means of sustained improvement.

Our findings suggest that current self-harm trial science rests on flawed assumptions about the relationship between mental states and behaviours and about our ability to measure both. Greater understanding of the outcomes that matter to people who self-harm is needed to inform both intervention development and trial design.

Postpartum depression is common in the perinatal period and poses a risk for the development of the infant and the mother–infant relationship. Infancy is a critical developmental period of life and supportive parenting is crucial for healthy development, however, the effects of interventions aimed at improving parenting among mothers with depression are uncertain.

To assess the effects of parenting interventions on parent–child relationship and child development among mothers with depressive symptoms with 0–12-month-old infants.

We conducted a systematic review with the inclusion criteria: (a) randomised controlled trials of structured psychosocial parenting interventions for women with depressive symptoms and a child aged 0–12 months in Western Organisation for Economic Co-operation and Development countries, (b) minimum three sessions with at least half of these delivered postnatally and (c) outcomes relating to the parent–child-relationship and/or child development. Publications were extracted from 10 databases in September 2018 and supplemented with grey search and hand search. We assessed risk of bias, calculated effect sizes and conducted meta-analysis.

Eight papers representing seven trials were included. We conducted meta-analysis on the post-intervention parent–child relationship. The analysis included six studies and showed no significant effect. For individual study outcomes, no significant effects on the majority of both the parent–child relationship and child development outcomes were reported.

No evidence of the effect of parenting interventions for mothers with depressive symptoms was found on the parent–child relationship and child development. Larger studies with follow-up assessments are needed, and future reviews should examine the effects in non-Western countries.

The common recommendation that adults with onset of mental illness after the age of 65 should receive specialised psychogeriatric treatment is based on limited evidence.

To compare factors related to psychiatric acute admission in older adults who have no previous psychiatric history (NPH) with that of those who have a previous psychiatric history (PPH).

Cross-sectional cohort study of 918 patients aged ≥65 years consecutively admitted to a general adult psychiatric acute unit from 2005 to 2014.

Patients in the NPH group (n = 526) were significantly older than those in the PPH group (n = 391) (77.6 v. 70.9 years P < 0.001), more likely to be men, married or widowed and admitted involuntarily. Diagnostic prevalence in the NPH and PPH groups were 49.0% v. 8.4% (P < 0.001) for organic mental disorders, 14.6% v. 30.4% (P < 0.001) for psychotic disorders, 30.2% v. 55.5% (P < 0.001) for affective disorders and 20.7% v. 13.3% (P = 0.003) for somatic disorders. The NPH group scored significantly higher on the Health of the Nation Outcome Scale (HoNOS) items agitated behaviour; cognitive problems; physical illness or disability and problems with activities of daily living, whereas those in the PPH group scored significantly higher on depressed mood. Although the PPH group were more likely to report suicidal ideation, those in the NPH group were more likely to have made a suicide attempt before the admission.

Among psychiatric patients >65 years, the subgroup with NPH were characterised by more physical frailty, somatic comorbidity and functional and cognitive impairment as well as higher rates of preadmission suicide attempts. Admitting facilities should be appropriately suited to manage their needs.

Major depressive episodes (MDEs) show diverse cortisol level alterations. Heterogeneity in symptom profiles, symptom severity and cortisol specimens may explain these heterogeneous results. Less severely ill out-patients with a non-melancholic MDE (NM-MDE) may have a variation in the rhythm of cortisol secretion rather than in its concentration.

Cortisol measures were taken (a) over a short-term period (12 h) by measuring daily salivary output using the area under the curve with respect to the ground (AUCg) and (b) over a long-term period (3 months) in hair. Additionally, cortisol reactivity measures in saliva – the cortisol awakening response and the 30 min delta cortisol secretion after awakening (DELTA) – were investigated in 19 patients with a melancholic MDE (M-MDE) and 52 with a NM-MDE, and in 40 matched controls who were recruited from the UK and Chile. Depression severity scores were correlated with different cortisol measures.

The NM-MDE group showed a decreased AUCg in comparison with controls (P = 0.02), but normal cortisol reactivity and long-term cortisol levels. The M-MDE group did not exhibit any significant cortisol alterations nor an association with depression severity scores. Higher Hamilton Rating Scale for Depression score was linked with decreased hair cortisol concentration (HCC, P = 0.05) and higher DELTA (P = 0.04) in NM-MDEs, whereas decreased HCC was the sole alteration associated with out-patients with severe M-MDEs.

The contrasting short- and long-term cortisol output results are compatible with an alteration in the rhythm of cortisol secretion in NM-MDEs. This alteration may consist of large and/or intense episodes of hypercortisolaemia in moderate NM-MDEs and frequent, but brief and sharp early-morning DELTAs in its severe form. These changes may reflect the effects of environmental factors or episodes of nocturnal hypercortisolaemia that were not measured by the short-term samples used in this study.

The steep rise in the rate of psychiatric hospital detentions in England is poorly understood.

To identify explanations for the rise in detentions in England since 1983; to test their plausibility and support from evidence; to develop an explanatory model for the rise in detentions.

Hypotheses to explain the rise in detentions were identified from previous literature and stakeholder consultation. We explored associations between national indicators for potential explanatory variables and detention rates in an ecological study. Relevant research was scoped and the plausibility of each hypothesis was rated. Finally, a logic model was developed to illustrate likely contributory factors and pathways to the increase in detentions.

Seventeen hypotheses related to social, service, legal and data-quality factors. Hypotheses supported by available evidence were: changes in legal approaches to patients without decision-making capacity but not actively objecting to admission; demographic changes; increasing psychiatric morbidity. Reductions in the availability or quality of community mental health services and changes in police practice may have contributed to the rise in detentions. Hypothesised factors not supported by evidence were: changes in community crisis care, compulsory community treatment and prescribing practice. Evidence was ambiguous or lacking for other explanations, including the impact of austerity measures and reductions in National Health Service in-patient bed numbers.

Better data are needed about the characteristics and service contexts of those detained. Our logic model highlights likely contributory factors to the rise in detentions in England, priorities for future research and potential policy targets for reducing detentions.

Depression is characterised by negative views of the self. Antidepressant treatment may remediate negative self-schema through increasing processing of positive information about the self. Changes in affective processing during social interactions may increase expression of prosocial behaviours, improving interpersonal communications.

To examine whether acute administration of citalopram is associated with an increase in positive affective learning biases about the self and prosocial behaviour.

Healthy volunteers (n = 41) were randomised to either an acute 20 mg dose of citalopram or matched placebo in a between-subjects double-blind design. Participants completed computer-based cognitive tasks designed to measure referential affective processing, social cognition and expression of prosocial behaviours.

Participants administered citalopram made more cooperative choices than those administered placebo in a prisoner's dilemma task (β = 20%, 95% CI: 2%, 37%). Exploratory analyses indicated that participants administered citalopram showed a positive bias when learning social evaluations about a friend (β = 4.06, 95% CI: 0.88, 7.24), but not about the self or a stranger. Similarly, exploratory analyses found evidence of increased recall of positive words and reduced recall of negative words about others (β = 2.41, 95% CI: 0.89, 3.93), but not the self, in the citalopram group.

Participants administered citalopram showed greater prosocial behaviours, increased positive recall and increased positive learning of social evaluations towards others. The increase in positive affective bias and prosocial behaviours towards others may, at least partially, be a mechanism of antidepressant effect. However, we found no evidence that citalopram influenced self-referential processing.

There are a number of candidates as useful outcomes in self-harm research – repetition of self-harm; symptom states; quality of life, social participation. Repetition of self-harm has been the predominant choice of researchers, not least because of its status as a risk for eventual suicide. Use of alternatives would respond to the preferences of study participants, but there are substantial methodological constraints. Nonetheless more attention could be given to the use of outcomes other than repetition. Another option is to broaden the approach to evaluation design, incorporating advances in the use of observational data either alone or linked to data from trials.

The density of information in digital health records offers new potential opportunities for automated prediction of cost-relevant outcomes.

We investigated the extent to which routinely recorded data held in the electronic health record (EHR) predict priority service outcomes and whether natural language processing tools enhance the predictions. We evaluated three high priority outcomes: in-patient duration, readmission following in-patient care and high service cost after first presentation.

We used data obtained from a clinical database derived from the EHR of a large mental healthcare provider within the UK. We combined structured data with text-derived data relating to diagnosis statements, medication and psychiatric symptomatology. Predictors of the three different clinical outcomes were modelled using logistic regression with performance evaluated against a validation set to derive areas under receiver operating characteristic curves.

In validation samples, the full models (using all available data) achieved areas under receiver operating characteristic curves between 0.59 and 0.85 (in-patient duration 0.63, readmission 0.59, high service use 0.85). Adding natural language processing-derived data to the models increased the variance explained across all clinical scenarios (observed increase in r2 = 12–46%).

EHR data offer the potential to improve routine clinical predictions by utilising previously inaccessible data. Of our scenarios, prediction of high service use after initial presentation achieved the highest performance.