Depression is among the most common psychiatric disorders in Alzheimer's disease (AD). Nevertheless, given the overlap between the symptoms of depression and the symptoms of dementia, diagnosing depression is still problematic. Several depression rating scales have been validated for use in AD. Both the Hamilton Depression Rating Scale and the Montgomery-Asberg Depression Rating Scale have been used for screening purposes, to measure the severity of depression, and for assessing response to treatment. The recommendation to diagnose depression in AD is by using structured psychiatric interviews, such as the Structured Clinical Diagnostic Interview for DSM-IV or the Mini International Neuropsychiatric Interview. Based on information obtained from the structured interviews, depression is diagnosed using DSM-IV criteria. Consensus groups suggested specific changes to the diagnostic criteria to account for the overlap of symptoms between depression and dementia, and recent studies validated the DSM-IV criteria for major depression for use in AD.

Reliable assessment of suicide and self-harm risk in emergency medicine is critical for effective intervention and treatment of patients affected by mental health disorders. Teams of clinicians are faced with the challenge of rapidly integrating medical history, wide-ranging psychosocial factors, and real-time patient observations to inform diagnosis, treatment and referral decisions. Patient outcomes therefore depend on the reliable flow of information though networks of clinical staff and information systems. We studied information flow at a systems-level in a tertiary hospital emergency department using network models and machine learning. Data were gathered by mapping trajectories and recording clinical interactions for patients at suspected risk of suicide or self-harm. A network model constructed from the data revealed communities closely aligned with underlying clinical team structure. By analysing connectivity patterns in the network model we identified a vulnerability in the system with the potential to adversely impact information flow. We then developed an algorithmic strategy to mitigate this risk by targeted strengthening of links between clinical teams. Finally, we investigated a novel application of machine learning for distinguishing specific interactions along a patient’s trajectory which were most likely to precipitate a psychiatric referral. Together, our results demonstrate a new framework for assessing and reinforcing important information pathways that guide clinical decision processes and provide complimentary insights for improving clinical practice and operational models in emergency medicine for patients at risk of suicide or self-harm.

No abstract