Personalized Depression Treatment

Traditional therapies and medications don't work for a majority of patients suffering from depression. The individual approach to treatment could be the answer.

Cue is an intervention platform that converts passively acquired sensor data from smartphones into personalized micro-interventions for improving mental health. We examined the most effective-fitting personalized ML models to each person using Shapley values to discover their feature predictors. This revealed distinct features that deterministically changed mood over time.

Predictors of Mood

postnatal depression treatment is a major cause of mental illness around the world.1 Yet the majority of people affected receive treatment. To improve the outcomes, healthcare professionals must be able to identify and treat patients who have the highest chance of responding to particular treatments.

Personalized depression treatment is one method to achieve this. Utilizing sensors for mobile phones and an artificial intelligence voice assistant, and other digital tools, researchers at the University of Illinois Chicago (UIC) are developing new methods to predict which patients will benefit from the treatments they receive. Two grants worth more than $10 million will be used to discover biological and behavior indicators of response.

So far, the majority of research into predictors of depression treatment effectiveness has been focused on sociodemographic and clinical characteristics. These include demographics like age, gender, and education, and clinical characteristics like severity of symptom and comorbidities as well as biological markers.

While many of these variables can be predicted by the information in medical records, only a few studies have employed longitudinal data to determine predictors of mood in individuals. They have not taken into account the fact that mood varies significantly between individuals. Therefore, it is crucial to create methods that allow the identification of the individual differences in mood predictors and treatments effects.

The team's new approach uses daily, in-person evaluations of mood and lifestyle variables using a smartphone app called AWARE, a cognitive evaluation with the BiAffect app and electroencephalography -- an imaging technique that monitors brain activity. The team will then create algorithms to recognize patterns of behavior and emotions that are unique to each person.

In addition to these modalities the team created a machine learning algorithm to model the changing predictors of each person's depressed mood. The algorithm integrates the individual differences to produce a unique "digital genotype" for each participant.

This digital phenotype was associated with CAT DI scores, a psychometrically validated symptom severity scale. The correlation was low, however (Pearson r = 0,08; P-value adjusted for BH = 3.55 10 03) and varied significantly between individuals.

Predictors of symptoms

Depression is one of the most prevalent causes of disability1 yet it is often underdiagnosed and undertreated2. In addition an absence of effective treatments and stigmatization associated with depressive disorders stop many individuals from seeking help.

To help with personalized treatment, it is essential to identify predictors of symptoms. However, the methods used to predict symptoms are based on the clinical interview, which is unreliable and only detects a tiny number of symptoms that are associated with depression.2

Machine learning can improve the accuracy of the diagnosis and treatment of depression by combining continuous, digital behavioral phenotypes collected from smartphone sensors along with a verified mental health tracker online (the Computerized Adaptive Testing Depression Inventory CAT-DI). Digital phenotypes permit continuous, high-resolution measurements. They also capture a wide variety of unique behaviors and activity patterns that are difficult to capture through interviews.

The study involved University of California Los Angeles students with moderate to severe depression symptoms who were taking part in the Screening and Treatment for Anxiety and Depression program29 developed as part of the UCLA Depression Grand Challenge. Participants were referred to online support or in-person clinical treatment depending on their depression severity. Patients with a CAT DI score of 35 65 students were assigned online support with a coach and those with a score 75 patients were referred to in-person clinics for psychotherapy.

At the beginning of the interview, participants were asked a series of questions about their personal demographics and psychosocial characteristics. The questions asked included education, age, sex and gender and marital status, financial status and whether they were divorced or not, current suicidal thoughts, intentions or attempts, as well as how often they drank. Participants also rated their level of depression severity on a 0-100 scale using the CAT-DI. The CAT DI assessment was carried out every two weeks for participants who received online support, and weekly for those who received in-person care.

Predictors of the Reaction to Treatment

Research is focusing on personalized treatment for depression. Many studies are focused on finding predictors that can aid clinicians in identifying the most effective drugs for each person. Particularly, pharmacogenetics can identify genetic variations that affect how the body's metabolism reacts to antidepressants. This allows doctors select medications that are likely to be the most effective for each patient, while minimizing the time and effort needed for trials and errors, while avoiding any side consequences.

Another promising approach is building models of prediction using a variety of data sources, combining data from clinical studies and neural imaging data. These models can be used to determine the most appropriate combination of variables that is predictors of a specific outcome, like whether or not a particular medication will improve symptoms and mood. These models can be used to determine the patient's response to treatment, allowing doctors to maximize the effectiveness of their treatment.

A new generation of studies uses machine learning methods, such as supervised learning and classification algorithms (like regularized logistic regression or tree-based techniques) to combine the effects of multiple variables and improve the accuracy of predictive. These models have been demonstrated to be effective in predicting treatment outcomes for example, the response to antidepressants. These approaches are becoming more popular in psychiatry, and are likely to become the standard of future clinical practice.

In addition to ML-based prediction models research into the underlying mechanisms of depression continues. Recent research suggests that the disorder is linked with neurodegeneration in particular circuits. This suggests that an individual depression treatment will be focused on therapies that target these circuits in order to restore normal function.

One way to do this is by using internet-based programs which can offer an individualized and tailored experience for patients. One study found that a web-based program improved symptoms and improved quality of life for MDD patients. A controlled study that was randomized to a customized treatment resistant anxiety And depression for depression found that a significant percentage of patients experienced sustained improvement as well as fewer side negative effects.

Predictors of adverse effects

A major obstacle in individualized depression treatment involves identifying and predicting which antidepressant medications will have the least amount of side effects or none at all. Many patients take a trial-and-error approach, using a variety of medications being prescribed before settling on one that is effective and tolerable. Pharmacogenetics offers a new and exciting method to choose antidepressant medications that is more efficient and targeted.

Many predictors can be used to determine which antidepressant is best drug to treat anxiety and depression to prescribe, including gene variants, phenotypes of patients (e.g. sexual orientation, gender or ethnicity) and the presence of comorbidities. To determine the most reliable and accurate predictors for a specific treatment, random controlled trials with larger numbers of participants will be required. This is because the identifying of moderators or interaction effects may be much more difficult in trials that consider a single episode of treatment options for depression per participant, rather than multiple episodes of treatment over time.

Furthermore, the prediction of a patient's response to a particular medication is likely to require information on the symptom profile and comorbidities, as well as the patient's prior subjective experience with tolerability and efficacy. Currently, only a few easily measurable sociodemographic variables as well as clinical variables seem to be consistently associated with response to MDD. These include age, gender and race/ethnicity, SES, BMI and the presence of alexithymia.

Many challenges remain in the use of pharmacogenetics to treat depression. First is a thorough understanding of the underlying genetic mechanisms is needed as well as a clear definition of what is a reliable predictor of treatment response. Ethics, such as privacy, and the responsible use of genetic information are also important to consider. In the long-term the use of pharmacogenetics could be a way to lessen the stigma that surrounds mental health treatment and improve the treatment outcomes for patients with depression. But, like all approaches to psychiatry, careful consideration and implementation is required. At present, the most effective option is to offer patients an array of effective depression medications and encourage them to speak openly with their doctors about their concerns and experiences.