Relay model for recruiting alcohol dependent patients in general hospitals--a single-blind pragmatic randomized trial

The identification and treatment of alcohol problems in primary care (iTAPP) study: protocol for a stepped wedge cluster randomized control trial testing the 15-method in a primary care setting

BACKGROUND

The 15-method is a targeted screening and treatment approach for alcohol problems in primary care. The 15-method used in primary care has proven as effective as specialized treatment for mild to moderate alcohol dependence in Sweden. A feasibility study of the 15-method in Danish primary care found the method acceptable and feasible.

AIMS

To evaluate the effectiveness of the 15-method in a Danish primary care setting in (1) lowering the proportion of patients exceeding the Danish low-risk alcohol consumption limit of ten standard units per week and a maximum of four standard units on a single day for men and women, and (2) increasing the likelihood of alcohol use being addressed during a consultation in general practice. Further, the rate of prescribed pharmacological treatment for alcohol problems (Disulfiram, Naltrexone, Acamprosate, and Nalmefene) will be measured along with the use of the biomarkers Alanine Transaminase and Gamma-Glutamyl Transferase.

METHODS

Stepped wedge cluster randomized controlled trial in sixteen general practices in the Region of Southern Denmark. Following a three-month baseline, the practices are randomly assigned to launch dates in one of four clusters. General practitioners and nurses receive three hours of training in the 15-method before launch. Patient questionnaires will collect data on alcohol consumption levels among patients affiliated with the practices. The healthcare professionals will register consultations in which alcohol is addressed in their patient filing system. Pharmacological treatment rates and the use of biomarkers will be collected through Danish national registries. The study follows the Medical Research Council's guidelines for developing and evaluating complex interventions.

DISCUSSION

From the patient's perspective, the 15-method may help identify alcohol-related problems at an earlier stage with flexible treatment offers in a familiar setting. For healthcare professionals, it addresses a traditionally challenging topic by equipping them with concrete tools, communication training, and clear treatment directives. From a societal perspective, primary care holds a unique position to identify hazardous and harmful alcohol use across different age groups, with potential public health and economic benefits through early identification and intervention.

TRIAL REGISTRATION

Clinicaltrials.gov NCT05916027. Retrospectively registered 22 June 2023.

AUD-DSS: a decision support system for early detection of patients with alcohol use disorder

BACKGROUND

Alcohol use disorder (AUD) causes significant morbidity, mortality, and injuries. According to reports, approximately 5% of all registered deaths in Denmark could be due to AUD. The problem is compounded by the late identification of patients with AUD, a situation that can cause enormous problems, from psychological to physical to economic problems. Many individuals suffering from AUD never undergo specialist treatment during their addiction due to obstacles such as taboo and the poor performance of current screening tools. Therefore, there is a lack of rapid intervention. This can be mitigated by the early detection of patients with AUD. A clinical decision support system (DSS) powered by machine learning (ML) methods can be used to diagnose patients' AUD status earlier.

METHODS

This study proposes an effective AUD prediction model (AUDPM), which can be used in a DSS. The proposed model consists of four distinct components: (1) imputation to address missing values using the k-nearest neighbours approach, (2) recursive feature elimination with cross validation to select the most relevant subset of features, (3) a hybrid synthetic minority oversampling technique-edited nearest neighbour approach to remove noise and balance the distribution of the training data, and (4) an ML model for the early detection of patients with AUD. Two data sources, including a questionnaire and electronic health records of 2571 patients, were collected from Odense University Hospital in the Region of Southern Denmark for the AUD-Dataset. Then, the AUD-Dataset was used to build ML models. The results of different ML models, such as support vector machine, K-nearest neighbour, decision tree, random forest, and extreme gradient boosting, were compared. Finally, a combination of all these models in an ensemble learning approach was selected for the AUDPM.

RESULTS

The results revealed that the proposed ensemble AUDPM outperformed other single models and our previous study results, achieving 0.96, 0.94, 0.95, and 0.97 precision, recall, F1-score, and accuracy, respectively. In addition, we designed and developed an AUD-DSS prototype.

CONCLUSION

It was shown that our proposed AUDPM achieved high classification performance. In addition, we identified clinical factors related to the early detection of patients with AUD. The designed AUD-DSS is intended to be integrated into the existing Danish health care system to provide novel information to clinical staff if a patient shows signs of harmful alcohol use; in other words, it gives staff a good reason for having a conversation with patients for whom a conversation is relevant.

Identification of clinical factors related to prediction of alcohol use disorder from electronic health records using feature selection methods

BACKGROUND

High dimensionality in electronic health records (EHR) causes a significant computational problem for any systematic search for predictive, diagnostic, or prognostic patterns. Feature selection (FS) methods have been indicated to be effective in feature reduction as well as in identifying risk factors related to prediction of clinical disorders. This paper examines the prediction of patients with alcohol use disorder (AUD) using machine learning (ML) and attempts to identify risk factors related to the diagnosis of AUD.

METHODS

A FS framework consisting of two operational levels, base selectors and ensemble selectors. The first level consists of five FS methods: three filter methods, one wrapper method, and one embedded method. Base selector outputs are aggregated to develop four ensemble FS methods. The outputs of FS method were then fed into three ML algorithms: support vector machine (SVM), K-nearest neighbor (KNN), and random forest (RF) to compare and identify the best feature subset for the prediction of AUD from EHRs.

RESULTS

In terms of feature reduction, the embedded FS method could significantly reduce the number of features from 361 to 131. In terms of classification performance, RF based on 272 features selected by our proposed ensemble method (Union FS) with the highest accuracy in predicting patients with AUD, 96%, outperformed all other models in terms of AUROC, AUPRC, Precision, Recall, and F1-Score. Considering the limitations of embedded and wrapper methods, the best overall performance was achieved by our proposed Union Filter FS, which reduced the number of features to 223 and improved Precision, Recall, and F1-Score in RF from 0.77, 0.65, and 0.71 to 0.87, 0.81, and 0.84, respectively. Our findings indicate that, besides gender, age, and length of stay at the hospital, diagnosis related to digestive organs, bones, muscles and connective tissue, and the nervous systems are important clinical factors related to the prediction of patients with AUD.

CONCLUSION

Our proposed FS method could improve the classification performance significantly. It could identify clinical factors related to prediction of AUD from EHRs, thereby effectively helping clinical staff to identify and treat AUD patients and improving medical knowledge of the AUD condition. Moreover, the diversity of features among female and male patients as well as gender disparity were investigated using FS methods and ML techniques.

Making a bridge between general hospital and specialised community-based treatment for alcohol use disorder-A pragmatic randomised controlled trial

AIM

To investigate if more patients in the intervention group attended specialized alcohol treatment compared with a treatment-as-usual group (TAU).

METHODS

Pragmatic randomized controlled trial where consecutive patients, admitted to somatic hospitals, filled out a lifestyle questionnaire with the Alcohol Use Disorder Identification Test (AUDIT) embedded. Patients scoring 8+ on AUDIT were included in the study. Included patients were randomized to either a Danish screening brief intervention and referral to treatment (SBIRT) called the Relay model or TAU depending on date of admission. The Relay group was offered a brief alcohol intervention by an outreach alcohol therapist. Patients scoring 16 points and above on the AUDIT test also received referral to alcohol treatment. Outcome was attendance at specialized outpatient alcohol treatment centres after discharge from hospital. Information on patients was gathered from municipal databases at 18 months follow-up.

RESULTS

A total of 3534 patients completed the questionnaire, and 609 patients (17%) scored AUDIT 8+. 48 patients were lost to follow-up, and the final sample had 561 patients. Only 33 patients (6%) attended outpatient treatment at 18-months follow-up, but significantly more patients in the Relay group sought alcohol treatment than in the TAU group (OR = 2.5 [1.2;5.2] (p = 0.017)). Number needed to treat (NNT) was 20 [95% CI 11.2;112.3].

CONCLUSION

The Relay intervention was associated with more patients attending specialized treatment, but further research is needed to establish if general hospitals are an excellent platform for performing SBIRT.

Factors influencing whether nurses talk to somatic patients about their alcohol consumption

Aim

Many Danes drink so much that it is detrimental to their health. As they are at risk of suffering diseases which can lead to hospitalisation on somatic wards, hospitals are ideal arenas for identifying individuals whose alcohol consumption is excessive. However, literature points out that this identification rarely takes place in hospitals, and literature further suggests that the staff experience barriers to talking about alcohol use with their patients. The primary aim of this study is to identify potential factors that influence whether or not nurses talk to patients about their alcohol consumption on somatic wards. Secondarily, we wish to examine whether a screening project may affect the nurses' readiness to talk about alcohol use with their patients.

Methods

A Glaserian Grounded Theory Method was used to collect and analyse data in this qualitative study. Semi-structured one-to-one interviews were conducted with seven nurses from somatic departments at two Danish hospitals. All seven nurses were already taking part in an alcohol screening project.

Results

In the analysis of the interview material, four categories emerged: The Nurse, The Patient, The Ward and The Relay Study.

Conclusion

We identified a series of barriers and promoting factors for nurses to talk about alcohol use with patients in a hospital setting. The barriers and promoting factors emerged within four categories: The Nurse, The Patient, The Ward, and The Relay Study. The most important barrier to talking to patients about alcohol seemed to be factors within the nurses themselves, in particular personal experiences, lack of knowledge and lack of confidence. We found, however, that by participating in a screening project the nurses seemed to overcome some of these barriers.