Identifying Risk Factors for Drug Use in an Iranian Treatment Sample: A Prediction Approach Using Decision Trees

Prevalence of smoking, alcohol consumption, and drug abuse in Iranian adults: Results of Azar Cohort Study

Background

Substance abuse has significant health impacts on families and society as a whole. We aimed to provide detailed prevalence estimates of substance abuse among the Azar Cohort Study respondents in Tabriz, Iran.

Methods

Data on 15006 participants of Azar Cohort Study were analyzed. The variables included tobacco smoking, alcohol use, drug abuse, and socio-demographic characteristics. The prevalence of substance abuse (with a 95% confidence interval) was calculated using the direct standardization method.

Results

Overall, 9.3% and 6.2% of the participants were regular and heavy cigarette smokers, respectively. Also, 1.9% and 2.1% of participants reported a history of using illicit drugs and alcohol, respectively. Substance abuse was more prevalent among males than females. Substance abuse varied significantly with age and socioeconomic variables.

Conclusion

We identified specific demographic and socioeconomic groups with a higher prevalence of all studied behaviors. Such high-risk groups should be targeted when designing substance abuse prevention programs.

Comparison of decision tree with common machine learning models for prediction of biguanide and sulfonylurea poisoning in the United States: an analysis of the National Poison Data System

BACKGROUND

Biguanides and sulfonylurea are two classes of anti-diabetic medications that have commonly been prescribed all around the world. Diagnosis of biguanide and sulfonylurea exposures is based on history taking and physical examination; thus, physicians might misdiagnose these two different clinical settings. We aimed to conduct a study to develop a model based on decision tree analysis to help physicians better diagnose these poisoning cases.

METHODS

The National Poison Data System was used for this six-year retrospective cohort study.The decision tree model, common machine learning models multi layers perceptron, stochastic gradient descent (SGD), Adaboosting classiefier, linear support vector machine and ensembling methods including bagging, voting and stacking methods were used. The confusion matrix, precision, recall, specificity, f1-score, and accuracy were reported to evaluate the model's performance.

RESULTS

Of 6183 participants, 3336 patients (54.0%) were identified as biguanides exposures, and the remaining were those with sulfonylureas exposures. The decision tree model showed that the most important clinical findings defining biguanide and sulfonylurea exposures were hypoglycemia, abdominal pain, acidosis, diaphoresis, tremor, vomiting, diarrhea, age, and reasons for exposure. The specificity, precision, recall, f1-score, and accuracy of all models were greater than 86%, 89%, 88%, and 88%, respectively. The lowest values belong to SGD model. The decision tree model has a sensitivity (recall) of 93.3%, specificity of 92.8%, precision of 93.4%, f1_score of 93.3%, and accuracy of 93.3%.

CONCLUSION

Our results indicated that machine learning methods including decision tree and ensembling methods provide a precise prediction model to diagnose biguanides and sulfonylureas exposure.

A comparison of risk factors for relapse in opiate-related and stimulant-related substance use disorders: A cross-sectional multicenter study

INTRODUCTION

Recognizing the specific factors in relapse disorders related to each substance can help improve treatment methods and adopt more effective preventive strategies. This study aimed to compare the situational factors associated with relapse in opiate-related disorders with stimulant-related disorders (SRDs) of those referred to substance misuse treatment centers.

DESIGN

This study was a cross-section type.

METHODS

The study participants were 150 clients with SRDs and 150 with opiate-related disorders. Samples were selected using two stages random sampling method. Data were collected through a demographic questionnaire and the Inventory of Drug-Taking Situations (IDTS).

RESULTS

The mean score of IDTS in the two groups was significantly different (X̄₁  = 45.93 ± 11.12 vs. X̄₂  = 48.34 ± 15.07; t = 3.32, p < 0.01). The mean scores of 'unpleasant emotions,' 'physical discomfort,' 'conflict with others,' and 'social pressure to use and urge/temptations' subscales were significantly higher in the stimulant group than in the opiate group (p < 0.05). However, the mean of the testing' personal control' subscale was higher in the opiate group than in the stimulant group (p < 0.05).

CONCLUSION

This study reveals that despite some similarities, relapse-related situational factors in opiates and stimulants differ. Some situational factors, such as social pressure and coping with unpleasant emotions, play a more critical role in relapse to both stimulant and opiate groups.

Application of Artificial Intelligent Approach to Predict the Normal Boiling Point of Refrigerants

Due to the desirable and interesting applications of refrigerants in organic Rankine cycles, heat pumps, and refrigeration, engineers and researchers are becoming more interested in refrigerant properties. One of the most dominant thermophysical properties of these fluids is their normal boiling point (Tb). In the current study, a novel extreme learning method (ELM) and ensemble decision tree boosted algorithm (EDT Boosted) are proposed to forecast the normal boiling point from 16 different molecular groups and one topological index. To this end, a total of 334 data points of Tb are gathered to prepare and test ELM and EDT boosted algorithms. The visual and mathematical comparisons of model outputs and real Tb express that proposed models have great potential to predict Tb of refrigerant. Moreover, sensitivity analysis is applied to explain the effectiveness of input parameters on the determination of Tb for refrigerants.

Evaluation of the Negative Effects of Opium Tincture on Memory and Hippocampal Neurons in the Presence of Chicory Extract

Background

Due to the high prevalence of addiction in society and the need to its attention, various methods are used for addiction withdrawal. The side effects of some methods restrict their use and increase the risk of recurrence. One of the Iranian useful methods is consumption of opium tincture (OT) that may cause brain structure and memory defects. Hence, this study aimed the effects of different doses of OT on memory and hippocampal neurons with the use of an antioxidant such as various concentrations chicory.

Materials and Methods

In the present study, 70 Wistar rats were randomly divided into 10 groups and the effect of various doses of chicory extract and OT were assessed on memory by the passive avoidance test. The neurons and astrocyte cells numbers in dentate gyrus were investigated, using histological examination.

Results

In passive avoidance test, the total time in dark compartment was significantly more in groups with 100 and 75 μl OT compared with control and normal saline groups (P < 0.001). Traffic number results showed that there was a significant difference between T100 and control groups (P > 0.05). Moreover, initial latency time was significantly shorter in groups with 75 and 100 μl of OT compared with control and normal saline groups (P < 0.05). However, the presence 250 mg/kg of chicory increases granular layer thickness of dentate gyrus and number of neurons.

Conclusion

The use of 250 mg/kg of chicory extract may be promising strategy for inducing neurogenesis and this dose could prevent neural damage.

Utility of support vector machine and decision tree to identify the prognosis of metformin poisoning in the United States: analysis of National Poisoning Data System

BACKGROUND

With diabetes incidence growing globally and metformin still being the first-line for its treatment, metformin's toxicity and overdose have been increasing. Hence, its mortality rate is increasing. For the first time, we aimed to study the efficacy of machine learning algorithms in predicting the outcome of metformin poisoning using two well-known classification methods, including support vector machine (SVM) and decision tree (DT).

METHODS

This study is a retrospective cohort study of National Poison Data System (NPDS) data, the largest data repository of poisoning cases in the United States. The SVM and DT algorithms were developed using training and test datasets. We also used precision-recall and ROC curves and Area Under the Curve value (AUC) for model evaluation.

RESULTS

Our model showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, and renal failure are the most important determinants in terms of outcome prediction of metformin poisoning. The average negative predictive value for the decision tree and SVM models was 92.30 and 93.30. The AUC of the ROC curve of the decision tree for major, minor, and moderate outcomes was 0.92, 0.92, and 0.89, respectively. While this figure of SVM model for major, minor, and moderate outcomes was 0.98, 0.90, and 0.82, respectively.

CONCLUSIONS

In order to predict the prognosis of metformin poisoning, machine learning algorithms might help clinicians in the management and follow-up of metformin poisoning cases.

Decision tree outcome prediction of acute acetaminophen exposure in the United States: A study of 30,000 cases from the National Poison Data System

Acetaminophen is one of the most commonly used analgesic drugs in the United States. However, the outcomes of acute acetaminophen overdose might be very serious in some cases. Therefore, prediction of the outcomes of acute acetaminophen exposure is crucial. This study is a 6-year retrospective cohort study using National Poison Data System (NPDS) data. A decision tree algorithm was used to determine the risk predictors of acetaminophen exposure. The decision tree model had an accuracy of 0.839, an accuracy of 0.836, a recall of 0.72, a specificity of 0.86 and an F1_score of 0.76 for the test group and an accuracy of 0.848, a recall of 0.85, a recall of 0.74, a specificity of 0.87 and an F1_score of 0.78 for the training group. Our results showed that elevated serum levels of liver enzymes, other liver function test abnormality, anorexia, acidosis, electrolyte abnormality, increased bilirubin, coagulopathy, abdominal pain, coma, increased anion gap, tachycardia and hypotension were the most important factors in determining the outcome of acute acetaminophen exposure. Therefore, the decision tree model is a reliable approach in determining the prognosis of acetaminophen exposure cases and can be used in an emergency room or during hospitalization.

Prevalence of Substance Use among Psychotic Patients and Determining Its Strongest Predictor

Objective: Although comorbidity of psychotic disorders and substance use can lead to increase in mortality, less is known about the outbreak and predictors. Psychotic patients tend to be overlooked during assessment; hence, the possibility of an undertreated or missed condition such as increasing substance use. This investigation aimed to measure the prevalence of substance use in psychotic patients and to survey the powerful predictors.

Method: In a 1-year cross-sectional study, 311 psychotic patients were assessed using the Structured Interview Based on DSM-5 for diagnostic confirmation as well as questions surveying prevalence and possible predictors of substance use.

Results: Prevalence of substance use among psychotic patients was 37.9%. Several variables were identified as factors associated with drug abuse among the psychotic patients. These included male gender, younger age, being currently homeless, a history of imprisonment, and having family history of drug use. The strongest predictors of substance use, however, were family history of drug use, male gender, and being currently homelessness.

Conclusion: Policymakers should note the importance of substance use among psychotic patients. Developing active screening strategies and comprehensive preventive plans, especially in the high-risk population, is suggested.

The effect of crocin (the main active saffron constituent) on the cognitive functions, craving, and withdrawal syndrome in opioid patients under methadone maintenance treatment

Individuals under methadone maintenance treatment (MMT) programs are susceptible to several complications, including withdrawal syndrome, craving, and cognitive deficits. This study was designed to elevate the effect of crocin administration on withdrawal syndrome, craving, and cognitive function in subjects under MMT programs. It was a clinical trial that was conducted among 60 patients referred to Soltan Mirahmad Clinic for addict patients in Kashan, Iran. The patients were allocated to two groups including placebo and intervention groups. The intervention group received 30 mg/day crocin (n = 30) and placebo (n = 30) once a day, in 12 weeks. Withdrawal syndrome, craving, and cognitive function parameters were measured before and after the intervention in subjects under MMT programs. Compared with the placebo group, crocin resulted in a significant improvement in craving score (p = .03), and withdrawal symptoms score (p = .01) in the intervention group. In addition, crocin supplementation did not affect cognitive function parameters (e.g., TMT, FAS test, and DGSP score). Overall, crocin supplementation for 12 weeks to patients under MMT programs had beneficial effects on craving and withdrawal symptoms score, but did not affect the cognitive function parameters.

Prevalence of drug use, alcohol consumption, cigarette smoking and measure of socioeconomic-related inequalities of drug use among Iranian people: findings from a national survey

Background

Drug use can lead to several psychological, medical and social complications. The current study aimed to measure and decomposes socioeconomic-related inequalities in drug use among adults in Iran.

Methods

This was a cross-sectional study The PERSIAN Cohort is the largest and most important cohort among 18 distinct areas of Iran. This study was conducted on 130,570 adults 35 years and older. A structured questionnaire was applied to collect data. The concentration index (C) was used to quantify and decompose socioeconomic inequalities in drug use.

Results

The prevalence experience of drug use was 11.9%. The estimated C for drug use was − 0.021. The corresponding value of the C for women and men were − 0.171 and − 0.134, respectively. The negative values of the C suggest that drug use is more concentrated among the population with low socioeconomic status in Iran (p < 0.001). For women, socioeconomic status (SES) (26.37%), province residence (− 22.38%) and age (9.76%) had the most significant contribution to socioeconomic inequality in drug use, respectively. For men, SES (80.04%), smoking (32.04%) and alcohol consumption (− 12.37%) were the main contributors to socioeconomic inequality in drug use.

Conclusions

Our study indicated that drug use prevention programs in Iran should focus on socioeconomically disadvantaged population. Our finding could be useful for health policy maker to design and implement effective preventative programs to protect Iranian population against the drug use.

Loneliness, hopelessness and suicide in later life: a case-control psychological autopsy study in rural China

AIMS

Loneliness is increasingly recognised as a serious public health issue worldwide. However, there is scarce research addressing the association between loneliness and suicide in older adults in rural China. We set out to examine loneliness and other psychosocial factors in elderly suicide cases and explore their interaction effects.

METHODS

Using a 1 : 1 matched case-control design, data were collected from 242 elderly suicide cases and 242 living community controls by psychological autopsy method in rural China, including demographic characteristics, loneliness, depression, hopelessness and social support. The chi-square automatic interaction detection (CHAID) tree model and multivariable logistic regression analysis were used to explore the relationships of these factors and suicide.

RESULTS

The CHAID tree model showed that loneliness, hopelessness and depressive symptoms were closely associated with completed suicide and that loneliness and hopelessness interacted with each other. The result of multivariable logistic regression showed that individuals who were unemployed [odds ratio (OR) = 2.344; 95% confidence interval (CI): 1.233-4.457], living alone (OR = 2.176; 95% CI: 1.113-4.254), had lower levels of subjective social support (OR = 2.185; 95% CI: 1.243-3.843), experienced depressive symptoms (OR = 6.700; 95% CI: 3.405-13.182), showed higher levels of hopelessness (OR = 7.253; 95% CI: 3.764-13.974) and felt higher levels of hopelessness × higher levels of loneliness (OR = 2.446; 95% CI: 1.089-5.492) were significantly associated with an elevated suicide risk in older people in rural China.

CONCLUSIONS

Regular evaluation of loneliness, hopelessness and depression can help detect older adults who are at risk of committing suicide. Interventions should target social support systems, particularly among people living alone, to alleviate feelings of loneliness and hopelessness. Treating depression is also key to preventing suicide among elderly people in rural China.

The Effect of Transcranial Direct Current Stimulation on Relapse, Anxiety, and Depression in Patients With Opioid Dependence Under Methadone Maintenance Treatment: A Pilot Study

Background and Objective

Patients under methadone maintenance therapy (MMT) are susceptible to several complications including mental disturbances and risk of relapse. The present study was designed to evaluate the effects of tDCS on relapse, depression, and anxiety of opioid-dependent patients under methadone maintenance treatment (MMT).

Methods

It was a randomized-clinical trial that conducted among 27 male patients referred to the outpatient addiction clinic of Ibn-e-Sina psychiatric hospital in Mashhad from July 2018 to May 2019. Participants were allocated to two treatment groups including intervention and sham groups. The intervention group received seven sessions of tDCS, in the F3 (cathode) and F4 (anode) areas of the brain, each one lasts 20 min, in two consecutive weeks. Depression, anxiety, and stress scale-21 (DASS-21) were measured before, during, and after the intervention in patients under MMT. Relapse on the morphine, cannabis, and methamphetamine was screened by urine dipstick tests of morphine, cannabis, and methamphetamine.

Results

Depression, anxiety, and stress of participants were significantly reduced in the intervention group compared with the control after the seventh session of tDCS (P < 0.001, P=0.01, and P=0.01, respectively). In addition, the relapse rate showed no significant changes between the two groups (P=0.33).

Conclusion

Overall, our study demonstrated that depression, anxiety, and stress of participants were significantly reduced after the seventh session of tDCS, but did not affect on the relapse rate. Therefore, it can be applied as a safe and effective technique to relieve mental disorder among receiving MMT.

Clinical Trial Registration

http://www.irct.ir, identifier IRCT20180604039979N1.

Prediction and Sensitivity Analysis of Bubble Dissolution Time in 3D Selective Laser Sintering Using Ensemble Decision Trees

The presence of defects like gas bubble in fabricated parts is inherent in the selective laser sintering process and the prediction of bubble shrinkage dynamics is crucial. In this paper, two artificial intelligence (AI) models based on Decision Trees algorithm were constructed in order to predict bubble dissolution time, namely the Ensemble Bagged Trees (EDT Bagged) and Ensemble Boosted Trees (EDT Boosted). A metadata including 68644 data were generated with the help of our previously developed numerical tool. The AI models used the initial bubble size, external domain size, diffusion coefficient, surface tension, viscosity, initial concentration, and chamber pressure as input parameters, whereas bubble dissolution time was considered as output variable. Evaluation of the models’ performance was achieved by criteria such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE) and coefficient of determination (R²). The results showed that EDT Bagged outperformed EDT Boosted. Sensitivity analysis was then conducted thanks to the Monte Carlo approach and it was found that three most important inputs for the problem were the diffusion coefficient, initial concentration, and bubble initial size. This study might help in quick prediction of bubble dissolution time to improve the production quality from industry.

Drug-induced prolonged corrected QT interval in patients with methadone and opium overdose

BACKGROUND

Iran is a country with the highest rate of opioid addiction in the world. The most commonly used opioid in Iran is opium, and methadone is in second place. The trend of drug use has changed from opium to methadone from 2006 to 2011. Presence of a large number of addicted people and methadone maintenance therapy clinics make methadone readily available in Iran. Therefore, evaluation of the epidemiological characteristic of methadone toxicity and its effects on the heart is essential.

METHODS

In This cross-sectional, retrospective, descriptive, analytical study all patients with methadone or opium toxicity who had been admitted to Vasei hospital, Sabzevar, Iran, during the years 2015 and 2016 were included, and their records were evaluated. Demographic data, addiction history, underlying diseases, and the outcome of admission were recorded. Then, corrected QT interval (QTc) of the first ECG of the patients after admission was evaluated.

RESULTS

The Majority of toxicities occurred in those above 30 years of age (71.4%), who lived in cities (62.8%), and were married (69.2%). A positive history of addiction was considerably higher in the opium group (72.3% versus 43.3%). There was no significant difference regarding QTc prolongation between patients with methadone and opium toxicity (p = 0.3).

CONCLUSION

QTc prolongation is one of the adverse effects of methadone or opium overdose. It seems that significant QTc prolongation is not uncommon among patients with opium overdose.

Comparing Three Data Mining Algorithms for Identifying the Associated Risk Factors of Type 2 Diabetes

Background:

Increasing the prevalence of type 2 diabetes has given rise to a global health burden and a concern among health service providers and health administrators. The current study aimed at developing and comparing some statistical models to identify the risk factors associated with type 2 diabetes. In this light, artificial neural network (ANN), support vector machines (SVMs), and multiple logistic regression (MLR) models were applied, using demographic, anthropometric, and biochemical characteristics, on a sample of 9528 individuals from Mashhad City in Iran.

Methods:

This study has randomly selected 6654 (70%) cases for training and reserved the remaining 2874 (30%) cases for testing. The three methods were compared with the help of ROC curve.

Results:

The prevalence rate of type 2 diabetes was 14% in our population. The ANN model had 78.7% accuracy, 63.1% sensitivity, and 81.2% specificity. Also, the values of these three parameters were 76.8%, 64.5%, and 78.9%, for SVM and 77.7%, 60.1%, and 80.5% for MLR. The area under the ROC curve was 0.71 for ANN, 0.73 for SVM, and 0.70 for MLR.

Conclusion:

Our findings showed that ANN performs better than the two models (SVM and MLR) and can be used effectively to identify the associated risk factors of type 2 diabetes.