Survey of awareness and beliefs about cancer (ABC) in Tehran Province, Iran.

INTRODUCTION: Knowledge, attitudes, and practices are essential measures for planning and evaluating cancer control programs. Little is known about these in Iran. METHODS: We conducted a population-based interview survey of adults aged 30-70 using the Farsi version of the Awareness and Beliefs about Cancer questionnaire in the capital province of Tehran, Iran, 2019. We calculated weighted estimates of levels of cancer knowledge, attitudes, and practices to allow for different selection probabilities and nonresponse. We used multivariate logistic regression to understand demographic factors associated with bowel, cervix, and breast screening practices. RESULTS: We interviewed 736 men and 744 women. The mean number of recalled cancer warning signs was less than one; 57.7% could not recall any cancer warning signs. Participants recognized 5.6 out of 11 early cancer warning signs and 8.8 of 13 cancer risk factors. Most (82.7%) did not know that HPV infection was a cancer risk factor. Approximately, half had negative attitudes towards cancer treatment, but over 80% had positive attitudes towards the effectiveness of screening for improving survival. Colorectal, breast, and cervical screening rates were 24%, 42%, and 49%, respectively. Higher socioeconomic status increased the odds of taking up screening for cancer. Women aged 60-70 were less likely to report taking up breast and cervical screening than younger women. DISCUSSION: The Iranian population has poor awareness and negative attitudes about cancer, and participation in screening programs is low. Public awareness and early detection of cancer should be promoted in Iran.

Development of a gastric cancer risk calculator for questionnaire-based surveillance of Iranian dyspeptic patients.

BACKGROUND: Gastric cancer (GC) is considered a silent killer, taking more than three quarters of a million lives annually. Therefore, prior to further costly and invasive diagnostic approaches, an initial GC risk screening is desperately in demand. METHODS: In order to develop a simple risk scoring system, the demographic and lifestyle indices from 858 GC and 1132 non-ulcer dyspeptic (NUD) patients were analysed. We applied a multivariate logistic regression approach to identify the association between our target predictors and GC versus NUD. The model performance in classification was assessed by receiver operating characteristic (ROC) analysis. Our questionnaire covering 64 predictors, included known risk factors, such as demographic features, dietary habits, self-reported medical status, narcotics use, and SES indicators. RESULTS: Our model segregated GC from NUD patients with the sensitivity, specificity, and accuracy rates of 85.89, 63.9, and 73.03%, respectively, which was confirmed in the development dataset (AUC equal to 86.37%, P < 0.0001). Predictors which contributed most to our GC risk calculator, based on risk scores (RS) and shared percentages (SP), included: 1) older age group [> 70 (RS:+ 241, SP:7.23), 60-70 (RS:+ 221, SP:6.60), 50-60 (RS:+ 134, SP:4.02), 2) history of gastrointestinal cancers (RS:+ 173, SP:5.19), 3) male gender (RS:+ 119, SP:3.55), 4) non-Fars ethnicity (RS:+ 89, SP:2.66), 5) illiteracy of both parents (RS:+ 78, SP:2.38), 6) rural residence (RS:+ 77, SP:2.3), and modifiable dietary behaviors (RS:+ 32 to + 53, SP:0.96 to 1.58). CONCLUSION: Our developed risk calculator provides a primary screening step, prior to the subsequent costly and invasive measures. Furthermore, public awareness regarding modifiable risk predictors may encourage and promote lifestyle adjustments and healthy behaviours.

Challenges and benefits of telepathology in education: lessons learned from COVID-19-a systematic review.

BACKGROUND: The COVID-19 pandemic in 2020 posed significant communication challenges, especially in the healthcare sector. Telepathology provides a valuable means for healthcare providers to communicate. This study investigated the key challenges and benefits of telepathology in education through a systematic review of relevant studies conducted during this period. METHODS: This systematic review was conducted in 2022. We utilized databases, including PubMed, Google Scholar and ScienceDirect. Our search was performed from 7 February 2022 to 13 February 2022. We selected articles based on inclusion criteria and used the Critical Appraisal Skills Program checklist to assess study strengths and limitations. We extracted data using a checklist and synthesized the results narratively. RESULTS: We initially identified 125 articles, and after screening, 15 were included in the study. These studies reported various challenges, including cost, technology, communication problems, educational difficulties, time wasting, legal issues and family distraction problems. Conversely, studies mentioned benefits, such as care improvement, better education, time efficiency, proper communication, cost and technology advancement. CONCLUSIONS: The results of this study will help future efforts and investigations to implement and set up telepathology. Based on our review, despite the challenges, the benefits of telepathology in education are greater than these obstacles, indicating its potential for future use.

Population attributable proportion and number of cancer cases attributed to potentially modifiable risk factors in Iran in 2020.

In the current study, we aimed to calculate the fraction of cancer attributable to modifiable risk factors in Iran in 2020. Population attributable fractions (PAFs) were calculated for established cancer risk factors using three data sources: the national cancer incidence reports, relative risks extracted from global and national meta-analyses, and exposure prevalence from national/subnational population-based surveys. In addition to overall cancers, the PAFs were estimated separately for each cancer site among men and women. Overall, 32.6% of cancers in 2020 in Iran were attributable to known risk factors. The PAF in men (40.2%) was twice as high as in women (21.1%). Cigarette smoking (15.4%), being overweight (5.0%), opium use (3.9%) and H. pylori infection (3.8%) were the leading causes of cancers. For men, the highest PAFs belonged to cigarette smoking (26.3%), opium use (6.8%) and being overweight (3.1%), while for women, the highest PAFs belonged to being overweight (7.2%), H. pylori infection (2.7%) and cigarette smoking (2.7%). Among Iranian men and women, the PAFs of waterpipe smoking were 2% and 0.9%, respectively. A third of incident cancers in Iran are due to modifiable exposures, mainly cigarette smoking, being overweight, and H. pylori infection. Opium consumption and waterpipe smoking collectively accounted for 8.8% of cancer occurrence in men and 1.3% in women in Iran. These emerging risk factors should be taken into consideration in future PAF studies.

Machine learning-based mortality prediction models for smoker COVID-19 patients.

BACKGROUND: The large number of SARS-Cov-2 cases during the COVID-19 global pandemic has burdened healthcare systems and created a shortage of resources and services. In recent years, mortality prediction models have shown a potential in alleviating this issue; however, these models are susceptible to biases in specific subpopulations with different risks of mortality, such as patients with prior history of smoking. The current study aims to develop a machine learning-based mortality prediction model for COVID-19 patients that have a history of smoking in the Iranian population. METHODS: A retrospective study was conducted across six medical centers between 18 and 2020 and 15 March 2022, comprised of 678 CT scans and laboratory-confirmed COVID-19 patients that had a history of smoking. Multiple machine learning models were developed using 10-fold cross-validation. The target variable was in-hospital mortality and input features included patient demographics, levels of care, vital signs, medications, and comorbidities. Two sets of models were developed for at-admission and post-admission predictions. Subsequently, the top five prediction models were selected from at-admission models and post-admission models and their probabilities were calibrated. RESULTS: The in-hospital mortality rate for smoker COVID-19 patients was 20.1%. For "at admission" models, the best-calibrated model was XGBoost which yielded an accuracy of 87.5% and F1 score of 86.2%. For the "post-admission" models, XGBoost also outperformed the rest with an accuracy of 90.5% and F1 score of 89.9%. Active smoking was among the most important features in patients' mortality prediction. CONCLUSION: Our machine learning-based mortality prediction models have the potential to be adapted for improving the management of smoker COVID-19 patients and predicting patients' chance of survival.

Use of sentiment analysis for capturing hospitalized cancer patients' experience from free-text comments in the Persian language.

PURPOSE: Today, the Internet provides access to many patients' experiences, which is crucial in assessing the quality of healthcare services. This paper introduces a model for detecting cancer patients' opinions about healthcare services in the Persian language, both positive and negative. METHOD: To achieve the objectives of this study, a combination of sentiment analysis (SA) and topic modeling approaches was employed. All pertinent comments made by cancer patients were collected from the patient feedback form of the Tehran University of Medical Science (TUMS) Cancer Institute (CI) in Iran, from March to October 2021. Conventional evaluation metrics such as accuracy, precision, recall, and F-measure were utilized to assess the performance of the proposed model. RESULT: The experimental findings revealed that the proposed SA model achieved accuracies of 89.3%, 92.6%, and 90.8% in detecting patients' sentiments towards general services, healthcare services, and life expectancy, respectively. Based on the topic modeling results, the topic "Metastasis" exhibited lower sentiment scores compared to other topics. Additionally, cancer patients expressed dissatisfaction with the current appointment booking service, while topics such as "Good experience," "Affable staff", and "Chemotherapy" garnered higher sentiment scores. CONCLUSION: The combined use of SA and topic modeling offers valuable insights into healthcare services. Policymakers can utilize the knowledge obtained from these topics and associated sentiments to enhance patient satisfaction with cancer institution services.

National surveillance of cancer survival in Iran (IRANCANSURV): Analysis of data of 15 cancer sites from nine population-based cancer registries.

Cancer survival is a key indicator for the national cancer control programs. However, survival data in the East Mediterranean region (EMR) are limited. We designed a national cancer survival study based on population-based cancer registries (PBCRs) from nine provinces in Iran. The current study reports 5-year net survival of 15 cancers in Iranian adults (15-99 years) during 2014 to 2015 in nine provinces of Iran. We used data linkages between the cancer registries and the causes of death registry and vital statistics and active follow-up approaches to ascertain the vital status of the patients. Five-year net survival was estimated through the relative survival analysis. We applied the international cancer survival standard weights for age standardization. Five-year survival was highest for prostate cancer (74.9%, 95% CI 73.0, 76.8), followed by breast (74.4%, 95% CI 72.50, 76.3), bladder (70.4%, 95% CI 69.0, 71.8) and cervix (65.2%, 95% CI 60.5, 69.6). Survival was below 25% for cancers of the pancreas, lung, liver, stomach and esophagus. Iranian cancer patients experience a relatively poor prognosis as compared to those in high-income countries. Implementation of early detection programs and improving the quality of care are required to improve the cancer survival among Iranian patients. Further studies are needed to monitor the outcomes of cancer patients in Iran and other EMR countries.

A cost-effectiveness modeling study of treatment interventions for stage I to III esophageal squamous cell carcinoma.

BACKGROUND: Esophageal cancer causes considerable costs for health systems. Appropriate treatment options for patients with esophageal squamous cell carcinoma (ESCC) can reduce medical costs and provide more improved outcomes for health systems and patients. This study evaluates the cost-effectiveness of treatment interventions for patients with ESCC according to the Iranian health system. MATERIAL AND METHODS: A five-state Markov model with a 15-year time horizon was performed to evaluate the cost-effectiveness of treatment interventions based on stage for ESCC patients. Costs ($US 2021) and outcomes were calculated from the Iranian health system, with a discount rate of 3%. One-way sensitivity analyses were performed to assess the potential effects of uncertain variables on the model results. RESULTS: In stage I, the Endoscopic Mucosal Resection (EMR) treatment yielded the lowest total costs and highest total QALY for a total of $1473 per QALY, making it the dominant strategy compared with esophagectomy and EMR followed by ablation. In stages II and III, chemoradiotherapy (CRT) followed by surgery dominated esophagectomy. CRT followed by surgery was also cost-effective with an incremental cost-effectiveness ratio (ICER) of $2172.8 per QALY compared to CRT. CONCLUSION: From the Iranian health system's perspective, EMR was the dominant strategy versus esophagectomy and EMR followed by ablation for ESCC patients in stage I. The CRT followed by surgery was a cost-effective intervention compared to CRT and esophagectomy in stages II and III.

Smartphone-based application for self-management of patients with colorectal cancer: development and usability evaluation.

PURPOSE: Self-management is considered essential for improving the treatment and management of colorectal cancer patients. This study was conducted to develop and evaluate the usability of a smartphone-based application for the self-management of patients with colorectal cancer. METHODS: We developed an Android-based application called "Colorectal Cancer Along (ColorectAlong)" for the self-management of colorectal cancer. Seventeen patients, from the Cancer Institute of Iran, evaluated the usability of this application for 2 weeks using a standard questionnaire for user interaction satisfaction (QUIS) version 5.5. The obtained data was analyzed using SPSS software version 26. RESULTS: The ColorectAlong application was developed with features covering several areas of colorectal cancer self-management. The areas available for evaluation in the app are patient profile; medication management; nutrition and diet management; pain management; mental health; smoking cessation and alcohol reduction; educational content; and adding reminders, questions, and notes. Our results indicate that the usability of this application is good with an average QUIS score of 8.03 out of 9. CONCLUSION: The ColorectAlong application can improve self-management, assist patients, and facilitate access to required information without the need for internet access.

Cost per DALY averted in low, middle- and high-income countries: evidence from the global burden of disease study to estimate the cost-effectiveness thresholds.

BACKGROUND: Determining the cost-effectiveness thresholds for healthcare interventions has been a severe challenge for policymakers, especially in low- and middle-income countries. This study aimed to estimate the cost per disability-adjusted life-year (DALY) averted for countries with different levels of Human Development Index (HDI) and Gross Domestic Product (GDP). METHODS: The data about DALYs, per capita health expenditure (HE), HDI, and GDP per capita were extracted for 176 countries during the years 2000 to 2016. Then we examined the trends on these variables. Panel regression analysis was performed to explore the correlation between DALY and HE per capita. The results of the regression models were used to calculate the cost per DALY averted for each country. RESULTS: Age-standardized rate (ASR) DALY (DALY per 100,000 population) had a nonlinear inverse correlation with HE per capita and a linear inverse correlation with HDI. One percent increase in HE per capita was associated with an average of 0.28, 0.24, 0.18, and 0.27% decrease on the ASR DALY in low HDI, medium HDI, high HDI, and very high HDI countries, respectively. The estimated cost per DALY averted was $998, $6522, $23,782, and $69,499 in low HDI, medium HDI, high HDI, and very high HDI countries. On average, the cost per DALY averted was 0.34 times the GDP per capita in low HDI countries. While in medium HDI, high HDI, and very high HDI countries, it was 0.67, 1.22, and 1.46 times the GDP per capita, respectively. CONCLUSIONS: This study suggests that the cost-effectiveness thresholds might be less than a GDP per capita in low and medium HDI countries and between one and two GDP per capita in high and very high HDI countries.

Automated Misspelling Detection and Correction in Persian Clinical Text.

Accurate electronic health records are important for clinical care, research, and patient safety assurance. Correction of misspelled words is required to ensure the correct interpretation of medical records. In the Persian language, the lack of automated misspelling detection and correction system is evident in the medicine and health care. In this article, we describe the development of an automated misspelling detection and correction system for radiology and ultrasound's free texts in the Persian language. To achieve our goal, we used n-gram language model and three different types of free texts related to abdominal and pelvic ultrasound, head and neck ultrasound, and breast ultrasound reports. Our system achieved the detection performance of up to 90.29% for radiology and ultrasound's free texts with the correction accuracy of 88.56%. Results indicated that high-quality spelling correction is possible in clinical reports. The system also achieved significant savings during the documentation process and final approval of the reports in the imaging department.

Cancer incidence in the East Azerbaijan province of Iran in 2015-2016: results of a population-based cancer registry.

BACKGROUND: Few countries in the Middle East have a population-based cancer registry, despite a clear need for accurate cancer statistics in this region. We therefore established a registry in the East Azerbaijan province, the sixth largest province in northwestern Iran. METHODS: We actively collected data from 20 counties, 62 cities, and 44 districts for the period between 20th March 2015 and 19th March 2016 (one Iranian solar year). The CanReg5 software was then used to estimate age-standardized incidence rates (ASRs) per 100,000 for all cancers and different cancer types. RESULTS: Data for 11,536 patients were identified, but we only analyzed data for 6655 cases after removing duplicates and non-residents. The ASR for all cancers, except non-melanoma skin cancer, was 167.1 per 100,000 males and 125.7 per 100,000 females. The most common cancers in men were stomach (ASR 29.7), colorectal (ASR 18.2), bladder (ASR 17.6), prostate (ASR 17.3), and lung (ASR 15.4) cancers; in women, they were breast (ASR 31.1), colorectal (ASR 13.7), stomach (13.3), thyroid (ASR 7.8), and esophageal (ASR 7.1) cancers. Both the death certificate rate (19.5%) and the microscopic verification rate (65%) indicated that the data for the cancer registry were of reasonable quality. CONCLUSION: The results of the East Azerbaijan Population-based Cancer Registry show a high incidence of cancer in this province, especially gastrointestinal cancers.

Economic burden of colorectal cancer in Iran in 2012.

Background: Colorectal cancer is one of the most common cancers in Iran. However little is known about the economic burden associated with this cancer in Iran. The aim of this study was to estimate the economic burden of colorectal cancer in Iran in the year 2012. Methods: We used the prevalence-based approach and estimated direct and indirect costs of all colorectal cancer cases in 2012. To estimate the total direct costs, we model the treatment process of colorectal cancer patients in initial and continuing phase in Iran. Then the average cost of each treatment in each phase was multiplied by the number of patients who received the treatment in the country in 2012. We used the human capital method to estimate the indirect costs. We extracted data from several sources, including national cancer registry reports, hospital records, literature review, occupational data, and interviews with experts. Results: The incidence and 5-year prevalence of colorectal cancer in Iran in 2012 were 7,163 and 22,591 individuals respectively. The economic burden of colorectal cancer in Iran was US$298,148,718 in 2012. Most of the cost (58%) was attributed to the mortality cost, and the direct medical cost accounted for 32.14 percent of the estimated total cost. The majority of the direct medical cost was associated with chemotherapy costs (50%). Conclusion: The economic burden of colorectal cancer in Iran is substantial and will increase in the future years.

Finding the PSA-based screening stopping age using prostate cancer risk.

BACKGROUND: Prostate Cancer screening was not rational for people who were suffered from other serious diseases and had a low quality of life. Biopsy and Prostate-Specific Antigen based screening also had imperfect information, pain, and costs. Finding the Prostate Cancer screening stopping age was important because after an age, Prostate-Specific Antigen test was not recommended and patients should not perform subsequent procedures. It could reduce the economic burden of Prostate Cancer. In this study, we modeled the effects of Prostate Cancer risk and comorbidities on the Prostate Cancer screening stopping age. METHODS: first, using a Markov model for PC progression, we provided a model for optimal Prostate Cancer screening stopping age. Second, we explored the relationship between comorbidities effects, Prostate Cancer risk and the stopping age. RESULTS: Our results suggest that the stopping age was an increasing function of PC risk and comorbidities effects. Screening should be stopped before 70 years. Finding showed that for men with diseases such as stroke or heart diseases, screening should not be performed at any age. CONCLUSIONS: Personalizing PC screening through paying more attention to PC risk can improve efficiency of screening. The role of personal characteristics such as race, family history, and previous PSA in PC screening decision-making was highlighted by stratifying men in different PC risk groups to find their stopping age. Incorporating comorbidity effects shows that severity of comorbidity was a crucial factor in PC screening stopping age decision-making process.

Determining the Content of a Melanoma Prevention and Care Mobile Application for Melanoma Patients: A Survey Study.

OBJECTS: Skin cancer is one of the most common cancers and melanoma is the deadliest type of skin cancer. Since the Mobile based health interventions have significant effects in the management of diseases. This study aimed to determining the content of melanoma prevention and care application. METHODS: This is a descriptive study that was conducted in three stages. at first, in order to identify the features of the application, search was conducted based on PubMed, Google Scholar ، Scopus databases.  In the second step, in order to determine the content of information, articles and guidelines related to melanoma were searched. In the third stage, in order to ensure the validity of researcher-made questionnaire, a survey was conducted by 51 experts based on targeted sampling. The collected data were analyzed. RESULTS: demographic and clinical information and melanoma prevention training was common in 14 reviewed applications; The possibility of sending reminders for patients was less considered. Most of the articles have paid attention to the data elements: age, gender, history of sunburn, skin color and history of cancer, exposure to ultraviolet rays, how to care for the skin. The content of the melanoma prevention and care application consists of three general sections: 1- Demographic, clinical information and melanoma risk factors; 2- Educational needs; 3- Features and capabilities of the application for prevention and care  were formed. The results of the expert survey showed that 81% of the section1, 80% of the section2, and 91% of the section3 were approved. CONCLUSIONS: Mobile applications has an effective role in disease management and accelerating the provision of health services and reducing the costs of providing services. Melanoma prevention and care application can be an aid in patient education and ultimately better disease management in the field of prevention and care during the current crisis.

Neoadjuvant Chemotherapy in Patients with HER2-Negative Breast Cancer: A Report from Clinical Breast Cancer Registry of Iran.

BACKGROUND: Neoadjuvant chemotherapy (NCT) has become an increasingly popular approach in management of breast cancer (BC). This study was conducted to evaluate the pathologic response and 36-month recurrence and survival rates of patients with human epidermal growth factor receptor 2 (HER2)-negative BC treated with different NCT regimens. METHODS: A total of 163 female patients with HER2-negative BC who received NCT during 2017-2020 were identified from the Clinical Breast Cancer Registry of Iran and entered the study. The prescribed NCT regimens included 4 cycles of doxorubicin plus cyclophosphamide, 4 cycles of doxorubicin plus cyclophosphamide followed by 4 cycles of paclitaxel, 4 cycles of doxorubicin plus cyclophosphamide followed by 4 cycles of docetaxel or 6 cycles of doxorubicin plus cyclophosphamide plus docetaxel (TAC). RESULTS: Thirty-two patients (19.6%) experienced pathologic complete response (pCR). TAC regimen, triple negative-BC and ki67>10% were significantly associated with increased pCR. The recurrence, overall survival (OS) and disease-free survival (DFS) rate at 36 months for all patients were 16.6%, 84.7% and 79.8%, respectively. Type of neoadjuvant regimen as well as age, hormone receptor status, Ki67, grade, clinical stage, type of surgery and pathologic response to chemotherapy did not significantly influence the survival and recurrence; however, TAC results in improved recurrence, OS and DFS rates. CONCLUSION: This study provides further evidence that NCT is a viable treatment option for patients with HER2-negative BC. The TAC regimen resulted in a significantly higher pCR rate compared to other regimens, but did not result in a significant improvement in recurrence, OS and DFS and rates.

Optimal age to stop prostate cancer screening and early detection.

BACKGROUND: Prostate Cancer screening should be discontinued at older ages because competing mortality risks eventually dominate the risk of Prostate Cancer and harms exceed benefits. We explored the Prostate Cancer screening stopping age from the patient, healthcare system, and social perspectives in Iran. METHODS: We applied Bellman Equations to formulate the net benefits biopsy and "do nothing". Using difference between the net benefits of two alternatives, we calculated the stopping age. The cancer states were without cancer, undetected cancer, detected cancer, metastatic cancer, and death. To move between states, we applied Markov property. Transition probabilities, rewards, and costs were inferred from the medical literature. The base-case scenario estimated the stopping age from the patient, healthcare system, and social perspectives. A one-way sensitivity used to find the most influential parameters on the stopping age. RESULTS: Our results suggested that Prostate Cancer screening stopping ages from the patient, healthcare system, and social were 70, 68, and 68 respectively. The univariate sensitivity analysis showed that the stopping ages were sensitive to the disutility of treatment, discount factor, the disutility of metastasis, the annual probability of death from other causes, and the annual probability of developing metastasis from the hidden cancer state. CONCLUSIONS: Men should not be screened for Prostate Cancer beyond 70 years old, as this results in the net benefit of "do nothing" above the biopsy. Nevertheless, this finding needs to be further studied with more detailed cancer progression models (considering re-biopsy, comorbidities, and more complicated states transition) and using local utility and willingness to pay value information.

Using Electronic Health Record System to Establish a National Patient's Registry : Lessons learned from the Cancer Registry in Iran.

BACKGROUND: In Iran, the Integrated Electronic Health Record system, called SEPAS, has been established to store all patient encounters of individuals referring to healthcare facilities. OBJECTIVE: We aimed to develop a model for cleaning SEPAS and applying its data in other databases. METHODS: We used cancer data from SEPAS as the sample. We developed a guideline to identify codes for cancer-related diagnoses and services in the database. Furthermore, we searched the SEPAS database based on ICD-10 and the diagnosis description in English and Farsi in an Excel sheet. We added codes and descriptions of pharmaceuticals and procedures to the list. We applied the above database and linked it to the patient records to identify cancer patients. A dashboard was designed based on this information for every cancer patient. RESULTS: We selected 5,841 diagnostic codes and phrases, 9,300 cancer pharmaceutics codes, and 452 codes from cancer-specific items related to the diagnostic procedures and treatment methods. Linkage of this list to the patient list generated a database of about 197,164 cancer patients for linkage in the registry database. CONCLUSIONS: Patient registries are one of the most important sources of information in healthcare systems. Data linkage between Electronic Health Record Systems (EHRs) and registries, despite its challenges, is profitable. EHRs can be used for case finding in any patient registry to reduce the time and cost of case finding.

Unraveling complex relationships between COVID-19 risk factors using machine learning based models for predicting mortality of hospitalized patients and identification of high-risk group: a large retrospective study.

Background: At the end of 2019, the coronavirus disease 2019 (COVID-19) pandemic increased the hospital burden of COVID-19 caused by the SARS-Cov-2 and became the most significant health challenge for nations worldwide. The severity and high mortality of COVID-19 have been correlated with various demographic characteristics and clinical manifestations. Prediction of mortality rate, identification of risk factors, and classification of patients played a crucial role in managing COVID-19 patients. Our purpose was to develop machine learning (ML)-based models for the prediction of mortality and severity among patients with COVID-19. Identifying the most important predictors and unraveling their relationships by classification of patients to the low-, moderate- and high-risk groups might guide prioritizing treatment decisions and a better understanding of interactions between factors. A detailed evaluation of patient data is believed to be important since COVID-19 resurgence is underway in many countries. Results: The findings of this study revealed that the ML-based statistically inspired modification of the partial least square (SIMPLS) method could predict the in-hospital mortality among COVID-19 patients. The prediction model was developed using 19 predictors including clinical variables, comorbidities, and blood markers with moderate predictability (Q2 = 0.24) to separate survivors and non-survivors. Oxygen saturation level, loss of consciousness, and chronic kidney disease (CKD) were the top mortality predictors. Correlation analysis showed different correlation patterns among predictors for each non-survivor and survivor cohort separately. The main prediction model was verified using other ML-based analyses with a high area under the curve (AUC) (0.81-0.93) and specificity (0.94-0.99). The obtained data revealed that the mortality prediction model can be different for males and females with diverse predictors. Patients were classified into four clusters of mortality risk and identified the patients at the highest risk of mortality, which accentuated the most significant predictors correlating with mortality. Conclusion: An ML model for predicting mortality among hospitalized COVID-19 patients was developed considering the interactions between factors that may reduce the complexity of clinical decision-making processes. The most predictive factors related to patient mortality were identified by assessing and classifying patients into different groups based on their sex and mortality risk (low-, moderate-, and high-risk groups).

Feasibility of measuring comorbidity indices based on clinical breast cancer records.

Background: Comorbidities have a significant impact on the treatment and outcome of breast cancer. However, data on such comorbidities from low-income countries are limited. Aim: To evaluate the feasibility and accuracy of comorbidity data extracted from medical records for estimating the prevalence of comorbidities among patients registered in the clinical breast cancer registry of the Islamic Republic of Iran. Methods: We collected data from the medical records of 400 patients on 30 comorbidities included in the Charlson Comorbidity Index (CCI) and Elixhauser Comorbidity Index (ECI). The sensitivity and specificity of comorbidity data extracted from medical records were calculated using data from interviews with 97 randomly selected patients. We studied the prevalence of comorbidities using the CCI and ECI. Data were analysed using SPSS version 24. Results: The mean age of patients was 51.69 SD 12.28 years. The sensitivity and specificity of medical records for detecting any comorbidity data contained in CCI versus non-comorbidity were 93.2% and 69.8%, respectively. However, for the comorbidity data included in ECI, both sensitivity (86.9%) and specificity (44.4%) were lower than in CCI. Hypertension (n = 144, 36.0%) and diabetes without chronic complications (n = 77, 19.3%) were the most prevalent comorbidities. A higher proportion of patients had no comorbidity with CCI (72.2%) than with ECI (44.8%). Conclusion: It is feasible to construct a comorbidity index using medical records with high accuracy, especially when we extract comorbidities using the CCI. Further studies are needed to understand the association between comorbidity index and breast cancer survival among Iranian patients.