|Year : 2020 | Volume
| Issue : 6 | Page : 2995-3004
A minimum data set of user profile or electronic health record for chemical warfare victims' recommender system
Elham Aalipour1, Marjan Ghazisaeedi2, Mohamad Reza Sedighi Moghadam3, Leila Shahmoradi4, Batool Mousavi3, Hamid Beigy5
1 Department of Health Information Management, School of Allied Medical Sciences, Tehran University of Medical Sciences, Tehran; Department of Health Information Technology, School of Allied Medical Sciences, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
2 Department of Health Information Management, School of Allied Medical Sciences, Tehran University of Medical Sciences; Evidence Based Medicine Research Center, Endocrinology and Metabolism Clinical Sciences Institute, Tehran University of Medical Sciences, Tehran, Iran
3 Janbazan Medical and Engineering Research Center, Tehran, Iran
4 Department of Health Information Management, School of Allied Medical Sciences, Tehran University of Medical Sciences; Halal Research Center of IRI, FDA, Tehran, Iran
5 Department of Computer Engineering, Sharif University of Technology, Tehran, Iran
|Date of Submission||15-Feb-2020|
|Date of Decision||13-Mar-2020|
|Date of Acceptance||07-Apr-2020|
|Date of Web Publication||30-Jun-2020|
Dr. Marjan Ghazisaeedi
Department of Health Information Management, School of Allied Medical Sciences, Keshavarz Blvd, Tehran University of Medical Sciences, Tehran
Source of Support: None, Conflict of Interest: None
Background: There are many people who are suffering from a variety of physical and mental illnesses due to the chemical attacks. There are various technologies such as recommender systems that can identify the main concerns related to health and make efforts to address them. To design and develop a recommender system, preparation of data source of this system should be considered. The aim of this study was to determine the minimum data set for user profile or user's electronic health record in chemical warfare victims' recommender system. Methods: This applied descriptive, cross-sectional study which was conducted in 2017. A questionnaire was developed by the authors from the data elements that were collected using the data extraction form from the studied sources. Content validity of the questionnaire was confirmed by using the experts. Test–retest method was used to determine the reliability of the questionnaire. The reliability of the questionnaire with Cronbach's alpha coefficient was confirmed as 84%. The questionnaire were submitted for related experts based on Delphi method by email or in person. Data resulting from the Delphi technique with descriptive statistics methods in SPSS software were analyzed. Results: Forty-seven nonclinical data elements and 181 clinical data elements were classified. Conclusion: Determining minimum data set of user profile or electronic health record in the recommender system for chemical warfare victims helps the health authorities to implement the recommender system which demonstrates chemical warfare victims' needs.
Keywords: Chemical warfare victim, minimum data set, recommender system
|How to cite this article:|
Aalipour E, Ghazisaeedi M, Sedighi Moghadam MR, Shahmoradi L, Mousavi B, Beigy H. A minimum data set of user profile or electronic health record for chemical warfare victims' recommender system. J Family Med Prim Care 2020;9:2995-3004
|How to cite this URL:|
Aalipour E, Ghazisaeedi M, Sedighi Moghadam MR, Shahmoradi L, Mousavi B, Beigy H. A minimum data set of user profile or electronic health record for chemical warfare victims' recommender system. J Family Med Prim Care [serial online] 2020 [cited 2020 Sep 25];9:2995-3004. Available from: http://www.jfmpc.com/text.asp?2020/9/6/2995/287858
| Introduction|| |
Using toxic and chemical substances as an effective weapon to succumb the enemy in wars has a long history. Some species of these weapons were used in ancient wars for hundreds of years BC.,, More than 90,000 people were killed and almost 1,300,000 ones were wounded in chemical operations based on new ways in the World War I (1914–1918)., Italy used chemical agents in the war against Ethiopia from 1935 to 1936., The Japanese dropped chemical bombs on Chinese soldiers around 1937–1943.,, In the Vietnam War in 1960s and the early years of 1970s, the Americans inflicted irreparable damages to Vietnam rangelands, forests, and the environment as well as its people through the use of hazardous chemical agents, especially Agent Orange, that despite major restoration plans, its ominous effects were not resolved after decades later.,
However, the most widespread and heinous event is the use of chemical agents recorded after World War I against the human race used by the Saddam's Baathist regime against Iran in the 1980s., There are many people in Iran who are suffering from a variety of physical and mental illnesses due to the chemical attacks that followed the 8-year war between Iran and Iraq. Based on the study by Taebi et al. in 2015, it was found that about 63,417 chemical veterans have records in the Foundation of Martyrs and Veterans Affairs in Iran.
Mustard gas was one of the main chemical agents used in the war imposed by Iraq against Iran, which its late-onset toxic effects occur several months to a few years later. Mustard gas causes many complications in the tissues of the gastrointestinal tract, endocrine, bone marrow, nervous, immune, especially respiratory, skin and eye. In addition, it has mutagenic and carcinogenic effects, the severity of these complications depends on the type of the exposure, the amount of exposure to mustard gas, the age of the individual, and the degree of one's immunity and resistance.,
Further, chemical bombardment had adverse psychological and social consequences on its victims. Psychological symptoms such as anxiety and depression among chemical warfare victims are highly prevalent. These victims need benefiting from certain services due to their special conditions. Chronic diseases affect all economic, social, financial, and emotional aspects of the individual, their family and society then only the pharmacological treatment and periodic control of the disease among these individuals are not enough.
Today, there are various technologies that can identify the main concerns related to health and make efforts to address them. Recommender system is software that its user preferences can be determined by data analysis and give the best recommendation to the user based on his/her status. Components of a recommender system such as data source, recommendations database, filtering techniques, and data analysis can mention.
Different data are daily produced at healthcare centers to show the status of the clients. Using all these data is time-consuming and is not cost-effective. Thus, the issue of preparing, adjusting, and standardizing a minimum dataset are propounded. Providing a minimum dataset in a manual or computer-based system can be the basis for unifying and integrating documented data from different institutions and systems for easy data comparison.,
The mandatory collection and reporting of integrated, standardized data as well as data exchange between organizations and individuals at national and even international level are among the objectives of minimum dataset. Providing a minimum data set helps to collect an appropriate and relevant data based on related goals of a wealth of data, and these standardized, key and core data allow comparability between the data needed to report and present the results of various organizations, institutions, and systems.,
Recommender systems are among the technologies that, if properly designed and implemented, can help health authorities manage physical and mental health of chemical warfare victims. To design and develop a recommender system for chemical warfare victims, preparation of data source of this system should be considered. In this research, data source is actually the user profile, which is considered as the user's electronic health record in the recommender system. The present collection points to the formation of minimum data set for user profile or user's electronic health record in health recommender system for chemical warfare victims.
| Methods|| |
This applied descriptive, cross-sectional study was conducted in 2017. In the first step, the print sources available and the websites of organizations and associations related to the health and general status of chemical warfare victims and war survivors were reviewed. In the second step, the articles in the SID, Magiran, PubMed, Google Scholar, ScienceDirect, Web of Science, Scopus databases were searched using chemical warfare victims or chemical war veterans or chemical war survivors and health recommender system or medical recommender system or medicine recommender system or healthcare recommender system and electronic health record or profile user and minimum data set or core data set or core data elements or essential data set keywords as well as the equivalent Persian words were searched without time limit. At this step, sampling was not performed and the retrieved sources were considered based on the inclusion criteria that the valid sources should be Farsi and English full text.
In the third step, the chemical warfare victims' records at Chemical Warfare Victims Affairs Center and Janbazan Medical and Engineering Research Center of Foundation of Martyrs and Veterans Affairs in Tehran were reviewed. In the fourth step, a questionnaire was developed by the authors from the data elements that were collected using the data extraction form from the studied sources. The questionnaire included a section on the identity information of the person completing the questionnaire and a section on questions about the importance of existing data elements in a five-point Likert scale (very high, high, medium, low, very low). Data elements are divided into two main sections, nonclinical and clinical. The nonclinical data elements section consisted of 47 data elements and the clinical data elements section contained of 181 data elements with 16 subclasses. The questionnaire also contained an open-ended question in the nonclinical and clinical data elements sections for data elements proposed by experts.
In the fifth step, content validity of the questionnaire was confirmed by using the opinions of available 4 health information management experts, 4 pulmonologists, 4 dermatologists, and 4 ophthalmologists. The pulmonologists, dermatologists, and ophthalmologists were selected because chemical warfare victims suffered from the most damage in their lung, skin, and eye based on the reliable sources.,, All available physicians were experienced in the treatment chemical warfare victims. In the sixth step, test–retest method was used to determine the reliability of the questionnaire and in fact the stability of tool. Thus, 10 homogeneous professionals with the ones who performed content validity were asked to review the questionnaire and after 10 days, the questionnaire was returned to them to review. The reliability of the questionnaire with Cronbach's alpha coefficient to determine internal consistency was confirmed as 84%. Chi-square test at a significance level of 5% was used to examine the significance of the data elements by experts in different disciplines, and no significant opinion difference was observed between experts.
In the seventh step, the questionnaire were submitted for faculty members with master degree in health information technology, medical record, and Ph.D. degree in health information management available in health information technology and management departments of the universities of medical sciences in Iran based on Delphi method by email or in person in Tehran to determine and approve the required minimum dataset of user profile or electronic health record of the recommender system for chemical warfare victims. The reason for selecting the faculty members was their skills in selecting the medical records data elements. In addition, the health authorities and physicians who had the most contact with chemical warfare victims were introduced by Chemical Warfare Victims Affairs Center of Foundation of Martyrs and Veterans Affairs in Tehran were polled in person.
It was decided that if 75% or more of the respondents (mean of 3.75 to 5) chose very high and high options for the importance of any data element in the electronic health record, this data element would be considered. If 50–75% of respondents (mean of 2.5–3.75) chose very high and high options, the proposed data element to be considered as the Delphi second step for the survey. If less of 50% of respondents (mean below 2.5 of 5), chose very high and high options for the data element, removing that data element. In the eighth step, analysis of the data resulting from the Delphi technique with descriptive statistics methods in SPSS version 25 was performed. In the ninth step, the information was presented in tables.
| Results|| |
Of 70 faculty members, only 52 filled questionnaires were returned. Of the total number of 13 health authorities and 19 physicians, 13 and 18 completed questionnaires were received, respectively, whose demographic characteristics are shown in [Table 1].
None of the data elements in the nonclinical class and clinical class were not entered the second round of Delphi polls and were not removed. The mean of the research population's views on the nonclinical and clinical data elements are shown in [Table 2] and [Table 3], respectively.
|Table 2: Nonclinical data elements of profile user or electronic health record in chemical warfare victims' recommender system|
Click here to view
|Table 3: Clinical data elements of profile user or electronic health record in chemical warfare victims' recommender system|
Click here to view
According to [Table 2], among 47 nonclinical data elements, the highest mean was for the cell phone number (5) and the lowest mean for ethnicity (3.79), religion (3.79), and sect (3.79). According to [Table 3], 181 clinical data elements were classified into 16 subclasses. Among clinical data elements in health status records, the highest mean is related to drug history (4.9) and the lowest mean belongs to address (3.8) and phone number of healthcare centers (3.8). Among the total body examinations data elements, the highest mean is related to the chief complaint (4.8), current disease history (4.8) and the lowest mean belongs to endocrine (4.2) and urinary tract system (4.2).
Among specialized dermatological examinations data elements, the highest mean is related to the chief complaint (4.8), treatment plan (4.8), dermatologist's final diagnosis (4.8), and the lowest mean belongs to the dermatologist's medical council number (4). Among the specialized ophthalmological examinations data elements, the highest mean is related to treatment plan (4.8), ophthalmologist's final diagnosis (4.8) and the lowest mean belongs to ophthalmologist's medical council number (4.1). Among the specialized pulmonary examinations data elements, the highest mean is related to chief complaint (4.9) and the lowest mean belongs to pulmonologist's medical council number (4.1).
Among the specialized cardiac examinations data elements, the highest mean is related to chief complaint (4.8), primary diagnosis (4.8), treatment plan (4.8), cardiologist's final diagnosis (4.8) and the lowest mean belongs to cardiologist's medical council number (4.1) and signature (4.1). Among the specialized psychiatric examinations data elements, the highest mean is related to chief complaint (4.8), treatment plan (4.8), psychiatrist's final diagnosis (4.8) and the lowest mean belongs to psychiatrist's medical council number (4.1) and signature (4.1). Among the specialized dental examinations data elements, the highest mean is related to the specialized dentist's final diagnosis (4.8) and the lowest mean belongs to the specialized dentist's signature (3.97). Among the specialized sports medical examinations data elements, the highest mean is related to chief complaint (4.7), primary diagnosis (4.7), treatment plan (4.7), sports medicine specialist's final diagnosis (4.7), and the lowest mean belongs to sports medicine specialist's medical council number (3.96).
Among specialized rehabilitation examinations data elements, the highest mean is related to chief complaint (4.8), primary diagnosis (4.8), treatment plan (4.8), rehabilitation specialist's final diagnosis (4.8), and the lowest mean belongs to rehabilitation specialist's medical council number (4.1) and signature (4.1). Among the nutrition counseling data elements, the highest mean is related to the main nutritional complaint (4.4), weight (4.4), body mass index (4.4), nutritional sensitivity record (4.4), final nutritional diagnosis (4.4), nutritional advice (4.4), and the lowest mean belongs to nutrition counselor's full name (3.97) and signature (3.97).
Among the medications data elements, the highest mean is related to the medication name of prescribed (4.8), and the lowest mean belongs to medication form (4.5) and the time of medication use (4.5). Among the laboratory tests data elements, the highest mean is related to the test name (4.7) and the test result (4.7), and the lowest mean belongs to date of the test run (4.6). Among the surgeries data elements, the highest mean is related to postoperative diagnosis (4.9) while the lowest mean belongs to the surgeon assistant's medical council number (3.79), signature (3.79), and operating room nurse's nursing council number (3.79). Among the injuries data elements, the highest mean is related to the type of injury (4.8), the nature of the injury (4.8), the severity of the injury (4.8), and the lowest mean belongs to the hour of injury occurred (4.2). Among the vaccinations data elements, the highest mean is related to the vaccine name (4.7) and the lowest mean belongs to injector person's name (3.79) and signature (3.79).
| Discussion|| |
The late effects of chemical warfare on the militaries and civilians' bodies and souls were confirmed. Electronic technologies have emerged in the health domain, for example, recommender systems can be used to monitor the health status of people instantly. Recommender systems are in fact electronic systems containing filtering techniques along with other components like data source, recommendations database, and data analysis techniques that offer recommendations according to the users' needs.,, Recommender systems also model users' preferences, needs, and behaviors to predict future users' preferences, needs, and behaviors in order to recommend useful and appropriate services to users based on it. Recommender systems can be divided into different types based on the techniques and approaches used, the most popular of which are content-based, collaborative, knowledge-based, and hybrid. In a content-based recommender system, when a particular user is working with the system, his or her activity history is stored in a part of the system called the user profile. Then, the system uses these records in its knowledge repositories to find items which are similar to items in the user profile and recommends them to a specific user. But, in the collaborative filtering recommender system, the system provides recommendations to a given user based on interests, needs, and behaviors of other users that are similar to those of the given user. The knowledge-based recommender system provides recommendations based on the perception they have of the needs, interests, and behaviors of the particular user and the characteristics of the specific user's items features. Knowledge-based systems use a variety of methods that are applicable to knowledge analysis such as genetic, fuzzy, neural network algorithms. Finally, hybrid recommender systems use a combination of various types of techniques mentioned above and offer recommendations.,,,
The authors could not find a study to determine the essential data for a user profile or electronic health record for chemical warfare victims' health recommender system. Therefore, the authors compared the data elements of health information systems intended for war victims or military affairs as well as the recommender systems designed and developed for the health field with the results of this study.
In the studies by Kraft MR, Hynes DM, and Bouhaddou O, et al., VistA (veterans health information systems and technology architecture) was introduced that maintains records of the outpatient and inpatient services of US military personnel and war veterans. In this study, pointed to data elements such as demographic characteristics, test result, radiology report, admission, discharge, transfer, visit, and medication that were similar with the results of the present study.,
In the studies by Perlin JB, Kolodner RM, Roswell RH., and Rajeevan N, et al., a portion of the American survivors' electronic health record of health management system, CPRS (Computerized patient record system), was introduced that enables health care providers to view and update war survivors' medical records. In the study pointed to data elements such as test, radiology, medication, and medical order that were similar with the results of the current study., In a study, MiCARE (Military Care) was introduced as a portal to manage of the US military electronic personal health record. In this study referred to data elements such as test results, allergies, medications, radiology reports, appointments, medical practice, medical problems lists, counseling reports, hospitalizations, and outpatient reporting which were similar with the results of the present study.
In another study, AHLTA (Armed Forces Health Longitudinal Technology Application) was introduced as an US military electronic health record for outpatient clinics. The study referred to data elements such as test results, radiology and medication reports, physician orders, client acceptance registrations that were similar with the results of the current study. In a study, Graber et al. pointed to data elements such as date of birth, gender, and weight, which were similar with the results of the present study, but Graber et al.' recommender system was for the treatment of psoriasis. Agapito et al. in a study referred to data elements such as age, sex, race, and weight that were similar with the results of the current study. But Agapito et al.' recommender system was to monitor nutrients for healthy people and patients with chronic diseases.
| Conclusion|| |
In the present world of electronic technologies, the presence of a recommender system may be useful for continuous monitoring of chemical warfare victims' health status and timely reflection of their needs to health authorities to plan for supplying chemical warfare victims' demands in less time and cost. Hence, determination of essential data for user profile or electronic health record of the recommender system is a starting point to achieve this importance. The basis of an efficient and effective health recommender system is the existence of appropriate and accurate essential data that can be used to design user profile.
This study was part of a PhD dissertation registered with IR.TUMS.SPH.REC.2017.1873 ethical code number in Tehran University of Medical Sciences. The authors express their gratitude to cooperate in this study specialty physicians, health authorities, and personnel of Sasan hospital, Chemical Warfare Victims Affairs Center, Janbazan Medical and Engineering Research Center of the Martyrs and Victims Foundation in Tehran, and faculty members of health information technology and management departments in universities of medical sciences in Iran.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Coleman K. A History of Chemical Warfare. New York: Palgrave Macmillan; 2005.
Delfino RT, Ribeiro TS, Figueroa-Villar JD. Organophosphorus compounds as chemical warfare agents: A review. J Braz Chem Soc 2009;20:407-28.
Haines DD, Fox SC. Acute and long-term impact of chemical weapons: Lessons from the Iran-Iraq war. Forensic Sci Rev 2014;26:97-114.
World Health Organization. Public Health Response to Biological and Chemical Weapons: WHO Guidance. 2nd
ed. Geneva: WHO; 2004.
Mangerich A, Esser C. Chemical warfare in the First World War: Reflections 100 years later. Arch Toxicol 2014;88:1909-11.
Pitschmann V. Overall view of chemical and biochemical weapons. Toxins 2014;6:1761-84.
Szinicz L. History of chemical and biological warfare agents. Toxicology 2005;214:167-81.
Grinberg D. Tracing toxic legacies: GIS and the dispersed violence of Agent Orange. J War Cult Stud 2018;11:38-57.
Kovačević D, Afrimadona, Claar M. Gas, power, and norms: Competing logics for the declining use of chemical warfare. Nonproliferation Rev 2019;26:251-66.
Ghasemi-Broumand MR, Amiri Z. Delayed ocular complications of mustard gas on 500 veterans. Arch Rehabil 2007;8:67-74.
Schummer J. Ethics of chemical weapons research: Poison gas in World War One. HYLE–Int J Philos Chem 2018;24:5-28.
Jarahi L, Habibzadeh Shojaie SR. Long-term effects of sulfur mustard poisoning in Iranian chemical warfare victims: A systematic review. J Isfahan Med Sch 2013;30:2353-66.
Taebi Gh, Soroush MR, Modirian E, Khateri S, Mousavi B, Ganjparvar Z, et al
. Human costs of Iraq's chemical war against Iran; An epidemiological study. Iranian J War Public Health 2015;7:115-21.
Chenary R, Noroozi A, Tahmasebi R. Survey of effective factors on health promotion behaviors based on health promotion model in chemically injured veterans of Ilam Province in 2012-13. Sci J Ilam Univ Med Sci 2013;21:257-67.
Hamta A, Fouladi Mansouri H. Studies of relationships between angiotensin converting enzyme (ACE) genotype and produced late complications by Sulfur mustard in Kermanshah Zardeh Village chemical exposed people. J Cell Tissue 2016;7:191-200.
Malkari B, Karimian N. Psychological- social consequences chemical bombardment of the Sardasht victims (a qualitative study). Iran J War Public Health 2013;5:7-13.
Roshan R, Rahnama P, Ghazanfari Z, Montazeri A, Soroush MR, Naghizadeh MM, et al
. Long-term effects of sulfur mustard on civilians'mental health 20 years after exposure (The Sardasht-Iran Cohort Study). Health Qual Life Outcomes 2013;11:69.
Salarian A, Mohammadi HR, Dargahi H. Evaluation of the quality of health care services of the injured and chemical veterans during the holy defense period since the time of injury in Kashan in 2017. Iran J War Public Health 2019;11:15-21.
Hors-Fraile S, Schneider F, Fernandez-Luque L, Luna-Perejon F, Civit A, Spachos D, et al
. Tailoring motivational health messages for smoking cessation using an mHealth recommender system integrated with an electronic health record: A study protocol. BMC Public Health 2018;18:698.
Sanchez Bocanegra CL, Sevillano Ramos JL, Rizo C, Civit A, Fernandez-Luque L. HealthRecSys: A semantic content-based recommender system to complement health videos. BMC Med Inform Decis Mak 2017;17:63.
Zavyalova YV, Kuznetsova TY, Korzun DG, Borodin AV, Meigal AY. Designing a mobile recommender system for treatment adherence improvement among hypertensives. In 22nd
Conference of Open Innovations Association (FRUCT), 2018. p. 290-6.
Sheykhotayefeh M, Safdari R, Ghazisaeedi M, Khademi SH, Seyed Farajolah SS, Maserat E, et al
. Development of a minimum data set (MDS) for C-section anesthesia information management system (AIMS). Anesth Pain Med 2017;7:e44132.
Davis J, Morgans A, Burgess S. Information management for aged care provision in Australia: Development of an aged care minimum dataset and strategies to improve quality and continuity of care. Health Inf Manag J 2016;45:27-35.
Choquet R, Maaroufi M, de Carrara A, Messiaen C, Luigi E, Landais P. A methodology for a minimum data set for rare diseases to support national centers of excellence for healthcare and research. J Am Med Inform Assoc 2015;22:76-85.
Nokhodian Z, ZareFarashbandi F, Shoaei P. Mustard gas exposure in Iran–Iraq war – A scientometric study. J Educ Health Promot 2015;4:56.
Biat Saeed K, Parandeh A, Alhani F, Salaree MM. Health-related quality of life of chemical warfare victims: An assessment with the use of a specific tool. Trauma Mon 2014;19:e13800.
Yang, S, Zhou P, Duan K, Hossain MS, Alhamid MF. emHealth: Towards emotion health through depression prediction and intelligent health recommender system. Mob Netw Appl 2018;23:216-26.
Hu S, Lu L, Jin X, Jiang Y, Zheng H, Xu Q, et al
. The recommender system for a cloud-based electronic medical record system for regional clinics and health centers in China. In 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2017. p. 1027-4.
Hu H, Elkus A, Kerschberg L. A personal health recommender system incorporating personal health records, modular ontologies, and crowd-sourced data. In 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2016. p. 1027-33.
Kamran M, Javed A. A survey of recommender systems and their application in healthcare. Tech J Univ Eng Technol Taxila 2015;20:111-9.
Bedi P, Agarwal SK. Aspect-Oriented trust based mobile recommender system. Int J Comput Inf Syst Ind Manag Appl 2013;5:354-64.
Uçar T, Karahoca A. Personalizing trip recommendations: A framework proposal. Glob J Comput Sci 2015;5:30-5.
Martinez-Cruz C, Porcel C, Bernabé-Moreno J, Herrera-Viedma E. A model to represent users trust in recommender systems using ontologies and fuzzy linguistic modeling. Inf Sci 2015;311:102-18.
Kraft MR, Hynes DM. Decision support within the Veterans' Health Administration. Stud Health Technol Inform 2006;122:100-4.
Bouhaddou O, Bennett J, Teal J, Pugh M, Sands M, Fontaine F, et al
. Toward a virtual lifetime electronic record: The department of veterans affairs experience with the nationwide health information network. AMIA Annu Symp Proc 2012;2012:51-60.
Perlin JB, Kolodner RM, Roswell RH. The Veterans Health Administration: Quality, value, accountability, and information as transforming strategies for patient-centered care. Am J Manag Care 2004;10:828-36.
Rajeevan N, Niehoff KM, Charpentier P, Levin FL, Justice A, Brandt CA, et al
. Utilizing patient data from the Veterans Administration electronic health record to support web-based clinical decision support: Informatics challenges and issues from three clinical domains. BMC Med Inf Decis Mak 2017;17:111.
Do NV, Barnhill R, Heermann-Do KA, Salzman KL, Gimbel RW. The military health system's personal health record pilot with Microsoft HealthVault and Google Health. J Am Med Inform Assoc 2011;18:118-24.
Bohnsack KJ, Parker DP, Zheng K. Quantifying temporal documentation patterns in clinician use of AHLTA—The DoD's ambulatory electronic health record. AMIA Annu Symp Proc 2009;2009:50-4.
Gräßer F, Beckert S, Küster D, Schmitt J, Abraham S, Malberg H, et al
. Therapy decision support based on recommender system methods. J Healthc Eng 2017;2017:8659460.
Agapito G, Simeoni M, Calabrese B, Caré I, Lamprinoudi T, Guzzi PH, et al
. DIETOS: A dietary recommender system for chronic diseases monitoring and management. Comput Methods Programs Biomed 2018;153:93-104.
[Table 1], [Table 2], [Table 3]