Exploring data security risks associated with trauma medical data within hospitals in China: a qualitative study

Abstract

Objectives

To explore data security risks associated with trauma medical data within hospitals in China, identify and explore the potential underlying factors contributing to these risks and collect suggestions from different stakeholders.

Design

This is a qualitative study involving two types of hospital personnel. Data analysis was performed using thematic analysis.

Setting

The study was conducted in 21 tertiary hospitals from 12 provinces in China, distributed across the eastern, central and western regions of the country. The interviews were conducted between April and August 2022.

Participants

A total of 27 respondents (7 department heads and 20 doctors) were interviewed through stratified purposive sampling.

Results

Data security risks associated with trauma medical data might arise from trauma physicians’ inadequate management of trauma data, the absence of trauma data administrators and data management systems and the lack of security measures for trauma databases. Feasible suggestions included training trauma physicians, establishing a trauma data administrator and a trauma data management system and improving basic data security protection measures.

Conclusions

Determining the risks of trauma medical data security and providing tailored suggestions contribute to the development of healthcare data governance in China. This research establishes a foundation for addressing the current risks related to trauma medical data security and could contribute to efforts to improve the overall capacity of trauma data management in China.

Correspondence to Dr Wei Huang; [email protected] ; Dr Tianbing Wang; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• Using qualitative research methods, the study delves into the sources of data security risks associated with trauma medical data in China and explores improvement strategies.

• Through stratified purposive sampling, the study incorporates insights into and suggestions for trauma data security from researchers in different regions and roles across China, enhancing the representativeness of the research findings.

• Combining one-on-one interviews and focus groups enables deep exploration of individual experiences and group perspectives.

• The findings of this exploratory analysis should be considered preliminary due to the limitations of the methodology.

• The interpretation of the data may be affected by the cultural background and experiences of the authors.

Introduction

With recent advancements in information technology, electronic medical data-related systems have been gradually implemented worldwide. In China, tertiary hospitals exhibit a notably high penetration rate of hospital information system of 99%.¹ However, the vast amount of electronic medical data presents potential risks, especially with regard to leakage of patients’ personal privacy data and the potential abuse of these data. The security risks associated with electronic medical data have been recognised as early as 2013, emphasising deficiencies in relevant legal provisions for regulating electronic medical data in the USA.^{2 3} The Personal Information Protection Law, effective since November 2021, is the first privacy law in China. Driven by security and privacy concerns, this legislation has established a comprehensive legal framework and principles governing the processing of personal data for individuals residing within the territory of China. The law defines personal data as ‘information related to an identified or identifiable natural person recorded electronically or by other means, but excluding anonymized information’. Health data, classified as sensitive, must be processed through a rigorously regulated pathway, with clear justification for the proposed use and unambiguous consent.⁴ The General Data Protection Regulation (GDPR), the European Union’s new data privacy law, entered into force on 25 May 2016. Compliance with GDPR is especially relevant to the digital health domain, where it is common to process highly sensitive personal data regarding a person’s health.^5–7

Medical institutions and researchers need to develop an enhanced awareness of security considerations in the use of medical data and understand how to ensure compliance in its usage. However, in China, there is a notable deficiency in both compliance practices and security awareness of medical data.⁸ Trauma is one of the major global public health issues. The Global Burden of Disease, Injury, and Risk Factors Study 2019⁹ showed that trauma accounted for 8% of the total annual deaths. In addition, road injuries were the leading cause of disability-adjusted life years among the population aged 10–49. Time is one crucial factor affecting the prognosis of trauma patients, which refers to the duration from the occurrence of trauma to the receipt of treatment.¹⁰ Research has found that providing emergency medical care within a short period of time after trauma (such as road traffic accidents, falls and violent injuries) can reduce the mortality rate and incidence of complications in trauma patients.^11–13 Therefore, in the prehospital emergency and in-hospital treatment stages, clinicians are racing against time to save patients. Doctors on ambulances take necessary rescue measures on receiving the patient and promptly communicate the patient’s condition to the doctors waiting in the hospital, ensuring timely treatment of the patient on arrival at the hospital.

However, this poses substantial challenges to the collection and transmission of trauma data. Despite the information transmission system on the ambulance being able to comprehensively transmit the condition of trauma patients to in-hospital doctors, communication via phone or WeChat is often required due to the limited time available to doctors on ambulances. In-hospital doctors often report difficulties recording the patient’s condition timely in the electronic medical record (EMR) system during rescue operations. Following the completion of a rescue, they need to supplement patient information by recalling or consulting checklists,^14–17 introducing security risks to trauma medical data. Based on ISO27001 standards, the National Institute of Standards and Technology risk management framework and China’s data security/risk management standards/guidelines, security risk refers specifically to events or situations that may occur during the processing of trauma data and pose a threat to data security. These risks include data breaches, data tampering, data misuse, illegal transmission and unauthorised access. These risks can lead to serious consequences, such as patient privacy breaches and data integrity damage, necessitating effective measures to identify and address these risks to ensure trauma data security and compliance. Therefore, the current study aimed to explore the security risks associated with trauma medical data during clinical diagnosis and treatment and to collect suggestions from different stakeholders through qualitative interviews in order to develop corresponding improvement measures for the identified data safety risks.

Methods

Study design

Qualitative research enables understanding of how trauma data are collected and used, as well as understanding of participants’ experiences.¹⁸ The current study was reported in compliance with the Consolidated Criteria for Reporting Qualitative Research guidelines.¹⁹ The framework approach uses a systematic procedure to analyse qualitative data via thematic analysis.²⁰ The interview was based on deep exploration of known stages in medical data processing. Therefore, a qualitative research strategy guided by thematic analysis was employed. Indepth interviews and focus groups were conducted in 21 tertiary hospitals across 12 provinces in China, distributed in the eastern, central and western regions of the country. The interviews were conducted between April and August 2022.

Participants

To explore the security risks associated with trauma medical data from different perspectives and to develop a comprehensive understanding, stakeholders with different roles within hospitals were invited to participate,^{21 22} which included doctors and directors related to trauma departments. Stratified purposive sampling method²³ was used to select representatives of different roles. TW and WH contacted the participants via email or WeChat, along with a brief introduction to the research topic in the invitation. Participants were able to understand the main purpose of the study and the estimated duration of the interview. Recruitment of interviewees concluded when the study reached information saturation, the point when the interviews with participants in each role no longer yielded new coding information.²⁴

The selection criteria for the interviewees were as follows:

Inclusion criteria

• Hospital staff related to trauma departments.

• Consent to record interview and voluntary signing of informed consent.

• Have experience in medical data processing.

Exclusion criteria

• Unable to provide at least 20 min of interview time.

Data collection

Face-to-face, semistructured interviews were conducted.^25–27 Depending on the doctor’s time and willingness, a focus group interview was arranged instead of a one-on-one interview to promote discussion and communication.²⁸ For each interview, the interview time and the basic information of the interviewees were collected. Each participant received a consent form to provide written informed consent for the study. The interview outline (online supplemental material) was jointly developed by the research team, incorporating literature and professional experience. A pilot study was conducted with preinterviews with two doctors to test the feasibility and effectiveness of the interview guides. Both the interviews and the focus group discussions were arranged in the participants’ consulting rooms or meeting rooms according to their preferences. Interviewers encouraged participants to elaborate on their viewpoints as much as possible during the interviews.

Three interviewers conducted the interviews, all of whom hold doctoral degrees. FJ is female, while BW and WH are male. The interviewers were trained in qualitative research, with previous experiences in conducting interviews.

Analysis

All interviews were audio-recorded and transcribed verbatim by two coders (XL, BW). No private details of the respondents were recorded. Coding and memoing started with the first interview. An iterative process continued, involving successive rounds of data collection, coding and analysis, until saturation was reached. Data saturation was demonstrated in the last few interviews, where findings aligned with existing patterns and no new ideas were generated. Two trained qualitative coders (XL, BW) employed the techniques of thematic analysis.^29–31 QSR NVivo V.14 software was used for coding. The coders thoroughly reviewed the interview transcripts and developed a structured coding tree that started with inductive open coding. Once the core categories emerged, deductive selective coding was performed. Open coding was employed independently by two coders, and the derived core categories were compared through multiple rounds of discussions until a consensus was reached. To verify the reliability of the analysis results, an analyst independent of the research team was invited to conduct validation analysis on the research data. The analyst (KW) is a male with a PhD in clinical research methodology and possesses extensive experience in data analysis of qualitative interviews.

Patient and public involvement

None.

Results

31 participants were invited, of whom 4 declined to participate in the interviews, primarily due to their busy clinical schedules leaving them no time for interviews. A total of 27 respondents participated, including 7 department heads and 20 doctors from 21 tertiary hospitals (table 1). The study involved 20 indepth interviews and 7 focus group interviews. The average age of the respondents was 44.70±8.56 years (range 30–60 years), with an average work experience of 18.22±9.5 years.

Table 1

Participants’ demographic information

Variable	n (%)
Gender
Male	24 (88.89)
Female	3 (11.11)
Role
Doctors	20 (74.07)
Department heads	7 (25.93)
Interview form
Indepth interviews	20 (74.07)
Focus group interviews	7 (25.93)
Province
Beijing	8 (29.63)
Guangdong	2 (7.41)
Shaanxi	2 (7.41)
Jiangsu	2 (7.41)
Shanxi	1 (3.70)
Gansu	1 (3.70)
Shandong	1 (3.70)
Zhejiang	1 (3.70)
Chongqing	1 (3.70)
Henan	1 (3.70)
Hebei	1 (3.70)
Hainan	1 (3.70)
Total	27 (100)

The trauma medical data security risk themes and suggested themes identified by the study are shown in table 2. Quotes supporting these key themes and subthemes are shown in online supplemental table 1.

Table 2

Overview of key themes and subthemes

Risks	Suggestions
Lack of standardisation in the management of diagnosis and treatment data by trauma doctors.	Provide training to trauma doctors for improved awareness of data security.
Data collection using non-EMR systems for trauma patients.
Lack of anonymisation and encryption of data transmission for trauma patients.
Failure to promptly destroy trauma patient information in personal software.
Unregular application of diagnostic and treatment data.
Incomplete trauma data management.	Establish a security management system for the entire life-cycle of trauma data.
Lack of dedicated trauma data management personnel.	Set up dedicated trauma data management personnel.
Lack of trauma data management system.	Establish a system for collecting, transmitting, storing, applying and destroying trauma data.
Inadequate security measures for trauma data.	Improve software, hardware and physical environment related to trauma data.
Need for updated technology in data security protection.

EMR, electronic medical record.

Risks

Lack of standardisation in the management of diagnosis and treatment data by trauma doctors

Data collection using non-EMR systems for trauma patients

To facilitate data transmission, trauma doctors often used personal mobile phones to record and take photos of patient information during prehospital emergency treatment and in-hospital communication. This practice was often perceived by doctors as a routine diagnostic and treatment procedure, neglecting the protection of patients’ personal information.

Lack of anonymisation and encryption of data transmission for trauma patients

Trauma doctors frequently transmitted sensitive patient information to in-hospital doctors through private mobile phones, WeChat and phone calls when receiving patients on ambulances. While this approach aimed to ensure timely and clear transmission of patient information, facilitating prompt treatment of the patient on arrival at the hospital, there were still potential safety risks without anonymisation and encryption of data transmission.

Failure to promptly destroy trauma patient information in personal software

Trauma doctors often faced difficulties in promptly and thoroughly destroying patient information from their personal phones after recording and transmitting it, introducing a potential risk for patient information protection.

Unregular application of diagnostic and treatment data

When trauma doctors used EMR data for scientific research, they often manually transcribed data directly from EMR. Due to the complexity of data export programs in most hospitals, doctors highlighted challenges in rapidly exporting data from the information department. As a result, many doctors opted for manual transcription. However, this method lacked the capability to desensitise patient information and posed data security risks.

Incomplete trauma data management

Lack of dedicated trauma data management personnel

There was a lack of professional trauma data management personnel for centralised management of data collection, storage, transmission, application and destruction. A dedicated data manager in the trauma department was crucial, but most hospitals currently lacked this role.

Lack of trauma data management system

Due to the lack of a data management system tailored to the characteristics of trauma diagnosis and treatment, trauma doctors faced difficulties in understanding the standardised data management process, leading to greater data security risks.

Inadequate security measures for trauma database

Need for updated technology in data security protection

Higher technical requirements were required for data encryption, desensitisation and traceability of trauma data. Basic security measures for trauma data are currently insufficient, and the adoption of the latest data security technologies is not widespread.

Suggestions

Provide training to trauma doctors for improved awareness of data security

These trauma data were scattered and stored in multiple information systems. Trauma doctors were the initial personnel to interact with trauma patients and to collect data. Hence, it is essential to provide data security training to trauma doctors to promote understanding of the security risks at each stage of the trauma data process.

Establish a security management system for the entire life-cycle of trauma data

Set up dedicated trauma data management personnel

The trauma department needs to employ dedicated data management personnel. Data administrators need to possess professional skills in trauma data security governance, establish clear classification and grading standards and implement management systems for trauma data.

Establish a system for collecting, transmitting, storing, applying and destroying trauma data

It is crucial to establish data security management standards for the entire life-cycle of trauma data and to develop standardised data operation procedures. For trauma data, classification standards can be developed based on factors such as sensitivity, importance and scope of use. For example, data can be categorised as public data, internal data and sensitive data in order to adopt different security management measures for each data category.

Improve software, hardware and physical environment related to trauma data

In terms of software technology, employing advanced encryption algorithms is necessary for encrypting trauma data and ensuring the security of data during transmission and storage. Similarly, it is fundamental to implement strict access control mechanisms and regularly update data backup and recovery technologies.

Discussion

This study investigated the security risks associated with trauma medical data and explored practical recommendations for implementation. The participating hospitals were located in the eastern, central and western regions of China, and these hospitals are recognised as top trauma hospitals in their respective regions, with highly skilled trauma physicians and a strong commitment to information management and scientific research.

The interviews suggested that the security risks associated with trauma data existed beyond the information system to include front-line activities in clinical diagnosis and treatment. This finding has rarely been reported in existing medical data security studies.^32–34 During pre-hospital trauma care, patient information was occasionally documented in non-electronic systems during handoffs to in-hospital teams. Integrating frontline clinical workflows is critical to identifying trauma data security risks.

The General Data Protection Regulation (GDPR) and clinical practice guidelines summarise the rights and obligations of data controllers. In clinical practice, the following must be adhered to^35–37:

• Make privacy policies freely available to patients.

• Keep a record of activities in processing medical data.

• Take appropriate technical and organisational measures to ensure data security.

• Set up a data protection commissioner.

• Have data breach risk identification techniques.

• Train all relevant employees in data security.

• Adequately protect patient privacy when sharing data.

• Ensure that information system suppliers possess recognised security certifications.

This study found that, in addition to the risks mentioned in the GDPR, there are some special risks in trauma medical data processing in China. The first risk identified by the current study was the lack of standardisation of trauma physicians’ management of diagnosis and treatment data, including (1) data collection using a non-EMR system for trauma patients, (2) lack of anonymisation and encryption of data transmission for trauma patients, (3) failure to promptly destroy trauma patient information in personal software and (4) unregulated application of diagnostic and treatment data. Time is one crucial factor associated with the prognosis of trauma patients. Paramedics often use personal mobile devices (eg, WeChat) to communicate with hospital teams, despite some ambulances having telemedicine systems that are time-consuming to activate. This preference prioritises efficiency in data collection and transmission. Additionally, in-hospital clinicians frequently rely on patient recall during emergency medical record (EMR) documentation. Unfortunately, these doctors also frequently fail to promptly delete and destroy sensitive patient information stored on their private cell phones. As previously reported, EMRs are mainly used for clinical practice,³⁸ posing challenges for trauma physicians in accessing patient data for the purpose of scientific research.

A survey highlighted that ambulance-based telemedicine, though an emerging concept, could improve care quality by reducing time-sensitive treatment delays. However, there were ongoing concerns regarding its implementation, including issues such as audio and video connection problems, interference with patient care, workflow disruptions and patient confidentiality.³¹ It is important to note that sharing patient data over personal mobile phones is a clear violation of the regulations of Health Insurance Portability and Accountability Act.^{39 40} In the process of trauma data processing, the use of personal devices represents a significant risk point. Personal devices often lack necessary security measures, such as encryption and access control, which can result in risks such as data breaches, tampering or unauthorised access. Therefore, training for trauma physicians need to be prioritised to raise awareness about data security,^41–43 preventing leakage of trauma patients’ information due to lack of awareness about security risks among doctors.

Trauma data management in China currently remains inadequate, characterised by (1) lack of dedicated trauma data personnel and (2) lack of trauma data management system. The absence of a dedicated data personnel indicates that no individual is responsible for monitoring and securing trauma data, potentially leading to unauthorised access and misuse of sensitive data associated with data breaches.^{44 45} In addition, the lack of a security management system for trauma data can lead to chaotic data management processes, absence of consistent standards and norms and inefficient data management, thereby increasing the risk of errors and omissions.

Therefore, it is recommended to establish a full-time trauma data manager position in the hospital, with clearly defined responsibilities. In addition, standardised and comprehensive data management specifications need to be implemented that specify the requirements for data collection, storage, use and protection. The National Trauma Data Bank is the largest trauma registry data in the USA and has implemented measures to ensure the security and privacy of data.⁴¹ From the perspectives of security and privacy, medical big data security involves the entire life-cycle. Yazan et al developed a big data security life-cycle model that provides fundamental theoretical basis for implementing security measures for trauma medical data.^45–48

Furthermore, there were insufficient security measures implemented for trauma database. Physical security is the foundation of data security. However, the existing access control measures for trauma data lack stringency due to inadequate data encryption, allowing clinicians easy access to sensitive patient information. Redspin’s annual breach report released by CynergisTek revealed a 320% increase in hacking attacks against healthcare providers in 2016. Ransomware, a type of malware that encrypts data and holds them hostage until a ransom demand is met, has been identified as the most substantial threat to hospitals.^{49 50}

The establishment of a well-defined data access process and the utilisation of data encryption algorithms for processing, along with regular technology updates, are critical. For sensitive and important data, strict access control measures need to be implemented to ensure that only authorised personnel have access. Simultaneously, data access logs need to be recorded to trace data access behaviours. Although various encryption algorithms have been relatively well developed, selecting the appropriate cryptographic algorithm remains a challenge.^51–53 The difficulty in employing multiple data masking algorithms primarily lies in anonymising high-dimensional data sets.^54–56 The balance between access control and data security introduces additional challenges to the algorithms.^57–59

It is worth noting several constraints of this study. Recruitment took place exclusively in 21 tertiary hospitals, potentially introducing selection bias as non-participating hospitals might hold differing perspectives.^{60 61} To mitigate this bias, a stratified and purposeful sampling strategy was employed, encompassing hospitals from 12 provinces and presuming information saturation.⁶² Although the interviewers possessed experience and were trained in qualitative research, the interpretation of data could still have been influenced by the authors’ cultural backgrounds and prior experiences. The research team comes from the National Trauma Medical Center, which is the only national-level centre in the field of trauma medicine in China, with its medical level, hospital management and scientific research capabilities among the best in the country. However, trauma centres in different regions and hospitals may have significant differences in medical level, scientific research capabilities and management. Furthermore, the research team generally has higher academic qualifications, which may not reflect the educational profiles of clinical staff in non-specialised regions. To minimise this impact, we maintained an open and objective attitude throughout the research process, striving to understand participants’ experiences and perspectives. At the same time, we also carefully interpreted the research results to ensure accuracy and reliability.

Conclusion

This qualitative study investigated the data security risks associated with trauma medical data within hospitals in China. Trauma data security risk might arise from various factors. Feasible recommendations to mitigate trauma data security risks included training trauma physicians in data security awareness and standardised data operation procedures, setting up a full-time trauma data administrator and trauma data management system and improving data security protection measures. Further research is warranted to assess and mitigate trauma data security risks, as identified in this qualitative study.

We thank all the people who took time to participate in our interviews.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Obtained.

Ethics approval

This study involves human participants and was approved by the Peking University Institutional Review Board (no: IRB00001052-19052) in 2 May 2019. Participants gave informed consent to participate in the study before taking part.

Footnote

FJ, BW, XL and KW contributed equally.

Contributors FJ and TW designed the study. FJ, WH and BW collected the data. XL, KW and BW analysed the data. FJ and XL wrote the first draft of the manuscript. WH revised the manuscript. All authors contributed to the interpretation of the data, edited the manuscript and approved the final manuscript. FJ is the guarantor.

Funding This study was supported by Peking University People's Hospital Project 'Research on Data Security Assessment in the Construction of National Trauma Informatization' (grant no: 2159000064) and Peking University People's Hospital Project 'Research on Comprehensive Intervention of Data Security in the Construction of National Trauma Informatization' (grant no: 2159000065).

Competing interests None declared.

Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Wu Y, Yin D, Abbasi K. China’s medical research revolution. BMJ 2018; 360: k547. doi:10.1136/bmj.k547

² Taitsman JK, Grimm CM, Agrawal S. Protecting patient privacy and data security. N Engl J Med 2013; 368: 977–9. doi:10.1056/NEJMp1215258

³ Mohammadi D. Data security in the brave new world of eHealth. Lancet Respir Med 2016; 4: 611. doi:10.1016/S2213-2600(16)30197-7

⁴ Yin D, Li X, Liu R, et al. China’s Personal Information Protection Law. BMJ 2022; 379: e072619. doi:10.1136/bmj-2022-072619

⁵ Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). 2016. Available: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32016R0679 [Accessed 16 Dec 2018 ].

⁶ Bentzen HB, Høstmælingen N. Balancing Protection and Free Movement of Personal Data: The New European Union General Data Protection Regulation. Ann Intern Med 2019; 170: 335–7. doi:10.7326/M18-2782

⁷ Hussein R, Wurhofer D, Strumegger E-M, et al. General Data Protection Regulation (GDPR) Toolkit for Digital Health. Stud Health Technol Inform 2022; 290: 222–6. doi:10.3233/SHTI220066

⁸ Matsumura Y, Hattori A, Manabe S, et al. Interconnection of electronic medical record with clinical data management system by CDISC ODM. Stud Health Technol Inform 2014; 205: 868–72. doi:10.3233/978-1-61499-432-9-868

⁹ Vos T, Lim SS, Abbafati C. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2020; 396: 1204–22. doi:10.1016/S0140-6736(20)30925-9

¹⁰ Boyd DR, Cowley RA. Comprehensive regional trauma/emergency medical services (EMS) delivery systems: the United States experience. World J Surg 1983; 7: 149–57. doi:10.1007/BF01655923

¹¹ Esposito TJ, Maier RV, Rivara FP, et al. The impact of variation in trauma care times: urban versus rural. Prehosp Disaster Med 1995; 10: 161–6. doi:10.1017/s1049023x00041947

¹² Feero S, Hedges JR, Simmons E, et al. Does out-of-hospital EMS time affect trauma survival? Am J Emerg Med 1995; 13: 133–5. doi:10.1016/0735-6757(95)90078-0

¹³ Gervin AS, Fischer RP. The importance of prompt transport of salvage of patients with penetrating heart wounds. J Trauma 1982; 22: 443–8. doi:10.1097/00005373-198206000-00001

¹⁴ Grossman DC, Kim A, Macdonald SC, et al. Urban-rural Differences in Prehospital Care of Major Trauma. J Trauma 1997; 42: 723–9. doi:10.1097/00005373-199704000-00024

¹⁵ Porter A, Badshah A, Black S, et al. Electronic health records in ambulances: the ERA multiple-methods study. Health Serv Deliv Res 2020; 8: 1–140. doi:10.3310/hsdr08100

¹⁶ Adler NE, Stead WW. Patients in context--EHR capture of social and behavioral determinants of health. N Engl J Med 2015; 372: 698–701. doi:10.1056/NEJMp1413945

¹⁷ Zhang L, Wang H, Li Q, et al. Big data and medical research in China. BMJ 2018; 360: j5910. doi:10.1136/bmj.j5910

¹⁸ Grossoehme DH. Overview of qualitative research. J Health Care Chaplain 2014; 20: 109–22. doi:10.1080/08854726.2014.925660

¹⁹ Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care 2007; 19: 349–57. doi:10.1093/intqhc/mzm042

²⁰ Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol 2006; 3: 77–101. doi:10.1191/1478088706qp063oa

²¹ Bentley T, Rizer M, McAlearney AS, et al. The journey from precontemplation to action: Transitioning between electronic medical record systems. Health Care Manage Rev 2016; 41: 22–31. doi:10.1097/HMR.0000000000000041

²² Denny E, Weckesser A. Qualitative research: what it is and what it is not: Study design: qualitative research. BJOG 2019; 126: 369. doi:10.1111/1471-0528.15198

²³ Andrade C. The Inconvenient Truth About Convenience and Purposive Samples. Indian J Psychol Med 2021; 43: 86–8. doi:10.1177/0253717620977000

²⁴ Kordzadeh N, Young DK. Exploring Hospitals’ Use of Facebook: Thematic Analysis. J Med Internet Res 2018; 20: e190. doi:10.2196/jmir.9549

²⁵ Whiting LS. Semi-structured interviews: guidance for novice researchers. Nurs Stand 2008; 22: 35–40. doi:10.7748/ns2008.02.22.23.35.c6420

²⁶ Peters K, Halcomb E. Interviews in qualitative research. Nurse Res 2015; 22: 6–7. doi:10.7748/nr.22.4.6.s2

²⁷ Britten N. Qualitative interviews in medical research. BMJ 1995; 311: 251–3. doi:10.1136/bmj.311.6999.251

²⁸ Rabiee F. Focus-group interview and data analysis. Proc Nutr Soc 2004; 63: 655–60. doi:10.1079/pns2004399

²⁹ Castleberry A, Nolen A. Thematic analysis of qualitative research data: Is it as easy as it sounds? Curr Pharm Teach Learn 2018; 10: 807–15. doi:10.1016/j.cptl.2018.03.019

³⁰ Vaismoradi M, Jones J, Turunen H, et al. Theme development in qualitative content analysis and thematic analysis. JNEP 2016; 6. doi:10.5430/jnep.v6n5p100

³¹ Vaismoradi M, Turunen H, Bondas T. Content analysis and thematic analysis: Implications for conducting a qualitative descriptive study. Nurs Health Sci 2013; 15: 398–405. doi:10.1111/nhs.12048

³² Coventry L, Branley D. Cybersecurity in healthcare: A narrative review of trends, threats and ways forward. Maturitas 2018; 113: 48–52. doi:10.1016/j.maturitas.2018.04.008

³³ Abouelmehdi K, Beni-Hessane A, Khaloufi H. Big healthcare data: preserving security and privacy. J Big Data 2018; 5. doi:10.1186/s40537-017-0110-7

³⁴ Houlding D, CISSP. Health information at risk: successful strategies for healthcare security and privacy. Healthcare IT program of CE Intel corporation, white paper. 2011.

³⁵ Carmi L, Zohar M, Riva GM. The European General Data Protection Regulation (GDPR) in mHealth: Theoretical and practical aspects for practitioners’ use. Med Sci Law 2023; 63: 61–8. doi:10.1177/00258024221118411

³⁶ Molnár-Gábor F, Sellner J, Pagil S, et al. Harmonization after the GDPR? Divergences in the rules for genetic and health data sharing in four member states and ways to overcome them by EU measures: Insights from Germany, Greece, Latvia and Sweden. Semin Cancer Biol 2022; 84: 271–83. doi:10.1016/j.semcancer.2021.12.001

³⁷ Yuan B, Li J. The Policy Effect of the General Data Protection Regulation (GDPR) on the Digital Public Health Sector in the European Union: An Empirical Investigation. Int J Environ Res Public Health 2019; 16: 1070. doi:10.3390/ijerph16061070

³⁸ Drolet BC, Marwaha JS, Hyatt B, et al. Electronic Communication of Protected Health Information: Privacy, Security, and HIPAA Compliance. J Hand Surg Am 2017; 42: 411–6. doi:10.1016/j.jhsa.2017.03.023

³⁹ Kavoussi SC, Huang JJ, Tsai JC, et al. HIPAA for physicians in the information age. Conn Med 2014; 78: 425–7.

⁴⁰ Markota M, Kern J, Svab I. Attitudes towards information system security among physicians in Croatia. Methods Inf Med 2001; 40: 221–4. doi:10.1055/s-0038-1634156

⁴¹ Zamosky L. Avoid the breach: put data security measures in place. Physician Exec 2014; 40: 82–4.

⁴² Kamoun F, Nicho M. Human and organizational factors of healthcare data breaches: The swiss cheese model of data breach causation and prevention. IJHISI 2014; 9: 42–60. doi:10.4018/ijhisi.2014010103

⁴³ Yang PK, Yang B, Snekkenes EA. Healthcare staffs’ information security practices towards mitigating data breaches: a literature survey. pHealth 2019; 239–45.

⁴⁴ Hornor MA, Hoeft C, Nathens AB. Quality Benchmarking in Trauma: from the NTDB to TQIP. Curr Trauma Rep 2018; 4: 160–9. doi:10.1007/s40719-018-0127-1

⁴⁵ Yazan A, Yong W, Raj Kumar N. Big data life cycle: threats and security model. 21st Americas conference on information systems; 2015

⁴⁶ Xu L, Jiang C, Wang J, et al. Information Security in Big Data: Privacy and Data Mining. IEEE Access 2014; 2: 1149–76. doi:10.1109/ACCESS.2014.2362522

⁴⁷ General dynamics health solutions white paper UK [ Securing Big Health Data ]. 2015. Available: http://gdhealth.com/globalassets/health-solutions/documents/brochures/securing-big-health-data_-white-paper_UK.pdf

⁴⁸ WHO. Mobile phones help people with diabetes to manage fasting and feasting during Ramadan. Features. 2014.

⁴⁹ CynergisTek. Breach report 2016: protected health information (PHI). 2017.

⁵⁰ Shafer J, Rixner S, Cox AL. The hadoop distributed filesystem: balancing portability and performance. Proceedings of 2010 IEEE International Symposium on Performance Analysis of Systems & Software (ISPASS); White Plain, NY, 122–33.

⁵¹ Yang C, Lin W, Liu M. A novel triple encryption scheme for hadoop-based cloud data security. In: emerging intelligent data and web technologies (eidwt). 2013 fourth international conference on. 2013; 2013: 437–42. doi:10.1109/EIDWT.2013.80

⁵² Specification for the advanced encryption standards (AES). In: Federal information processing standards publication 197. 2001.

⁵³ Somu N, Gangaa A, Sriram VS. Authentication Service in Hadoop using One Time Pad. Indian J Sci Technol 2014; 7: 56–62. doi:10.17485/ijst/2014/v7sp4.16

⁵⁴ Sweeney L. Achieving k-anonymity privacy protection using generalization and suppression. Int J Unc Fuzz Knowl Based Syst 2002; 10: 571–88. doi:10.1142/S021848850200165X

⁵⁵ Samarati P. Protecting respondents identities in microdata release. IEEE Trans Knowl Data Eng 2001; 13: 1010–27. doi:10.1109/69.971193

⁵⁶ Truta TM, Vinay B. Privacy protection: p-sensitive k-anonymity property. Proceedings of 22nd International Conference on Data Engineering Workshops. 94.

⁵⁷ Science Applications International Corporation (SAIC). Role-based access control (RBAC) role engineering process version 3.0. 2004.

⁵⁸ Mohan A, Blough DM. An attribute-based authorization policy framework with dynamic conflict resolution. Proceedings of the 9th symposium on identity and trust on the internet; 2010 doi:10.1145/1750389.1750395

⁵⁹ Zhou H, Wen Q. Data security accessing for hdfs based on attribute-group in cloud computing. International conference on logistics engineering, management and computer science (LEMCS 2014); 2014 doi:10.2991/lemcs-14.2014.255

⁶⁰ Sofaer S. Qualitative methods: what are they and why use them? Health Serv Res 1999; 34: 1101–18.

⁶¹ Tenny S, Brannan JM, Brannan GD. Qualitative study. Treasure Island: StatPearls, 2022.

⁶² Hamilton JB. Rigor in Qualitative Methods: An Evaluation of Strategies Among Underrepresented Rural Communities. Qual Health Res 2020; 30: 196–204. doi:10.1177/1049732319860267