Automatic de-identification of electronic health records : an Australian perspective

We present an approach to automatically de-identify health records. In our approach, personal health information is identified using a Conditional Random Fields machine learning classifier, a large set of linguistic and lexical features, and pattern matching techniques. Identified personal information is then removed from the reports. The de-identification of personal health information is fundamental for the sharing and secondary use of electronic health records, for example for data mining and disease monitoring. The effectiveness of our approach is first evaluated on the 2007 i2b2 Shared Task dataset, a widely adopted dataset for evaluating de-identification techniques. Subsequently, we investigate the robustness of the approach to limited training data; we study its effectiveness on different type and quality of data by evaluating the approach on scanned pathology reports from an Australian institution. This data contains optical character recognition errors, as well as linguistic conventions that differ from those contained in the i2b2 dataset, for example different date formats. The findings suggest that our approach compares to the best approach from the 2007 i2b2 Shared Task; in addition, the approach is found to be robust to variations of training size, data type and quality in presence of sufficient training data.

De-Identification of health records using Anonym: Effectiveness and robustness across datasets

Objective Evaluate the effectiveness and robustness of Anonym, a tool for de-identifying free-text health records based on conditional random fields classifiers informed by linguistic and lexical features, as well as features extracted by pattern matching techniques. De-identification of personal health information in electronic health records is essential for the sharing and secondary usage of clinical data. De-identification tools that adapt to different sources of clinical data are attractive as they would require minimal intervention to guarantee high effectiveness. Methods and Materials The effectiveness and robustness of Anonym are evaluated across multiple datasets, including the widely adopted Integrating Biology and the Bedside (i2b2) dataset, used for evaluation in a de-identification challenge. The datasets used here vary in type of health records, source of data, and their quality, with one of the datasets containing optical character recognition errors. Results Anonym identifies and removes up to 96.6% of personal health identifiers (recall) with a precision of up to 98.2% on the i2b2 dataset, outperforming the best system proposed in the i2b2 challenge. The effectiveness of Anonym across datasets is found to depend on the amount of information available for training. Conclusion Findings show that Anonym compares to the best approach from the 2006 i2b2 shared task. It is easy to retrain Anonym with new datasets; if retrained, the system is robust to variations of training size, data type and quality in presence of sufficient training data.

eHealth-as-a-Service (eHaaS) : towards universal stakeholder engagement

With the introduction of the Personally Controlled Health Record (PCEHR), the Australian public is being asked to accept greater responsibility for their healthcare by taking an active role in the management of personal health information. Although well designed, constructed and intentioned, policy and privacy concerns have resulted in an eHealth model that may impact future health sharing requirements. Hence, as a case study for a consumer eHealth initative in the Australian context, eHealth-as-a-Service (eHaaS) serves as a disruptive step in in the aggregation and transformation of health information for use as real-world knowledge. The strategic value of extending the community Health Record Bank (HRB) model lies in the ability to automatically draw on a multitude of relevant data repositories and sources to create a single source of the truth and to engage market forces to create financial sustainability. The opportunity to transform the beleaguered Australian PCEHR into a realisable and sustainable technology consumption model for patient safety is explored. Moreover, the current clerical focus of healthcare practitioners acting in the role of de facto record keepers is renegotiated to establish a shared knowledge creation landscape of action for safer patient interventions. To achieve this potential however requires a platform that will facilitate efficient and trusted unification of all health information available in real-time across the continuum of care. eHaaS provides a sustainable environment and encouragement to realise this potential.

Sharing of clinical data in a maternity setting: How do paper hand-held records and electronic health records compare for completeness?

Background Historically, the paper hand-held record (PHR) has been used for sharing information between hospital clinicians, general practitioners and pregnant women in a maternity shared-care environment. Recently in alignment with a National e-health agenda, an electronic health record (EHR) was introduced at an Australian tertiary maternity service to replace the PHR for collection and transfer of data. The aim of this study was to examine and compare the completeness of clinical data collected in a PHR and an EHR. Methods We undertook a comparative cohort design study to determine differences in completeness between data collected from maternity records in two phases. Phase 1 data were collected from the PHR and Phase 2 data from the EHR. Records were compared for completeness of best practice variables collected The primary outcome was the presence of best practice variables and the secondary outcomes were the differences in individual variables between the records. Results Ninety-four percent of paper medical charts were available in Phase 1 and 100% of records from an obstetric database in Phase 2. No PHR or EHR had a complete dataset of best practice variables. The variables with significant improvement in completeness of data documented in the EHR, compared with the PHR, were urine culture, glucose tolerance test, nuchal screening, morphology scans, folic acid advice, tobacco smoking, illicit drug assessment and domestic violence assessment (p = 0.001). Additionally the documentation of immunisations (pertussis, hepatitis B, varicella, fluvax) were markedly improved in the EHR (p = 0.001). The variables of blood pressure, proteinuria, blood group, antibody, rubella and syphilis status, showed no significant differences in completeness of recording. Conclusion This is the first paper to report on the comparison of clinical data collected on a PHR and EHR in a maternity shared-care setting. The use of an EHR demonstrated significant improvements to the collection of best practice variables. Additionally, the data in an EHR were more available to relevant clinical staff with the appropriate log-in and more easily retrieved than from the PHR. This study contributes to an under-researched area of determining data quality collected in patient records.

Paper and electronic antenatal records: What do women and health care providers tell us about their use?

The antenatal paper hand-held record (PHR) has been used extensively in general practice (GP) shared-care management of pregnant women, but recently the antenatal electronic health record (EHR) was introduced. This study aimed to examine the experiences of women and health care providers who use the PHR and the EHR, and find out the relative role of these records in the integration of care. Purposive homogenous samples of women and health care providers were interviewed as users of the PHR in phase 1 and the EHR in phase 2 of the study. Qualitative data were collected via interview with women and GPs and focus groups held with hospital health care providers. Interviews were coded manually and analysed using qualitative content analysis. Fifteen women participated in phase 1 and 12 in phase 2. Seventeen GPs participated in phase 1 and 15 in phase 2. Five focus groups with hospital health care providers were conducted in each phase. Results were categorised into four themes: 1. Record purpose; 2. Perception of the record; 3. Content of the record, and; 4. Sharing information in the record. Both women and health care providers were familiar with the PHR, but identified that some information was missing or not utilised well, and reported underuse of the EHR. The study identified continued widespread use of the PHR and several issues concerning the use of the EHR. An improvement in the strategic implementation of the EHR is suggested as a mechanism to facilitate its wider adoption.

From maternity paper hand-held records to electronic health records: What do women tell us about their use?

The paper hand-held record (PHR) has been used extensively in general practice (GP) shared care management of pregnant women, and recently, the first Mater Shared Electronic Health Record (MSEHR) was introduced. The aim of this qualitative study was to examine women’s experiences using the records and the contribution of the records to integrate care. At the 36-week antenatal visit in a maternity tertiary centre clinic, women were identified as a user of either the PHR or the MSEHR and organised into Phase 1 and Phase 2 studies respectively. Fifteen women were interviewed in Phase 1 and 12 women in Phase 2. Semi-structured interviews were used for data collection, and analysed using qualitative content analysis. Four main themes were identified: (1) purpose of the record; (2) perceptions of the record; (3) content of the record, and; (4) sharing the record. Findings indicate that the PHR is a well-liked maternity tool. The findings also indicate there is under-usage of the MSEHR due to health-care providers failing to follow up and discuss the option of using the electronic health record option or if a woman has completed the log-in process. This paper adds to an already favourable body of knowledge about the use of the PHR. It is recommended that continued implementation of the MSEHR be undertaken to facilitate its use.

Truckies and workplace health promotion: Targeting a hard-to-reach group

- Introduction Research identifies truck drivers as being at high risk of chronic disease. For most truck drivers their workplace is their vehicle. Truck drivers’ health is impacted by the limitations of this unique working environment, including reduced opportunities for physical activity and the intake of healthy foods. Workplaces are widely recognised as effective platforms for health promotion. However, the effectiveness of traditional and contemporary health promotion interventions in truck drivers’ novel workplace is unknown. - Methods This project worked with six transport industry workplaces in Queensland, Australia over a two-year period. Researchers used Participatory Action Research (PAR) processes to engage truck drivers and workplace managers in the implementation and evaluation of six workplace health promotion interventions. These interventions were designed to support truck drivers to increase their physical activity and access to healthy foods at work. They included traditional health promotion interventions such as a free fruit initiative, a ten thousand steps challenge, personal health messages and workplace posters, and a contemporary social media intervention. Participants were engaged via focus groups, interviews and mixed-methods surveys. - Results The project achieved positive changes in truck drivers’ health knowledge and health behaviours, particularly related to nutrition. There were positive changes in truck drivers’ self-reported health rating, body mass index (BMI) and readiness to make health-related lifestyle changes. There were also positive changes in truck drivers reporting their workplace as a key source of health information. These changes were underpinned by a positive shift in the culture of participating workplaces. Truck drivers’ perceptions of their workplace valuing, encouraging, modelling and facilitating healthy nutrition and physical activity behaviours improved. PAR processes enabled researchers to develop relationships with workplace managers, contextualise interventions and deliver rigorous outcomes. Despite the novelty of truck drivers’ mobile workplace, traditional health promotion interventions were more effective than contemporary ones. - Conclusion In this workplace health promotion project targeting a ‘hard-to-reach’ group of truck drivers, a combination of well-designed traditional workplace interventions and the PAR process resulted in positive health outcomes.

An eroding social justice agenda: The case of physical education and health edu-business in schools

In this paper we draw on current research to explore notions of a socially just Health and Physical Education (HPE), in light of claims that a neoliberal globalisation promotes markets over the states, and a new individualism that privileges self-interest over the collective good. We also invite readers to consider United Nations Educational, Scientific and Cultural Organization’s ambition for PE in light of preliminary findings from an Australian led research project exploring national and international patterns of outsourcing HPE curricula. Data were sourced from this international research project through a mixed method approach. Each external provider engaged in four phases of research activity: (a) Web-audits, (b) Interviews with external providers, (c) Network diagrams, and (d) School partner interviews and observations. Results We use these data to pose what we believe to be three emerging lines of inquiry and challenge for a socially just school HPE within neoliberal times. In particular our data indicates that the marketization of school HPE is strengthening an emphasis on individual responsibility for personal health, elevating expectations that schools and teachers will “fill the welfare gap” and finally, influencing the nature and purchase of educative HPE programs in schools. The apparent proliferation of external providers of health work, HPE resources and services reflects the rise and pervasiveness of neoliberalism in education. We conclude that this global HPE landscape warrants attention to investigate the extent to which external providers’ resources are compatible with schooling’s educative and inclusive mandates.

HealthMap: A cluster randomised trial of interactive health plans and self-management support to prevent coronary heart disease in people with HIV

Background The leading causes of morbidity and mortality for people in high-income countries living with HIV are now non-AIDS malignancies, cardiovascular disease and other non-communicable diseases associated with ageing. This protocol describes the trial of HealthMap, a model of care for people with HIV (PWHIV) that includes use of an interactive shared health record and self-management support. The aims of the HealthMap trial are to evaluate engagement of PWHIV and healthcare providers with the model, and its effectiveness for reducing coronary heart disease risk, enhancing self-management, and improving mental health and quality of life of PWHIV. Methods/Design The study is a two-arm cluster randomised trial involving HIV clinical sites in several states in Australia. Doctors will be randomised to the HealthMap model (immediate arm) or to proceed with usual care (deferred arm). People with HIV whose doctors are randomised to the immediate arm receive 1) new opportunities to discuss their health status and goals with their HIV doctor using a HealthMap shared health record; 2) access to their own health record from home; 3) access to health coaching delivered by telephone and online; and 4) access to a peer moderated online group chat programme. Data will be collected from participating PWHIV (n = 710) at baseline, 6 months, and 12 months and from participating doctors (n = 60) at baseline and 12 months. The control arm will be offered the HealthMap intervention at the end of the trial. The primary study outcomes, measured at 12 months, are 1) 10-year risk of non-fatal acute myocardial infarction or coronary heart disease death as estimated by a Framingham Heart Study risk equation; and 2) Positive and Active Engagement in Life Scale from the Health Education Impact Questionnaire (heiQ). Discussion The study will determine the viability and utility of a novel technology-supported model of care for maintaining the health and wellbeing of people with HIV. If shown to be effective, the HealthMap model may provide a generalisable, scalable and sustainable system for supporting the care needs of people with HIV, addressing issues of equity of access. Trial registration Universal Trial Number (UTN) U111111506489; ClinicalTrial.gov Id NCT02178930 submitted 29 June 2014

Veterans health administration hepatitis B testing and treatment with anti-CD20 antibody administration.

AIM: To evaluate pretreatment hepatitis B virus (HBV) testing, vaccination, and antiviral treatment rates in Veterans Affairs patients receiving anti-CD20 Ab for quality improvement. METHODS: We performed a retrospective cohort study using a national repository of Veterans Health Administration (VHA) electronic health record data. We identified all patients receiving anti-CD20 Ab treatment (2002-2014). We ascertained patient demographics, laboratory results, HBV vaccination status (from vaccination records), pharmacy data, and vital status. The high risk period for HBV reactivation is during anti-CD20 Ab treatment and 12 mo follow up. Therefore, we analyzed those who were followed to death or for at least 12 mo after completing anti-CD20 Ab. Pretreatment serologic tests were used to categorize chronic HBV (hepatitis B surface antigen positive or HBsAg+), past HBV (HBsAg-, hepatitis B core antibody positive or HBcAb+), resolved HBV (HBsAg-, HBcAb+, hepatitis B surface antibody positive or HBsAb+), likely prior vaccination (isolated HBsAb+), HBV negative (HBsAg-, HBcAb-), or unknown. Acute hepatitis B was defined by the appearance of HBsAg+ in the high risk period in patients who were pretreatment HBV negative. We assessed HBV antiviral treatment and the incidence of hepatitis, liver failure, and death during the high risk period. Cumulative hepatitis, liver failure, and death after anti-CD20 Ab initiation were compared by HBV disease categories and differences compared using the χ(2) test. Mean time to hepatitis peak alanine aminotransferase, liver failure, and death relative to anti-CD20 Ab administration and follow-up were also compared by HBV disease group. RESULTS: Among 19304 VHA patients who received anti-CD20 Ab, 10224 (53%) had pretreatment HBsAg testing during the study period, with 49% and 43% tested for HBsAg and HBcAb, respectively within 6 mo pretreatment in 2014. Of those tested, 2% (167/10224) had chronic HBV, 4% (326/7903) past HBV, 5% (427/8110) resolved HBV, 8% (628/8110) likely prior HBV vaccination, and 76% (6022/7903) were HBV negative. In those with chronic HBV infection, ≤ 37% received HBV antiviral treatment during the high risk period while 21% to 23% of those with past or resolved HBV, respectively, received HBV antiviral treatment. During and 12 mo after anti-CD20 Ab, the rate of hepatitis was significantly greater in those HBV positive vs negative (P = 0.001). The mortality rate was 35%-40% in chronic or past hepatitis B and 26%-31% in hepatitis B negative. In those pretreatment HBV negative, 16 (0.3%) developed acute hepatitis B of 4947 tested during anti-CD20Ab treatment and follow-up. CONCLUSION: While HBV testing of Veterans has increased prior to anti-CD20 Ab, few HBV+ patients received HBV antivirals, suggesting electronic health record algorithms may enhance health outcomes.

An evaluation of the implementation of hand held health records with adults with learning disabilities: a cluster randomized controlled trial

Background: Personal health records were implemented with adults with learning disabilities (AWLD) to try to improve their health-care. Materials and Method: Forty GP practices were randomized to the Personal Health Profile (PHP) implementation or control group. Two hundred and one AWLD were interviewed at baseline and 163 followed up after 12 months intervention (PHP group). AWLD and carers of AWLD were employed as research interviewers. AWLD were full research participants. Results: Annual consultation rates in the intervention and control groups at baseline were low (2.3 and 2.6 visits respectively). A slightly greater increase occurred over the year in the intervention group 0.6 ()0.4 to 1.6) visits ⁄ year compared with controls. AWLD in PHP group reported more health problems at follow-up 0.9 (0.0 to 1.8). AWLD liked their PHP (92%) but only 63% AWLD and 55% carers reported PHP usage. Carers had high turnover (34%). Conclusions: No significant outcomes were achieved by the intervention.

Tackling antibiotic resistance: a dose of common antisense?

Resistance to antimicrobial agents undermines our ability to treat bacterial infections. It attracts intense media and political interest and impacts on personal health and costs to health infrastructures. Bacteria have developed resistance to all licensed antibacterial agents, and their ability to become resistant to unlicensed agents is often demonstrated during the development process. Conventional approaches to antimicrobial development, involving modification of existing agents or production of synthetic derivatives, are unlikely to deliver the range or type of drugs that will be needed to meet all future requirements. Although many companies are seeking novel targets, further radical approaches to both antimicrobial design and the reversal of resistance are now urgently required. In this article, we discuss ‘antisense’ (or ‘antigene’) strategies to inhibit resistance mechanisms at the genetic level. These offer an innovative approach to a global problem and could be used to restore the efficacy of clinically proven agents. Moreover, this strategy has the potential to overcome critical resistances, not only in the so-called ‘superbugs’ (methicillin-resistant Staphylococcus aureus, glycopeptide-resistant enterococci and multidrug-resistant strains of Acinetobacter baumannii, and Pseudomonas aeruginosa), but in resistant strains of any bacterial species.

A comparison of self-report and health care provider data to assess surveillance definitions of influenza-like illness in outpatients

Objective: Several surveillance definitions of influenza-like illness (ILI) have been proposed, based on the presence of symptoms. Symptom data can be obtained from patients, medical records, or both. Past research has found that agreements between health record data and self-report are variable depending on the specific symptom. Therefore, we aimed to explore the implications of using data on influenza symptoms extracted from medical records, similar data collected prospectively from outpatients, and the combined data from both sources as predictors of laboratory-confirmed influenza. Methods: Using data from the Hutterite Influenza Prevention Study, we calculated: 1) the sensitivity, specificity and predictive values of individual symptoms within surveillance definitions; 2) how frequently surveillance definitions correlated to laboratory-confirmed influenza; and 3) the predictive value of surveillance definitions. Results: Of the 176 participants with reports from participants and medical records, 142 (81%) were tested for influenza and 37 (26%) were PCR positive for influenza. Fever (alone) and fever combined with cough and/or sore throat were highly correlated with being PCR positive for influenza for all data sources. ILI surveillance definitions, based on symptom data from medical records only or from both medical records and self-report, were better predictors of laboratory-confirmed influenza with higher odds ratios and positive predictive values. Discussion: The choice of data source to determine ILI will depend on the patient population, outcome of interest, availability of data source, and use for clinical decision making, research, or surveillance. © Canadian Public Health Association, 2012.

Who's minding the shop?:the role of Canadian research ethics boards in the creation and uses of registries and biobanks

Abstract. Background. The amount of research utilizing health information has increased dramatically over the last ten years. Many institutions have extensive biobank holdings collected over a number of years for clinical and teaching purposes, but are uncertain as to the proper circumstances in which to permit research uses of these samples. Research Ethics Boards (REBs) in Canada and elsewhere in the world are grappling with these issues, but lack clear guidance regarding their role in the creation of and access to registries and biobanks. Methods. Chairs of 34 REBS and/or REB Administrators affiliated with Faculties of Medicine in Canadian universities were interviewed. Interviews consisted of structured questions dealing with diabetes-related scenarios, with open-ended responses and probing for rationales. The two scenarios involved the development of a diabetes registry using clinical encounter data across several physicians' practices, and the addition of biological samples to the registry to create a biobank. Results. There was a wide range of responses given for the questions raised in the scenarios, indicating a lack of clarity about the role of REBs in registries and biobanks. With respect to the creation of a registry, a minority of sites felt that consent was not required for the information to be entered into the registry. Whether patient consent was required for information to be entered into the registry and the duration for which the consent would be operative differed across sites. With respect to the creation of a biobank linked to the registry, a majority of sites viewed biobank information as qualitatively different from other types of personal health information. All respondents agreed that patient consent was needed for blood samples to be placed in the biobank but the duration of consent again varied. Conclusion. Participants were more attuned to issues surrounding biobanks as compared to registries and demonstrated a higher level of concern regarding biobanks. As registries and biobanks expand, there is a need for critical analysis of suitable roles for REBs and subsequent guidance on these topics. The authors conclude by recommending REB participation in the creation of registries and biobanks and the eventual drafting of comprehensive legislation.