Medical care commonly involves the apprehension of complex patterns of patient derangements to which the practitioner responds with patterns of interventions, as opposed to single therapeutic maneuvers. This complexity renders the objective assessment of practice patterns using conventional statistical approaches difficult.
Combinatorial approaches drawn from symbolic dynamics are used to encode the observed patterns of patient derangement and associated practitioner response patterns as sequences of symbols. Concatenating each patient derangement symbol with the contemporaneous practitioner response symbol creates “words” encoding the simultaneous patient derangement and provider response patterns and yields an observed vocabulary with quantifiable statistical characteristics.
A fundamental observation in many natural languages is the existence of a power law relationship between the rank order of word usage and the absolute frequency with which particular words are uttered. We show that population level patterns of patient derangement: practitioner intervention word usage in two entirely unrelated domains of medical care display power law relationships similar to those of natural languages, and that–in one of these domains–power law behavior at the population level reflects power law behavior at the level of individual practitioners.
Our results suggest that patterns of medical care can be approached using quantitative linguistic techniques, a finding that has implications for the assessment of expertise, machine learning identification of optimal practices, and construction of bedside decision support tools.
Power law; Natural languages; Mechanical ventilation; Dialysis; Medical practices
The objective of this study was to ascertain the performance of syndromic algorithms for the early detection of patients in healthcare facilities who have potentially transmissible infectious diseases, using computerised emergency department (ED) data.
A retrospective cohort in an 810-bed University of Lyon hospital in France was analysed. Adults who were admitted to the ED and hospitalised between June 1, 2007, and March 31, 2010 were included (N=10895). Different algorithms were built to detect patients with infectious respiratory, cutaneous or gastrointestinal syndromes. The performance parameters of these algorithms were assessed with regard to the capacity of our infection-control team to investigate the detected cases.
For respiratory syndromes, the sensitivity of the detection algorithms was 82.70%, and the specificity was 82.37%. For cutaneous syndromes, the sensitivity of the detection algorithms was 78.08%, and the specificity was 95.93%. For gastrointestinal syndromes, the sensitivity of the detection algorithms was 79.41%, and the specificity was 81.97%.
This assessment permitted us to detect patients with potentially transmissible infectious diseases, while striking a reasonable balance between true positives and false positives, for both respiratory and cutaneous syndromes. The algorithms for gastrointestinal syndromes were not specific enough for routine use, because they generated a large number of false positives relative to the number of infected patients. Detection of patients with potentially transmissible infectious diseases will enable us to take precautions to prevent transmission as soon as these patients come in contact with healthcare facilities.
Emergency service; Hospital; Syndromic surveillance; Detection algorithm; Infection control; Sensitivity and specificity; Population surveillance
Methods for linking real-world healthcare data often use a latent class model, where the latent, or unknown, class is the true match status of candidate record-pairs. This commonly used model assumes that agreement patterns among multiple fields within a latent class are independent. When this assumption is violated, various approaches, including the most commonly proposed loglinear models, have been suggested to account for conditional dependence.
We present a step-by-step guide to identify important dependencies between fields through a correlation residual plot and demonstrate how they can be incorporated into loglinear models for record linkage. This method is applied to healthcare data from the patient registry for a large county health department.
Our method could be readily implemented using standard software (with code supplied) to produce an overall better model fit as measured by BIC and deviance. Finding the most parsimonious model is known to reduce bias in parameter estimates.
This novel approach identifies and accommodates conditional dependence in the context of record linkage. The conditional dependence model is recommended for routine use due to its flexibility for incorporating conditional dependence and easy implementation using existing software.
Poor adherence to the Integrated Management of Childhood Illness (IMCI) protocol reduces the potential impact on under-five morbidity and mortality. Electronic technology could improve adherence; however there are few studies demonstrating the benefits of such technology in a resource-poor settings. This study estimates the impact of electronic technology on adherence to the IMCI protocols as compared to the current paper-based protocols in Tanzania.
In four districts in Tanzania, 18 clinics were randomly selected for inclusion. At each site, observers documented critical parts of the clinical assessment of children aged 2 months to 5 years. The first set of observations occurred during examination of children using paper-based IMCI (pIMCI) and the next set of observations occurred during examination using the electronic IMCI (eIMCI). Children were re-examined by an IMCI expert and the diagnoses were compared. A total of 1221 children (671 paper, 550 electronic) were observed.
For all ten critical IMCI items included in both systems, adherence to the protocol was greater for eIMCI than for pIMCI. The proportion assessed under pIMCI ranged from 61% to 98% compared to 92% to 100% under eIMCI (p < 0.05 for each of the ten assessment items).
Use of electronic systems improved the completeness of assessment of children with acute illness in Tanzania. With the before-after nature of the design, potential for temporal confounding is the primary limitation. However, the data collection for both phases occurred over a short period (one month) and so temporal confounding was expected to be minimal. The results suggest that the use of electronic IMCI protocols can improve the completeness and consistency of clinical assessments and future studies will examine the long-term health and health systems impact of eIMCI.
Recommendations from international task forces on geriatric assessment
emphasize the need for research including validation of cancer-specific
geriatric assessment (C-SGA) tools in oncological settings. This study was
to evaluate the feasibility of the SAKK Cancer-Specific Geriatric Assessment
(C-SGA) in clinical practice.
A cross sectional study of cancer patients ≥65 years old
(N = 51) with pathologically confirmed cancer presenting for
initiation of chemotherapy treatment (07/01/2009-03/31/2011) at two oncology
departments in Swiss canton hospitals: Kantonsspital Graubünden (KSGR
N = 25), Kantonsspital St. Gallen (KSSG N = 26).
Data was collected using three instruments, the SAKK C-SGA plus physician
and patient evaluation forms. The SAKK C-SGA includes six measures covering
five geriatric assessment domains (comorbidity, function, psychosocial,
nutrition, cognition) using a mix of medical record abstraction (MRA) and
patient interview. Five individual domains and one overall SAKK C-SGA score
were calculated and dichotomized as below/above literature-based cut-offs.
The SAKK C-SGA was evaluated by: patient and physician estimated time to
complete, ease of completing, and difficult or unanswered questions.
Time to complete the patient questionnaire was considered acceptable by
almost all (≥96%) patients and physicians. Patients reported slightly
shorter times to complete the questionnaire than physicians
(17.33 ± 7.34 vs.
20.59 ± 6.53 minutes, p = 0.02). Both
groups rated the patient questionnaire as easy/fairly easy to complete (91%
vs. 84% respectively, p = 0.14) with few difficult
or unanswered questions. The MRA took on average
8.32 ± 4.72 minutes to complete. Physicians (100%)
considered time to complete MRA acceptable, 96% rated it as easy/fairly easy
to complete. Individual study site populations differed on health-related
characteristics (excellent/good physician-rated general health KSGR 71%
vs. KSSG 32%, p = 0.007). The overall mean C-SGA
score was 2.4 ± 1.12. Patients at KSGR had lower C-SGA
scores (2.00 ± 1.19 vs.
2.81 ± 0.90, p = 0.009) and a smaller
proportion (28% vs.65%, p = 0.008) was above the C-SGA
cut-off score compared to KSSG.
These results suggest the SAKK C-SGA is a feasible practical tool for use in
clinical practice. It demonstrated discriminative ability based on objective
geriatric assessment measures, but additional investigations on use for
clinical decision-making are warranted. The SAKK C-SGA also provides
important usable domain information for intervention to optimize outcomes in
older cancer patients.
Assessment; Cancer-specific geriatric assessment; Decision-making; Geriatric assessment; Older cancer patients; Older adults
The forces which affect homelessness are complex and often interactive in nature. Social forces such as addictions, family breakdown, and mental illness are compounded by structural forces such as lack of available low-cost housing, poor economic conditions, and insufficient mental health services. Together these factors impact levels of homelessness through their dynamic relations. Historic models, which are static in nature, have only been marginally successful in capturing these relationships.
Fuzzy Logic (FL) and fuzzy cognitive maps (FCMs) are particularly suited to the modeling of complex social problems, such as homelessness, due to their inherent ability to model intricate, interactive systems often described in vague conceptual terms and then organize them into a specific, concrete form (i.e., the FCM) which can be readily understood by social scientists and others. Using FL we converted information, taken from recently published, peer reviewed articles, for a select group of factors related to homelessness and then calculated the strength of influence (weights) for pairs of factors. We then used these weighted relationships in a FCM to test the effects of increasing or decreasing individual or groups of factors. Results of these trials were explainable according to current empirical knowledge related to homelessness.
Prior graphic maps of homelessness have been of limited use due to the dynamic nature of the concepts related to homelessness. The FCM technique captures greater degrees of dynamism and complexity than static models, allowing relevant concepts to be manipulated and interacted. This, in turn, allows for a much more realistic picture of homelessness. Through network analysis of the FCM we determined that Education exerts the greatest force in the model and hence impacts the dynamism and complexity of a social problem such as homelessness.
The FCM built to model the complex social system of homelessness reasonably represented reality for the sample scenarios created. This confirmed that the model worked and that a search of peer reviewed, academic literature is a reasonable foundation upon which to build the model. Further, it was determined that the direction and strengths of relationships between concepts included in this map are a reasonable approximation of their action in reality. However, dynamic models are not without their limitations and must be acknowledged as inherently exploratory.
Homelessness; Complex social system; Fuzzy logic; Fuzzy Cognitive Map; Network analysis
Decisions to adopt a particular innovation may vary between stakeholders because individual stakeholders may disagree on the costs and benefits involved. This may translate to disagreement between stakeholders on priorities in the implementation process, possibly explaining the slow diffusion of innovations in health care. In this study, we explore the differences in stakeholder preferences for innovations, and quantify the difference in stakeholder priorities regarding costs and benefits.
The decision support technique called the analytic hierarchy process was used to quantify the preferences of stakeholders for nine information technology (IT) innovations in hospital care. The selection of the innovations was based on a literature review and expert judgments. Decision criteria related to the costs and benefits of the innovations were defined. These criteria were improvement in efficiency, health gains, satisfaction with care process, and investments required. Stakeholders judged the importance of the decision criteria and subsequently prioritized the selected IT innovations according to their expectations of how well the innovations would perform for these decision criteria.
The stakeholder groups (patients, nurses, physicians, managers, health care insurers, and policy makers) had different preference structures for the innovations selected. For instance, self-tests were one of the innovations most preferred by health care insurers and managers, owing to their expected positive impacts on efficiency and health gains. However, physicians, nurses and patients strongly doubted the health gains of self-tests, and accordingly ranked self-tests as the least-preferred innovation.
The various stakeholder groups had different expectations of the value of the nine IT innovations. The differences are likely due to perceived stakeholder benefits of each innovation, and less to the costs to individual stakeholder groups. This study provides a first exploratory quantitative insight into stakeholder positions concerning innovation in health care, and presents a novel way to study differences in stakeholder preferences. The results may be taken into account by decision makers involved in the implementation of innovations.
Implementation; Information technology; Innovation; Hospital care; Stakeholders
Prior studies demonstrate the suitability of natural language processing (NLP) for identifying pneumonia in chest radiograph (CXR) reports, however, few evaluate this approach in intensive care unit (ICU) patients.
From a total of 194,615 ICU reports, we empirically developed a lexicon to categorize pneumonia-relevant terms and uncertainty profiles. We encoded lexicon items into unique queries within an NLP software application and designed an algorithm to assign automated interpretations (‘positive’, ‘possible’, or ‘negative’) based on each report’s query profile. We evaluated algorithm performance in a sample of 2,466 CXR reports interpreted by physician consensus and in two ICU patient subgroups including those admitted for pneumonia and for rheumatologic/endocrine diagnoses.
Most reports were deemed ‘negative’ (51.8%) by physician consensus. Many were ‘possible’ (41.7%); only 6.5% were ‘positive’ for pneumonia. The lexicon included 105 terms and uncertainty profiles that were encoded into 31 NLP queries. Queries identified 534,322 ‘hits’ in the full sample, with 2.7 ± 2.6 ‘hits’ per report. An algorithm, comprised of twenty rules and probability steps, assigned interpretations to reports based on query profiles. In the validation set, the algorithm had 92.7% sensitivity, 91.1% specificity, 93.3% positive predictive value, and 90.3% negative predictive value for differentiating ‘negative’ from ‘positive’/’possible’ reports. In the ICU subgroups, the algorithm also demonstrated good performance, misclassifying few reports (5.8%).
Many CXR reports in ICU patients demonstrate frank uncertainty regarding a pneumonia diagnosis. This electronic tool demonstrates promise for assigning automated interpretations to CXR reports by leveraging both terms and uncertainty profiles.
Pneumonia; Intensive care unit; Natural language processing; Chest imaging; Electronic tool
The editors of BMC Medical Informatics and Decision Making would like to thank all our reviewers who have contributed to the journal in Volume 12 (2012).
Adopting mobile electronic medical record (MEMR) systems is expected to be one of the superior approaches for improving nurses’ bedside and point of care services. However, nurses may use the functions for far fewer tasks than the MEMR supports. This may depend on their technological personality associated to MEMR acceptance. The purpose of this study is to investigate nurses’ personality traits in regard to technology readiness toward MEMR acceptance.
The study used a self-administered questionnaire to collect 665 valid responses from a large hospital in Taiwan. Structural Equation modeling was utilized to analyze the collected data.
Of the four personality traits of the technology readiness, the results posit that nurses are optimistic, innovative, secure but uncomfortable about technology. Furthermore, these four personality traits were all proven to have a significant impact on the perceived ease of use of MEMR while the perceived usefulness of MEMR was significantly influenced by the optimism trait only. The results also confirmed the relationships between the perceived components of ease of use, usefulness, and behavioral intention in the Technology Acceptance Model toward MEMR usage.
Continuous educational programs can be provided for nurses to enhance their information technology literacy, minimizing their stress and discomfort about information technology. Further, hospital should recruit, either internally or externally, more optimistic nurses as champions of MEMR by leveraging the instrument proposed in this study. Besides, nurses’ requirements must be fully understood during the development of MEMR to ensure that MEMR can meet the real needs of nurses. The friendliness of user interfaces of MEMR and the compatibility of nurses’ work practices as these will also greatly enhance nurses’ willingness to use MEMR. Finally, the effects of technology personality should not be ignored, indicating that hospitals should also include more employees’ characteristics beyond socio-demographic profiles in their personnel databases.
Mobile electronic medical record (MEMR); Nurses; Technology readiness index (TRI); Technology acceptance model (TAM)
The causes of the underutilization of disease modifying anti-rheumatic drugs (DMARDS) for rheumatoid arthritis (RA) are not fully known, but may in part, relate to individual patient factors including risk perception. Our objective was to identify the determinants of risk perception (RP) in RA patients and predictors of their willingness to take a proposed DMARD (DMARD willingness).
A cross-sectional mail survey of RA patients in a community rheumatology practice. Patients were presented a hypothetical decision scenario where they were asked to consider switching DMARDs. They evaluated how risky the proposed medication was and how likely they would be to take it.
The completed sample included 1009 RA patients. The overall survey response rate was 71%. Patient characteristics: age 61.6 years (range 18-93), 75% female, minority 6.5%, low or marginal health literacy 8.8%, depression 15.0%, duration RA 13.1 years (range 0.5 – 68). Regression models demonstrated that health literacy, independent of low educational achievement or other demographic (including race), was a common predictor of both RP and DMARD willingness. There was partial mediation of the effects of HL on DMARD willingness through RP. Depression and happiness had no significant effect on RP or DMARD willingness. RP was influenced by negative RA disease and treatment experience, while DMARD willingness was affected mainly by perceived disease control.
Risk aversion may be the result of potentially recognizable and correctable cognitive defect. Heightened clinician awareness, formal screening for low health literacy or cognitive impairment in high-risk populations, may identify patients could benefit from additional decision support.
Decision-making; Risk perception; Depression; Health disparity; Disease-modifying anti-rheumatic drugs; Rheumatoid arthritis
This study was performed to investigate the usefulness of clinical pathway (CP) using an electronic medical record (EMR) in pediatric patients undergoing closed pinning for supracondylar fracture of the humerus, by analyzing the length of hospital stay, hospital cost and satisfaction of the medical teams.
This before and after comparative study included consecutive children who underwent closed pinning for supracondylar fracture of the humerus since 2009. The pre-CP group consists of 90 patients with the mean age of 5.7 years, and the post-CP group consists of 32 patients with the mean age of 6.2 years. Multidisciplinary work-team developed CP using an EMR system in March 2011. The length of hospital stay was the primary outcome variable, and hospital cost and medical team’s satisfaction score were secondary outcome variables. The non-inferiority test was used to demonstrate the efficiency of the pathway.
The length of hospital stay decreased from 2.9 ± 0.7 days to 2.4 ± 0.7 days by 15.0%, after the implementation of CP, and the lower bound of the 95% CI of the difference (0.14 day) was within the non-inferiority margin of −0.3 days. The hospital cost decreased from 1162.2 ± 236.7 US$ to 1139.8 ± 291.1 US$ by 1.9% and the lower bound of the 95% CI of the difference was −81.3 US$, which did not exceed the non-inferiority margin of −116.2 US$. Therefore, the post-CP group was not inferior compared with the pre-CP group in term of the length of hospital stay and total hospital cost. There was significant increase in the satisfaction score for doctors after implementation of CP (p < 0.001), but, no change in the satisfaction score for nursing staffs (p = 0.793).
The development and implementation of CP, using an EMR, in pediatric patients undergoing closed pinning for supracondylar fracture of the humerus enhances the treatment efficiency by streamlining the treatment process with no increases of the length of the hospital stay and total hospital costs.
Clinical pathway; Supracondylar fracture; Length of hospital stay; Hospital cost
Despite considerable financial incentives for adoption, there is little evidence available about providers’ use and satisfaction with key functions of electronic health records (EHRs) that meet “meaningful use” criteria.
We surveyed primary care providers (PCPs) in 11 general internal medicine and family medicine practices affiliated with 3 health systems in Texas about their use and satisfaction with performing common tasks (documentation, medication prescribing, preventive services, problem list) in the Epic EHR, a common commercial system. Most practices had greater than 5 years of experience with the Epic EHR. We used multivariate logistic regression to model predictors of being a structured documenter, defined as using electronic templates or prepopulated dot phrases to document at least two of the three note sections (history, physical, assessment and plan).
146 PCPs responded (70%). The majority used free text to document the history (51%) and assessment and plan (54%) and electronic templates to document the physical exam (57%). Half of PCPs were structured documenters (55%) with family medicine specialty (adjusted OR 3.3, 95% CI, 1.4-7.8) and years since graduation (nonlinear relationship with youngest and oldest having lowest probabilities) being significant predictors. Nearly half (43%) reported spending at least one extra hour beyond each scheduled half-day clinic completing EHR documentation. Three-quarters were satisfied with documenting completion of pneumococcal vaccinations and half were satisfied with documenting cancer screening (57% for breast, 45% for colorectal, and 46% for cervical). Fewer were satisfied with reminders for overdue pneumococcal vaccination (48%) and cancer screening (38% for breast, 37% for colorectal, and 31% for cervical). While most believed the problem list was helpful (70%) and kept an up-to-date list for their patients (68%), half thought they were unreliable and inaccurate (51%).
Dissatisfaction with and suboptimal use of key functions of the EHR may mitigate the potential for EHR use to improve preventive health and chronic disease management. Future work should optimize use of key functions and improve providers’ time efficiency.
Electronic health record (EHR); Attitude of health personnel; Attitude to computers; Primary care; Efficiency; Quality of care; Medical informatics/utilization
Privacy and information security are important for all healthcare services, including home-based services. We have designed and implemented a prototype technology platform for providing home-based healthcare services. It supports a personal electronic health diary and enables secure and reliable communication and interaction with peers and healthcare personnel. The platform runs on a small computer with a dedicated remote control. It is connected to the patient’s TV and to a broadband Internet. The platform has been tested with home-based rehabilitation and education programs for chronic obstructive pulmonary disease and diabetes. As part of our work, a risk assessment of privacy and security aspects has been performed, to reveal actual risks and to ensure adequate information security in this technical platform.
Risk assessment was performed in an iterative manner during the development process. Thus, security solutions have been incorporated into the design from an early stage instead of being included as an add-on to a nearly completed system. We have adapted existing risk management methods to our own environment, thus creating our own method. Our method conforms to ISO’s standard for information security risk management.
A total of approximately 50 threats and possible unwanted incidents were identified and analysed. Among the threats to the four information security aspects: confidentiality, integrity, availability, and quality; confidentiality threats were identified as most serious, with one threat given an unacceptable level of High risk. This is because health-related personal information is regarded as sensitive. Availability threats were analysed as low risk, as the aim of the home programmes is to provide education and rehabilitation services; not for use in acute situations or for continuous health monitoring.
Most of the identified threats are applicable for healthcare services intended for patients or citizens in their own homes. Confidentiality risks in home are different from in a more controlled environment such as a hospital; and electronic equipment located in private homes and communicating via Internet, is more exposed to unauthorised access. By implementing the proposed measures, it has been possible to design a home-based service which ensures the necessary level of information security and privacy.
Privacy; Confidentiality; Information security; Risk assessment; Pulmonary rehabilitation; Diabetes self-management education; Video conference; Tele-homecare
Audit Trails (AT) are fundamental to information security in order to guarantee access traceability but can also be used to improve Health information System’s (HIS) quality namely to assess how they are used or misused. This paper aims at analysing the existence and quality of AT, describing scenarios in hospitals and making some recommendations to improve the quality of information.
The responsibles of HIS for eight Portuguese hospitals were contacted in order to arrange an interview about the importance of AT and to collect audit trail data from their HIS. Five institutions agreed to participate in this study; four of them accepted to be interviewed, and four sent AT data. The interviews were performed in 2011 and audit trail data sent in 2011 and 2012. Each AT was evaluated and compared in relation to data quality standards, namely for completeness, comprehensibility, traceability among others. Only one of the AT had enough information for us to apply a consistency evaluation by modelling user behaviour.
The interviewees in these hospitals only knew a few AT (average of 1 AT per hospital in an estimate of 21 existing HIS), although they all recognize some advantages of analysing AT. Four hospitals sent a total of 7 AT – 2 from Radiology Information System (RIS), 2 from Picture Archiving and Communication System (PACS), 3 from Patient Records. Three of the AT were understandable and three of the AT were complete. The AT from the patient records are better structured and more complete than the RIS/PACS.
Existing AT do not have enough quality to guarantee traceability or be used in HIS improvement. Its quality reflects the importance given to them by the CIO of healthcare institutions. Existing standards (e.g. ASTM:E2147, ISO/TS 18308:2004, ISO/IEC 27001:2006) are still not broadly used in Portugal.
Patients with prostate cancer face the difficult decision between a wide range of therapeutic options. These men require elaborate information about their individual risk profile and the therapeutic strategies´ risks and benefits to choose the best possible option. In order to detect time trends and quality improvements between an early patient population (2003/2004) and a later reference group (2007/2008) data was analysed with regards to epidemiologic parameters, differences in diagnostics and the type and ranking of the recommended therapies taking into account changes to Gleason Grading System and implementation of new therapeutic strategies, particularly Active surveillance, in 2005.
Data from all 496 consecutive patients who received consultation in 2003/2004 (n = 280) and 2007/2008 (n = 216) was retrospectively evaluated. Categorical variables were compared using the Chi-square test. Dependent variables were analysed using the unpaired Students´ t-test and the Mann–Whitney U-test.
The cohorts were comparable concerning clinical stage, initial PSA, prostate volume, comorbidities and organ confined disease. Patients in Cohort I were younger (66.44 vs. 69.31y; p < .001) and had a longer life expectancy (17.22 vs. 14.75y; p < .001). 50.9%, 28.2% and 20.9% in Cohort I and 37.2%, 39.6% and 23.2% in Cohort II showed low-, intermediate- and high-risk disease (D´Amico) with a trend towards an increased risk profile in Cohort II (p = .066). The risk-adapted therapy recommended as first option was radical prostatectomy for 91.5% in Cohort I and 69.7% in Cohort II, radiation therapy for 83.7% in Cohort I and 50.7% in Cohort II, and other therapies (brachytherapy, Active surveillance, Watchful waiting, high-intensity focused ultrasound) for 6.5% in Cohort I and 6.9% in Cohort II (p < .001). Radiation therapy was predominant in both cohorts as second treatment option (p < .001). Time trends showing quality improvement involved an increase in biopsy cores (9.95 ± 2.38 vs. 8.43 ± 2.29; p < .001) and an increased recommendation for bilateral nerve sparing (p < .001).
In the earlier years, younger patients with a more favourable risk profile presented for interdisciplinary consultation. A unilateral recommendation for radical prostatectomy and radiation therapy was predominant. In the later years, the patient population was considerably older. However, this group may have benefitted from optimised diagnostic possibilities and a wider range of treatment options.
Prostate cancer; Interdisciplinary consultation; Medical decision making; Time trends
Electronic decision support is commonplace in medical practice. However, its adoption at the point-of-care is dependent on a range of organisational, patient and clinician-related factors. In particular, level of clinical experience is an important driver of electronic decision support uptake. Our objective was to examine the way in which Australian doctors at different stages of medical training use a web-based oncology system (http://www.eviq.org.au).
We used logfiles to examine the characteristics of eviQ registrants (2009–2012) and patterns of eviQ use in 2012, according to level of medical training. We also used a web-based survey to evaluate the way doctors at different levels of medical training use the online system and to elicit perceptions of the system’s utility in oncology care.
Our study cohort comprised 2,549 eviQ registrants who were hospital-based medical doctors across all levels of training. 65% of the cohort used eviQ in 2012, with 25% of interns/residents, 61% of advanced oncology trainees and 47% of speciality-qualified oncologists accessing eviQ in the last 3 months of 2012. The cohort accounted for 445,492 webhits in 2012. On average, advanced trainees used eviQ up to five-times more than other doctors (42.6 webhits/month compared to 22.8 for specialty-qualified doctors and 7.4 webhits/month for interns/residents). Of the 52 survey respondents, 89% accessed eviQ’s chemotherapy protocols on a daily or weekly basis in the month prior to the survey. 79% of respondents used eviQ at least weekly to initiate therapy and to support monitoring (29%), altering (35%) or ceasing therapy (19%). Consistent with the logfile analysis, advanced oncology trainees report more frequent eviQ use than doctors at other stages of medical training.
The majority of the Australian oncology workforce are registered on eviQ. The frequency of use directly mirrors the clinical role of doctors and attitudes about the utility of eviQ in decision-making. Evaluations of this kind generate important data for system developers and medical educators to drive improvements in electronic decision support to better meet the needs of clinicians. This end-user focus will optimise the uptake of systems which will translate into improvements in processes of care and patient outcomes.
Clinical decision support systems; Evidence-based practice; Medical education; Cancer chemotherapy protocols; Health personnel; ‘Medical staff; Hospital’
Effective population management of patients with diabetes requires timely recognition. Current case-finding algorithms can accurately detect patients with diabetes, but lack real-time identification. We sought to develop and validate an automated, real-time diabetes case-finding algorithm to identify patients with diabetes at the earliest possible date.
The source population included 160,872 unique patients from a large public hospital system between January 2009 and April 2011. A diabetes case-finding algorithm was iteratively derived using chart review and subsequently validated (n = 343) in a stratified random sample of patients, using data extracted from the electronic health records (EHR). A point-based algorithm using encounter diagnoses, clinical history, pharmacy data, and laboratory results was used to identify diabetes cases. The date when accumulated points reached a specified threshold equated to the diagnosis date. Physician chart review served as the gold standard.
The electronic model had a sensitivity of 97%, specificity of 90%, positive predictive value of 90%, and negative predictive value of 96% for the identification of patients with diabetes. The kappa score for agreement between the model and physician for the diagnosis date allowing for a 3-month delay was 0.97, where 78.4% of cases had exact agreement on the precise date.
A diabetes case-finding algorithm using data exclusively extracted from a comprehensive EHR can accurately identify patients with diabetes at the earliest possible date within a healthcare system. The real-time capability may enable proactive disease management.
The present study aimed to develop an artificial neural network (ANN) based prediction model for cardiovascular autonomic (CA) dysfunction in the general population.
We analyzed a previous dataset based on a population sample consisted of 2,092 individuals aged 30–80 years. The prediction models were derived from an exploratory set using ANN analysis. Performances of these prediction models were evaluated in the validation set.
Univariate analysis indicated that 14 risk factors showed statistically significant association with CA dysfunction (P < 0.05). The mean area under the receiver-operating curve was 0.762 (95% CI 0.732–0.793) for prediction model developed using ANN analysis. The mean sensitivity, specificity, positive and negative predictive values were similar in the prediction models was 0.751, 0.665, 0.330 and 0.924, respectively. All HL statistics were less than 15.0.
ANN is an effective tool for developing prediction models with high value for predicting CA dysfunction among the general population.
Cardiovascular autonomic dysfunction; Artificial neural network; Prediction model; Chinese population
Although usage and acceptance are important factors for a successful implementation of clinical decision support systems for medication, most studies only concentrate on their design and outcome. Our objective was to comparatively investigate a set of traditional medication safety measures such as medication safety training for physicians, paper-based posters and checklists concerning potential medication problems versus the additional benefit of a computer-assisted medication check. We concentrated on usage, acceptance and suitability of such interventions in a busy emergency department (ED) of a 749 bed acute tertiary care hospital.
A retrospective, qualitative evaluation study was conducted using a field observation and a questionnaire-based survey. Six physicians were observed while treating 20 patient cases; the questionnaire, based on the Technology Acceptance Model 2 (TAM2), has been answered by nine ED physicians.
During field observations, we did not observe direct use of any of the implemented interventions for medication safety (paper-based and electronic). Questionnaire results indicated that the electronic medication safety check was the most frequently used intervention, followed by checklist and posters. However, despite their positive attitude, physicians most often stated that they use the interventions in only up to ten percent for subjectively “critical” orders. Main reasons behind the low usage were deficits in ease-of-use and fit to the workflow. The intention to use the interventions was rather high after overcoming these barriers.
Methodologically, the study contributes to Technology Acceptance Model (TAM) research in an ED setting and confirms TAM2 as a helpful diagnostic tool in identifying barriers for a successful implementation of medication safety interventions. In our case, identified barriers explaining the low utilization of the implemented medication safety interventions - despite their positive reception - include deficits in accessibility, briefing for the physicians about the interventions, ease-of-use and compatibility to the working environment.
Evaluation; Medication safety; Emergency department; Clinical decision support systems; Technology acceptance; TAM2; Patient safety
Despite the increasing pervasiveness of mobile computational technologies, knowledge about psychiatric patients’ preferences regarding the design and utility of mobile applications is very poor. This paper reports on a pilot-study that involved 120 psychiatric patients in the development of a mobile application (app) that is being used for data entry into the Signature Project data bank at the Institut universitaire en santé mentale de Montréal (IUSMM), Canada. Participants were invited to comment on the ‘look and feel’ of the Signature App. Their input also extended the procedures for data collection. These suggestions may contribute to increased mental health literacy and empowerment of persons with mental illness receiving services at the IUSMM.
Participants were recruited to fill out a questionnaire on a tablet computer while waiting at the Emergency Room (ER, n = 40), Psychotic Disorders outpatient clinic (n = 40) or Anxiety and Mood Disorders outpatient clinic (n = 40) of IUSMM. Nine patients from each of these sub-groups participated in a focus group to review the results and to discuss how the design and use of the Signature App could be improved to better meet the needs of patients.
This study (n = 120) indicated that psychiatric patients are clearly capable of using a tablet computer to fill out questionnaires for quantitative data entry, and that they enjoyed this experience. Results from the focus groups (n = 27) highlight that the app could also be used by patients to communicate some personal and contextual qualitative information. This would support a holistic and person-centered approach, especially at the ER where people acutely need to describe their recent history and receive emotional support.
This pilot-study has confirmed the necessity of involving patients not only in the testing of a new mobile application, but also as active contributors in the entire research and development process of a person-centered information and communication technology infrastructure. The input of participants was essential in designing the Signature Project computational procedure and making use of the app a positive and empowering experience. Participants also gave critical feedback remarks that went beyond the initial scope of the pilot-study, for example they suggested the addition of a client-clinician component.
Personalization; Patient participation; Strategy for patient-oriented research; Information and communication technology; Experiential knowledge translation; Person- centeredness; eMental health; Mobile application; Signature project; Mixed methods
Translational medical research literature has increased rapidly in the last few decades and played a more and more important role during the development of medicine science. The main aim of this study is to evaluate the global performance of translational medical research during the past few decades.
Bibliometric, social network analysis, and visualization technologies were used for analyzing translational medical research performance from the aspects of subject categories, journals, countries, institutes, keywords, and MeSH terms. Meanwhile, the co-author, co-words and cluster analysis methods were also used to trace popular topics in translational medical research related work.
Research output suggested a solid development in translational medical research, in terms of increasing scientific production and research collaboration. We identified the core journals, mainstream subject categories, leading countries, and institutions in translational medical research. There was an uneven distribution of publications at authorial, institutional, and national levels. The most commonly used keywords that appeared in the articles were “translational research”, “translational medicine”, “biomarkers”, “stroke”, “inflammation”, “cancer”, and “breast cancer”.
The subject categories of “Research & Experimental Medicine”, “Medical Laboratory Technology”, and “General & Internal Medicine” play a key role in translational medical research both in production and in its networks. Translational medical research and CTS, etc. are core journals of translational research. G7 countries are the leading nations for translational medical research. Some developing countries, such as P.R China, also play an important role in the communication of translational research. The USA and its institutions play a dominant role in the production, collaboration, citations and high quality articles. The research trends in translational medical research involve drug design and development, pathogenesis and treatment of disease, disease model research, evidence-based research, and stem and progenitor cells.
Translational medical research; Web of science; Informetrics; Social network analysis
Following the completion of treatment and as they enter the follow-up phase, breast cancer patients (BCPs) often recount feeling ‘lost in transition’, and are left with many questions concerning how their ongoing care and monitoring for recurrence will be managed. Family physicians (FPs) also frequently report feeling ill-equipped to provide follow-up care to BCPs. In this three-phase qualitative pilot study we designed, implemented and evaluated a multi-faceted survivorship care plan (SCP) to address the information needs of BCPs at our facility and of their FPs.
In Phase 1 focus groups and individual interviews were conducted with 35 participants from three stakeholder groups (BCPs, FPs and oncology specialist health care providers (OHCPs)), to identify specific information needs. An SCP was then designed based on these findings, consisting of both web-based and paper-based tools (Phase 2). For Phase 3, both sets of tools were subsequently evaluated via focus groups and interviews with 26 participants. Interviews and focus groups were audio taped, transcribed and content analysed for emergent themes and patterns.
In Phase 1 patients commented that web-based, paper-based and human resources components were desirable in any SCP. Patients did not focus exclusively on the post-treatment period, but instead spoke of evolving needs throughout their cancer journey. FPs indicated that any tools to support them must distill important information in a user-friendly format. In Phase 2, a pilot SCP was subsequently designed, consisting of both web-based and paper-based materials tailored specifically to the needs of BCPs as well as FPs. During Phase 3 (evaluation) BCPs indicated that the SCP was effective at addressing many of their needs, and offered suggestions for future improvements. Both patients and FPs found the pilot SCP to be an improvement from the previous standard of care. Patients perceived the quality of the BCP-FP relationship as integral to their comfort with FPs assuming follow-up responsibilities.
This pilot multi-component SCP shows promise in addressing the information needs of BCPs and the FPs who care for them. Next steps include refinement of the different SCP components, further evaluation (including usability testing), and planning for more extensive implementation.
Breast cancer; Survivorship; Qualitative methods; Pilot study; Survivorship care plan; Information needs
The usage of patient data for research poses risks concerning the patients’ privacy and informational self-determination. Next-generation-sequencing technologies and various other methods gain data from biospecimen, both for translational research and personalized medicine. If these biospecimen are anonymized, individual research results from genomic research, which should be offered to patients in a clinically relevant timeframe, cannot be associated back to the individual. This raises an ethical concern and challenges the legitimacy of anonymized patient samples. In this paper we present a new approach which supports both data privacy and the possibility to give feedback to patients about their individual research results.
We examined previously published privacy concepts regarding a streamlined de-pseudonymization process and a patient-based pseudonym as applicable to research with genomic data and warehousing approaches. All concepts identified in the literature review were compared to each other and analyzed for their applicability to translational research projects. We evaluated how these concepts cope with challenges implicated by personalized medicine. Therefore, both person-centricity issues and a separation of pseudonymization and de-pseudonymization stood out as a central theme in our examination. This motivated us to enhance an existing pseudonymization method regarding a separation of duties.
The existing concepts rely on external trusted third parties, making de-pseudonymization a multistage process involving additional interpersonal communication, which might cause critical delays in patient care. Therefore we propose an enhanced method with an asymmetric encryption scheme separating the duties of pseudonymization and de-pseudonymization. The pseudonymization service provider is unable to conclude the patient identifier from the pseudonym, but assigns this ability to an authorized third party (ombudsman) instead. To solve person-centricity issues, a collision-resistant function is incorporated into the method. These two facts combined enable us to address essential challenges in translational research. A productive software prototype was implemented to prove the functionality of the suggested translational, data privacy-preserving method. Eventually, we performed a threat analysis to evaluate potential hazards connected with this pseudonymization method.
The proposed method offers sustainable organizational simplification regarding an ethically indicated, but secure and controlled process of de-pseudonymizing patients. A pseudonym is patient-centered to allow correlating separate datasets from one patient. Therefore, this method bridges the gap between bench and bedside in translational research while preserving patient privacy. Assigned ombudsmen are able to de-pseudonymize a patient, if an individual research result is clinically relevant.
Pseudonymization; Pseudonymisation; Pseudonym; Anonyms and pseudonyms; Translational research; Biobanking; Data privacy; Individual research results; Record linkage; Re-identification; De-pseudonymization; De-pseudonymisation
In previous work, we described the development of an 81-item video-animated tool for assessing mobility. In response to criticism levied during a pilot study of this tool, we sought to develop a new version built upon a flexible framework for designing and administering the instrument.
Rather than constructing a self-contained software application with a hard-coded instrument, we designed an XML schema capable of describing a variety of psychometric instruments. The new version of our video-animated assessment tool was then defined fully within the context of a compliant XML document. Two software applications—one built in Java, the other in Objective-C for the Apple iPad—were then built that could present the instrument described in the XML document and collect participants’ responses. Separating the instrument’s definition from the software application implementing it allowed for rapid iteration and easy, reliable definition of variations.
Defining instruments in a software-independent XML document simplifies the process of defining instruments and variations and allows a single instrument to be deployed on as many platforms as there are software applications capable of interpreting the instrument, thereby broadening the potential target audience for the instrument. Continued work will be done to further specify and refine this type of instrument specification with a focus on spurring adoption by researchers in gerontology and geriatric medicine.