PMCC PMCC

Search tips
Search criteria

Advanced
Results 1-25 (976)
 

Clipboard (0)
None

Select a Filter Below

Journals
Year of Publication
more »
1.  Tailored e-Health services for the dementia care setting: a pilot study of ‘eHealthMonitor’ 
Background
The European eHealthMonitor project (eHM) developed a user-sensitive and interactive web portal for dementia care: the eHM Dementia Portal (eHM-DP). It aims to provide targeted and personalized support for informal caregivers of people with dementia in a home-based care setting. The objective of the pilot study was to obtain feedback on the eHM-DP from two user perspectives (caregivers and medical professionals), focusing on caregiver empowerment, decision aid, and the perceived benefits of the eHM-DP.
Methods
The study on the eHM-DP was conducted from March 2014 to June 2014. The methodological approach followed a user-participatory design with a total number of 42 participants. The study included caregivers of people with dementia and medical professionals (MPs) from the metropolitan region of Erlangen-Nürnberg (Bavaria, Germany). Study participants were interviewed face-to-face with semi-structured, written interviews.
Results
Caregivers indicated a high degree of perceived support by the eHM-DP and of provided decision aid. In total, 89 % of caregivers and 54 % of MPs would use the eHM-DP if access were provided. The primary benefits participants perceived were the acquisition of individualized information, computerized interaction between caregivers and MPs, empowerment in health-related decisions and comprehensive insights into the progress of the disease. Major recommendations for improving the eHM-DP encompassed: an active search functionality based on predefined terms, the implementation of a chatroom for caregivers, an upload function and alerts for MPs, as well as the overall design.
Conclusions
Our study is the first to have provided new insights and results on an interactive and needs-oriented web portal, endeavouring towards empowerment and assistance in decision making for caregivers as well as MPs within the realm of caring for patients with dementia. The acceptance and willingness to use the eHM-DP emphasizes the potential of eHealth services for community-based dementia care settings.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0182-2) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0182-2
PMCID: PMC4517387  PMID: 26215731
e-health; Web portal; Dementia; Caregiver
2.  Details of development of the resource for adults with asthma in the RAISIN (randomized trial of an asthma internet self-management intervention) study 
Background
Around 300 million people worldwide have asthma and prevalence is increasing. Self-management can be effective in improving a range of outcomes and is cost effective, but is underutilised as a treatment strategy. Supporting optimum self-management using digital technology shows promise, but how best to do this is not clear. We aimed to develop an evidence based, theory informed, online resource to support self-management in adults with asthma, called ‘Living well with Asthma’, as part of the RAISIN (Randomized Trial of an Asthma Internet Self-Management Intervention) study.
Methods
We developed Living well with Asthma in two phases.
Phase 1: A low fidelity prototype (paper-based) version of the website was developed iteratively through input from a multidisciplinary expert panel, empirical evidence from the literature, and potential end users via focus groups (adults with asthma and practice nurses). Implementation and behaviour change theories informed this process.
Phase 2: The paper-based designs were converted to a website through an iterative user centred process. Adults with asthma (n = 10) took part in think aloud studies, discussing the paper based version, then the web-based version. Participants considered contents, layout, and navigation. Development was agile using feedback from the think aloud sessions immediately to inform design and subsequent think aloud sessions. Think aloud transcripts were also thematically analysed, further informing resource development.
Results
The website asked users to aim to be symptom free. Key behaviours targeted to achieve this include: optimising medication use (including inhaler technique); attending primary care asthma reviews; using asthma action plans; increasing physical activity levels; and stopping smoking. The website had 11 sections, plus email reminders, which promoted these behaviours. Feedback on the contents of the resource was mainly positive with most changes focussing on clarification of language, order of pages and usability issues mainly relating to navigation difficulties.
Conclusions
Our multifaceted approach to online intervention development underpinned by theory, using evidence from the literature, co-designed with end users and a multidisciplinary panel has resulted in a resource which end users find relevant to their needs and easy to use. Living well with Asthma is undergoing evaluation within a randomized controlled trial.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0177-z) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0177-z
PMCID: PMC4517557  PMID: 26215651
Asthma; Self-management; Adherence; E-health; Randomized controlled trial; Complex intervention; Inhaled corticosteroids; Internet; Behaviour change; Lifeguide
3.  Usability evaluation and adaptation of the e-health Personal Patient Profile-Prostate decision aid for Spanish-speaking Latino men 
Background
The Personal Patient Profile-Prostate (P3P), a web-based decision aid, was demonstrated to reduce decisional conflict in English-speaking men with localized prostate cancer early after initial diagnosis. The purpose of this study was to explore and enhance usability and cultural appropriateness of a Spanish P3P by Latino men with a diagnosis of prostate cancer.
Methods
P3P was translated to Spanish and back-translated by three native Spanish-speaking translators working independently. Spanish-speaking Latino men with a diagnosis of localized prostate cancer, who had made treatment decisions in the past 24 months, were recruited from two urban clinical care sites. Individual cognitive interviews were conducted by two bilingual research assistants as each participant used the Spanish P3P. Notes of user behavior, feedback, and answers to direct questions about comprehension, usability and perceived usefulness were analyzed and categorized.
Results
Seven participants with a range of education levels identified 25 unique usability issues in navigation, content comprehension and completeness, sociocultural appropriateness, and methodology. Revisions were prioritized to refine the usability and cultural and linguistic appropriateness of the decision aid.
Conclusions
Usability issues were discovered that are potential barriers to effective decision support. Successful use of decision aids requires adaptation and testing beyond translation. Our findings led to revisions further refining the usability and linguistic and cultural appropriateness of Spanish P3P.
doi:10.1186/s12911-015-0180-4
PMCID: PMC4513952  PMID: 26204920
Usability; Adaptation; Spanish; Localized prostate cancer; Disparities; Decision aid
4.  Identifying the components of clinical vignettes describing Alzheimer’s disease or other dementias: a scoping review 
Background
Clinical vignettes are often used to elicit information about health conditions in research studies. This review summarizes the components of clinical vignettes describing Alzheimer’s disease (AD) or other dementias. The purpose is to provide recommendations for the development of standardized vignettes that may be used in future studies.
Methods
MEDLINE, EMBASE, PsycINFO, ASSIA, CINAHL were searched from their inception to June 2014. Primary English-language studies employing vignettes to describe AD or similar disorders (including other dementias and Parkinson’s disease) were included in the review. Included studies had to describe the content of the vignettes in the published manuscripts. The characteristics of the included studies and the vignettes were extracted in tabular form and summarized qualitatively.
Results
Forty-two studies were included in the review. Twenty-four of the studies contained at least one AD vignette, 11 had vignettes focusing on non-AD dementias, and seven contained vignettes describing conditions other than dementia. In total, 58 vignettes were obtained from the 42 included studies.
Conclusions
Key aspects to consider when constructing vignettes for AD or other dementias include writing the vignettes from a third-person perspective and presenting hypothetical patients as being at least 65 years of age. Researchers should develop standardized vignettes for use across studies.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0179-x) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0179-x
PMCID: PMC4502543  PMID: 26174660
Vignette; Alzheimer’s disease; Dementia
5.  Assessing measures of comorbidity and functional status for risk adjustment to compare hospital performance for colorectal cancer surgery: a retrospective data-linkage study 
Background
Comparing outcomes between hospitals requires consideration of patient factors that could account for any observed differences. Adjusting for comorbid conditions is common when studying outcomes following cancer surgery, and a commonly used measure is the Charlson comorbidity index. Other measures of patient health include the ECOG performance status and the ASA physical status score. This study aimed to ascertain how frequently ECOG and ASA scores are recorded in population-based administrative data collections in New South Wales, Australia and to assess the contribution each makes in addition to the Charlson comorbidity index in risk adjustment models for comparative assessment of colorectal cancer surgery outcomes between hospitals.
Methods
We used linked administrative data to identify 6964 patients receiving surgery for colorectal cancer in 2007 and 2008. We summarised the frequency of missing data for Charlson comorbidity index, ECOG and ASA scores, and compared patient characteristics between those with and without these measures. The performance of ASA and ECOG in risk adjustment models that also included Charlson index was assessed for three binary outcomes: 12-month mortality, extended length of stay and 28-day readmission. Patient outcomes were compared between hospital peer groups using multilevel logistic regression analysis.
Results
The Charlson comorbidity index could be derived for all patients, ASA score was recorded for 78 % of patients and ECOG performance status recorded for only 24 % of eligible patients. Including ASA or ECOG improved the predictive ability of models, but there was no consistently best combination. The addition of ASA or ECOG did not substantially change parameter estimates for hospital peer group after adjusting for Charlson comorbidity index.
Conclusions
While predictive ability of regression models is maximised by inclusion of one or both of ASA score and ECOG performance status, there is little to be gained by adding ASA or ECOG to models containing the Charlson comorbidity index to address confounding. The Charlson comorbidity index has good performance and is an appropriate measure to use in risk adjustment to compare outcomes between hospitals.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0175-1) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0175-1
PMCID: PMC4502567  PMID: 26174550
Risk adjustment; Comorbidity; Surgical outcomes; Administrative data; Charlson comorbidity index; ECOG performance status; ASA score
6.  Automatic classification of diseases from free-text death certificates for real-time surveillance 
Background
Death certificates provide an invaluable source for mortality statistics which can be used for surveillance and early warnings of increases in disease activity and to support the development and monitoring of prevention or response strategies. However, their value can be realised only if accurate, quantitative data can be extracted from death certificates, an aim hampered by both the volume and variable nature of certificates written in natural language. This study aims to develop a set of machine learning and rule-based methods to automatically classify death certificates according to four high impact diseases of interest: diabetes, influenza, pneumonia and HIV.
Methods
Two classification methods are presented: i) a machine learning approach, where detailed features (terms, term n-grams and SNOMED CT concepts) are extracted from death certificates and used to train a set of supervised machine learning models (Support Vector Machines); and ii) a set of keyword-matching rules. These methods were used to identify the presence of diabetes, influenza, pneumonia and HIV in a death certificate. An empirical evaluation was conducted using 340,142 death certificates, divided between training and test sets, covering deaths from 2000–2007 in New South Wales, Australia. Precision and recall (positive predictive value and sensitivity) were used as evaluation measures, with F-measure providing a single, overall measure of effectiveness. A detailed error analysis was performed on classification errors.
Results
Classification of diabetes, influenza, pneumonia and HIV was highly accurate (F-measure 0.96). More fine-grained ICD-10 classification effectiveness was more variable but still high (F-measure 0.80). The error analysis revealed that word variations as well as certain word combinations adversely affected classification. In addition, anomalies in the ground truth likely led to an underestimation of the effectiveness.
Conclusions
The high accuracy and low cost of the classification methods allow for an effective means for automatic and real-time surveillance of diabetes, influenza, pneumonia and HIV deaths. In addition, the methods are generally applicable to other diseases of interest and to other sources of medical free-text besides death certificates.
doi:10.1186/s12911-015-0174-2
PMCID: PMC4502908  PMID: 26174442
Syndromic surveillance; Machine learning; Death certificates
7.  Public preferences for engagement in Health Technology Assessment decision-making: protocol of a mixed methods study 
Background
Much attention in recent years has been given to the topic of public engagement in health technology assessment (HTA) decision-making. HTA organizations spend substantial resources and time on undertaking public engagement, and numerous studies have examined challenges and barriers to engagement in the decision-making process however uncertainty remains as to optimal methods to incorporate the views of the public in HTA decision-making. Little research has been done to ascertain whether current engagement processes align with public preferences and to what extent their desire for engagement is dependent on the question being asked by decision-makers or the characteristics of the decision. This study will examine public preferences for engagement in Australian HTA decision-making using an exploratory mixed methods design.
Methods/Design
The aims of this study are to: 1) identify characteristics about HTA decisions that are important to the public in determining whether public engagement should be undertaken on a particular topic, 2) determine which decision characteristics influence public preferences for the extent, or type of public engagement, and 3) describe reasons underpinning these preferences. Focus group participants from the general community, aged 18–70 years, will be purposively sampled from the Australian population to ensure a wide range of demographic groups. Each focus group will include a general discussion on public engagement as well as a ranking exercise using a modified nominal group technique (NGT). The NGT will inform the design of a discrete choice study to quantitatively assess public preferences for engagement in HTA decision-making.
Discussion
The proposed research seeks to investigate under what circumstances and how the public would like their views and preferences to be considered in health technology assessments. HTA organizations regularly make decisions about when and how public engagement should occur but without consideration of the public’s preferences on the method and extent of engagement. This information has the potential to assist decision-makers in tailoring engagement approaches, and may be particularly useful in decisions with potential for conflict where clarification of public values and preferences could strengthen the decision-making process.
doi:10.1186/s12911-015-0176-0
PMCID: PMC4499948  PMID: 26166149
Health technology assessment; Discrete choice study; Decision-making; Mixed methods; Public engagement
8.  Psychometric properties of a brief measure of autonomy support in breast cancer patients 
Background
The Health Care Climate Questionnaire measures patient perceptions of their clinician’s autonomy supportive communication. We sought to evaluate the psychometric properties of a modified brief version of the Health Care Climate Questionnaire (mHCCQ) adapted for breast cancer patients.
Methods
We surveyed 235 women aged 20–79 diagnosed with breast cancer within the previous 18 months at two cancer specialty centers using a print questionnaire. Patients completed the mHCCQ for their surgeon, medical oncologist, and radiation oncologist separately, as well as the overall treatment experience. Exploratory factor analysis (EFA) using principal components was used to explore the factor structure.
Results
One hundred sixty out of 235 (68.1 %) women completed the survey. Mean age was 57 years and time since diagnosis was 12.6 months. For surgeon, medical oncologist, and radiation oncologist ratings separately, as well as overall treatment, women rated 6 dimensions of perceived physician autonomy support. Exploratory factor analysis indicated a single factor solution for each clinician type and for the overall experience. Further, all six items were retained in each clinician subscore. Internal consistency was 0.93, 0.94, 0.97, and 0.92 for the overall, surgeon, medical oncologist, and radiation oncologist scales, respectively. Hierarchical factor analysis demonstrated that a summary score of the overall treatment experience accounts for only 52 % of the total variance observed in ratings of autonomy support for the three provider types.
Conclusions
These results describe the first use of the mHCCQ in cancer patients. Ratings of the overall treatment experience account for only half of the variance in ratings of autonomy support, suggesting that patients perceive and report differences in communication across provider types. Future research is needed to evaluate the relationship between physician communication practices and the quality of decision making, as well as other outcomes among cancer patients.
doi:10.1186/s12911-015-0172-4
PMCID: PMC4496811  PMID: 26155944
Autonomy support; Breast cancer; Health Care Climate Questionnaire
9.  Framing optional genetic testing in the context of mandatory newborn screening tests 
Background
Parents are increasingly faced with decisions about optional newborn bloodspot screening (NBS) despite no consistent policy for communicating information about such testing. We examined whether framing optional NBS alongside mandatory NBS influenced intention to participate in optional NBS.
Methods
For this Internet-administered study, 2,991 adults read a hypothetical vignette in which optional NBS for Duchenne muscular dystrophy (DMD) was either presented by itself (in isolation), alongside a description including the total number of mandatory NBS tests (“bundled” mandatory context), or alongside a listing of each mandatory NBS test (“unbundled” mandatory context). We assessed associations with participants’ intended participation using ordered logistic regression models, and associations with attitudes towards optional DMD NBS and subjective norms using Analysis of Variance.
Results
Participants were more likely to choose optional DMD NBS if they also read information about mandatory NBS (either bundled or unbundled) versus when DMD NBS was presented in isolation. Participants who read about optional DMD NBS in isolation also reported such testing to be less important and that they would worry more about the results than those who also saw mandatory NBS information.
Conclusions
Future NBS programs should pay attention to the framing of optional testing communication, as it influences parental behavior. Predictors of NBS uptake will become increasingly important as NBS programs continue expanding.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0173-3) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0173-3
PMCID: PMC4485334  PMID: 26123051
Newborn screening; Decision making; Duchenne muscular dystrophy; Optional newborn screening
10.  Improving prevalence estimation through data fusion: methods and validation 
Background
Estimation of health prevalences is usually performed with a single survey. Some attempts have been made to integrate more than one source of data. We propose here to validate this approach through data fusion. Data Fusion is the process of integrating two sources of data into one combined file. It allows us to take even greater advantage of existing information collected in databases. Here, we use data fusion to improve the estimation of health prevalences for two primary health factors: cardiovascular diseases and diabetes.
Methods
We use a real data fusion operation on population health, where the imputation of basic health risk factors is used to enrich a large-scale survey on self-reported health status. We propose choosing the imputation methodology for this problem through a suite of validation statistics that assess the quality of the fused data. The compared imputation techniques have been chosen from among the main imputation methodologies: k-nearest neighbor, probabilistic modeling and regression. We use the 2006 Health Survey of Catalonia, which provides a complete report of the perceived health status. In order to deal with the uncertainty problem, we compare these methodologies under the single and multiple imputation frames.
Results
A suite of validation statistics allows us to discern the strengths and weaknesses of studied imputation methods. Multiple outperforms single imputation by providing better and much more stable estimates, according to the computed validation statistics. The summarized results indicate that the probabilistic methods preserve the multivariate structure better; sequential regression methods deliver greater accuracy of imputed data; and nearest neighbor methods end up with a more realistic distribution of imputed data.
Conclusions
Data fusion allows us to integrate two sources of information in order to take grater advantage of the available data. Multiple imputed sequential regression models have the advantage of grater interpretability and can be used for health policy. Under certain conditions, more accurate estimates of the prevalences can be obtained using fused data (the original data plus the imputed data) than just by using only the observed data.
doi:10.1186/s12911-015-0169-z
PMCID: PMC4478714  PMID: 26104747
Population surveys; Prevalences; Diabetes; Cardio vascular diseases; Multiple imputation; Sequential regression
11.  Exploring the relationship between patients’ information preference style and knowledge acquisition process in a computerized patient decision aid randomized controlled trial 
Background
We have shown in a randomized controlled trial that a computerized patient decision aid (P-DA) improves medical knowledge and reduces decisional conflict, in early stage papillary thyroid cancer patients considering adjuvant radioactive iodine treatment. Our objectives were to examine the relationship between participants’ baseline information preference style and the following: 1) quantity of detailed information obtained within the P-DA, and 2) medical knowledge.
Methods
We randomized participants to exposure to a one-time viewing of a computerized P-DA (with usual care) or usual care alone. In pre-planned secondary analyses, we examined the relationship between information preference style (Miller Behavioural Style Scale, including respective monitoring [information seeking preference] and blunting [information avoidance preference] subscale scores) and the following: 1) the quantity of detailed information obtained from the P-DA (number of supplemental information clicks), and 2) medical knowledge. Spearman correlation values were calculated to quantify relationships, in the entire study population and respective study arms.
Results
In the 37 P-DA users, high monitoring information preference was moderately positively correlated with higher frequency of detailed information acquisition in the P-DA (r = 0.414, p = 0.011). The monitoring subscale score weakly correlated with increased medical knowledge in the entire study population (r = 0.268, p = 0.021, N = 74), but not in the respective study arms. There were no significant associations with the blunting subscale score.
Conclusions
Individual variability in information preferences may affect the process of information acquisition from computerized P-DA’s. More research is needed to understand how individual information preferences may impact medical knowledge acquisition and decision-making.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0168-0) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0168-0
PMCID: PMC4474358  PMID: 26088605
Cancer; Patient decision aid; Behaviour; Health information; Decision making; Consumer health information; Information seeking behaviors
12.  Fuzzy association rule mining and classification for the prediction of malaria in South Korea 
Background
Malaria is the world’s most prevalent vector-borne disease. Accurate prediction of malaria outbreaks may lead to public health interventions that mitigate disease morbidity and mortality.
Methods
We describe an application of a method for creating prediction models utilizing Fuzzy Association Rule Mining to extract relationships between epidemiological, meteorological, climatic, and socio-economic data from Korea. These relationships are in the form of rules, from which the best set of rules is automatically chosen and forms a classifier. Two classifiers have been built and their results fused to become a malaria prediction model. Future malaria cases are predicted as LOW, MEDIUM or HIGH, where these classes are defined as a total of 0–2, 3–16, and above 17 cases, respectively, for a region in South Korea during a two-week period. Based on user recommendations, HIGH is considered an outbreak.
Results
Model accuracy is described by Positive Predictive Value (PPV), Sensitivity, and F-score for each class, computed on test data not previously used to develop the model. For predictions made 7–8 weeks in advance, model PPV and Sensitivity are 0.842 and 0.681, respectively, for the HIGH classes. The F0.5 and F3 scores (which combine PPV and Sensitivity) are 0.804 and 0.694, respectively, for the HIGH classes. The overall FARM results (as measured by F-scores) are significantly better than those obtained by Decision Tree, Random Forest, Support Vector Machine, and Holt-Winters methods for the HIGH class. For the MEDIUM class, Random Forest and FARM obtain comparable results, with FARM being better at F0.5, and Random Forest obtaining a higher F3.
Conclusions
A previously described method for creating disease prediction models has been modified and extended to build models for predicting malaria. In addition, some new input variables were used, including indicators of intervention measures. The South Korea malaria prediction models predict LOW, MEDIUM or HIGH cases 7–8 weeks in the future. This paper demonstrates that our data driven approach can be used for the prediction of different diseases.
doi:10.1186/s12911-015-0170-6
PMCID: PMC4472166  PMID: 26084541
Malaria; Prediction; Association rule mining; Fuzzy logic; Classification; Environmental data; Socio-economic data; Epidemiological data
13.  Using a mobile health application to support self-management in chronic obstructive pulmonary disease: a six-month cohort study 
Background
Self-management strategies have the potential to support patients with chronic obstructive pulmonary disease (COPD). Telehealth interventions may have a role in delivering this support along with the opportunity to monitor symptoms and physiological variables. This paper reports findings from a six-month, clinical, cohort study of COPD patients’ use of a mobile telehealth based (mHealth) application and how individually determined alerts in oxygen saturation levels, pulse rate and symptoms scores related to patient self-initiated treatment for exacerbations.
Methods
The development of the mHealth intervention involved a patient focus group and multidisciplinary team of researchers, engineers and clinicians. Individual data thresholds to set alerts were determined, and the relationship to exacerbations, defined by the initiation of stand-by medications, was measured. The sample comprised 18 patients (age range of 50–85 years) with varied levels of computer skills.
Results
Patients identified no difficulties in using the mHealth application and used all functions available. 40 % of exacerbations had an alert signal during the three days prior to a patient starting medication. Patients were able to use the mHealth application to support self- management, including monitoring of clinical data. Within three months, 95 % of symptom reporting sessions were completed in less than 100 s.
Conclusions
Home based, unassisted, daily use of the mHealth platform is feasible and acceptable to people with COPD for reporting daily symptoms and medicine use, and to measure physiological variables such as pulse rate and oxygen saturation. These findings provide evidence for integrating telehealth interventions with clinical care pathways to support self-management in COPD.
doi:10.1186/s12911-015-0171-5
PMCID: PMC4472616  PMID: 26084626
Chronic obstructive pulmonary disease; Chronic condition; Self-management; Telehealth; Digital health; E-health; Mobile health; Alerts
15.  Detection of sentence boundaries and abbreviations in clinical narratives 
Background
In Western languages the period character is highly ambiguous, due to its double role as sentence delimiter and abbreviation marker. This is particularly relevant in clinical free-texts characterized by numerous anomalies in spelling, punctuation, vocabulary and with a high frequency of short forms.
Methods
The problem is addressed by two binary classifiers for abbreviation and sentence detection. A support vector machine exploiting a linear kernel is trained on different combinations of feature sets for each classification task. Feature relevance ranking is applied to investigate which features are important for the particular task. The methods are applied to German language texts from a medical record system, authored by specialized physicians.
Results
Two collections of 3,024 text snippets were annotated regarding the role of period characters for training and testing. Cohen's kappa resulted in 0.98. For abbreviation and sentence boundary detection we can report an unweighted micro-averaged F-measure using a 10-fold cross validation of 0.97 for the training set. For test set based evaluation we obtained an unweighted micro-averaged F-measure of 0.95 for abbreviation detection and 0.94 for sentence delineation. Language-dependent resources and rules were found to have less impact on abbreviation detection than on sentence delineation.
Conclusions
Sentence detection is an important task, which should be performed at the beginning of a text processing pipeline. For the text genre under scrutiny we showed that support vector machines exploiting a linear kernel produce state of the art results for sentence boundary detection. The results are comparable with other sentence boundary detection methods applied to English clinical texts. We identified abbreviation detection as a supportive task for sentence delineation.
doi:10.1186/1472-6947-15-S2-S4
PMCID: PMC4474545  PMID: 26099994
clinical narrative; natural language processing; machine learning
16.  SNOMED CT in a language isolate: an algorithm for a semiautomatic translation 
Background
The Systematized Nomenclature of Medicine - Clinical Terms (SNOMED CT) is officially released in English and Spanish. In the Basque Autonomous Community two languages, Spanish and Basque, are official. The first attempt to semi-automatically translate the SNOMED CT terminology content to Basque, a less resourced language is presented in this paper.
Methods
A translation algorithm that has its basis in Natural Language Processing methods has been designed and partially implemented. The algorithm comprises four phases from which the first two have been implemented and quantitatively evaluated.
Results
Results are promising as we obtained the equivalents in Basque of 21.41% of the disorder terms of the English SNOMED CT release. As the methods developed are focused on that hierarchy, the results in other hierarchies are lower (12.57% for body structure descriptions, 8.80% for findings and 3% for procedures).
Conclusions
We are in the way to reach two of our objectives when translating SNOMED CT to Basque: to use our language to access rich multilingual resources and to strengthen the use of the Basque language in the biomedical area.
doi:10.1186/1472-6947-15-S2-S5
PMCID: PMC4474582  PMID: 26100112
SNOMED CT translation; Basque Language Isolate; Natural Language Processing; Finite State Transducers
17.  Exploring Spanish health social media for detecting drug effects 
Background
Adverse Drug reactions (ADR) cause a high number of deaths among hospitalized patients in developed countries. Major drug agencies have devoted a great interest in the early detection of ADRs due to their high incidence and increasing health care costs. Reporting systems are available in order for both healthcare professionals and patients to alert about possible ADRs. However, several studies have shown that these adverse events are underestimated. Our hypothesis is that health social networks could be a significant information source for the early detection of ADRs as well as of new drug indications.
Methods
In this work we present a system for detecting drug effects (which include both adverse drug reactions as well as drug indications) from user posts extracted from a Spanish health forum. Texts were processed using MeaningCloud, a multilingual text analysis engine, to identify drugs and effects. In addition, we developed the first Spanish database storing drugs as well as their effects automatically built from drug package inserts gathered from online websites. We then applied a distant-supervision method using the database on a collection of 84,000 messages in order to extract the relations between drugs and their effects. To classify the relation instances, we used a kernel method based only on shallow linguistic information of the sentences.
Results
Regarding Relation Extraction of drugs and their effects, the distant supervision approach achieved a recall of 0.59 and a precision of 0.48.
Conclusions
The task of extracting relations between drugs and their effects from social media is a complex challenge due to the characteristics of social media texts. These texts, typically posts or tweets, usually contain many grammatical errors and spelling mistakes. Moreover, patients use lay terminology to refer to diseases, symptoms and indications that is not usually included in lexical resources in languages other than English.
doi:10.1186/1472-6947-15-S2-S6
PMCID: PMC4474583  PMID: 26100267
18.  Care episode retrieval: distributional semantic models for information retrieval in the clinical domain 
Patients' health related information is stored in electronic health records (EHRs) by health service providers. These records include sequential documentation of care episodes in the form of clinical notes. EHRs are used throughout the health care sector by professionals, administrators and patients, primarily for clinical purposes, but also for secondary purposes such as decision support and research. The vast amounts of information in EHR systems complicate information management and increase the risk of information overload. Therefore, clinicians and researchers need new tools to manage the information stored in the EHRs. A common use case is, given a - possibly unfinished - care episode, to retrieve the most similar care episodes among the records. This paper presents several methods for information retrieval, focusing on care episode retrieval, based on textual similarity, where similarity is measured through domain-specific modelling of the distributional semantics of words. Models include variants of random indexing and the semantic neural network model word2vec. Two novel methods are introduced that utilize the ICD-10 codes attached to care episodes to better induce domain-specificity in the semantic model. We report on experimental evaluation of care episode retrieval that circumvents the lack of human judgements regarding episode relevance. Results suggest that several of the methods proposed outperform a state-of-the art search engine (Lucene) on the retrieval task.
doi:10.1186/1472-6947-15-S2-S2
PMCID: PMC4474584  PMID: 26099735
19.  Using text mining techniques to extract phenotypic information from the PhenoCHF corpus 
Background
Phenotypic information locked away in unstructured narrative text presents significant barriers to information accessibility, both for clinical practitioners and for computerised applications used for clinical research purposes. Text mining (TM) techniques have previously been applied successfully to extract different types of information from text in the biomedical domain. They have the potential to be extended to allow the extraction of information relating to phenotypes from free text.
Methods
To stimulate the development of TM systems that are able to extract phenotypic information from text, we have created a new corpus (PhenoCHF) that is annotated by domain experts with several types of phenotypic information relating to congestive heart failure. To ensure that systems developed using the corpus are robust to multiple text types, it integrates text from heterogeneous sources, i.e., electronic health records (EHRs) and scientific articles from the literature. We have developed several different phenotype extraction methods to demonstrate the utility of the corpus, and tested these methods on a further corpus, i.e., ShARe/CLEF 2013.
Results
Evaluation of our automated methods showed that PhenoCHF can facilitate the training of reliable phenotype extraction systems, which are robust to variations in text type. These results have been reinforced by evaluating our trained systems on the ShARe/CLEF corpus, which contains clinical records of various types. Like other studies within the biomedical domain, we found that solutions based on conditional random fields produced the best results, when coupled with a rich feature set.
Conclusions
PhenoCHF is the first annotated corpus aimed at encoding detailed phenotypic information. The unique heterogeneous composition of the corpus has been shown to be advantageous in the training of systems that can accurately extract phenotypic information from a range of different text types. Although the scope of our annotation is currently limited to a single disease, the promising results achieved can stimulate further work into the extraction of phenotypic information for other diseases. The PhenoCHF annotation guidelines and annotations are publicly available at https://code.google.com/p/phenochf-corpus.
doi:10.1186/1472-6947-15-S2-S3
PMCID: PMC4474585  PMID: 26099853
20.  Understanding older women’s decision making and coping in the context of breast cancer treatment 
Background
Primary endocrine therapy (PET) is a recognised alternative to surgery followed by endocrine therapy for a subset of older, frailer women with breast cancer. Choice of treatment is preference-sensitive and may require decision support. Older patients are often conceptualised as passive decision-makers. The present study used the Coping in Deliberation (CODE) framework to gain insight into decision making and coping processes in a group of older women who have faced breast cancer treatment decisions, and to inform the development of a decision support intervention (DSI).
Methods
Semi-structured interviews were carried out with older women who had been offered a choice of PET or surgery from five UK hospital clinics. Women’s information and support needs, their breast cancer diagnosis and treatment decisions were explored. A secondary analysis of these interviews was conducted using the CODE framework to examine women’s appraisals of health threat and coping throughout the deliberation process.
Results
Interviews with 35 women aged 75-98 years were analysed. Appraisals of breast cancer and treatment options were sometimes only partial, with most women forming a preference for treatment relatively quickly. However, a number of considerations which women made throughout the deliberation process were identified, including: past experiences of cancer and its treatment; scope for choice; risks, benefits and consequences of treatment; instincts about treatment choice; and healthcare professionals’ recommendations. Women also described various strategies to cope with breast cancer and their treatment decisions. These included seeking information, obtaining practical and emotional support from healthcare professionals, friends and relatives, and relying on personal faith. Based on these findings, key questions were identified that women may ask during deliberation.
Conclusions
Many older women with breast cancer may be considered involved rather than passive decision-makers, and may benefit from DSIs designed to support decision making and coping within and beyond the clinic setting.
doi:10.1186/s12911-015-0167-1
PMCID: PMC4461993  PMID: 26058557
Breast cancer treatment; Old age; Decision making; Coping; Deliberation
21.  Predicting postoperative complications of head and neck squamous cell carcinoma in elderly patients using random forest algorithm model 
Background
Head and Neck Squamous Cell Carcinoma (HNSCC) has a high incidence in elderly patients. The postoperative complications present great challenges within treatment and they're hard for early warning.
Methods
Data from 525 patients diagnosed with HNSCC including a training set (n = 513) and an external testing set (n = 12) in our institution between 2006 and 2011 was collected. Variables involved are general demographic characteristics, complications, disease and treatment given. Five data mining algorithms were firstly exploited to construct predictive models in the training set. Subsequently, cross-validation was used to compare the different performance of these models and the best data mining algorithm model was then selected to perform the prediction in an external testing set.
Results
Data from 513 patients (age > 60 y) with HNSCC in a training set was included while 44 variables were selected (P < 0.05). Five predictive models were constructed; the model with 44 variables based on the Random Forest algorithm demonstrated the best accuracy (89.084 %) and the best AUC value (0.949). In an external testing set, the accuracy (83.333 %) and the AUC value (0.781) were obtained by using the random forest algorithm model.
Conclusions
Data mining should be a promising approach used for elderly patients with HNSCC to predict the probability of postoperative complications. Our results highlighted the potential of computational prediction of postoperative complications in elderly patients with HNSCC by using the random forest algorithm model.
doi:10.1186/s12911-015-0165-3
PMCID: PMC4459053  PMID: 26054335
Head and neck squamous cell carcinoma (HNSCC); Postoperative complications; Predictive model; Data mining (DM); Elderly patients
22.  PubMed-supported clinical term weighting approach for improving inter-patient similarity measure in diagnosis prediction 
Background
Similarity-based retrieval of Electronic Health Records (EHRs) from large clinical information systems provides physicians the evidence support in making diagnoses or referring examinations for the suspected cases. Clinical Terms in EHRs represent high-level conceptual information and the similarity measure established based on these terms reflects the chance of inter-patient disease co-occurrence. The assumption that clinical terms are equally relevant to a disease is unrealistic, reducing the prediction accuracy. Here we propose a term weighting approach supported by PubMed search engine to address this issue.
Methods
We collected and studied 112 abdominal computed tomography imaging examination reports from four hospitals in Hong Kong. Clinical terms, which are the image findings related to hepatocellular carcinoma (HCC), were extracted from the reports. Through two systematic PubMed search methods, the generic and specific term weightings were established by estimating the conditional probabilities of clinical terms given HCC. Each report was characterized by an ontological feature vector and there were totally 6216 vector pairs. We optimized the modified direction cosine (mDC) with respect to a regularization constant embedded into the feature vector. Equal, generic and specific term weighting approaches were applied to measure the similarity of each pair and their performances for predicting inter-patient co-occurrence of HCC diagnoses were compared by using Receiver Operating Characteristics (ROC) analysis.
Results
The Areas under the curves (AUROCs) of similarity scores based on equal, generic and specific term weighting approaches were 0.735, 0.728 and 0.743 respectively (p < 0.01). In comparison with equal term weighting, the performance was significantly improved by specific term weighting (p < 0.01) but not by generic term weighting. The clinical terms “Dysplastic nodule”, “nodule of liver” and “equal density (isodense) lesion” were found the top three image findings associated with HCC in PubMed.
Conclusions
Our findings suggest that the optimized similarity measure with specific term weighting to EHRs can improve significantly the accuracy for predicting the inter-patient co-occurrence of diagnosis when compared with equal and generic term weighting approaches.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0166-2) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0166-2
PMCID: PMC4450834  PMID: 26032596
23.  Design and evaluation of a software for the objective and easy-to-read presentation of new drug properties to physicians 
Background
When new pharmaceutical products appear on the market, physicians need to know whether they are likely to be useful in their practices. Physicians currently obtain most of their information about the market release and properties of new drugs from pharmaceutical industry representatives. However, the official information contained in the summary of product characteristics (SPCs) and evaluation reports from health agencies, provide a more complete view of the potential value of new drugs, although they can be long and difficult to read. The main objective of this work was to design a prototype computer program to facilitate the objective appraisal of the potential value of a new pharmaceutical product by physicians. This prototype is based on the modeling of pharmaceutical innovations described in a previous paper.
Methods
The interface was designed to allow physicians to develop a rapid understanding of the value of a new drug for their practices. We selected five new pharmaceutical products, to illustrate the function of this prototype. We considered only the texts supplied by national or international drug agencies at the time of market release. The perceived usability of the prototype was evaluated qualitatively, except for the System Usability Scale (SUS) score evaluation, by 10 physicians differing in age and medical background.
Results
The display is based on the various axes of the conceptual model of pharmaceutical innovations. The user can select three levels of detail when consulting this information (highly synthetic, synthetic and detailed). Tables provide a comparison of the properties of the new pharmaceutical product with those of existing drugs, if available for the same indication, in terms of efficacy, safety and ease of use.
The interface was highly appreciated by evaluators, who found it easy to understand and suggested no other additions of important, internationally valid information. The mean System Usability Scale score for the 10 physicians was 82, corresponding to a “good” user interface.
Conclusions
This work led us to propose the selection, grouping, and mode of presentation for various types of knowledge on pharmaceutical innovations in a way that was appreciated by evaluators. It provides physicians with readily accessible objective information about new drugs.
Electronic supplementary material
The online version of this article (doi:10.1186/s12911-015-0158-2) contains supplementary material, which is available to authorized users.
doi:10.1186/s12911-015-0158-2
PMCID: PMC4460682  PMID: 26025025
24.  Patient-reported outcomes in a large community-based pain medicine practice: evaluation for use in phenotype modeling 
Background
An academic, community medicine partnership was established to build a phenotype-to-outcome model targeting chronic pain. This model will be used to drive clinical decision support for pain medicine in the community setting. The first step in this effort is an examination of the electronic health records (EHR) from clinics that treat chronic pain. The biopsychosocial components provided by both patients and care providers must be of sufficient scope to populate the spectrum of patient types, treatment modalities, and possible outcomes.
Methods
The patient health records from a large Midwest pain medicine practice (Michigan Pain Consultants, PC) contains physician notes, administrative codes, and patient-reported outcomes (PRO) on over 30,000 patients during the study period spanning 2010 to mid-2014. The PRO consists of a regularly administered Pain Health Assessment (PHA), a biopsychosocial, demographic, and symptomology questionnaire containing 163 items, which is completed approximately every six months with a compliance rate of over 95 %. The biopsychosocial items (74 items with Likert scales of 0–10) were examined by exploratory factor analysis and descriptive statistics to determine the number of independent constructs available for phenotypes and outcomes. Pain outcomes were examined both in the aggregate and the mean of longitudinal changes in each patient.
Results
Exploratory factor analysis of the intake PHA revealed 15 orthogonal factors representing pain levels; physical, social, and emotional functions; the effects of pain on these functions; vitality and health; and measures of outcomes and satisfaction. Seven items were independent of the factors, offering unique information. As an exemplar of outcomes from the follow-up PHAs, patients reported approximately 60 % relief in their pain. When examined in the aggregate, patients showed both a decrease in pain levels and an increase in coping skills with an increased number of visits. When examined individually, 80-85 % of patients presenting with the highest pain levels reported improvement by approximately two points on an 11-point pain scale.
Conclusions
We conclude that the data available in a community practice can be a rich source of biopsychosocial information relevant to the phenotypes of chronic pain. It is anticipated that phenotype linkages to best treatments and outcomes can be constructed from this set of records.
doi:10.1186/s12911-015-0164-4
PMCID: PMC4446111  PMID: 26017305
Chronic pain; Patient-reported outcomes; Factor analysis; Pain severity
25.  ODM2CDA and CDA2ODM: Tools to convert documentation forms between EDC and EHR systems 
Background
Clinical trials apply standards approved by regulatory agencies for Electronic Data Capture (EDC). Operational Data Model (ODM) from Clinical Data Interchange Standards Consortium (CDISC) is commonly used. Electronic Health Record (EHR) systems for patient care predominantly apply HL7 standards, specifically Clinical Document Architecture (CDA). In recent years more and more patient data is processed in electronic form.
Results
An open source reference implementation was designed and implemented to convert forms between ODM and CDA format. There are limitations of this conversion method due to different scope and design of ODM and CDA. Specifically, CDA has a multi-level hierarchical structure and CDA nodes can contain both XML values and XML attributes.
Conclusions
Automated transformation of ODM files to CDA and vice versa is technically feasible in principle.
doi:10.1186/s12911-015-0163-5
PMCID: PMC4494189  PMID: 26004011
EHR; EDC; CDA; ODM; Documentation form; Data model

Results 1-25 (976)